Digitized by the Internet Archive

in

2013

http://archive.org/details/dynamicsofbasesfOObark

DYNAMICS OF BASES AND FOUNDATIONS

McGRAW-HILL SERIES IN SOILS

ENGINEERING AND FOUNDATIONS

Barkan, Dynamics

of

Bases and Foundations

Leonards, Foundation Engineering

YNAM CS OF BASES
AND FOUNDATIONS
D

D.

D.

BARKAN

DOCTOR OF TECHNICAL SCIENCES

Translated from the Russian by L. Drashevska

Translation edited by G. P. Tschebotarioff

McGraw-Hill book company,

New York

San Francisco

Toronto

inc.
London

DYNAMICS OF BASES AND FOUNDATIONS

Copyright
Reserved.

1962 by the McGraw-Hill Book Company, Inc.

All Rights
Printed in the United States of America. This book, or
parts thereof, may not be reproduced in any form without permission
of the publishers.
Library of Congress Catalog Card Number 62-15743

03650

Revised for English translation by the author in 1959 and 1960.

First published in the Russian language

DYNAMIKA OSNOVANIY
and copyright

1948

under the

title

FUNDAMENTOV

by Stroyvoenmorizdat,

Moscow.

EDITOR'S

FOREWORD TO THE

ENGLISH TRANSLATION

Since the Fourth International Conference on Soil Mechanics and

Foundation Engineering in London (1957), Dr. D. D. Barkan has become
well known in the West for his work of recent years on the development
of vibratory driving equipment for piles, sheet piles, and cylindrical
However, the remarkable work on machine-foundation
bridge caissons.
problems of the Soil Dynamics Laboratory which he heads at the Institute
of Foundations and Underground Structures remains little known in the

West.

The development of rational and effective approaches to these problems

has been greatly hampered throughout the world by a lack of continuity
and organized coordination of efforts in the several disciplines involved.
Since its creation some 35 years ago Dr. Barkan's laboratory has made an
impressive start in remedying this situation.
I fully endorse the views expressed on the matter in the preface to the
Russian text by the late Professor N. M. Gersevanov, who had sponsored
the creation of the laboratory. Therefore I am very happy to have the
privilege of presenting to the attention of English-speaking engineers the
book "Dynamics
book is unique.

translation of Dr. Barkan's

In

many
The

respects this

relevant basic premises of

soil

of Bases

and Foundations."

mechanics, the theory of elasticity,

and the dynamics of vibratory motion are presented in a coordinated

manner, and the resulting findings are examined in the light of experimental evidence; the use of complex mathematical expressions is simplified by the provision of auxiliary charts and tables relevant coefficients
obtained from theoretical analyses and from laboratory and model tests
are checked against and corrected by the results of numerous measurements
;

| The term "base" which appears in the title of the book is used in Russian in the
sense of the mass of soil which supports the foundation of any type of structure.

EDITOR'S

vi

FOREWORD TO

THE ENGLISH TRANSLATION

on full-scale industrial installations; design recommendations are

made

in

the light of practical considerations of civil and mechanical engineering;

numerical examples illustrate the points made.

This book should therefore be of immediate use to practicing engineers

who have to deal with machine foundations
civil and mechanical

both

and the vibrations which machines cause it should also stimulate further
creative thought and research in the field.
After a formal agreement on the matter between the Soviet publishing
agency Mezhdunarodnaya Kniga (International Book) and the McGrawHill Book Company, Inc., Dr. Dominik D. Barkan in 1959 and 1960
reviewed and brought up to date the text of his original 1948 book; the
English translation was made from this revised Russian text.
Acknowledgments are due to my wife, Florence Bill TschebotariorT,
who helped me edit the English translation, and to several Princeton
;

colleagues

who helped me in the selection of proper mechanical termsW. Suppiger, Professor of Mechanical Engineering,

especially to Dr. E.

who

read the manuscript.

Gregory P.

T schebotarioff

Professor of Civil Engineering

Princeton University

EDITORS'

FOREWORD TO

THE

ORIGINAL RUSSIAN TEXT

This book

and vibratory properties of soil

under machines
with dynamic loads, questions of wave propagation from industrial
sources, and problems of the effect of these waves on structures.
This book was compiled mainly from results of investigations which
were carried out by the author and workers of the laboratory headed by
him. Included are data gained from experience in the performance of
foundations under machinery at industrial plants of the U.S.S.R.
This book is intended for civil engineers and scientific workers.
is

devoted to the

bases, the theory

and practice

elastic

of the design of foundations

the editors

G. I. Berdichevsky

D.

Tumarkin

PREFACE TO THE ORIGINAL

RUSSIAN TEXT

The

practice of

many problems

modern

industrial construction requires the solution of

concerning the performance of foundations subjected

dynamic loads, the propagation of waves in soil, and the effects of

waves on structures.
Some twenty years ago, on the initiative of the undersigned, a special
laboratory of dynamics of soil bases and foundations was organized for
The laboratory was headed by Professor
the study of these problems.

to

these

D. D. Barkan.

The laboratory work concentrated

and changes

chiefly

on the investigation of

physicomechanical characteristics (in particular, internal friction, cohesion, and void ratio) under
the influence of vibrations, on the study of vibrations of massive foundations, and on the study of exciting loads created by various machines.
In addition, much of the laboratory work was devoted to study of the
propagation in soils of elastic waves from industrial sources, the screening of these waves by various measures, and the investigation of the
influence of several other factors on wave propagation.
Considerable attention was paid by the laboratory to the study of
foundation vibrations and the action on structures of waves emanating
elastic properties of soils

in their

from foundations.

The
of

extensive research

dynamics

of soil bases

work carried out by the laboratory in the

and foundations made it possible for the

field
first

time to compile standard specifications for the computation, design, and

erection of foundations under

The use

machinery with dynamic

loads.

standard specifications in the practical design of

industrial projects was very effective, in both the engineering and
economic aspects of construction.
of these

PREFACE TO THE ORIGINAL RUSSIAN TEXT

The book, "Dynamics of Bases and Foundations," by Doctor of

Technical Sciences D. D. Barkan, is compiled mainly from materials
derived from investigations conducted by his laboratory.

guiding monograph in the

field

This

is

the

first

concerned.

The originality of the material and the subjects covered make this
book interesting not only for civil engineers who are designers but also
for wide circles of scientific and engineering workers who are studying the
dynamics of machines and structures.
Professor N.

Corresponding

Academy

M.

Gersevanov

Member

of the

of Sciences of the U.S.S.R.

PREFACE TO THE ENGLISH

TRANSLATION

is accompanied by an increase in the

and the growth of their power, as well as by
the introduction of blasting operations into construction practice and the
increasing speed, intensity, and tonnage of various kinds of transport.
All these factors lead to an increase in the influence of shocks and vibrations, the study of which is the subject of a new branch of mechanics
designated as industrial seismology.
One of the most important divisions
of industrial seismology is the dynamics of bases and foundations, which
studies the problems of shocks on and vibrations of foundations under

Technical and industrial progress

number

of

machines

in use

industrial installations, as well as the laws governing the propagation of

waves from these foundations through soil.

The problems solved by study of the dynamics of bases and foundations
are of great importance in engineering practice.
Vibrations of machine foundations are harmful to the operation of the
machines themselves. The following case may be quoted as an example:
due to an unsatisfactory design, the foundations of several piston gas
compressors installed in the same building underwent considerable
vibrations.
The compressors, being rigidly connected to the foundations, also underwent vibrations of large amplitude which were transmitted to the pipes leading from them. Vibrations frequently caused
damage to and even rupture of these pipes near the vibrating compressors.
This caused the seepage of explosive gas into the compressor house; as
a result the house blew up.
Measurements of vibrations of foundations and anvils under operating
hammers show that some 10 per cent of the work of the dropping parts
is spent in exciting vibrations of the foundation and anvil.
Vibrations induced by a vibrating foundation often cause poor performance of machine bearings, which may be heated up or may wear
dissymmetrically.

PREFACE TO THE ENGLISH TRANSLATION

xii

In addition, foundation vibrations under high-speed machines cause

premature fatigue of machine parts.
Vibrations and shocks also have a harmful effect on the foundations
themselves. Although very few cases have been recorded in which
machine foundations have been destroyed by their vibrations, some
instances are known where vibrations or shocks led to a rapid enlargement
of cracks which were initially caused by other factors such as stresses
imposed by differential settlements of the foundation. This led to the
destruction of the foundations.

Vibrations of and shocks to foundations under machinery lead to

considerable nonuniform settlements which necessitate periodic adjust-

ments and gauging

of machines.
foundation subjected to vibrations becomes a source of waves propagated through the soil which have a harmful effect on structures and

people within them.

These waves

also hinder

Several examples are presented in Chap.

normal operation of plants.

IX showing how waves

propagated from foundations under hammers caused damage to structures or fully destroyed them.
Waves from hammer foundations fairly frequently cause differential
settlements of columns or walls of forge shops. Shocks on hammer
foundations lead also to skewing of locomotive crane tracks, distortions
of

window sashes, damage of some connections of structural elements, etc.

Waves from machine foundations affect not only structures, but also

various technological processes and the operation of precision machines

devices.
Cases are known from industrial practice in which machinefoundation shocks hindered the operation of various devices and caused
an increase in the amount of rejects in molding and machine shops.
A harmful effect on structures, and especially on technological processes

and

within these structures, may be produced by waves propagated from

foundations under low-frequency machines with a stable regime of
operation such as compressors, diesels, and saw frames.
usually have comparatively low speed, of the

same order

These machines
magnitude as

of

the principal values characterizing the natural vibrations of structures.

Consequently, the frequency of waves propagated from the foundation

under a low-frequency machine sometimes coincides with the frequency
of natural vibrations of the structure.
Considerable vibrations then
develop in the superstructure of the building subjected to resonance.
If a building undergoes vibrations of considerable amplitude, many
technological processes within this building become difficult or even
impossible.
Such vibrations have an especially harmful effect on men if
living quarters are located in a vibrating building.
Interviews with
people

who

jected

day and night

lived in such a vibrating building

showed that persons subfrom

to the action of vibrations often suffered

PREFACE

TO THE ENGLISH TRANSLATION

xiii

headaches, insomnia, nervousness, etc. A thorough study of the dynamics

of bases and foundations is needed to successfully overcome the harmful
effects of

such vibratory processes.

Problems of the dynamics of bases and foundations are important from

an economic standpoint as well. Because the bearing capacity of soils
under machinery foundations was usually taken as 50 to 60 per cent of the
value corresponding to the action of static loads only, in many cases the
bases were reinforced by piles. This raised considerably the price of the

The

foundation.

clarification in recent years of

taining to the dynamics of bases under structures

some questions per-

made

it

possible to

increase design values of the bearing capacity of soils under machinery

foundations up to 80 to 100 per cent of the static-load values. Thus the

The cost of
installation of a pile base could be avoided in many cases.

such a base under powerful industrial machinery

may

reach hundreds of

not millions, of rubles for a single project.

Until recently it was believed by designers that the larger the mass of a
machine foundation, the smaller the amplitude of its vibrations. Therethousands,

fore

if

foundations under machines were

designed

as

massive

blocks

attaining sometimes a weight of several thousand tons and placed at

depths up to 10 to 12 m.
The erection of such foundations involved the expenditure of considerable funds, materials,

and time.

The study

of questions of the rational

design of foundations led to the conclusion that, from the standpoint of

dynamic

stability, it is often

expedient to

make

the foundations as light

as possible.

Owing to this fact, the cost of industrial projects has been considerably
reduced in many cases. Let us mention, for example, that foundations
under powerful horizontal compressors were formerly built with weights
of 2,000 to 2,500 tons; at present

one-half the above figure.

dynamic computations

As a

they are designed for weights reduced to

result of the

development

of foundations, the cost of a

of

methods

of

foundation under

one powerful horizontal compressor is now 150,000 to 200,000 rubles less

than it was previously.
Only a few years ago, the weight of a foundation under a hammer was
selected to be approximately 100 to 120 times the weight of dropping
parts; in addition, it was held necessary to increase the mass of the
foundation not by enlarging its area in plan, but by increasing its depth.

As a

result,

foundations under

hammers were built as heavy massive

The study of the behavior

blocks sometimes placed at depths up to 10 m.

under the action of impacts led to the conclusion that it is

without increasing the permissible amplitude of vibrations, to
design foundations with weights exceeding by not more than 10 to 50
times the weights of the dropping parts.
Thus the construction of one
of foundations

possible,

PREFACE TO THE ENGLISH TRANSLATION

xiv

foundation under a 3-ton hammer

rubles than what it cost previously.

is

now cheaper by

25,000 to 30,000

The great economic importance of questions of the dynamics of bases

and foundations will be still more evident if one realizes that annually
thousands of machines which require the construction of individual
foundations are built or moved from one place to another.
The vibrations of foundations under machines and the propagation of
waves from industrial sources depend on the elastic properties of soil.
It is therefore natural that the first chapter of this book is devoted to the
The experimental data which it presents
description of these properties.
on the dependence of elastic deformations of soil on stresses show that the
application of Hooke's law to soils may be considered as a first approximaDeviations from Hooke's law are caused by the influence of
tion only.
initial stresses in

capillary forces,

external loads.
its

modulus

(which are present because of friction forces,

on the elastic deformations induced by
Consequently, the elastic constants of soil, particularly
the

soil

and other

of shear,

factors)

depend on the values

The modern theory

of

normal pressure.

massive foundation vibrations is based on the

assumption that the hypothesis of linear relationship between settlements

and loads

is

of

Therefore, for the computation of forced or natural

valid.

vibrations of foundations
of elasticity of soils,

i.e.,

in the soil along the

it is

necessary to

know

the so-called coefficients

the coefficients of proportionality between stress

foundation area in contact with the soil and the

The first chapter presents

settlement or sliding of the foundation.

and experimental data which make it possible to establish the

above coefficients for different soils and the dependAt the end of the chapter,
these coefficients on several factors.

theoretical

limiting values of the

ence of

experimental data are given concerning the elastic properties of pile

A considerable part of the first chapter is devoted to the

foundations.

results of experimental investigations of elastic properties of soil bases

under structures.
The second chapter treats questions related to the formation of
irreversible (residual) settlements under the simultaneous action on the
soil of static pressure and vibrations.
This question is important, first
of all, for the estimation of the bearing value of the soil under the combined action of vibrations and static pressure and the clarification of
factors affecting residual settlements under the influence of vibrations;
second, it is important for the scientific justification of new methods for
the provision of artificial bases under structures by means of vibrations.
The results of experiments presented in the second chapter show that
basically all soil properties (including those which determine the resistance of soils to localized loading) may change considerably under the
action of vibrations.
Also very strongly affected are the compressive

PREFACE TO THE ENGLISH TRANSLATION

properties of

soils,

especially

sandy

soils.

xv

Computational problems of

residual (plastic) settlements of foundations subjected to vibrations or

shocks acquire considerable importance in this connection. The basic

theory for the computation of such settlements is given in the second
chapter, as are methods to decrease residual settlements.

Data presented in the first two chapters were obtained from investigaby the author or his associates in the laboratory headed
by him a considerable part of these data is being published here for the

tions carried out

;

first

time.

The

and results of experimental

massive foundations. This chapter
part of the book, which covers the main general

third chapter treats the general theory

investigations

of

concludes the

first

principles of the

vibrations

dynamics

of

of foundations.

Chapters IV to VII form the second part of the book and treat the
problems of dynamic computations and design of foundations for special
types of machines such as reciprocating engines, machinery with impact
The table
loads, rolling mills, turbodynamos, and crushing equipment.
We shall
of contents shows the scope of the subject matter presented.
merely mention that for these chapters data are widely used from
numerous large-scale instrumental observations of vibrations of existing
foundations which have been carried out by the author since 1939.
Finally, Chaps. VIII and IX are devoted to the question of dynamics
of bases, i.e., to the propagation through soil of waves from machinery
and the action of these waves on structures. These chapters form the
third part of the book.

special feature of this part lies in the fact that

waves has been cornumerous experimental data. The latter

show that in the solution of some problems (for example, when establishing a relationship between the amplitudes of soil vibrations and distance)
theoretical conclusions based on the assumption that soil is an ideal elastic

theoretical material related to the propagation of

rected here on the basis of

medium

In such cases the author has

based on experimental results.
In compiling this book the author widely used material obtained from
investigations conducted in the Soil Dynamics Laboratory headed by
are not confirmed experimentally.

tried to introduce the necessary corrections

himself.

The author expresses his cordial thanks to Prof. G.

who spent much time and energy on the editing of the
its

preparation for publication in the United States.

thanks 0. Ya. Shekhter,

who helped him

P. Tschebotarioff,

translation

and on

The author

also

in the revision of the text.

D. D. Barkan

CONTENTS

Editor's Foreword to the English Translation

Editors'

Foreword

to the

Original Russian Text

Preface

to the

Original Russian Text

Preface

to the

English Translation

vii
ix

xi

Notation

xxiii

CHAPTER

I.

1-1.

Elastic Properties of Soil

Elastic Deformations of Soil

a.

and

Law
Hooke's Law

Its Elastic

Constants

Generalized Hooke's

b. Applicability of

1
1

to Soil

Young's Modulus and Poisson's Ratio for Soils

Coefficient of Elastic Uniform Compression of Soil c u
a. The Coefficient of Subgrade Reaction
b. Coefficient of Elastic Uniform Compression c u
c. The Effect of the Area and Shape of a Foundation
Base on the Coefficient of Elastic Uniform Comc.

1-2.

11

17
17
19

21

pression

Experimental Investigations of the Coefficient of

Elastic Uniform Compression
e. Values of the Coefficient of Elastic Uniform Compression for Soils Used in Design Computations
Coefficient of Elastic Nonuniform Compression of Soil c<p
Coefficient of Elastic Uniform Shear of Soil c T
Coefficient of Elastic Nonuniform Shear of Soil c$ .'
Elastic Resistance of Pile Foundations
a. Elastic Resistance of Piles under a Vertical Load
b. Elastic Resistance of Pile Foundations in Cases of
Eccentric Vertical Loading
c. Elastic Resistance of Pile Foundations to Horizontal
Forces
d.

...

1-3.
1-4.
1-5.
1-6.

CHAPTER

II.

II- 1.

Effects of Vibrations

on Residual

Soil Settlements

Effect of Vibrations on the Dissipative Properties of Soils

a.

28
30
34
40
41
41

50
51

54
55

Internal Friction and Cohesion of Soil during Vibra-

55

tions
b. Vibro viscous Soil Resistance
c.

25

Damping

Properties of Soils
xvii

61

65

CONTENTS

xviii

II-2.

Effect of Vibrations

on the Porosity and Hydraulic

Properties of Soils

Elements of the Theory of Residual Settlements of

Foundations Induced by Dynamic Loads

II-4.

Methods
of a

CHAPTER

68

II-3.

III.

III-l.

to Decrease the Residual

Foundation

Theory

82

of Vibrations of

Massive Machine Foundations

85

Vertical Vibrations of Foundations

c.

Basic Assumptions
Vibrations of Foundations Neglecting the
Damping Effect of Soil Reactions
Vertical Vibrations of Foundations Considering the

d.

Damping Effect of Soil Reactions

The Effect of Soil Inertia on Forced

a.

75

Dynamic Settlement

85
85

b. Vertical

III-2.

91
Vertical Vibra-

tions of Foundations
Rocking Vibrations of Foundations
a. Foundation Weight
b. Soil

94
103
104
104
106

Reaction

III-3.

Vibrations of Pure Shear

III-4.

Vibrations of Foundations Accompanied by Simultaneous

Rotation, Sliding, and Vertical Displacement
.

Equations of Vibration
b. Free Vibrations
c. Forced Vibrations
d. The Effect on the Natural Frequencies of Eccentric
Distribution of the Foundation and Machine Mass
Experimental Investigations of Vibrations of Massive
Foundations
a.

III-5.

Theory of Vertical Vibrations

a.

Verification of the

b.

Experimental Investigations of the Coefficient of

c.

Investigations of Vibrations of Pile Foundations

d.

Experimental Investigations of Complicated Forms of

Foundation Vibrations

Damping

CHAPTER

IV.

IV-1.

....

General Directives for the Design of Foundations

a. Design Values of Permissible Amplitudes of Founda.

for

Foundation Design

Foundation Material
d. Comments in Regard to Design
Unbalanced Inertial Forces in Reciprocating Engines
a. Single-line Machines
b. Multicylinder Engines
c. Vertical Two-cylinder Engines
d. Reciprocating Horizontal Compressors
e. Vertical Three-cylinder Engine
f. Vertical Four-cylinder Engine
g. Vertical Six-cylinder Engine
Stresses Imposed by Belt Pull

c.

IV-2.

....

IV-3.

110
114

118
122
122

129
129
131

131
131

tion Vibrations

Machine Data Required

107
107

126

Foundations under Reciprocating Engines

b.

86

...

132
132
133
135
135
142
144
145
146
146
146
147

CONTENTS
IV-4.

Examples

xix

Dynamic Analyses

of

Foundations for

of

Reciprocating Engines

Example

1.

149

Dynamic computations

vertical compressor coupled

of foundation for a

on a shaft with an

electro-

motor

Example

149
2.

Dynamic

analysis of a foundation for a

reciprocating horizontal compressor

IV-5.

Methods

154

for Decreasing Vibrations of Existing

Founda160

tions

Counterbalancing of Exciting
Engines
b. Chemical Stabilization of Soils
a.

IV-6.

CHAPTER

V.

V-l.

c.

Structural Measures

d.

Dynamic Vibration Dampers

Loads Imposed by
160
161
162
167

Analysis and Design of Foundations with Vibration

Absorbers
Example. Design computations for a foundation with
absorbers under a vertical compressor

Foundations for Machines Producing Impact Loads.

General Directives for the Design of Foundations for

Forge Hammers
a. Classification of Forge Hammers
b. Design Data for a Hammer Foundation
c. Material of the Foundation and Pad under the Anvil
d. Directives for Design
e. Remarks on Construction Procedures
Initial Conditions of Foundation Motion under Impact
Action
a. The Velocity of Dropping Parts at the Beginning of
.

V-2.

b.

Impact
Experimental
Coefficient

V-3.

of

the

77

Foundation Motion

c.

Initial Velocities of

d.

Coefficient of Restitution e

185

185
185
185
187
187
189
190

Correction

....

Natural Vibrations of a Hammer and Its Foundation as

a Result of a Centered Impact
a. The Main Assumptions Involved in Design Computations

Equations of the Vibrations of Foundation and Anvil

Experimental Studies of Vibrations of Foundations
Under Forge Hammers

Introduction
b. Description of Bases and Foundations
c. Results of Measurements of Foundation and Anvil
Vibrations
d. Amplitudes of Foundation Vibrations
a.

e.

Amplitudes of Anvil Vibrations

f.

The Determination

of

Elastic

Constants

of

Comparison of Different Formulas

tion of Amplitudes

192

193
195
196

196
197

200
200
202
202
205
205

the

206

"Anvil-Foundation" System
g.

181

190

Determination

b.

V-4.

172

for the

Computa209

CONTENTS

:x

V-5.
V-6.

Weight and Base Area of a Hammer

Foundation
Design of a Hammer Foundation
Example. Dynamic computations for the design of a

Selection of the

....

hammer

foundation

Standard Illustrative Designs of Hammer Foundations

Computation and Design of Hammer Foundations with
Vibration Absorbers
a. General Directives on Computation and Design
Example. Computations for a stamping-hammer
foundation with absorbers
b. Construction
Procedure for a Foundation with
Absorbers
Pressures in the Base under a Foundation Subjected to
Horizontal Impacts
Foundations (Bases) for Drop Hammers to Break Scrap
.

V-7.

V-8.

V-9.

Iron

Location of Drop Hammer within the Steelworks

b. Design of Crushing Platforms under Drop Hammers
Example. Dynamic computations for a cylindrical
base under a crushing platform
a.

CHAPTER

VI.

VI- 1.

Frame Foundations

for

Machinery

Instructions for the Design

Frame Foundations

b.

Design Assignment

c.

Instructions for the Design

Computations

of

....

Loads Imposed by
Electrical Machines

a.

Exciting

b.

Modulus

of Elasticity of Reinforced Concrete

d.

Rigidity of the

e.
f.

Computation
Computation

g.

Design

Upper Platform

of

of the

Foundation

Forced Vertical Vibrations

of Horizontal Transverse Vibrations

Values
Vibrations

of

the

Permissible

Amplitude

VII.

233
233
234
240

242
242
243
243
249

250
255

258
259
259
264

of

268

Dynamic Computations of Foundations of

the Frame Type
Example 1. Illustrative design of a frame foundation

270

under a 1,500-kw turbogenerator

270

Examples

of

2.

Illustrative design of a foundation for a

278

500-kw generator

VII-I.

232

Design Dimensions of the Upper Part of the Founda-

Example

CHAPTER

231

Turbodynamos and

tion

VI-3.

227

250

tions

c.

220
220

Frame

Construction Operations
Forced Vibrations of Frame Founda-

d. Instructions for

VI-2.

of

Field of Application of

216
219

242

and Construction

Foundations
a.

211
216

Massive Foundations

283

....

Massive Foundations for Motor Generators

General Remarks in Regard to Design and Design
Computations
Example. Dynamic computation

283

283
284

CONTENTS

xx

....

VII-2.

Massive Foundations under Turbodynamos

VII-3.

Foundations for Rolling Mills

a. Dynamic Loads Imposed on the Foundation by the
Driving Roll Motor
b. Dynamic Action on the Foundation by the Ilgner
c.

d.
e.
f.

Power System
Dynamic Loads on the Common Foundation of
Working and Gear Stands
Remarks Concerning Design
Design Loads
Data on Performance of Existing Foundations
under Rolling Mills

VII-4.

VIII.

VIII-1.

291

292
296
298
299
303

307
307

..."

307

Propagation of Elastic Waves in Soil

311

Introduction

311

an Infinite Solid Body

Longitudinal and Transverse Waves
Propagation Velocities of Longitudinal and Trans-

Elastic
a.

b.

Waves

verse
c.

VIII-2.

288

304

Foundations for Crushing Equipment

Computations of Foundations under Jaw
a. Design
Crushers
b. Computations of a Foundation under a Gyratory
Crusher.

CHAPTER

in

Waves

Propagation of Waves
Explosion: Camouflet

312
312

315

Induced by

Changes with Depth in Amplitudes

Produced by Surface Waves

Underground
316

of Soil Vibrations

318

General Expressions for Components of Soil Displacements Induced by Waves Caused by a Vertical
Exciting Force Acting along a Surface Line
b. Propagation Velocity of Surface Waves
c. Approximate Expressions for Soil Surface Displacements Induced by Waves Excited by a Concentrated
Vertical Force
d. The Variation with Depth of Surface-wave Amplitudes
e. Experimental Investigations of Variation with Depth
of Surface-wave Amplitudes
a.

....

VIII-3.

Dependence of the Amplitude of

Distance from the Source of Waves
a.

b.

c.

d.

VIII-4.

Soil

322
323

325

Vibrations on

331

General Expressions for Soil Displacements Produced

by Concentric Waves
Displacement of Soil at Small Distances from the
Source of Waves
Displacement of Soil at Large Distances from the
Source of Waves
Experimental Investigations

Waves on the
Amplitude of Vibrations of the Soil Surface
Example. Computation of amplitudes of the vertical
component of soil oscillations produced by shocks to a
foundation under a forge hammer
Influence of the Area of the Source of

318
321

....

331

333

336
338
349

354

CONTENTS

xxii

VIII-5.

Dependence

of Soil Vibration

of the Source of

VIII-6.

Amplitudes on the Depth

Waves

The Spreading of Surface Waves in Layered Soils

a. The Influence of a Layered Soil Structure on the
tribution of Surface Waves
Surface Waves in Two-layered
b. Transverse

Dis-

365
Soil

Systems
VIII-7.

Screening of Elastic
a.

c.

Soil

Waves

Experimental Screening of Elastic Waves in Loess

Experimental Screening of Waves in Water-saturated
.

Some Examples from

the Practice of Screening Waves;

General Conclusions

CHAPTER

IX.

IX-1.

Effect on Structures of

402

Waves from

Industrial Sources

Classification of Industrial Sources of Elastic

Waves

407

Action on Structures of Seismic Waves Propagating from

Foundations under Low-frequency Machinery
Main Cause of Vibrations of
a. Resonance as the
.

Structures
b.

d.

e.

The Action

of

Foundations
References

Name

Index

Subject Index

413

415
416

418

tion

IX-3.

409
409

Measures to Decrease Vibrations of a Power-station

Wall Induced by Operating Saw Frames
Vibrations of a Boiler-shop Wall Induced by Operating Compressors
Measures to Decrease Vibrations Induced by Waves
from Compressor Foundations
Measures to Decrease Vibrations of a Refrigerator
Plant Induced by Waves from a Compressor Founda.

c.

407

in

Soil

IX-2.

374
377
383

Soils
d.

370
374

Elementary Concepts of Physics Concerning the Diffraction of Elastic

b.

Waves Propagating through

356
365

Shocks Caused by Waves from

Hammer
419

425
429
431

NOTATION

Many symbols used in the original Russian text were also translated
and were replaced by symbols customary in the United States. Both sets
of symbols are given in the list which follows.
This list does not include
all subscripts to the main symbols which are defined in the text.
The metric system was used in the Russian text and has been kept
The

in the English translation.

may

following conversion factors

be

helpful to the reader

1

ton
1

m =

cm =
=
1 m
=
1 m
Xm =

ton

10 3 kg

cm =

100

0.395

X
X
X

2,205 lb

3.28 ft

39.4

in.

in.

10 4
10 6
10 5

cm = 10.35 ft
cm = 35.3 ft
kg X cm = 7.25 X 10 ft X
= 8.7 X 10 in. X
2

kg
1

X cm

ton/m 2

ton/m 3 =

5.61 lb
0.1

in.

kg/cm 2

10

206 lb /ft 2 (or

1.42 psi

kg/cm 3 =

=
1

ton

sec 2 /m
1

675 lb

lb

lb

0.1 U.S. ton/ft 2 )

56.6 lb /ft 3

3.28

10- 2 lb/in. 3

sec 2 /ft

kw

1.36

hp

ig

9.81

m/sec 2

32.2 ft/sec 2

Symbols

Used

Used

in

English

Definition

translation

text

Area

of foundation or attached slab in contact with soil;

hammer piston in Eq. (V-2-2)

Amplitude of vibrations in Eq. (III-I-7); constant; coefficient in Eq. (VIII-4-1)
Amplitude of soil vibrations where source of waves is placed
at a depth / below soil surface in Eq. (VIII-5-13)
cross-sectional area of

in

Russian

A
Af

NOTATION

Symbols (Continued)
Used

Used

in

English

Definition

translation

A
A

A
Ax

st

Ay

text

Permissible amplitude of vibrations in Eq. (V-5-3)

Amplitude of soil vibrations where source of waves is placed
on soil surface in Eq. (VIII-5-13); ratio in Eq. (VIII-3-24);

^0

At

Au
A<p

A Ron
A

amplitude of soil vibration at distance r from source

Amplitude of vertical soil vibrations at distance r from A r
source in Eq. (VIII-3-42)
Static vertical displacement of foundation in Eq. (III-1-15) A C m
Amplitude of horizontal vibrations in shear in Eq. (III-3-3) A x
Amplitude of horizontal vibrations in shear of center of Ay
gravity of whole system in Example IV-4-1
Amplitude of vertical vibrations
Az
Amplitude of forced vertical vibrations in Eq. (III-1-21)
a:
Amplitude of vertical soil vibrations in Eq. (VIII-4-9)
Ao,
Amplitude of rocking vibrations in Eqs. (III-2-7) and A,
(I V-5-3)

Ax
a

<p

Combined amplitude produced by rocking and

tions in Example IV-4-2

sliding vibra-

Half length of a foundation; full length of a foundation in

Art. III-2; coefficient in Eqs. (II-1-5), (VIII-1-8), and
(VIII-4-8); amplitude of forced vibrations of a foundation
in Eq. (III-1-29); ratio of vibration amplitudes in Eqs.

Ax

<p

(IV-5-7) and (IV-5-8); distance'in Eq. (VI-2-17); velocity

waves in soil in Table VIII-2

Amplitude of anvil vibrations in Eq. (V-4-7)
Am
Amplitude of forced vibrations at resonance in Eq. (III-1-34) a r
Constant in Eq. (II-3-14); amplitude coefficient in Eq. (III- B
1-7); constant in Art. IV-2-a and Eq. (VIII-4-1); constant
of propagation of longitudinal

aa
ar

Ca
ck
c

Ca

in

Russian

of elastic aftereffect in Art. VIII-3-d

Half width of a foundation; full width of a foundation in
Art. III-2; dimensionless value in Eq. (III-1-31); thickness
of pad under anvil in Art. V-3; distance in Fig. VI-8 and
Eq. (VI-2-17); ratio in Eq. (VIII-1-7); velocity of propagation of transverse waves in soil in Eq. (VIII-6-22)
Coefficient in Eq. (1-2-13); arbitrary constant in Eqs.
(II-3-14), (III-2-5) and (VIII-6-3); total rigidity in Eq.
(VI-2-14); constant in Eq. (VIII-3-21)
Arbitrary constant in Eq. (III-4-17)
Coefficient of elastic resistance of one pile in a group
Damping constant in Eq. (III-1-17)
Coefficient in Eqs. (III-4-16), (IV-6-2), and (V-3-5); velocity
of surface wave propagation
Velocity of surface waves when there is another soil layer
above that surface in Fig. VIII-23

Cb

Velocity of transverse

Reduced
(1-6-3)

soil

waves

coefficient of elastic resistance of a pile in

Ca
ck
n
c

Ca

Cb

Eq.

NOTATION

Symbols {Continued)
Used

Used

in

text

translation

Coefficient of proportionality variable along the base of a

C(A)

C(F)

bearing plate
Coefficient of proportionality; also called

cp

modulus

of sub-

cp

grade reaction
Cr

Coefficient of rigidity of soil base in Eq. (III-1-2)

Cs

Coefficient in Eqs. (1-2-12)

and

and Table

1-3 for

kz

1-3 for rigid rectangular

(1-2-14)

flexible rectangular footing

i

Coefficient in Eq. (1-2-12)

and Table

footing
Csp

Rigidity of one spring in example in Art. IV-6

Cu

Coefficient

of

Cz

C 'u

Coefficient of elastic uniform vertical compression of soil

Cz

of

elastic

uniform

"

vertical

Cup

compression

soil

under hammer foundations

Cx

cr

Cz

cu

A
A

in

Eq. (V-4-5)

in Art. III-4-d

Cx

Eq. (III-4-16)
Cz
Coefficient of elastic nonuniform shear of soil
Cf
C<p
Coefficient of elastic uniform shear of soil
Coefficient of elastic nonuniform compression of soil
C
C
Coefficient of elastic resistance of a pile in Eq. (1-6-1)
Arbitrary constant in Eqs. (II-3-14) and (VIII-6-3); diam- D
eter of spring coil in Eq. (IV-6-3); expressions in Eqs.

Cf
Cr
C(p

Cs

in

(VIII-3-29)

Diameter of a pile; diameter of a spring in Eq. (IV-6-3)

Young's modulus; arbitrary constant in Eq. (VIII-6-3)
Young's modulus for concrete
Void ratio; coefficient of restitution in Eq. (V-2-5)
Base of natural logarithms

6min

Minimum

E
E

value of void ratio corresponding to densest pos-

E
E
e

sible state in Art. II-2

Maximum

fimax

value of void ratio corresponding to loosest pos-

eo

sible state

e= x -\-y-\- z Volume expansion

value of void ratio prior to start of vibrations in Art.

function; centrifugal force in Eq. (VI-2-1); arbitrary con-

Initial

II-2

stant in Eq. (VIII-6-3)

F X ,F

Horizontal, vertical components of centrifugal force in Eq.

P X ,P*

(VI-2-2)

Depth below the

soil

surface of the source of

waves

in

Eq. J

(VIII-5-1)
fa

h
/

in

Russian

Definition

English

Frequency of natural vibrations of foundation above the Xp

springs in example of Art. V-7
Frequency of natural vibrations of foundation below the Xz
springs in example of Art. V-7
X
Natural frequency of vibrations o)/2ir

NOTATION

Symbols (Continued)
Used

Used

in

English

Definition

translation

text

Frequency

Jna

of natural vibrations of attached

masses in Eq.

XP

(IV-5-9); frequency of natural vibrations of the anvil in

fnd
J nx

J nz

fn<p

Ui>

fhft

u,u
/nl,

Eq. (V-3-6)
Natural frequenc}' of damped vibrations in Eq. (III-1-18)
Xi
Natural frequency of horizontal vibrations in Fig. 1-31 and x*
Eq. (III-3-2)
Natural frequency of vertical vibrations in Eq. (III-1-5)
xz
Natural frequency of free rocking vibrations of a foundation \<p
in Eq. (III-2-6)
Natural frequency of torsional vibrations around a vertical \f
axis in Eq. (III-3-5)
Functions in Eqs. (III-1-26) and (VIII-1-13); frequencies of fuf*
forced vibrations in example of Art. VII-1
First two main natural frequencies of a foundation, equal to Xi, x 2
the two positive roots of Eq. (III-4-8) [see Table (111-7)1
Main natural frequencies of a foundation for limiting case Xi, X2
6 =
in Eq. (III-4-20)

G,n

Modulus

Acceleration of gravity

Height; distance; thickness; coefficient (with various sub-

of

elasticity

in shear,

i.e.,

modulus

of rigidity

n,G

scripts)

Height; ratio co/a in Eq. (VIII-2-5)

Moment of inertia of a bearing plate with respect to one of
its axes; moment of inertia of a mass in Eq. (VII-3-6)
\/ 1 radius of gyration of mass of foundation and machine
in Art. III-4; moment of inertia per unit of mass in Eq.

h
I

J
i

(III-4-21)
J\,

J'2

Bessel's functions for solution of Eq. (III-1-26); used in Eq.

Ji,J*

(VIII-5-7)

Expressions in Eq. (VIII-3-16)

J:

Polar

moment of

inertia of the base contact area of a founda-

J,

tion
3

Coefficient in equation defining v* in Art. II-3

Coefficient of lateral earth pressure (ratio

and

K
K

Kinetic energy in Art. V-7

Coefficient of elastic resistance of a pile group

fc.

lateral

vertical pressures); coefficient in Eq. (1-3-1)

ia

kT

between

K(p

in

Russian

Coefficient of elastic

nonuniform resistance

of pile founda-

tions in Eq. (1-6-11)

Constant in Eq. (III-1-34); number of engine cylinders in
Eq. (IV-2-9); dynamic correction coefficient in Eq. ( V-4-5);
ratio in Eq. (VI-2-9); ratio co/b in Eq. (VIII-2-5); number
in Eq. (VIII-6-28)
Coefficient in Eq. (1-3-5)
Coefficient in Eq. (1-4-6)

K<p

Kip

NOTATION

Symbols {Continued)
Used

Used

in

text

translation

Distance between a pile and the axis of rotation of the foun-

dation in Eq. (1-6-10); distance; length
Distance between the axis of rotation and the center of h
gravity of the vibrating mass in Fig. III-9
L
Wavelength in Eq. (VIII-6-27)
Length of transverse waves in Eq. (VIII-5-15)
Wavelength in Eq. (VIII-2-25)
Lc
Length of shear waves propagated through soil from founda-

L
L

U
u
L

in

Russian

Definition

English

tion [see solution of Eq. (III-1-26)]

I

M
Mi
Mm

Length; depth of pile penetration in Fig. 1-28; constant in I

Eq. (III-1-34)
Distance in Fig. III-9
i
Depth of pile fixation in soil; span of beam in Eq. (VI-2-4a) u
External moment; mass; moment of inertia in Eq. (VI-2-13)
External exciting moment in Eqs. (III-2-1) and (III-4-12)

M
M

Moment of inertia of

the mass in respect to the y axis, which

passes through the center of gravity of foundation and

machine

M m0

Moment

in Fig. III-ll

and Eq.

Mr

M
M
M
M

(<)

/If,
2
3

m
Wo
W]

w
m
m
N

Total mass of

mass (foundation e

through center of gravity

of base contact area of foundation in Art. III-4
Moment of soil reactions in Eq. (III-2-2)
Moment of external exciting forces in Eq. (III-4-4)
Exciting moment acting in a horizontal plane in Eq. (III-3-4)
Mass of the crank in Eq. (IV-2-3)
Mass of the rod, crosshead, and piston in Eq. (IV-2-3)
Mass of the connecting rod in Eq. (IV-2-3)
Mass; mass of foundation in Eq. (V-2-3); empirical coefficient in Eq. (VIII-5-14)
Mass of falling ram in Eq. (V-2-3)
Mass of foundation in Eqs. (V-3-5)

Mass
Mass

(III-4-2)

of inertia of the total vibrating

and machine) with respect

to axis

M
M
M
M
M
M

r
{t)

z
l

2
3

m
m
n%\

of anvil in Eqs. (V-3-5)

nia

of soil in Eq. (III-1-37)

m TP

and foundation
Number of oscillations per minute (speed) = 60a>/27r = 60n N
Number of piles in Eq. (1-6-4); number of oscillations per n
second (in hertz) = o/2ir; number of springs in example of
soil

Art. IV-6

n
na

Coefficient of vibroviscosity in Eq. (II-1-2)

nf

P
Pact

Ratio of weight of anvil and frame

a to weight of dropping
parts of hammer
in Eq. (V-5-9)
Ratio of weight of foundation Wf to weight of dropping
parts of hammer
in Eq. (V-5-9)
Exciting force
Actual load on a spring in example of Art. IV-6

nm
n<t>

P
P^eacTB.

NOTATION

Symbols (Continued)
Used

Used

in

English

Definition

translation

text

Pc

Centrifugal force in Art. IV-2-a

Wi

pp

Permissible load on a spring in examples of Arts. IV-6 and

Proii

Pr

Peripheral stress transmitted, to

V-7

by

pulley

Eq.

Pr

along foundation

PX

belt

in

(IV-3-2)

Pt

Total horizontal tangential pressure on

soil

base; resultant force of pull on belt in Eq. (IV-3-1); total

horizontal exciting force in Eq. (IV-5-1) and example of
Art. VII-1

Pit)

PX
P
Z

Exciting force which changes with time

Horizontal force; horizontal projection of a force
Vertical force; vertical projection of a force; exciting load in
Eq. (VIII-3-9)

Exciting force P(t) per unit of mass in Eqs. (III-l-ll) and

(IV-5-9); pressure on piston of hammer in Eq. (V-2-2)

Vdy

Pst

Dynamic pressure on soil beneath a foundation in Eqs.

(II-3-3)and(V-5-l)
Limit pressure on soil beneath a foundation in Art. II-3
Permissible pressure on soil if only static load is acting in
Eq. (V-5-1)
Static pressure on soil beneath a foundation in Eqs. (II-3-2),

P*

and (V-5-1)
Vertical pressure on horizontal

Coefficient of required pressure decrease

Qo

ence of vibrations in Eq. (II-3-3)

Radial component of soil surface displacement in Eq.

Pi

Po

Pit)

PX
P*
P
<?d

<Tl

Pern

O'cm

(III-1-35),

soil

surface

on

P*
soil in

the pres-

qo

(VIII-5-2)

Total reaction of

soil

against a bearing plate in Eqs. (1-2-3)

and (III-1-25); radius of a plate in Eq. (1-2-6); force acting

anywhere in Eq. (II-1-2); distance between a center of
gravity and an axis of rotation in Art. IV-2-a; ratio in Eq.
(VI-2-19); radius in Eq. (VIII-4-3)

Ri

Elastic force in Eq. (VI-2-13)

Rdy

Maximum dynamic

Rst

Maximum

Ri
exceeded, causes large

Rd

static pressure which, if exceeded, causes large

settlements in Eq. (II-3-2)
Permissible torsional stress of the spring material in example

Rs

pressure which,

if

settlements in Eq. (II-3-3)

RP

of Art.
r

To

in

Russian

Rroii

IV-6

Radius of a sphere in Eq. (II-1-2); ratio in Eq. (II-3-11);

eccentricity of impact in Eq. (V-2-4); radius of gyration in
Eq. (VI-2-13); radius in Eqs. (VIII-1-13)
Radius of a circle in Eq. (III-1-26); eccentricity of machine
rotor in Eq. (VI-2-2)
Settlement; perimeter of a pile in Eq. (1-6-3); load coefficient in Eq. (II-1-3)

z,

NOTATION

Symbols {Continued)
Used

Used

in

translation

in

Russian

Definition

English

text

Coefficient defining slope of settlement-time curve during

vibration in Eq. (II-3-13)

Average settlement of a flexible plate in Eq. (1-2-10)

Average value of horizontal sliding of a foundation

Say

$av

Se

Total elastic settlement of a bearing plate

Total elastic sliding of a bearing plate
Total settlement of a bearing plate in Eq. (1-2-1)
Total sliding of a bearing plate in Eq. (1-4-2)
Shearing strength of soil; distance in Fig. VIII-12
Period of vibration; tension in a machine belt in Eq. (IV-3-1)

Time; distance between

s:
St
$tot
s

u
U, V,

Ua
UT

Va

piles

Cp

u cp
z

ux

Zn
U

T
t

Slope of a curve in Art. II-2

Components of displacements

U,

Displacements on soil surface in Chap. VIII

Radial displacement of soil in Eq. (VIII-1-15)

Uz
Ur

Volume

Velocity; peripheral speed of engine belt in Art. IV-3

Vo

Initial velocity

Maximum

Vo
*
Vo

value of

initial velocity of

vibratory foundation

motion at which no residual settlements

V,

Vh

will occur in Art.

II-3
Vs

W
w
Wm

Velocity of shear waves in soil [see solution of Eq. (III-1-26)]

Work; engine power

Weight

Moment

Qm

mass of the whole system with

an axis passing through common center of

of inertia of the

respect to

gravity perpendicular to the plane of vibrations

Moment

Wo

w
Wo
X, F,

of inertia of the mass of foundation and machine

with respect to horizontal axis of rotation through the
center of gravity of foundation in Eq. (III-2-1)
Moment of inertia of the mass of foundation and machine
with respect to vertical axis of rotation in Eq. (III-3-4)
Water content (by weight)
Water content in Eq. (1-1-19)
Components of mass forces in Eq. (VIII-1-1)

Xi

Projection on x axis of

Horizontal displacement of the center of gravity of foundation in Eq. (III-3-1); horizontal displacement of the common center of gravity of foundation and machine in Art.
III-4; projection on a horizontal axis of the displacement
of the center of gravity of foundation in Eq. (V-8-1)
Horizontal displacement of center of gravity of contact area
of foundation in Art. III-4

on foundation

Yu Y u

in

all

external

and

ez
p or

or q

Po

Y,Z

X,

Xi

Eq. (III-1-4)

Bessel functions in Eqs. (VIII-3-27)

symbol J)

inertial forces acting

and (VIII-4-4)

(see also

Xo

Fi,

Yu

NOTATION

Symbols (Continued)
Used

Used

in

English

Definition

translation

text

Projection on z axis of

on foundation
x, y, z
x%, yt, Si

in

Russian

in

all

external

and

inertial forces acting

Zi

Eq. (III-4-1)

Rectangular coordinates
Coordinates of a point i of moving machine parts in Eq.

x, y, z
Xi,

yi}

Zi

Xq,

2c

(IV-2-1); coordinates of centers of gravity of separate ele-

yo, ZQ

z
z

z
Zst

ments of a vibrating system in Example IV-4-1

Coordinates of the common center of gravity of the vibrating
system in Example IV-4-1
Vertical displacement of foundation
First derivative of vertical displacement
Second derivative of vertical displacement
Elastic settlement of foundation under the action of the
weight of foundation and machinery
Ratio of half length a to half width b of a foundation in Eq.
(1-2-11); coefficient in Eqs. (1-1-16) and (I-l-18a); coefficient of vibratory compaction in Eq. (II-2-1); angle of
phase shift between displacement and soil reaction =
e in Eq. (III-1-30); coefficient defining c in Eqs.
v?
(III-1-17) and (III-1-37); arbitrary constant in solution of
Eq. (III-4-5); characteristic number of crank mechanisms
in Eq. (IV-2-7); pressure reduction coefficient in Eq.
(V-5-1); hammer coefficient in Eq. (V-7-2); coefficient in
Eq. (VI-2-4a) for frame foundations; coefficient in Eq.
(VI-2-24); charge coefficients in Eq. (VIII-1-19); coefficient in Eq. (VIII-2-20); coefficient in Eq. (VIII-6-1);
coefficient in Eq. (VIII-6-13); coefficient of absorption of
wave energy in Eq. (VIII-3-43)
Coefficient in Eqs. (1-1-11) and (II-l-l); coefficient of
increase of foundation mass due to participation of soil in
Eq. (III-1-40); ratio of foundation frequencies in Art.
III-4, Fig. 111-13, and Eq. (IV-5-14); angle between cranks

z
z

Eq. (IV-2-9); ratio of dynamic displacements of foundation and anvil in Eq. (V-4-2); charge
coefficient in Eq. (VIII-1-19); coefficient in Eqs. (VIII-2-6),

of different cylinders in

(VIII-2-20),

and (VIII-6-26)

Unit weight in Eqs. (II-1-4) and (VIII-1-1); angle of phase

shift between exciting force and the displacement it induces
in Eq. (III-1-22); ratio of moments of inertia following Eq.
(III-4-7); angle with the horizontal of straight line through
axes of rotation of pulleys in Art. IV-3; coefficient depending on parameters of the engine in Eqs. (IV-6-9) and
(VIII-3-26); ratio of frequencies in Eq. (V-4-3); moment
causing rotation of the beam of a frame foundation by an
angle equaling unity in Eq. (VI-2-17)
Ixy, Jxz, Jyz

Shearing-strain components in rectangular coordinates

Gxyy *xz, *yz

NOTATION

Symbols (Continued)
Used

Used

in

Definition

English
translation

text

Ratio of damping coefficient c to natural frequency / 2 relative change in volume in Eq. (VIII-1-1)
Coefficient in Eq. (IV-6-8)
Elastic settlement of a pile; effect of vibrations on decrease
of internal friction in soil in Art. II-1-a; ratio in Eq.
(IV-5-7); lateral displacement of frame foundation in Eq.
(VI-2-15)
Laplace operator
Phase shift between exciting force and soil reaction in Eq.
(III-1-25); eccentricity in distribution of the mass of
foundation and machine in Eq. (III-4-15); coefficient of
nonuniformity of engine speed in Eq. (IV-5-14); distance in
Eq. (VI-2-14) and Fig. VI-8
Unit elongations in x, y, z directions
Elastic unit deformations
Horizontal elastic displacement of a pile head
Element of width of a foundation base area in Eq. (1-4-3);
ratio of the acceleration of vibrations to the acceleration of
gravity in Eq. (II-l-l); dynamic modulus (magnification
factor) of undamped vibrations in Eq. (III-1-15); ratio of
amplitudes (degree of vibration absorption) in Eq.
(IV-6-11); hammer coefficient in Eq. (V-2-1) and Art. V-7
Coefficient of viscosity in Art. VIII-3-d
Dynamic modulus of damped vibrations in Eq. (III-1-23)
Threshold of vibratory compaction in Eq. (II-2-5)
Angle in cylindrical system of coordinates in Eq. ( VIII-3-16)
Coefficient in an Eq. of Art. III-1-d and Fig. III-5; root
of Eq. (VIII-2-17) defined by Eq. (VIII-2-18)
Ratio in Eq. (VIII-3-16)
Lame's constant = E/[(l + v)(l - 2v)]
Coefficient in Eq. (1-6-5); ratio of limiting frequencies in Art.
III-4; ratio of natural horizontal and rocking frequencies
;

A(co 2 )
8

V2
e

r, *y,

in Art. 111-13; coefficient of friction of belt

Mo

Go

in

Russian

on pulley

in

Eq. (IV-3-4); ratio of masses of damper and foundation in

Eq. (IV-5-14); ratio of masses above and below springs in
Eq. (IV-6-8); ratio of masses of foundation and ram in Eq.
(V-2-6); ratio of masses of anvil and foundation in Eqs.
(V-3-8) and (V-4-4); dynamic coefficient in Eq. (VII-3-5)
Lame's constant ( = G)
Modulus of rigidity of soil in its initial state when no
external stresses are acting thereon
Poisson's ratio

Element

of length of a foundation base area in Eq. (1-4-3);

reduced coefficient of damping in Art. II-l-c and Eq.
(III-1-39); ratio of actual to natural frequencies (frequency

A
A()

V2

e
ex

NOTATION

xxxii

Symbols (Continued)
Used

Used

in

Definition

English

in

Russian

Translation

text

Eq. (III-1-16); reduced damping

Eq. (III-1-39); ratio of the limit natural frequency of a foundation to the exciting frequency in Eq.
(IV-6-9); constant in Eq. (VIII-2-6)
c/fnd, coefficient in Eq. (II-3-7); ratio of constants expressed
by Eq. (III-4-9); ratio of amplitudes in Eq. (III-4-13);
square root of the ratio of moment of hammer and foundation mass inertia to the mass of hammer in Eq. (V-2-6);
density of soil in Table VIII-1; kr in Eq. (VIII-3-16)
Radius vector of a point within the area of a bearing plate
ratio) of vibration in
coefficient in

P
<r
0"

z , (Ty,

<T Z

T
Ta v

T xy> Txzi Tyz

*
<t>

Normal stress
Normal components

V
of stress parallel to x, y, z axes

Shearing stress; relative time in Eq. (II-3-7)

Average shear stress in soil along plane of contact with
foundation
Shear-stress components in rectangular coordinates
Function in Eq. (VIII-6-3)
Angle of internal friction of soil;

X, Y,
P,

Vx
J^ X,

<

= coefficient of internal friction of soil

tan
/
Angle of internal friction without vibrations;
tan s = coefficient of internal friction without vibrations Jem
Slope angle of a deflected plate along the line x =
in Eq. f
(1-3-1); angle of rotation of foundation in Eqs. (1-6-10)
and (III-2-2) and Fig. III-9; phase shift between exciting
force and displacement in Eq. (III-1-30); angle of rotation
<f>

<t>st

<f>

<P

of crankshaft; angle of belt contact in Fig. IV-6; function

in
<Pl

Eq. (VIII-3-1)

Arbitrary constant in Eq. (III-2-5);

initial velocity of rota-

<P0

tion of foundation in Eq. (V-2-4)

V
f

12
CO

Function in Eq. (VIII-6-1)

Angle of rotation of a foundation around its vertical axis in
Eq. (III-3-5); coefficient of energy absorption in Art. II-l;
angle in Fig. VIII-12
Function in Eq. (VIII-6-9)
Frequency of vibrations; frequency of exciting force; angular
velocity of machine rotation [in Eq. (IV-2-2)] = 2-n-N/QO =
2irn

^
<A

ft
CO

z x Zy
,

ELASTIC PROPERTIES

1-1.

a.

Deformations of

Elastic

Soil

Its

Elastic Constants

There are two kinds of external forces

forces and surface forces.

Generalized Hooke's Law.

acting on a deformable body:

and

OF SOIL

body

on volume elements. They comprise gravity forces,

and forces of mutual attraction. The value of the body
force per unit volume is called the mass force.
Classical examples are hydroSurface forces act on surface areas.
static pressure and forces of internal and external friction.
Forces of
interaction between the parts of a body belong to the category of internal

Body

forces act

inertial forces,

surface forces.

The magnitude
called

stress.

of the surface force per unit area of the surface is

Stresses induced

deformable bodies are called

by

internal

surface forces acting in

elastic stresses.
r

The magnitude

of the stress at a point

depends on the orientation of the stress

plane through the point. The stress acting
on any inclined plane through a point can
be obtained from stresses acting on three
planes passing through the point and perpendicular to the coordinate axes x, y, and
z.
Thus a stress acting on a plane perpendicular to the x axis

may

/
Tyz

'

'yp

'

/*.
Jxi

Cx
,

1
l

Ty X

*y

be resolved into

three components a x r xy and t xz (Fig. 1-1);

:

a x will be the normal

stress

and

r xy

and

component

of the

txz will be shearing

Fig. 1-1.

Components

of stress.

com-

ponents parallel to the y and z axes. Correspondingly, stresses acting

on planes perpendicular to the y and z axes may be resolved into the components r yx o-y, T yz and t zx T zy a z <j y and a z will be the normal components,
and the rest components of the shearing stress. The stress conditions of
,

DYNAMICS OF BASES AND FOUNDATIONS

the body at a point are then determined

where

Only
If

Tzy

six of the nine

<Tx

Txy

Txz

Tyx

(Ty

Tyz

Tzx

TZy

Tz x

components

by the

totality of nine values:

(T z

Txz

of stress are not interdependent.

a body undergoes deformations, then under the action of stresses the

body (i.e., its form) will change. Let

and w be the components of displacement along the coordinate axes
In the general case the values of these comx, y, and z, respectively.
ponents depend both on the coordinates of the point under consideration
and on time.
Assuming that deformations are small, we may represent the linear
strain of an element of length parallel to the x axis as
distances between points of the

u,

v,

du
dx
Similarly the linear strains of elements parallel to the y and z axes are
dv
y

_
z

~di

dw

~~dz~

However, the state of strain of a body is determined not only by linear

by the rotation of some linear elements in respect to
other elements.
As a result of deformations, a right angle between two
linear elements parallel to the x and y axes and originating at the same
point is changed by the value
dv
du
strains but also

lxy

~"d~x^"d~y

This deformation or distortion of the right angle

detrusion.

x and

called shear strain or

between the

axes and between the y and z axes will equal

dw

du

7xs

~dx" "dz'

yys

~Tz^d^

dw

dv

The

is

Similarly, the distortions of the right angles

quantities

and y xy y xz yyg which determine the deformation

components of strain.
the deformed body are of such magnitudes that

e x , eV} e 2

of the body, are called

If

the stresses in

(that is, no shearing occurs), then the deformation is

yXz = yVz =
accompanied only by a change in volume. The relative change in volume
Ixy

OF

ELASTIC PROPERTIES

SOIL

for small deformations equals

ex

ey

0, then the deformation leads to the sliding of some

elements of the body relative to other elements, and the volume of the
body will not change. According to the theory of elasticity, any small
deformation can be resolved into a sum of deformations accompanied
only by change in volume and of deformations accompanied only by shear.

If

ex

ev

If after

one

it

ez

the removal of the load a body resumes

had before the load was

shape

its initial

applied), then this

body

(i.e.,

the

is elastic.

truly solid body, as a rule, possesses only a partial elasticity, because,

the body does not exactly resume

after unloading,

its

initial

shape

have taken place). We shall consider here

elastic
deformations
which
only
can always be calculated by subtracting
the value of the residual deformation from the total deformation corre(i.e.,

residual deformations

sponding to the given loading.

It is

known from

experience that elastic deformations increase pro-

Assuming that internal

in external loads.
body are proportional to external loads, we may reach the
conclusion that deformations of a body depend on the stresses which act
within it. In general form this interrelationship between the stresses
and deformations can be analytically expressed by six functions, as
portionally

to

increases

stresses in a

follows:
(t x

= Fx (e x ,y,e
& Vz\x)y

Tyz

z ,y X y,y X2 ,y yz )

z,y X y,y X Z)yyz)

Let us assume that before external forces were applied to a body, it

was in such a state that no internal stresses were present (it was in an
undeformed state). Under these conditions the functions F must be
such that

if

= ev =
Fx = Fv =
x

then

We

=0
=

assume that deformations are small and therefore

functions

Fx FVf
,

of the first order only.

into Taylor's series

We

VX

Cns

Ci2y

Tyz

CeiCz

Cfi26y

we

in

expanding the
terms

restrict ourselves to

obtain

Cu z

C teg

C\Cfxy

Cl57x*

C\ dyz

(I-l-D

C^47*2/

4" ^657x2

C^67l/z

Equations (1-1-1) present the analytical expression of the generalized

Hooke's law, which can be formulated as follows: each of the six components of the stress is a linear function of the six components of the strain.

DYNAMICS OF BASES AND FOUNDATIONS

The

coefficients

are constants which depend on the

There are 36 constants in the general

CV,-

elastic properties

case.
However,
proved in the theory of elasticity that only 21 are independent; the
remaining 15 are fixed by the interrelationships Cy Cji.
The smallest number of independent constants is necessary to characterize an isotropic homogeneous body, i.e., one whose elastic properties

of the solid body.

it is

are identical in

all

directions

and

Two

at all points.

elastic

constants

describe such a body.

The

generalized Hooke's law for homogeneous isotropic bodies

may

be

written as follows:
cr x
<T y

<r 2

=
=
=

\e
\e

\e

Ty X
Tzx
Tzy

-\-

2/xe.c

+ 2jue
+ 2^

(I _ 1 _2)

MYzy

= MY*z
= Wyz

where X and n are elastic constants, called Lame's constants.

More than two elastic constants characterize anisotropic bodies.
For example, the elastic properties of a pyrite crystal are determined by
three constants, those of fluorite and rock salt by six, those of barite and
topaz by nine.
Equations (1-1-2) show directly that the elastic constant \i is the
coefficient of proportionality between the stress and the shearing strain;
therefore fi is called the modulus of elasticity in shear.
This modulus is
usually measured by observing the torsion of a sample or by the study of
pure shear.

which the
would be expressed
proportionality between the

It is difficult to create simple experimental conditions for

and the

relationships between the stresses

only by

X.

Therefore, the coefficient of

strains

compressive stress and the tensile or compressive strain is

commonly used as the second elastic constant. Its magnitude is determined directly from the testing of prismatic or cylindrical bars subjected
to elongation or compression.
Assuming that elongation induced by a
uniform stress a occurs in the direction of the z axis, we obtain for the
tensile or

stress

components
<?X

Since

we have

= t zv =
=
y yz =
y xz
<Ty

Tyx

y xy

=
=

<J 2

(T

r zx

Substituting the expressions for the normal stresses into the

Eqs. (1-1-2),

we

first

three of

obtain
\e
\e

\e

+
+
+

2fjie x

2fM y
2fj.e z

=
=
=

(1-1-3)

OF

ELASTIC PROPERTIES

From

these equations

we

SOIL

find directly

m(3X

(1-1-4)

2m)
A

2ju(3X

The quantity

E=

^ ++
X

(1-1-5)

2/i)

?m)

which defines the relationship between a tensile (or compressive) stress

and the elongation (or compression) caused by this stress, is called the
normal modulus of elasticity or Young's modulus.
Since the expressions for e x and e z have different signs, it follows that
axial elongation of a bar will be accompanied by lateral contraction.
From (1-1-4) and (1-1-5) we obtain
*

= - =

-=
ez

ez

o,

2(X

(1-1-7)

n)

Hence it follows that the ratio between the relative lateral contraction
and the relative axial elongation does not depend on the shape of the
cross section and is a constant for a given material.
The quantity v is called Poisson's ratio and lies within the limits
Young's modulus and Poisson's ratio are the two principal
< v <Y<l
They are usually
quantities defining the elastic properties of materials.
.

applied in engineering calculations.

Solving Eqs. (1-1-6) and (1-1-7) for X and

in terms of

Young's modulus
X

=
(1

"
b.

"

Applicability of Hooke's

n,

we obtain

and Poisson's

,0(1

2(nb
Law

ratio

(M "8)

2v)

=G

to Soil.

(I " 1 - 9)

Since Eqs. (1-1-2) are valid

only for a homogeneous isotropic body in which no

present,

it is

their expressions

v.

initial stresses are

evident that Young's modulus and Poisson's ratio

may

be

applied as elastic constants only to bodies satisfying these conditions.

Therefore

it is

soil satisfies

Some

necessary to clarify whether

it is

possible to consider that

the conditions of homogeneity and isotropy.

such as sands, clays with sand, and clays, consist of particles

surrounded by air or by a film of capillary water
containing solutions of various salts and gases.
The rigidity of separate
particles is much higher than that of the soil as a whole.
Therefore in
the investigation of elastic (and total) deformations of soils, the particles
composing the soil may be considered to be absolutely rigid.
soils,

of different materials

DYNAMICS OF BASES AND FOUNDATIONS

The dimensions of the particles of the soil skeleton vary and range
(gravelly soils) and several microns (i.e., thousandths of a
between 5
millimeter).
Suppose we have a small cubic element of soil, several
millimeters on a side.
It is clear that such a cube, if of clay, can be
considered, with a large degree of precision, to be homogeneous.
A cube
of similar dimensions consisting of gravelly soil must be considered to be
nonhomogeneous. However, if the sides of a cube of gravelly material
measure several tens of centimeters, then it is evident that it may be
assumed to consist of homogeneous material. Hence it follows that soil
may be considered homogeneous only in such volume elements as have
dimensions which are large in comparison with the dimensions of the soil
particles.
Therefore, by a stress or deformation at a selected point in
the soil is meant an average stress over an area or a volumetric deformation whose dimensions (i.e., of both the area and the volume) are large
in comparison with the dimensions of the soil particles.
With these
limitations, any discrete system, including soil, which consists of statistically uniform particles, may be considered a homogeneous body.
The
elastic properties of a volume element whose dimensions are large in
comparison with the dimensions of soil particles will be constant through-

mm

out the element.

more or less at random. Therefore the

same in all directions, and soil may be
body. The only exception is presented by
in which separate layers are characterized by

Soil particles are distributed

elastic properties of soil are the

considered an isotropic
layered systems of
different properties.

soils

In such cases a

soil

may

be considered to be

iso-

tropic only within the boundaries of a layer.

In addition to the forces of cohesion which are present to a lesser or

all solid bodies, soil is characterized by the presence of
the forces of internal friction and capillary forces which induce the

greater degree in

appearance of

The

(1-1-1), is

the

initial internal stresses.

generalized Hooke's law, formulated in the general case

initial

In reality,

by Eqs.

based on the assumption that the stresses and deformations in

state (i.e., before an external load is applied) are equal to zero.
initial stresses are

action can be neglected

if

present in

all

solid bodies.

However,

their

they are small in comparison with the induced

Then Eqs. (1-1-2) for a homogeneous isotropic body are valid.

Sometimes the initial stresses in a body may attain magnitudes comparable with working stresses or even surpassing them.
An example is
the earth, the interior of which is characterized by considerable stresses
caused by the mutual attraction of large masses. It is known that most
severe earthquakes are caused by internal stresses in the crust of the
earth.
In such cases the use of Hooke's law in the form (1-1-2) without
stresses.

considering initial stresses

may

lead to large errors.

ELASTIC PROPERTIES

OF

SOIL

If considerable initial stresses are present, then initial deformations

cannot be considered to be small, and the principle of superposition ceases
Therefore it is not possible to
to be valid in relation to deformations.
apply conventional methods, which take the state before deformation as

the unstressed state,

when computing

deformations.

induced by the action of

friction, by capillary
and by processes leading to the swelling and consolidation of soils.
In sandy soils, initial stresses result mostly from friction; in cohesive
soils the initial state of stress may be influenced by a number of physicochemical processes, the natures of which have not as yet been sufficiently

Initial stresses in soil are

forces,

investigated.

The

following

initial stresses in
filling

method can be employed

sandy

soil:

a load

is

a vessel with perfectly elastic walls.

a special device

is

The readings taken

to detect the presence of

soil sample
unloaded and

applied axially to a

The sample

is

used to measure radial deformations of the vessel.

from those taken before the

after the unloading differ

This shows that after the load which acted on the sand was
loading.
removed, stresses developed in the soil and induced elastic radial deformations of the vessel.
These initial stresses are explained by the wedging
action of grains and can exist only if forces of friction are acting between
sand particles.
The influence of initial stresses on elastic deformations of a soil may
be found only by comparing with experimental data some deductions
which follow from Eqs. (1-1-2).
Let us indicate some contradictions between theoretical conclusions
based on Eqs. (1-1-2) and experimental data.
Equations (1-1-2) were derived on the basis of the assumption that in
the absence of external forces the stresses equal zero, and thus deformations also equal zero.
Figure 1-2 presents typical shear-test results on
sand (Fig. I -2a) and on an artificially prepared sample of clay (Fig. 1-26).
Figure 1-3 shows a device which was used to investigate elastic shearing in sand.
The device consists of two metal walls 1 which rotate
freely around their supports.
The sharpened horizontal edges of the walls
rest on the upper and lower plates 2 and 3.
The load 4 was placed on the
plate 8.
This load pressed on the walls and thus increased the stability
of their sharp edges in grooves of the upper and lower plates.
The test sand 5 was placed between the two rigid walls when they were
in a vertical position.
Tamping was used to make the sand as dense as
possible.
The vertical edges of the walls were bent around a wooden
block to prevent sand from trickling out.
The application of the load 6 induced a horizontal displacement of the
upper plate and resulted in a distortion of the sample. These shear-test
conditions are very close to those of pure shear.

DYNAMICS OF BASES AND FOUNDATIONS

Shear tests on clay samples were performed by applying a horizontal

force to a metallic plate glued to the upper surface of the sample.

The

lower surface of the sample was glued to the base of the device. These
shear-test conditions are also close to those of pure shear when the
9.0
'

8.0
'

3.5

3.0

7 6.0

w
.^

jKo

2.5

>

Hj

I
r

5.0

p/

2.0

1.5

7.0

4.0

Oo

3.0

KXp>

&1.0

2.0

0.5

1.0

1.0

3.0

2.0

7xy X10"

4.0

>

10

5.0

7xy x10"
u>)

Ca)

Fig. 1-2. Shearing stress-strain relationships: (a) sand; (6) artificially compacted clay.

relationship between shear stresses

one of the last three of Eqs.

The

and shear

strains

determined by

is

(1-1-2).

shear-test results as plotted in Fig.

1-2 indicate that,

within

between shear
However, the
strain.
stress and
curves show that shearing deforma-

limits of test errors, there exists a linear relationship

only

develop

tions

in

cases

in

which the stress attains a certain

Apparently this value of
value.
the stress corresponds to the fricforce per unit area of the

tional

sand sample.
Shear-test

results

may

agreement with Hooke's law

be

in

if

the

influence of initial friction stresses

acting

Fig.

1-3.

shear,

Device for testing sand

in

between

soil

particles

is

taken into account in the last three

Then the relationof Eqs. (1-1-2).
ship

between

the

components

of

ELASTIC PROPERTIES

OF

shear stress and the corresponding strains

Tyx

Tyx

r zx
TZy

r zx
TZy

is

SOIL

determined by the equations

^fJ'*Yxy

2 M 7v.

(1-1-10)

ZfXy X z

and t zv are the components of the initial shearing stresses.

from Eqs. (1-1-2) that the shearing stresses do not
depend on the normal stresses. Results of shear tests on a clay sample
under different normal stresses show that the modulus of rigidity increases
with an increase of pressure on the sample. The results of field investigations of soils (see Art. 1-4) also show that the coefficient of elastic uniform
shear of the soil increases proportionally to the normal pressure on the
Thus the modulus of rigidity is not a constant, but depends upon
soil.
the magnitudes of the acting normal stresses.
In accordance with the above-mentioned experiments, let us assume
that the dependence of the modulus of rigidity of soil /x on the normal
components of stress is determined by the expression
where

t vx

t zx ,

It follows directly

where

/z

is

MO

P(<*x

the modulus of rigidity of the

external stresses are acting on

(Ty

(1-1-11)

<Tz)

soil in its initial state

(when no

it).

Substituting Eq. (1-1-11) into the right-hand parts of Eqs. (1-1-2),

we

obtain the following stress-strain relationships:

(T x

\e

Txy

2[/X

2[jU

P(<Tx

${<Fx

(Ty

(Ty

a z )]e

(1-1-12)
<T z

)]y X y

In these expressions, the stress components represent the differences

between the stresses induced by the acting loads and the initial stresses.
Equations (1-1-12) are not linear, although Eqs. (1-1-2) are.
Let us use Eqs. (1-1-12) to consider the problem of the compression of a
cube along the z axis. Let a z = a, and
(T X

(Ty

Substituting these expressions for normal stresses into Eqs. (1-1-12),

obtain the following relations:

+ 20*0 + Pa)u =
+ 2(/x + P<r)*v =
= -a
Xe + 2( M +

we

Xe
\e

|8r),

(1-1-13)

DYNAMICS OF BASES AND FOUNDATIONS

10

Adding these equations,

[3\

Then

2( M0

3X

fa)]e

2( M0

Substituting this expression for

we

(1-1-13),

= -a

jftr)

into the left-hand parts of Eqs.

find
Xer

e3

or

is

(1-1-15)

[3X

2( M o
(X

;0er)

of elasticity will be

E = - =
where

|fcr)]G*o
/?cr

=
[3X

The modulus

(1-1-14)

+ 2(mo +
+
X + mo +
+ 2( M0 + M]( M + #r)

2[3X

j8<r)](/io

Mo

jStr)

aEr

(1-1-16)

/8er)

Young's modulus determined by Eq. (1-1-6), and a is a

coefficient which takes into account the influence of
normal stresses on the shearing strength:

[3X

+ 2( M + ftr)](/xo + frr)(\ +
M o(3X + 2 M o)(X + mo + fa)
o

/io)

Tll7
V'^O
/

Thus it is seen that the modulus of elasticity

depends on the normal pressure a. For small
values of a-, the value of a is close to unity, and
the value of E is close to the value of E
With an
increase in pressure, the modulus of elasticity also
.

increases.

Fig. 1-4. Relationship

between

compressive

stresses

and

<r z

vertical

elastic

unit deforma-

tions

of clay

ez

and

cubes.

silt

These conclusions are confirmed by compression

tests on cubes of cohesive soils.
The results of all
tests on clays and silty clays with some sand reveal
an increase in the rigidity of the soil with an increase
of normal pressure.
Figure 1-4 presents a typical
curve of interrelationship between compressive
elastic deformations and normal pressure acting on
the sample.

Analysis of this curve leads to the

conclusion that elastic properties of samples change

considerably, even within the small range of pressures applied in testing.

Dividing (1-1-14) by (1-1-15), we obtain the following expression for

Poisson's ratio:
X

2(x

mo

(I "

M8)

This expression shows that the magnitude of Poisson's ratio is not

If the normal stresses (including their initial magnitudes) are

constant.

ELASTIC PROPERTIES

small, then the value of

elastic bodies for

v is

SOIL

11

by Eq. (1-1-7) for

With an increase in the

close to the value given

which Hooke's law

normal pressure, Poisson's

OF

is

valid.

ratio decreases in soils for

which Eqs. (1-1-12)

are valid.

In bodies for which the conditions expressed by Eqs. (1-1-12) are

modulus of elasticity E in tension will decrease with an increase

valid, the

same time Poisson's ratio v will increase.

Thus the hypothesis concerning the dependence of the modulus of shear

in the tensile stresses; at the

rigidity

on the normal

stresses leads to the conclusion that the elastic

and for tension.

depend on the time period during

properties of soil are different for compression

The

elastic

deformations of a

which the load

is

in cohesive soils.

such

soils

soil

The influence of time is particularly marked

known that settlements of structures erected on

applied.
It is

continue for a long period of time, sometimes reaching decades.

Similarly, elastic deformations do not disappear immediately after unloading,

but do so over an extended period.

aftereffect is observed in soils to a

much

This phenomenon of elastic

greater degree than in other

materials such as metals and concrete.

Owing

to the influence of the elastic aftereffect, elastic deformations

depend on the rate of load application. This is the reason why values of
elastic constants found in some soils under conditions of slow changes in
loads may differ from corresponding values obtained as a result of
investigations of elastic dynamic processes (for example, in studies of

wave propagation or natural frequencies of vibration of soil samples).

Young's Modulus and Poisson's Ratio for Soils. It was shown above
c.
that even when only elastic deformations of soil are considered, stressstrain relations in soil are much more complicated than those stipulated by
Hooke's law. However, the assumption of a relationship more complicated than Hooke's law will lead to the necessity of employing a nonlinear theory of elasticity operating with nonlinear differential equations.

The

solution of these equations, even in the simplest problems, leads to

considerable difficulties, and the application of relations analogous to

Eqs. (1-1-12) becomes practically impossible. Thus it is necessary to

restrict the analysis by the assumption that the soil strictly follows

Hooke's law.

Then

linear differential equations

found for many problems.

numerical values of elastic

may

be used

when

and solutions may be

It should, however, be kept in mind that the
soil constants should be selected with due

considering the equilibrium or

movement

of soils,

consideration of the influence of the simplifying assumptions.

In computations of settlements of structures under the action of

static nonrepetitional loads,

only residual settlements are usually determined, although elastic settlements under the structures are also observed.
However, elastic settlements are small in comparison with residual and are

DYNAMICS OF BASES AND FOUNDATIONS

12

not taken into account in computations. Therefore when computing

settlements or deformations resulting from the action of nonrepetitional
loads, one does not employ elastic constants, but rather constants

which determine the relationship between total deformations of soil

Thus instead of
(residual and elastic) and stresses acting therein.
Young's modulus, the modulus of total deformation or the compressibility
coefficient is used, sometimes determined from the first branch of the

The

compression curve.
laboratories usually

tioned

is

testing of strength properties of soils in field

limited to the determination of the above-men-

Up

soil properties.

to the present time, the elastic constants of

have seldom been investigated

in soil laboratories, and so no reliable

data are available concerning the interrelationship of these properties
with the physicomechanical properties of soils (grain size, porosity,
moisture content, etc.).
soils

N. N. Ivanov and T. P. Ponomarev, 21 'f in their investigations of

compression of clay cubes, did not find any lateral expansion. Hence
they came to the conclusion that Poisson's ratio equals zero for these
K. Terzaghi 45 investigated the coefficient of lateral earth pressure
soils.
in different soils by comparing the magnitudes of forces which had to be
applied to steel bands, inserted in vertical and horizontal positions into a
consolidometer, in order to remove them. As a result of these investigations he came to the conclusion that the coefficient of lateral earth pressure, i.e., the ratio a x/a z has a constant value for each soil; for sands it
,

equals 0.42, for clays 0.70 to 0.75. The coefficient of earth pressure
related to Poisson's ratio by the formula

K
1

is

from which

v
1

+K

From the above-quoted values of the lateral-earth-pressure

we obtain the values of Poisson's ratio: for sands, 0.3; for clays,

coefficient

0.41-0.43.

Pokrovsky and associates 34 investigated the coefficient of lateral

means of a specially designed device. They found
that the coefficient depends on the normal stress a z with an increase in
G.

I.

earth pressure by

the value of

<t z ,

terizing a certain

of

soil.

on the value

acting in

increases, asymptotically approaching a value charac-

of

N. M. Gersevanov 17 showed that

<j z

is

this

explained by the influence of

dependence

initial stresses

soils.

For samples of an artificially prepared clay which they studied, G. I.

Pokrovsky and his associates obtained an asymptotic value of the lateralf

Superscript

numbers

refer to the References at the

end

of the book.

ELASTIC PROPERTIES

OF

SOIL

13

earth-pressure coefficient equaling 0.62 to 0.65.

Correspondingly, the
value of Poisson's ratio was 0.38 to 0.40.
Ramspeck 36 determined values of Poisson's ratio on the basis of
measurements of the rate of propagation of longitudinal and shear waves.

For moist clay, he obtained a value of Poisson's ratio of 0.5, for loess,
For sandy soils, in three cases Poisson's ratio was found to equal
Finally, N. A. Tsytovich 46 recommends
0.42 to 0.47; in one case, 0.31.
for sandy soils ^ = 0.15 to 0.25; for clay with some sand and silt, 0.30 to
0.44.

0.35; for clays, 0.35 to 0.40.

This author and R. Z. Katsenelenbogen investigated in the laboratory

the effects of the moisture content of clay and of sand admixed thereto on
Poisson's ratio, determined from data obtained
from a sample tested both in compression
and in a consolidometer. It was found that
Poisson's ratio does not depend on the moisture
content, but an increase in the sand admixture

The

results in a decrease of Poisson's ratio.

pure clay was found to be

0.50; for clay with the admixture of 30 per cent
sand it was 0.42.
In spite of variations in the values of v obtained by authors who employed different
average value of

5.0

TV

4.0

3.0

2.0

for

methods and investigated various

1.0

/
2.0

1.0

3.0

4.0

x10" 3

can be

Fig. 1-5. Relationship be-

held as an established fact that Poisson's ratio

tween compressive stresses

<r 2
and elastic unit deformations e z obtained from

soils, it

smaller for sands than for clays.

mental data show that the value

is

All experiof Poisson's

Most

ratio for clays lies close to 0.5.

of the re-

hysteresis loop of loading

and unloading.

show that for sands it is about 0.30 to 0.35.

The author and R. Z. Katsenelenbogen performed laboratory investigations of Young's modulus for artificially prepared soils.
Since sand
cannot be tested by means of cube compression, the sand samples were
placed in a consolidometer. By means of repeated loading and unload-

sults

ing,

elastic

deformations corresponding to certain values of vertical

On the basis of these test results, curves showing

pressure were found.

the interrelationship between elastic deformation and vertical pressure on

soil

were plotted for each

teristic

diagrams.

The

modulus and Poisson's

soil.

ratio

v, it

is

related to Young's

by the following formula:

(1

Hence, knowing a and

Figure 1-5 presents one of these charac-

slope coefficient of this curve

is

v)(l

(I-l-18a)

2v)

easy to find Young's modulus.

In accord-

DYNAMICS OF BASES AND FOUNDATIONS

14

ance with the experimental data quoted above, we assumed v for all
sands investigated to be 0.35.
Figure 1-6 presents values of Young's modulus for the Lyuberetsky
sand, obtained as a result of investigations of samples with several
different moisture contents by means of the method just described.
It is seen from this figure that experimental values of E change only
slightly with changes in moisture content, within the range of testing
errors; hence it is possible to consider that Young's modulus for sand
1000
s

800<J>

o 6

600
400
200

10

12

w, moisture content (by weight),

Fig. 1-6. Young's

modulus

14

for sands with different

does not depend on moisture content.

16

18

percent

water contents.

Analogous results were obtained

for other sands.

Experiments also showed that the grain size of sand has comparatively
minor influence on Young's modulus. Table 1-1 presents values of
Young's modulus obtained for sands with different grain sizes.
Table

1-1.

Values of Young's Modulus for Different Grain Sizes of Sand

Grain

size of sand,

mm

1.25-1.50
1.00-1.25
0.60-0.80
0.35-0.60
0.30-0.35
20-0 30
.

E, kg/cm 2

450
520
620
480
480
620

Let us mention finally that, contrary to the opinion expressed in

publications, 46 the experiments established that Young's

modulus

for

sand does not depend much on the porosity of the sand.

Thus the results of experiments lead to the conclusion that Young's
modulus for a pure sand containing no silt or clay admixtures is a stable
characteristic which does not change much with changes in the moisture
content, grain

size,

or porosity of the sand.

However, the Young modulus

of a cohesive soil

depends to a large

ELASTIC PROPERTIES

OF

degree on the properties mentioned above.

SOIL

15

Figure 1-7 presents values of

Young's modulus obtained as a result of tests on artificially prepared clay

samples with different moisture contents. The diagram shows that for a
clay with moisture content close to 30 per cent, Young's modulus is very
This clay had the following characteristics: plastic limit = 16.16
small.

per cent; liquid limit

29.13 per cent; void ratio

0.70; molecular

500
400
300

T
T

Ni ^^
S

200

100

10

Fig. 1-7. Young's

modulus

15
#, percent

20

30

25

for a clay with varying water contents.

water absorption capacity = 13.32. The values of Young's modulus

corresponding to certain values of the moisture content were obtained
from compression tests of clay cubes.
Figure 1-7 also shows the Young' s-modulus-moisture-content relationship expressed by the empirical formula

E = E 0-5)
where

is

the value of Young's modulus for the clay sample with

zero moisture content

and

is

the
0.6

moisture content of the clay for

which

(1-1-19)

Young's modulus

is

i>
J

very

small (theoretically equal to zero).

If E = 430 kg/cm and w = 29
per cent, the graph plotted in Fig.
1-7 fully agrees with values of E
2

0.4

y,

)
<j

i)

^0.2
Uj

o
10

20

30

Sand content

40
in clay,

50

60

70

per cent

established experimentally.

The experiments showed

that,

Fig. 1-8. Young's modulus for a compacted clay with varying sand contents.

within a certain range, the addition

of Lyuberetsky white quartz sand to the clay does not induce large
changes in the values of E. This is demonstrated by Fig. 1-8, giving
values of E obtained for clay samples containing differing amounts of sand.

The limit value w

to the clay.

of the moisture content depends on the sand admixed

Experiments established that in soil samples containing less

DYNAMICS OF BASES AND FOUNDATIONS

16

than 50 per cent sand, iv decreases with an increase in the sand content.
For example, if a pure clay is characterized by w = 29 per cent, then
clay containing 20 per cent sand is characterized by w = 24 per cent,
clay containing 30 to 40 per cent sand has w = 20 per cent, and finally,
in clay containing 50 to 60 per cent sand the ultimate moisture content
decreases to 14 to 15 per cent.

The Young modulus for clays containing more than 60 per cent sand is
much by moisture content. Thus, for clays with sand

not influenced

content larger than 60 per cent, the

Young's-modulus-moisture-content relationship changes sharply as compared

2.0

to

oS
1.6

that

for

Figure 1-9 demonstrates the relation-

1.4

ship between the

1.2

clay

\
o

1.0

u;0-

with smaller sand

clays

admixtures.

and

its

Young modulus

porosity; the graph

is

of a

plotted

on the basis of test results obtained by

the author and R. Z. Katzenelenbogen.

Values of the void ratio are plotted along

8
!

\l
\o

0.6

0.4

the horizontal axis, values of

vertical

V
o%oo

V
V

0.2

0.4

0.2

0.6

0.8

While

axis.

the

E along the
void

ratio

changed from 0.32 to 0.93, the value of

Young's modulus decreased from 1,700
to 120

kg/cm 2

On the

i.e.,

fourteenfold.

basis of these data,

it is

possible

Young's modulus for

e
clayey soils depends to a large degree on
Fig. 1-9. Young's modulus for a
While
physicomechanical
properties.
clay with varying void ratios e.
the values of E for sands change comparatively little, those for clays and clays with silt and sand may change
greatly depending on the moisture content, the void ratio, and, to a lesser
degree, the grain size.

Table

to conclude that

Since the data

now available

are not sufficient for

Values of Young's Modulus for Different Types of Soils

1-2.

E, kg/cm 2

Soil description

Plastic

silty

organic

clay

with sand and

silt

Brown saturated

with sand
Dense silty clay with some sand
Medium moist sand
Gray sand with gravel
Fine saturated sand
Medium sand
Loess
silty clay

Loessial soil

310
440
2,950
540
540
850
830
1,000-1,300
1,200

ELASTIC PROPERTIES

OF

SOIL

17

generalizations concerning the effects of these characteristics on Young's

modulus,

not possible to recommend any values of E for cohesive soils

It is necessary to determine the design
in each case by direct testing of undisturbed soil taken from

it is

or use in design computations.

values of

the construction

site.

In conclusion, Table 1-2

puted from the

soils;

is

presented, giving Young's

coefficients of elastic

the values of

and from laboratory

modulus com-

uniform compression for different

were found as a result of

field testing (see Art. 1-2)

investigations.

Coefficient of Elastic Uniform Compression of Soil c u

1-2.

a.

The

Coefficient of

capacity of a

soil,

To determine

Subgrade Reaction.

the bearing

load tests are used during which a concentrated load

transferred to the soil through a rigid bearing plate.

Except

detailed soil investigations, plates not larger than 5,000

cm 2

is

for special

are used.

3.0
2.5

<2.0
a,-1.5

1.0

0.5

16
20
Settlement,

12

24

28

32

36

mm

Fig. 1-10. Results of load test of a 1.0-m 2 plate on sand.

By

by the weight of the bearing plate

and by measuring the settlement correspond-

increasing in steps the load formed

and the whole

installation,

ing to each load increment, one

may

dependence

on the magnitude

the

of plate settlement

derive data for a curve showing the

of external pressure

on

soil.

The

make

possible to draw some general conbetween settlement and load.

Figure 1-10 gives the relationship between vertical normal pressures
and settlements obtained from load tests on very moist medium sand
2
(a rigid bearing plate of 1
area was used). Figure 1-11 shows the
curve obtained from tests on loessial soils; a plate of 8 m 2 area was used.
results of load tests

it

clusions concerning the relationship

Two
The

distinct sections are observed in each of the graphs presented.

first

medium

For
observed for pressures smaller than

section reveals a linear settlement-pressure relationship.

sand, this relationship

is

DYNAMICS OF BASES AND FOUNDATIONS

18

1.5

kg/cm 2

pressures,

The second section, which corresponds to higher

characterized by a nonlinear settlement-pressure relationship

(Fig. 1-10).

is

expressed by a convex curve.

In this section, settlement increases at a

greater rate than pressure.

0.5
4

12

16
24
20
Settlement,

32

28

40

36

mm

Fig. 1-11. Results of load test of a 8.0-m 2 plate on loess.

For

loessial soil (Fig. 1-11) a linear relationship

plate settlement

is

between pressure and

observed up to a pressure of 1.25 kg/cm 2

then the

nonlinear section follows.

A similar relationship between settlement of the bearing plate and pressoil is observed in clayey soils.
This is confirmed by Fig. 1-12,
showing such a relationship in a clay with silt and sand (a bearing plate of
2
area was used).
The graph shows that the linear plate-settlement8

sure on the

1.00

'

10

12

14

<0.75
0.50

0.25
Q.

16

mm

Settlement,

Fig. 1-12. Results of load test of an 8.0-m 2 plate on clay.

load relationship exists up to 0.75

kg/cm 2 then settlement

;

increases at a

greater rate than pressure.

This characteristic settlement-pressure relationship may be generalized

A proportionality limit may be established, below
which the pressure-settlement relationship is linear; i.e., below a certain
for various soils.

point,
Pz

C p St

(1-2-1)

OF

ELASTIC PROPERTIES

where p z
cp

=
=

normal pressure on

SOIL

19

soil

coefficient of proportionality, also called

modulus

of

subgrade

reaction

S =
t

settlement of bearing plate resulting from external

total

pressure

The

proportionality limit

bearing value of

soil

under

is

a criterion in the selection of the permissible

static load, often

somewhat lower than the proportionality

taken as being equal to or

limit.

question arises as to whether the proportionality limit

physical property of a

or whether

soil

it

depends on such

is

a constant

test conditions

The experimental data available are

as the bearing-plate area or shape.
not sufficient to allow a fairly positive answer to this question. The
proportionality limit apparently does not depend on the area of the
bearing plate used in the investigations and apparently depends only on

This has been confirmed, for example, by D. E. Polshin's

performed on loessial soils. He studied settlements of six square
2
The experiplates having areas of 0.25, 0.50, 1.0, 2.0, 4.0, and 8.0
ments did not reveal any definite relationship between the bearing-plate
This leads to an assumption that
area and the proportionality limit.
the proportionality limit is a characteristic constant property for a given
soil properties.

tests

soil.

If

the validity of the law of linear relationship between bearing-plate

settlement and pressure thereon

is

confirmed by experiment, then, in the

range of pressures smaller than the proportionality limit, there exists also
a linear relationship between deformation at any point in the soil and
stress induced at that point by external forces acting on a limited area
Therefore, for stresses not exceeding a certain limit,

of the soil surface.

the

soil is

considered to be a linearly deformable body.

linearly deformable bodies

is

The theory

of

applied in studies of deformations and

stress conditions in soil.

b.

Coefficient of Elastic

Uniform Compression

the graph of the results of tests performed

who

cu

Figure 1-13 presents

by the author and

associates,

investigated the settlement of a 1.4-m 2 bearing plate on a loessial

under conditions of repeated loading and unloading. The maximum

was increased in each loop. In the first
loop, maximum stress reached only 0.34 kg/cm 2 i.e., it was much smaller
than the proportionality limit, which in loess is approximately equal to
1.25 to 1.5 kg/cm 2
In spite of the relatively small value of maximum
stress in this loop, residual settlement was about 90 per cent of total
settlement.
If one considers that the elastic limit corresponds to the
magnitude of stress at which residual settlement forms only a negligible
part (around 2 to 3 per cent) of total settlement, then, in the case concerned, the elastic limit of the soil evidently is much smaller than the
soil

stress transferred to the soil

DYNAMICS OF BASES AND FOUNDATIONS

20

proportionality limit.

This conclusion apparently

The experimental

to other soils.

may

also be applied

results are not sufficient to permit

conclusions concerning numerical values of the elastic limit for various

Probably for most

soils.

soils (not for preloaded soils and not for rocks)

only several tenths of a kilogram per square centimeter;
lower than the usually permissible working pressures on

this limit equals

i.e., it is

soil.

soils

much

Thus, for small pressures only, the assumption is possible that

of the clay and sand tj^pes satisfy the condition of reversibility of

deformations.

graph of the footing settlement-pressure relationship under condiand unloading permits easy separation of the

tions of repeated loading

16.8

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0 4.5 5.0

Settlement,

5.5

6.0

6.5 -7.0

7.5

8.0

8.5

9.0

mm

Fig. 1-13. Results of load test of a 1.4-m 2 plate on loess.

elastic part of the

deformation from the total settlement.

indicates the result of an analysis performed on Fig. 1-13.

indicate the magnitudes of elastic settlement for the

Figure 1-14

The

circles

maximum pressure in

The graph shows that within

a range of pressures up to
change proportionally to pressure.
Figure 1-15 presents graphs of elastic settlements obtained by the
author for 1,600- and 3,600-cm 2 bearing plates on medium sand of natural
moisture content. Similar graphs showing interrelationships between
settlement and pressure on the plate were obtained for other soils. On
the basis of these investigations, the statement can be made that within a
certain range there is a proportional relationship between elastic settlements of footings and external uniform pressure on the soil; i.e.,

each loop.
1.0

kg/cm 2

elastic settlements

V*

where

c u is

cu

Se

(1-2-2)

the coefficient of proportionality, called the coefficient of

uniform compression of soil, and S is the elastic settlement of the

bearing plate due to the external pressure.
The coefficient of elastic uniform compression c u of a soil differs from
the coefficient of proportionality c p [in Eq. (1-2-1)], but the difference is

elastic

ELASTIC PROPERTIES

OF

While

not often taken into consideration.

SOIL

21

c p relates

the

total settlement

to the external pressure, c u relates only the elastic part of the total settlement
Since the total settlement of a foundation is always
to the pressure.
larger than its elastic settlement, c u

is

larger than c v in all soils, without

exception.
1.3

1.2

1.12

1.1

0.98

i/

#/

1.0

#7

0.9

0.84

/
/

0.8

E0.70
'0.7

0.56

0.6

0.5

0.42
0.4

0.28

0.3
0.2

0.14

0.1

"//
f

0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00

Elastic settlement,

mm

Elastic settlement,

mm

Fig. 1-14. Evaluation of

Fig. 1-15. Evaluation of hysteresis loops obtained dur-

the

ing load tests on footings of two sizes.

hysteresis

loops

of

Fig. 1-13.

The

Area and Shape of a Foundation Base on the Coefficient

Uniform Compression. The coefficient of elastic uniform
compression and the coefficient of proportionality would be constant
values (characteristic of certain soils and independent of test conditions
and area and shape of the foundation base) if the stresses in the soil
under a uniformly loaded foundation remained constant at any point.
Otherwise, both coefficients would depend on the area of the footing
base and its shape.
If dR is the soil reaction on the element dA of the foundation base
contact area, then, assuming that at any point directly under dA the
c.

Effect of the

of Elastic

settlement

of the soil

is

proportional to the stress,

dR = c(A)SdA

we obtain

DYNAMICS OF BASES AND FOUNDATIONS

22

where c(A) is a variable coefficient of proportionality whose magnitude

depends on the elastic constants of the soil and the coordinates of the
element of area dA. The total reaction of the soil equals the load P 2
imposed on it. Therefore,

P =
2

JA

dR = S

JA

c{A)

dA

(1-2-3)

This integral extends over the entire area of the foundation base.
is assumed to be constant, and the foundation to be absolutely

Settlement
rigid.

But

P =
z

where p z
(1-2-3)

V*A

the magnitude of the normal pressure on the

is

soil,

and Eq.

can be rewritten as follows:

V*

j fA c(A) dA

(1-2-4)

If the normal stress in the soil under the foundation remains constant,
then c(A) will also remain constant and will equal c u
Hence its value
is determined only by the elastic properties of the soil, and
.

c(A)

dA =

cu

(1-2-5)

In this case the coefficient of uniform compression of the

size of the foundation base contact area.

soil

does not

depend on the

when a uniform

pressure acts on the foundation, the normal

under it are distributed irregularly.
Since machinery foundations consist of either a rigid block placed
directly on the soil or a framed elastic system supported by a rigid plate
placed on the soil, the problem of distribution of stresses under the base
contact area of a rigid plate which has undergone uniform settlement of a
certain magnitude is of great interest.
Sadovsky 38 gave a solution to this problem for a circular base contact
area of a rigid plate. For soil directly under the base contact area the
expression for c(A) may be written in the form

In reality,

stresses in the soil

o(A)

where

- T^>
i -

E =

Young's modulus

R =
p =

Poisson's ratio

(1-2-6)

*tVR

of soil

radius of plate
radius vector of point under consideration in
ing plate (p

<

R)

soil

under bear-

ELASTIC PROPERTIES

Equation

(1-2-6)

shows that the

OF

stress

SOIL

23

under the base contact area

The

of a rigid bearing plate is smallest at the center of the plate.

stresses at other points grow proportionally with distance from the center;
the maximum value is reached near the edges of the foundation, where
p

R.

Theoretically

maximum

this

However, no
and the soil at the

equals infinity.

actual material can stand infinitely large stresses,

foundation edges will undergo plastic deformations; the stresses in these

zones will be of finite magnitudes, although they will be much larger
than the stresses at points near the bearing-plate center.

Multiplying both parts of Eq. (1-2-6) by the area element

integrating over the entire foundation base contact area,

dA and

we

obtain,

according to Eq. (1-2-4),

p dp dip

-eM^
-a'Lf'+r'
- A
VR -P*
1

By

integration

we

'

Jo

2 .7)

( i.

Jo

find:

^=

1.13^^=

Comparing the right-hand parts

following expression for c u

d-2-8)

of Eqs. (1-2-2)

and

(1-2-8),

we

find the

cu

1.13

E
1

L.

(1-2-9)

Thus the equation

for the coefficient of elastic uniform compression

from the solution of the theory-of-elasticity problem
concerning the distribution of normal stresses in the soil under the base

of a soil, obtained

contact area of a rigid plate, leads to the conclusion that if the settlement
of the foundation is uniform, the stresses under its base are not distributed uniformly.
solution

is

The

coefficient of elastic

compression

given for a proportionally deformed

soil,

c u (or, if

proportionality c p ) depends not only on the elastic constants

which define the

the

the coefficient of

and

v,

but also on the size

The coefficient c u changes in

elastic or linear properties of a soil,

of the base contact area of the foundation.

inverse proportion to the square root of the base area of the foundation.
If the foundation consists of an absolutely flexible plate uniformly
loaded by a vertical pressure, then stresses in the soil under the foundation will be distributed uniformly, but settlement under the foundation

will

vary.

For an absolutely

flexible

foundation, the coefficient of

DYNAMICS OF BASES AND FOUNDATIONS

24

elastic

uniform compression

the ratio of the uniform pressure to the

is

average value of settlement

Cu

(1-2-10)

&

Schleicher 39 gave a solution for an absolutely flexible footing with

According to his solution, the average settlement

determined by the equation

rectangular base.

value

$av

is

VI +

\/A

7r

+ a In \/j +
Vi +
where a
2a, 26

=
=

-\/a L

H~ or

+ a")

(1

pz

(1-2-11)

2a/2b
(1-2-2)

we have:

S = p
Assuming S = Sav we obtain
,

for c w
Cu

z ct

CsC

(1-2-12)

where

(1-2-13)

a coefficient which depends only on the ratio a between the length

and the width

cs

<*y*

length, width of foundation

From Eq.

c s is

2 (1

of the foundation base.

It is equal to

* \/a

=
In

^j+_^ + + a
a

\/l

-f

V^+l
Vi + 2

%[(1

a , )K

Lt*
a:

(1-2-14)

Table 1-3 gives values of

c s for

various values of

a,

computed from Eq.

(1-2-14).

Table

1-3.

Auxiliary Values for the Computation of Settlements

of Rigid and Flexible Bearing Plates
ex.

1.5
2
3
5

10

cs

c[

1.06
1.07
1.09
1.13
1.22
1.41

1.08

1.10
1.15
1.24
1.41

ELASTIC PROPERTIES

OF

SOIL

25

The same table gives also values of c s computed by M. I. GorbunovPosadov 20 for an absolutely rigid rectangular foundation. The difference
between c s and c 8 is not larger than 3 per cent. Hence, there will be
little difference between the values of the coefficient of elastic uniform
compression as computed for absolutely rigid and absolutely flexible
Table 1-3 indicates that

rectangular plates.

if

the base area of a founda-

tion remains constant, the coefficient of elastic uniform compression

Foundations for machinery are seldom

increases with an increase in a.

constructed with a large ratio between the side lengths; in most cases

Values of \/T,

0.50

Fig. 1-16.

it

Dependence

0.71

1.00

on the loaded area under

of settlements

does not exceed -3.

Hence,

it

2.00

1.41

different pressures.

can be held that, in practice, the coeffimachine foundation does not

cient of elastic uniform compression for a

depend on

The

a.

foregoing conclusions concerning the effect of the area of the

foundation base on the settlement and on the coefficient of elastic uniform

compression will not change if we assume that the soil under the plate
behaves, not as an elastic body, but as a proportionally deformable body.
d.

Experimental

Compression.

The

Investigations

of the

Coefficient

of Elastic

Uniform

conclusions of the theory discussed above are fully

confirmed by experimental investigations performed within the range of

pressures smaller than the proportionality limit; then, residual settle-

ments appear in addition to elastic settlements, and the soil does not
behave like an elastic body. Figure 1-16 presents graphs obtained by
D. E. Polshin which show the settlement of rigid bearing plates on loess
plotted against the areas of the plates.

The square

roots of the bearing-

plate areas are plotted as abscissas, the settlements as ordinates.

were square.

The

Within the range of test errors, the relationship between settlement and the square root of base area may be contest footings

DYNAMICS OF BASES AND FOUNDATIONS

26

sidered to be directly proportional, which fully agrees with theoretical

conclusions.

number

of tests

on various

give us a chance to verify the

soils

validity of the theoretical conclusion that the coefficient of elastic

com-

pression increases in inverse proportion to the square root of bearing-

Table

1-4.

Computed and Measured Values of the Coefficient of

c u of a Medium Sand for
Different Areas of Loading

Elastic Uniform Compression

kg/cm

cu ,

A,

Experimental

Computed

30.0
18.0
11.4
8.9
6.6

30.0
18.8
13.4
7.7
4.9

200
500
1,000
3,000
7,500

plate area.

All tests

-'

cm 2

were performed on

rigid bearing plates

with varying

areas.

Gerner 16 performed laboratory

values of

c u for several plate

tests

on medium sands and obtained

areas (see Table 1-4).

Figure 1-17 gives curves, plotted on the basis of the data of Table 1-4,
The
in c plotted against changes in plate area.

which show changes

30
25
G

<20
\\

10

r2ZL*periment

'Om r ~'

2
Area

Fig. 1-17.

Dependence

'mputt ft/on

4cm 2 x10 3

of the coefficient of elastic

uniform compression of

soil c

on

the loaded area A.

were obtained from experiments and from theoretical conThe latter values were computed as follows: according to
Eq. (1-2-12), the coefficients of elastic compression c u \ and c w2 for two

values of

cu

siderations.

ELASTIC PROPERTIES

foundations with different areas A\ and

CU 2

OF

SOIL

27

are related

by the formula
(1-2-15)

Cu l

As a result of field tests performed by the author and the late Ya. N.
Smolikov on gray soft saturated clays with sand and silt, the following
values of c u were obtained for foundations with areas of 1.5, 1.0, and
3
Inserting the values A x = 0.5 m 2
0.5 m 2 2.1, 2.52, and 3.50 kg/cm
3
and Cui = 3.5 kg/cm we obtain from Eq. (1-2-15) the following values
of c u for the other two foundations investigated: 2.45 kg/cm 3 for
A = 1.0 m 2 and 2.02 kg/cm 3 for A = 1.5 m 2
Table 1-5 gives values of c u for foundations with three different base
areas obtained from tests performed by the author and A. I. Mikhalchuk
on saturated brown clays with silt and sand. When the values presented
in this table were computed, a value of c u \ equal to 4.4 kg/cm 3 and
corresponding to A = 2.0 m 2 was inserted in Eq. (1-2-15). The computed values of c u for areas of 4 and 8 m 2 were larger than those found
This difference was apparently caused by the diversity
experimentally.
.

of soil properties at the site of field investigations.

Table

1-5. Computed and Measured Values of the Coefficient of

Elastic Uniform Compression c u of a Saturated Clay
for Different Areas of Loading

cM ,

A,

Experimental

Computed

2.05
2.5
4.4

2.2
3.12
4.4

8
4
2

The

and Ya. N. Smolikov performed tests on

moisture content, using foundations with square bases
The results of these investigations are presented in

author, P. A. Saichev,

loess at its natural

of different sizes.

Table

kg/cm 3

1-6,

All the investigations confirm that the coefficient of elastic

compression of a

soil

uniform

decreases with an increase in the area of the founda-

The same experiments indicate, however, that values of

computed from Eq. (1-2-15) decrease at a higher rate than those
established experimentally.
Experimental studies of forced and free
tion base.
cu

vibrations of actual foundations with bases attaining 100

that experimentally established values of c u are

obtained from computations in which values of

much
cu

area indicate

larger than those

found by testing a

DYNAMICS OF BASES AND FOUNDATIONS

28

foundation of small area, were adjusted to a larger area. The author

studied free vibrations of the foundation for a horizontal compressor.
The foundation was erected on loessial soil and had a base area of 90 2
From the results of the studies the coefficient of elastic uniform compres-

was found

kg/cm 3

Another foundation had a

and an experimentally determined
base area of approximately 1.5
3
Using Eq. (1-2-15) and this data, we can find the
c u of 10.8 kg/cm
2
value of c u for A 2 = 90
sion of the soil

to be 4.7

cu

This computed value of

10.8

c u is

~=

1.4

kg/cm 3

almost 3.6 times smaller than that estab-

lished experimentally.

Similar results were obtained in other studies of actual machinery

foundations.

They

indicate that, for foundations with large base areas,

Table 1-6. Computed and Measured Values of the Coefficient of Elastic

Uniform Compression c v of Loess for Different Areas of Loading
cu ,

A,

kg/cm 3

Experimental

Computed

14.2
10.8
10.2
8.0

14.2
10.6
9.0
6.4

0.81
1.4
2.0
4.0

c u changes at a much lower rate than is indicated by

Eq. (1-2-15).
This disagreement between theory and experimental data can be
explained by the apparent dependence of the elastic constants of a soil,

the coefficient

particularly the

modulus

discussed in Art. 1-1.

of rigidity,

The

on the normal vertical pressure, as

properties of a uniform soil change with depth.

With an increase in base contact area, a greater depth of soil is affected

by the weight of the foundation, and the influence of deeper soil layers on
foundation settlement increases.
e.
Values of the Coefficient of Elastic Uniform Compression for Soils
Used in Design Computations. The true values of the coefficients of
elastic uniform compression, just as do Young's modulus and Poisson's
ratio, depend on a number of factors whose influence in each separate
case

is

very hard to evaluate.

Therefore,

involving the erection of a sufficiently large

machinery subjected

to

dynamic

in

construction practice

number

of foundations for

loads, special investigations of the elastic

ELASTIC PROPERTIES

OF

SOIL

29

properties of soils are necessary on the construction

these investigations cannot be carried out.

Sometimes

site.

Therefore

often happens

it

that designers of machinery foundations assume tentative values of the

coefficients of elastic

uniform compression selected on the basis of data

secured from tests performed on similar

soils at

other construction

sites.

Table 1-7. Summahy of Average Values of the Coefficient of Elastic

Uniform Compression c u of Different Soils Obtained under Varying
Conditions

Tentative value

Coefficient
of elastic

Description of

of permissible
soil

load on

uniform
compression

soil,

kg/cm 2

cu ,

Gray

Brown saturated

S =

kg/cm 3

2
;

static)

with

plastic silty clay

sand and organic

Contact areas of
foundation bases,
types of tests
(D = dynamic;

silt

silty

1.0

1.4

1.5, 1.0, 0.5 OS;

D)

1.5

2.0

8.0, 4.0, 2.0 (S;

D)

clay

with sand
Dense silty clay with some
sand (above ground-water

Up

level)

Medium

moist sand
Dry sand with gravel .......
Fine saturated sand

Medium sand

10.7
2.0
2.0
3.0-3.5

to 5

2
2
2.5
2.5

8.9(D)
1.5 (D)

0.25(D)
11.6 (D)

3.1

8.75(D)

2.5

3.4

up

4.5

0.81,

Gray

fine dense saturated

sand
Loess with natural moisture

content

to 15 (S;

(S;

Moist

4.7

loessial soil

1.4,

D)
2.0,

4.0

D)

90 (a foundation for
compressor) (D)

Table 1-7 gives a number of values of the coefficient of elastic uniform

compression c u established by special tests carried out under the supervision of the author.
These tests dealt with foundations having several
base areas built on various soils.

The values

of the coefficient of elastic

uniform compression, obtained

by testing foundations with base areas smaller than 10 m 2 were later

recalculated by means of Eq. (1-2-15) for an area of 10 m 2
From the data of Table 1-7, the author compiled Table 1-8, which gives
,

tentative values of

This table

is

c u for

four types of

included in the

soil.

official

Construction of Machinery Foundations.

Instructions for the Design

and

DYNAMICS OF BASES AND FOUNDATIONS

30

Table

1-8.

Recommended Design Values of the Coefficient of Elastic

Uniform Compression c
Permissible
load on

Soil

group

Coefficient of
soil

uniform
compression c u
elastic

under action of

group

Soil

static load only,

category

kg/cm 3

kg/cm 2

Weak

silty clays

with

and

soils (clays

sands; also

silty

and
and

soils of categories II

III with laminae of organic

silt

Up

of peat)

Soils of

II

and

sand, in a plastic state; clayey

medium

to 3

clays with sand, close to the

silty

Strong

soils (clays

3-5

1.5-3.5

plastic limit; sand)

III

Up

to 1.5

strength (clays and

and

silty clays

with

sand, of hard consistency; gravels

and gravelly sands;

loess

and

loessial

IV

1-3.

5-10

3.5-5

soils)

Rocks

greater than 5

greater than 10

Coefficient of Elastic Nonuniform Compression of Soil c 9

Let us consider the bending of a flexible rectangular plate with sides of

moment M, which bends the plate
with respect to the y axis (Fig. 1-18). Using the

length 2a and 26 acted upon by the

equations for settlement at any point produced

by

stresses in

an element of plate area situated

from the point, and apply-

at a certain distance

method

ing the

of superposition of settlements

induced by several loads, D. E. Polshin 35 derived

an equation for the slope angle <p of the deflected
plate at
Fig. 1-18. Induced tilting
of

a loaded plate causes

nonuniform compression of

any point

of its surface

along the straight line x

is

assumed to be absolutely

and at points

Since the plate

0.

flexible,

the slope

angle of the deflected plate at points along

soil.

depends on the y coordinate.

<p depending on the

this straight line

Polshin gives the following expression for the angle

value of y:
<p(y)

= K^(a

y)\n

Via + y y +
Via + y) +
2

+
where

(a

62

62

- b

y) In

tE

VJa
Via

M
I

y)
y)

+
+6

b'
5

3]

(1-3-1)

OF SOIL

ELASTIC PROPERTIES

the external

is

moment

31

acting on the plate in the zx plane and /

moment of inertia of the plate with respect to the y axis.

From Eq. (1-3-1), the author and B. M. Terenin derived an
which expresses the

both the

effect of

elastic properties of soil

is

the

equation

and the

size

foundation area on the coefficient of elasticity of the soil as a result

It was assumed that
of rotation of the foundation base by the angle <p.
foundation
tilted
base
of
the
(an absolutely rigid
angle
of
the
slope
the
<p
average
from
the
total
slope
angle
very
little
differs
plate)
<^ a v of tilt of an
of the

absolutely flexible plate.

Ay) dy

a Jo

Substituting under the integral the expression for the function

<p(y)

from

Eq. (1-3-1) and integrating, we obtain

K
On

2a 2 ln

4g2

&2

j + 6(^/4^ + h

(I

6) ]

_3_ 2 )

we may assume that the normal vertical soil reacany point under the base of an absolutely rigid bearing plate
depends only on the settlement of this point and is proportional to it.
Then we obtain an equation for the value of the soil reaction on the
foundation area element dA tilted through an angle <p by external bending
the other hand,

tion at

moment M.

This equation

is

as follows:

dR =
where

c,p

is

c^Lip

dA

the coefficient of elastic nonuniform compression of

the coefficient of proportionality, analogous to

when

The moment

of the

dA and

total reactive

moment

M
Assuming that

<^ av

<p,

we

c^L 2 ?

dR with

c^ip

is

the

L dA =
2

respect to this axis

is

dA

transmitted by the

the axis of rotation.

elementary reaction

dM =

or

soil, i.e.,

which was introduced

the uniform settlement of foundations was considered.

distance between the element

The

cu

soil to

the foundation

is

c v Ip

obtain:

c,

M:
VA

(1-3-4)

DYNAMICS OF BASES AND FOUNDATIONS

32

where

va

2tt(x

2a 2

[(V4

In

l)/(\/4a 2

1)]

->/4a 2

+ 1-1
(1-3-5)

and

<p,

as before, equals a/b)

It follows

from Eq.

is

by Eq.

defined

(1-3-4) that c v decreases

(1-2-13).

with an increase in the area

of the foundation base.

From

Eqs. (1-2-12) and (1-3-4)

we obtain

Equation
established

(1-3-6)

(1-3-6) indicates that the coefficient of elasticity of the soil c u

by means

of tests

with uniform loading,

is

not equal to the

coefficient c p established by means of tests with nonuniform loading.

Thus, the coefficients of elasticity of a soil, relating the external load to the
settlement of a foundation, depend not only on the elastic properties of
the soil and the size of the foundation, but also on the character of the
load transmitted to the soil.
Table 1-9 gives the computed values of the coefficient k v and the ratios
The data of this table lead to the concluCp/cu for different values of a.
sion that, with an increase in a, the coefficient of elastic nonuniform
compression c p increases faster than c.
,

Table

1-9.

Values of the Coefficient of Elastic Nonuniform Compression

for Different Ratios a of the Length to the
Width of a Foundation

1.0

1.5
2.0

*.

1.984
2.254
2.510

Cu

2.955
3.700
4.981

2.63
3.04
3.53

1.87
2.11
2.31

3
5
10

The author and A.I. Mikhalchuk were the

first

to verify experimentally

This was done during the experimental studies of foundations on saturated brown clays with silt and sand
described above. By means of these investigations, it was for the first
time established experimentally that the coefficient of elastic nonuniform
compression c v differs from the coefficient of elastic uniform compression
Prior to these experiments, the coefficients were considered equal in
cu
all design computations.
the theoretical findings related to c v

ELASTIC PROPERTIES

Values of

and

cv

OF SOIL

33

obtained experimentally for brown clays with silt and sand

Table 1-10. The tests were carried out

for other soils are given in

with square foundations both static and dynamic methods were employed
;

for determining the values of c v

The data of Table 1-10 confirm experimentally the basic theoretical
.

findings related to c v

The

test results agree

with the theoretical con-

an

siderations indicating a decrease in the value of the coefficient c v with

increase in area.

Table

1-10.

The

test results also indicate that the ratio c^/cu

Values of the Coefficient of Elastic Nonuniform Compression

c<p for Different Soils and Sizes of Foundations
Base contact
area of

Description of

soil

foundation,

m
Saturated brown

silty

clays

with

2.0
4.0
8.0

4.40
2.50
2.05

with

0.5

3.5
2.52
2.11

some sand

Saturated gray soft

silty clays

kg/cm 3

some sand

1.0
1.5

Loess at natural moisture content

0.81
1.4
2.0
4.0

Saturated gray fine dense sands

kg/cm 3

Cu

12.0
4.0
3.0

2.73
1.60
1.46

4.0
8.0
15.0

3.55
3.61
3.79

14.2

1.02
1.44
1.80

8.0

25.0
17.6
15.5
12.9

1.76
1.63
1.51
1.61

7.5
5.6
4.0

14.5
9.5
9.2

1.92
1.71
2.30

10.8
10.2

changes within the limits established by the theory for foundations with
square bases; experiments established the average value of this ratio to
be 1.73, which differs by only 8 per cent from that found on the basis of
theoretical considerations.

Thus

it

is

possible,

if

necessary, to limit

experimental investigations of the coefficients of elasticity of soil c u and c^

on the site of foundation construction. After one of these coefficients is

may be computed by means of

Figure 1-19 illustrates the results of

established experimentally, the second

theoretically derived formulas.
tests

on

elastic

nonuniform compression performed on

loessial soil.

It

presents one of the graphs of relationship between the slope angle of the
deflected foundation and the value of the moment, in cases where the

foundation was loaded on both sides of the axis of

tilting.

DYNAMICS OF BASES AND FOUNDATIONS

34

Fig. 1-19. Angle of rotation (tilting)

of a test plate as a function of the applied

<p

moment M.
1-4.

If

Coefficient of Elastic Uniform Shear of Soil

a test foundation

applied at the level of

is

its

subjected to the action of a horizontal force

base contact area,

The displacement

that force.

the force at that time.

cT

at

it

will slide in the direction of

any time depends on the magnitude

However,

of

after the force drops to zero, there

remains a residual displacement of the foundation. The sliding of a

foundation subjected to a horizontal force in principle does not in any
way differ from settlement. As shown by experience, under conditions
corresponding exactly to those established for soil compression (see
Art. 1-2),

it

between the

is

possible to consider that there

sliding of a foundation

is

a linear relationship

and the average shearing

developed along the foundation base contact area,

sliding stress

i.e.,

(1-4-1)

c,s:

where

r av is the average shearing stress in the soil at the plane of contact

with the foundation, and S'e is the elastic part of the total horizontal

sliding of the foundation base

Or,

by analogy with the

under the action of r av

compression

coefficient of elastic

c, is

called the

uniform shear of soil.

In order to clarify the dependence of c T on Young's modulus, Poisson's
ratio, and the size and shape of the foundation base, let us use the existing
solution 31 of the problem of deformation of a semi-infinite solid under the
coefficient of elastic

ELASTIC PROPERTIES

action of a horizontal force

tion of

is

made

applied to

its

35

surface (the point of applica-

the origin of the coordinate system)

Ptl

Si

OF SOIL

VJT+^2

v(x*

YA }

4-2)

In order to obtain the solution for a load uniformly distributed over a

rectangular area with sides 2a and 26,

we make

the following substitutions

in (1-4-2)

= T av dt] <li
X*
= v
$

Pt
X

7]

We

then integrate Eq. (1-4-2) over the entire loaded area.

Thus we

obtain
1

Si

-J-aV(v -

The average value

by 26 equals

x*)

tt-y*)
(v-x*) 2 dv

d-q

VRv - x*y +

(s

d^
(1-4-3)

y *y

of the horizontal sliding along the surface area

2a

&- a -**
Substituting the formula for

we
<?'

S tot from

and integrating,

(1-4-3) into (1-4-4)

obtain:

= 4a
T

~
E

(
a

sinh

[^

-1

sinh -1

(Vrr^
-j-sinh\a

a
i)

+ vt+~^ -

\/l

-.(Vi

a\

a'

(1-4-4)

i)

1)
where a

On

a/6.

the other hand, according to Eq. (1-4-1), the value of the elastic

sliding of the foundation base equals

s:
Cr

DYNAMICS OF BASES AND FOUNDATIONS

36

Assuming S'e = S'av we obtain the following expression

,

uniform shear of

of elastic

Cr

where the

K =
2

for the coefficient

soil:

d-4-5)

VA

coefficient k T equals

((l/) sinh- 1 a

\/l

+
Equation

HlVl +

sinh" 1 (l/)

a]

[v/(l

J^[(l/a 2 )(\/l

a2

1)

v)\{(l/a) sinh" 1 a

==
+ ~

(2/ 2 )(Vl

(1-4-6)
1)]})

(1-4-5) leads to the conclusion that the coefficient of shear c T ,

just as do the coefficients of compression c u

and

cv,

depends on the founda-

tion base area, decreasing proportionally to \/y/A.

Unlike

and k v

cs

[see

Eqs. (1-2-14) and

(1-3-5)], k T

depends not only on

the ratio a between the lengths of the sides, but also on the value of
Poisson's ratio.

Table 1-11 gives values of the

for different values of

Table

coefficient k T

computed

v.

Values of the Coefficient k T [Eq. (1-4-6)] for Varying

Values of the Poisson Ratio v and of the Ratio a
of the Length to the Width of a Foundation

1-11.

0.1
0.2
0.3

0.4
0.5

The

is

0.5

1.0

1.5

2.0

3.0

5.0

10.0

1.040
0.990
0.926
0.844
0.770

1.000
0.938
0.868
0.792
0.704

1.010
0.942
0.864
0.770
0.692

1.020
0.945
0.870
0.784
0.686

1.050
0.975
0.906
0.806
0.700

1.150
1.050
0.950
0.850
732

1.250
1.160
1.040
0.940
0.940

ratio

cu

and

cT

equaling

governed not only by the value of

the value of
it

between

v is

a,

but also by the value of v. Since

sandy soils (see Art. 1-1),

larger in cohesive soils than in

follows that under otherwise equal conditions the relative value of the

coefficient of elastic shear c T

soils.

must be smaller

in clayey soils

than

in

sandy

OF

ELASTIC PROPERTIES

SOIL

37

Experimental investigations of the coefficient of elastic shear c T were

on the same soils for which c u and c v were studied (see Arts. 1-2
and 1-3). Experiments were performed on gray saturated plastic silty
clays with some sand several test foundations of reinforced concrete with
2
were used. Some footings were square,
base areas of 1.5, 1.0, and 0.5
others were rectangular, but all had the same area.
The following method was employed during the tests: hooks were
embedded in the footings 5 to 10 cm above the base level. Wire ropes
were tied to the hooks and were connected by a turnbuckle to a dynamomTwisting the turnbuckle produced a horizontal force which acted
eter.
along the foundation base and induced its sliding. The shearing stress
carried out

was gradually increased and was measured by the dynamometer; the

600
/

500

-I

'400

300

/
E200
100

l-Jff

MXJl

yfl

-}/-/-J-

~/r

///
W\

/J

0.3

y~

rM

W
0.2

0-

9//

l$-

7?
0.1

/ ^~Xl_Z\m/_

/
7_

7~

0.4

0.5

0.6

Sliding,

0.7

0.8

0.9

1.0.

1.1

1.2

mm

Fig. 1-20. Results of a horizontal load test on saturated

silt.

displacement of the foundation was measured by a gauge and was

In each test, loading-unloading
stress.

recorded for certain values of

cycles were obtained.

Figure 1-20 presents curves of the relationship
between sliding and horizontal stress, plotted on the basis of data obtained
in one of the tests.
Experiments established that with an increase in normal pressure on
the soil, c T increases.
It was also found that the value of c T for a given soil
depends not only on the normal pressure, but also on the duration of its
action.
Thus, immediately after application of the vertical load, the
value of the coefficient of elastic shear is smaller than the corresponding

value established a

little later.

Under constant

vertical pressure, the rate

with time, and after several hours

changes in c T become so small that its value is practically constant.
For illustration let us analyze Fig. 1-21, which gives curves of the
relationship between c T and normal pressure, plotted on the basis of data
2
obtained in tests with footings having 1
base area. After c T was
of increase of c T gradually decreases

DYNAMICS OF BASES AND FOUNDATIONS

38

determined under a normal vertical pressure of 0.53 kg/cm 2 the test was
interrupted, and the footing remained under this load for 20 hr.
Then c T
was determined for the second time under the same pressure and was
found to be 30 per cent larger than before the interruption: c T was 1.10
,

kg/cm 3

before the interruption, and 1.45 after.

due to the interruption

Figure 1-21 reveals that,

curve of relationship between

in the test, the

cT

2.5
fr^-

2.0

rtV

1.5

ir\Q

JK

1.0

Interval

load ing

20 hr

0.5

0.2

0.4

0.6

0.8

Pressure/^, kg/cm'

Fig. 1-21. Variation during a horizontal load test on a 1.0-m 2 plate of the coefficient
of elastic uniform shear c T of clay soil.

12
10

_2>

h.
6

degi 0^Si
\\ooP

'

s\

Ob^

or"

is<**
4
2

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.8

1.6

2.0

2.2

2.4

Pressure/^, kg/cm 2

Fig. 1-22. Variation during a horizontal load test of the coefficient of elastic uniform

shear

c T of loessial soil.

at the point p z = 0.53 kg/cm 2

Later p z was increased
again at a regular rate, and c T also grew smoothly.
An interruption of 43 hr was made once more, after the normal pressure

and p z was broken

attained 0.83

kg/cm 2

It

interruption than before.

was found again that

cT

was larger

after the

regular decrease in the normal pressure

caused a smooth decrease in c T

For unloading, the curve relating c T and
normal pressure turned out to be parallel to the corresponding curve for
.

loading.

ELASTIC PROPERTIES

OF

SOIL

39

similar increase in c T with the time of normal pressure action

was

This is illustrated by Fig. 1-22, which presents

the curve of relationship between c T and normal pressure. Test conditions were the same as described for saturated plastic silty clays with
some sand. The normal pressure on soil was increased in approximately equal time intervals. The coefficient c T was determined for
After the maximum
each new value of the normal pressure p z
pressure of 2.0 kg/cm 2 had been attained, it was found that under
this pressure the horizontal force induced large settlements of the

found in

tests

on

loess.

foundation.

Ya. N. Smolikov and the author investigated the relationship between

the coefficient of elastic uniform shear and normal pressure for gray

saturated fine sands.

footing with an area of 4

was used.

The

4.0
">

3.0

^2.0

0.2

0.4
Pressure

0.6
,

0.8

1.0

kg/cm*

Fig. 1-23. Variation during a horizontal load test on a 4.0-m 2 plate of the coefficient
of elastic

uniform shear

cT

of fine saturated sand.

results of these investigations,

as in plastic silty clays with

elastic

uniform shear increases proportionally to the increase

pressure on the

The

shown in Fig. 1-23, indicate that in sands,

some sand and in loess, the coefficient of
in

normal

soil.

under consideration demonstrate that the elastic resistance

depends on the normal pressure. The fact that the
magnitude of elastic shear depends not only on the shearing stress, but
also on the normal pressure, indicates that the law of independent action
of forces cannot be applied in cases of wide ranges of changes in the
magnitudes of external loads, even when considering the elastic deformations of soils.
This deviation of soil from the properties of linear systems
tests

of a soil to shear

apparently explained by the fact, already discussed in Art. 1-1, that the
of elasticity in shear of a soil depends on the normal pressure.
Therefore, as noted before, there are limitations to the application of
Hooke's law to soils.

is

modulus

Investigations of test foundations with different base contact areas

indicate that according to Eq. (1-4-5) the coefficient of elastic uniform

shear of a

soil

decreases in inverse proportion to the square root of the

DYNAMICS OF BASES AND FOUNDATIONS

40

foundation base area. Figure 1-24 shows values of this coefficient

established experimentally for test foundations on loess.

The dotted

on the basis

line is plotted

of the

assumption that

cT

changes

in inverse proportion to the square root of the foundation base contact

area.

The conclusion can be

reached, as

it

was

in relation to the coefficient of

uniform compression, that the relationship expressed by Eq.

(1-4-5) is valid only for foundations with comparatively small contact
It can be held that for foundations with base areas larger than
areas.
2
10 to 12
c T does not depend on the area.
According to the data of Table 1-11, for foundations with square bases
and for cases in which Poisson's ratio
12
lies in the range 0.3 to 0.5, the ratio
10
c u /c T changes within the range 1.22

elastic

I 8

>H

~~

sult

test

pressure of 0.4 to 0.5

Base contact area 4,

loess, c u /cr

c u /c T

Dependence

of the coefficient

uniform shear
on the loaded area.

of elastic
soils

following values of the

were established as a reof experiments with the same

foundation and under a normal

ratio Cu/c T

?-

Fig. 1-24.

The

to 1.50.

cT

=1.9;

kg/cm 2

for

for gray fine sand,

2.20 to 2.40.

Since theoreti-

cally derived values of the ratio c u /c T

of loessial

do

not take into account the in-

fluence of normal pressure on the

magnitude

of c u

it is

advisable in design computations to use values of

this ratio close to those established experimentally.

assumed that
1-5.

If

a foundation

of a foundation

may

It

can be tentatively

2.

Coefficient of Elastic Nonuniform Shear of Soil c$

axis, it will rotate

yp

c u /c T

is

acted upon by a

around
is

moment with

respect to the vertical

Tests show that the angle of rotation

this axis.

proportional to the external

moment.

Therefore

we

write:

M
where

(1-5-1)

CfJjp

moment producing rotation of base of foundation

around a vertical axis to angle
external

\j/

Jz =

polar

moment

of inertia of contact base area of foundation

c$ = coefficient of elastic nonuniform shear

In the rotation of a foundation around a vertical axis, the base of the
foundation undergoes nonuniform sliding; hence the term "coefficient of
Experielastic nonuniform shear" may be applied to the coefficient c$.

ments show that

its

magnitude

is

somewhat

larger than that of the

ELASTIC PROPERTIES

coefficient of elastic

uniform shear

cT

OF SOIL

41

Table 1-12 gives values of

by tests performed on the same foundations on

clay with some sand and on gray sands.
c$ established

Table

1-12.

cT

and

plastic silty

Values of the Coefficients of Elastic Uniform Shear

and of Elastic Nonuniform Shear c^ for
Different Soils and Foundation Areas

cT

A,

Soil

Gray

plastic saturated silty clays

fine

kg/cm 3

1.5
1.0
0.5

1.27
1.64
1.88

2.72
2.33
2.08

2.14
1.42

1.0
4.0
15.0

2.54
2.20
1.90

3.90
2.85
2.20

1.54
1.30
1.16

with some

saturated sands

As established by the

kg/cm 3

sand

Gray

results of all tests, c$

is

larger than c T

CT

1.01

The

average value of c$/c T was in all cases found to equal 1.43. Therefore,
it can be assumed that a value of c$ = 1.5 c T when used for design com,

putations, will not differ greatly from the actual value.

1-6.

Elastic Resistance of Pile Foundations

a. Elastic Resistance of Piles under a Vertical Load.

The use of piles in
foundations subjected to vibrations and shocks may be necessary in four

cases: (1)

if

the total pressure on the

soil,

larger than the bearing capacity of the

action of the foundation on the

both

soil,

soil; (2) if it is

static

and dynamic,

is

considering the dynamic

necessary to increase the

natural frequency of vibrations of the foundation, which

is considered to
be a solid body resting on an elastic basis; (3) if it is necessary to decrease
the amplitude of natural or forced vibrations; (4) if it is necessary to

decrease the residual dynamic settlement of the foundation.

The second and third cases refer mainly to foundations for machines
with dynamic loads; the first and fourth cases are more common in the
design of foundations for structures subjected to the dynamic influence of
waves emanating from machine foundations.
foundation is to be used because the total pressure on the soil is
than its bearing capacity (taking into account the dynamic action
of the foundation on its base), then conventional methods of design, such
as are generally used in the design and computations of pile foundations
producing only static pressure on soil, are applied. In such cases the
practical procedure of pile-foundation design consists in the determination
of the number of piles needed from the known value of the bearing
If a pile

larger

DYNAMICS OF BASES AND FOUNDATIONS

42

is determined by load tests, and the

on the basis of test pile driving.
If the use of a pile foundation is due to the need for increasing the
natural frequency of vibrations and decreasing the amplitude of free or

capacity of a single

length of the piles

This value

pile.

is

selected

10

12

14

16

Total settlement,

20

18

22

24

26

28

mm

Fig. 1-25. Results of a vertical load test on a single reinforced concrete pile.

forced vibrations of the foundation, then the practical design procedure

has some peculiarities which will be examined in this article.

A pile foundation, like the natural soil base beneath it, is characterized

120

by

100

to a single reinforced-concrete pile.

80
o

is

^"60
.3

elastic properties.

Figure 1-25 shows the results of a test

in which a static vertical load was applied
clear

ment

mum

20

It

settle-

of the pile decreases with a decrease

in load.

40

from the graph that the

It is also seen that

value of load

is

if

the maxi-

constant, then the

decrease in the elastic settlement also

remains constant.
3
4
Elastic settlement S,
1

mm

loops of Fig. 1-25.

range of measurement errors,

after the first

was 3.5 mm after the second unloadwas 3.10 mm; after the subsequent
Within the
ones, 3.85 and 3.3 mm.
which for many reasons can be fairly signifipile

Fig. 1-26. Evaluation of hysteresis

Thus

unloading, the elastic settlement of the

cant in pile tests of this type,

ing

it is

it

possible to consider that the elastic

settlement of the pile up to a load of 75 tons remained constant and on

Figure 1-26 gives a graph showing the
the average was equal to 3.4 mm.
relationship between the elastic settlement of the pile and the magnitude

ELASTIC PROPERTIES

OF

SOIL

43

This graph shows that the elastic settlement of the

Thus, denoting by P
pile is proportional to the magnitude of the load.
by
elastic
the
pile,
and
8
the
settlement,
we obtain
the load acting on

of the load thereon.

P = C
where Cj

is

(1-6-1)

a coefficient of proportionality which

we

call

the coefficient of

elastic resistance of the pile; it represents the load required to

unit elastic settlement of the

The

induce a

pile.

depends on soil properties,

and the length of time the pile has been

coefficient of elastic resistance of the pile

pile characteristics (e.g., length),

For example, the

in the soil.

elastic resistance of

pile

has different

values during driving and some time later.

The problem under discussion was a subject of the author's investigaperformed by measuring vertical vibrations of the soil in the vicinity
of the pile being driven.
Soil vibrations induced by pile driving have the
same period as the vertical vibrations of the pile itself, the latter being
considered a solid body resting on an elastic base.
Therefore, by measurtions,

ing soil vibrations,

The

it is

possible to estimate the vibrations of the pile.

relationship between the period of vibrations

elastic resistance C& is given

JT

where m

is

T and

the coefficient of

by the expression

-&

(1-6-2)

the mass of the pile and the pile-driving mechanism.

of Eq. (1-6-2)

it is

By means

possible to establish the coefficient of elastic resistance

from the measured period.

Figure 1-27 shows the soil profile at the

of the pile

the test pile and gives

site of

diagrams illustrating the effect of the number of blows on the periods and
the depth of pile penetration.
The reinforced-concrete pile was 10 long
with cross section 30 by 30 cm. The driving was performed by means of a
3-ton steam hammer; the height of the drop was the same at each stroke
around 1.0 m.
A study of the diagrams illustrating the effect of depth of pile penetration on changes in periods of vibrations leads to the conclusion that an
increase in depth of penetration does not produce a decrease in period of
vibrations, as could be anticipated on the basis of the assumption that the
elastic resistance of a soil acts mostly along the lateral pile surfaces; on the
contrary, the period of vibrations even increases somewhat.
A pause of 10 days was made in the pile driving when it was resumed,
the average refusal of piles was found to be 10.5
per 11 blows.
During this additional pile driving, the periods of vibration were meas-

mm

ured.

The

results of these

measurements are

also given in Fig. 1-27;

_g

o
O

of

up

large

3:

o
to

sand

>>

sand,

larger

in

0^

contains

with

rocks

and

and

to E
t

composed

of
some

sand,
clayey

clay
fragments

oc

CO

gravel

_c

S'E
to

with

fine

fine silty

sand

"*"

of

If
a>

rock
various

of

O O

diam

"55

fragments

with

laminae

and

silty mica,

sand

of

Brownishryellow

Brownish-yellow

and

C
c

a>

> c

Grayish-brown

CD .2

to

TO -^

with

clay

c .
5 <u
">
o
1_ o

1=

glauconite

admixture

<1>

sizes,

crystalline

o
w o

cm

fragments

E
en

places grains

some

ro

flint

in

^r
u JO

N-

CO

>-

u_

000
000

to

a> to

some

to

to

grainsgravel

Green

with

O
O

CD

*u oiioj+auad

1J -0=^0

<
<!

*5

0
1

\ ;^ 5.^

>~-

'ft

i
f

o
c

T3

V_

>

<b

1
I

? * /
Q
-

r/

\i
I

-OC\J

tco

/
\

it

II

/
t

+4SL

-X
CO

ro

O
O

lo

O
d

CO

co

O
O

39S 'pOJJ9d

en

O
d

ELASTIC PROPERTIES

OF

SOIL

45

they indicate that, after the "rest," the period of natural vibrations of
During pile driving, the average period
the pile decreases somewhat.
(at a penetration depth of 6 to 9 m) was 0.060 sec during the additional
Since squares of periods are inversely
driving it decreased to 0.048 sec.
;

proportional to the coefficients of elastic resistance, this decrease in the

period after the pile "rest" attests to the fact that the elastic resistance
of the soil increased after the "rest"

by approximately 45 per

cent.

A decrease in the period of natural vertical vibrations of the pile resulting from a "rest"

may

occur only on account of an increase in the elastic

resistance of the soil acting along the lateral surface of the pile.

This conclusion fully agrees with the experience derived from conit is known that a few days after the completion of pile

struction practice

driving the dynamic resistance of the pile increases.

increase, the refusal of piles

driving, but

some time

later.

is

Because of

this

determined, not immediately after pile

The author found

in several cases that

during driving the period of vibrations of a pile is constant. This point

has also been confirmed by the experiments of F. A. Kirillov and S. V.
Puchkov, 23 who measured periods of soil vibrations during the driving of

two piles. When the first pile was driven, the period of vibration did not
change with an increase in the depth of penetration, but remained equal to
0.14 sec; observations on the second pile showed a decrease in the period
of vibrations from 0.17 sec at a penetration depth of 4.75 m to 0.13 sec at
a depth of 8.00 m. In this case, the decrease in period is explained by
changes in soil properties.
These experimental data lead to the conclusion that as the pile penetrates into the soil, the depth of its penetration has little effect on the
elastic resistance of the soil; therefore it should be assumed that elastic
soil reacts mostly on the pile tips.
As noted before, this conclusion is valid only when piles remain in the
soil for

a short time.

cohesion between the

When

soil

piles are in the soil for a long time, the

and the lateral

side surfaces of the piles increases,

The magnitude of elastic

from soil action along the lateral pile surfaces may be
considerable as compared to the magnitude of elastic resistance caused by
soil reactions against the tips of piles.
This is confirmed by experimental

resulting in an increase in elastic resistance.

resistance resulting

pullout tests of piles subjected to static loads.

Figure 1-28 gives values

wooden piles, established experimentally for pile driving

and for pulling out. The experiments were carried out by Ya. N.
Smolikov on saturated fine gray sands. Comparison of these values

of Cg for several

reveals that the elastic resistance of piles is approximately 30 per cent

higher during driving than it is during pullout.
Hence it follows that the
elastic resistance of a pile

conditioned mainly by

soil

which has been in the

soil for

a long time

reactions along the side surfaces of the

is

pile.

DYNAMICS OF BASES AND FOUNDATIONS

46

Since the magnitude of the soil reactions on these surfaces increases with
the depth of pile penetration, Ci should increase with an increase in pile
length.

This

confirmed by testing

is

1-28, which gives diagrams which

piles of different lengths (see Fig.

on the

illustrate the effect of pile length

coefficient of elastic resistance of the pile).

On the

basis of the results of experimental investigations, the coefficient

may

of elastic resistance of a pile Cs

be considered to be approximately

40

Pile^driving

Pulling-out

30

X
E

$if

J? 20

i
y
?
s
y

&

10

4.0

2.0

Pile length

Fig. 1-28. Variation of the coefficient

6.0
I,

8.0

d of a pile during a driving and pullout

proportional to the square of the depth

of pile penetration

and

test.

to its

perimeter S:

C
The reduced

coefficient c

--=

(1-6-3)

cSl*

has the dimensions tons per meter 4

Let us assume that the coefficient of elastic resistance of a pile group is

Thus if n denotes the
proportional to the number of piles in the group.
number of piles, the coefficient of elastic resistance of the group equals

K
An

increase in the

number

(1-6-4)

nC*

of piles

under the foundation leads to an

therefore the coefficient of

elastic uniform compression of the pile group increases proportionally to
increase in

its

area of contact with the

soil;

the contact area of the foundation.

This conclusion is confirmed by
experimental data. Figure 1-29 gives a diagram of the dependence of the

ELASTIC PROPERTIES

coefficient of elastic

OF

uniform compression

SOIL

47

for pile foundations

contact areas of these foundations, and consequently on the

piles

The
length was

under the foundation.

on the

number

distance between the piles in

all

of

cases

was 0.81 m; the pile

also constant, equaling 5.5 m.
Tests
were performed on saturated gray fine sands.
Curve 1 shows the relationship between K z and the foundation contact
area (or the number of piles).
It reveals that, unlike the case of spread
foundations resting on natural soils (curve 2), the coefficient of elastic
uniform compression for

foundations increases proportionally to the

pile

160

pile s

8
10
12
2
Area A, m

14

140

120
12 piles-

100

/
1

80
n 60

9f>iles

yz

40

/
20

16

18

Fig. 1-29. Variation with size of footing of the coefficient of elastic resistance
pile
soil

group (curve 1) and of the coefficient of

(curve 2) under a footing without piles.

elastic

uniform vertical compression

of a
c u of

increase in the foundation contact area, or, other conditions being equal,

proportionally to the

number

of piles.

Table 1-13 gives results of vertical load tests performed on six similar
foundations, each resting on nine piles.
The lengths of the piles were
The perimeters
identical, but the distances between them were different.
of all piles were approximately the same and equal to S = 0.9 m.
Tests
were carried out on saturated gray fine sands.
It follows from Table 1-13 that the distance between piles in a group
affects the value of the coefficient of elastic resistance Cs of the piles in

the group.

When

the

distance between piles increases,

resistance of each pile in the group increases;

when

the

elastic

this distance is suffi-

ciently large, the resistance of each pile in the group approaches the

resistance of a single pile.

Thus, in order to evaluate the

elastic resistance

of each pile in the group, a corrective influence coefficient should be

DYNAMICS OF BASES AND FOUNDATIONS

48

introduced to take into account the distance between piles. Then the
coefficient of elastic resistance of one pile in the group will equal:

Ck =

M C,

(1-6-5)

The experimental data available are insufficient to evaluate with an

adequate degree of precision the value of the influence coefficient ju for
Table

1-13.

Values of Coefficient Cs of Elastic Resistance of Piles

Pile

length,

5.5

3.5

between
piles,

Elastic

Load on

Distance

settlement

foundation,

of founda-

tons

tion,

c5
kg/cm
,

cm

0.81
1.22
1.62

30.2
28.8
28.8

0.049
0.028
0.026

6.9
11.5
12.4

X
X
X

10 4

0.81
1.22
1.62

28.8
30.2
30.2

0.063
0.047
0.048

5.1
7.2
7.0

X
X
X

10 4

10 4
10 4

10 4
10 4

piles.
A theoretical solution of this problem
would require the solution of the problem of distribution of soil deformaTherefore the
tion around each pile, but this has not yet been done.
influence coefficient necessarily has to be evaluated on the basis of those
insufficient experimental data which were obtained as a result of testing of
The values of the
pile groups with different distances between piles.

various distances between

coefficient of elastic resistance of piles of various lengths presented in

Fig. 1-28 give a basis for the determination of the value of the coefficient

With

of resistance of a single pile.

/*

for this pile, it is possible,

by

using the data of Table 1-13, to compute the value of the influence
coefficient for various ratios of the distance t between piles to the pile

diameter d. The values of influence coefficient established in this

are given in Table 1-14.

Table

way

Correction Coefficient fx (for Use with the C$ Values

of Table 1-13) Expressed as a Function of the Ratio
of t (Spacing) to d (Diameter) of Piles

1-14.

"

1.0
0.65

4.5
3

0.64
0.41

ELASTIC PROPERTIES

It follows

group

OF SOIL

49

from Table 1-14 that the mutual influence among

even when the distance between the piles

is significant,

the elastic resistance of one pile in the group

elastic resistance of a separate pile.

In the design of

pile

is

foundations, a value of t/d

times

then
only 40 per cent of the

If this ratio is three

larger than the pile diameter.

piles in the
is six

is

instead of

six,

usually selected within

According to the data of Table 1-14, the design value

corresponding to such distances is about 0.65.
of
As noted before, the value of the coefficient of elastic resistance Ck of
one pile in the group is the main characteristic of the elastic properties of
This value may be established by
the pile base under a foundation.
the range 5 to

6.

jli

Table

1-15.

Summary of Values of Coefficient Ck of Elastic

Resistance of a Pile in a Group
Founda-

Type

of

engine

Description of soil
beneath foundation

tion

Number

Distance

contact

of piles

between

area,

Vibrator

Saturated
silty clays

piles

t,

eter of
piles,

Ck,

kg/cm

gray
with some
1.5

sand
Soft plastic clay un-

Logsawing
frame

Saturated grayishbrown fine sand with

laminae of organic

Logsawing
frame
Logsawing
frame

Saturated fine sand

with admixture of

by saturated
medium sand

0.55

3.0

0.1

10*

derlaid

17.4

40

0.5

6.5

0.8

1.3

10*

24.8

30

0.7

6.0

0A

3.4

10*

24.8

20

0.8

7.7

10*

organic silt
Saturated fine sand
with admixture of

12.0

35

0.4

3.0

0.8

5.3

10*

organic silt
Loess and loessial

18.0
4.0

35

0.5
0.75

3.0
6.0

0.8
0.8

37.8

X
X

10*

Saturated sand with

admixture of organic
silt and blue plastic
clay

silt

means

Perim-

soft

Logsawing
frame
Logsawing
frame

Vibrator

Pile

length,

soil

1.1

10*

of experimental investigations of the elastic properties of single

There have not been many experiments of this kind. Ck may

be also determined from results of testing of pile groups or pile foundations.
Table 1-15 gives several values of Ck obtained by the author and
Ya. N. Smolikov as a result of investigations of test foundations and
foundations for log-sawing frames.
piles.

The data of Table 1-15 make it possible to select a tentative design

value of C k for the computation of vertical vibrations of machine foundations under soil conditions similar to those described in the table.

If

DYNAMICS OF BASES AND FOUNDATIONS

50

on the same construction site, then

should be established from special investigations of

several pile foundations are to be built

Ck

design values of
test foundations.

Elastic Resistance of Pile Foundations in Cases of Eccentric Vertical

b.

Let us assume that under the action of an external moment

its main vertical planes, the foundation rotates by a small
Then it will be under the action of the reactive moment of a pile

Loading.

acting in one of

angle

<p.

L from

located at a distance

CL

the axis of rotation; the value of this

By

equating the external moment

moments transmitted to the foundation by the

is

<p.

M
Here, as before, Cs
[Eq. (1-6-3)].

Cap

moment

and the sum

piles,

of the

we obtain

(1-6-6)

the coefficient of elastic resistance of one pile

is

Equation

can be rewritten as

(1-6-6)

K&
V

where

(1-6-7)

may

be called, as is done with a natural soil base, the coefficient of elastic

nonuniform resistance of pile foundations; its units are kilograms X

centimeters or tons

meters.

Table 1-16 gives the^ results of an investigation of three test foundations

under the action of nonuniform vertical loading; nine piles were driven
under each of the foundations.

Resistance of Pile Foundations

Dis-

Perimeter of

tance
be-

2L

piles,

tween

piles,

Pile

length,

External

2
,

moment,
tons

Xm

Angle of
rotation of

the founda-

tons/m

kg/cm

tion, radians

m
3.6
3.6
7.2

0.9
0.9
0.9

1.62
1.22
1.22

15.2
9.0
9.0

14.4
12.6
18.0

1.17
1.92
1.10

X
x
X

10~ 4

12.3

i<r 4
10~ 4

6.6
16.3

X
X
X

10 4

8.1

10 4

7.3
18.2

10 4

X
X
X

10 4
10 4
10 4

Comparison of the values of C& in Table 1-16 with the values obtained
from testing foundations under a uniform vertical load (Table 1-13) shows
that there is comparatively little difference between the values of the
reduced coefficient of elastic resistance obtained by these two methods.

ELASTIC PROPERTIES
c.

OF

SOIL

51

Elastic Resistance of Pile Foundations to Horizontal Forces.

If

the

main exciting forces act on a foundation in a horizontal direction, then in

some cases pile foundations are provided in order to increase the natural
frequencies or decrease the amplitudes of
1.2

excited horizontal vibrations.

When a pile is designed to take into account the action of a horizontal force, it is

1.0

assumed that the

0.8

pile is fixed in the soil at

The
depth greater than a certain value Z
depth of fixation is taken on the order of 1 to
The influence of soil reactions on pile
1.5 m.
deformation is disregarded, and it is assumed

//

4y

0.6

0.4

that the resistance of the pile to the horizontal

0.2

^
/

determined only by the properties

and depth of fixation of the pile.
0.5
1.0
1.5
2.0
Elastic lateral
On the basis of these assumptions, it can
displacement e'x cm
be stated that an increase in length to a value
Fig. 1-30. Dependence of elasgreater than the depth of fixation will have
tic lateral displacement z'x of
no effect on the elastic resistance of the pile.
pile head on the lateral load
This is confirmed by experiments. Figure
and length of pile.
1-30 gives two graphs of the relationship between the elastic displacement of a pile head and the magnitude of the
horizontal force acting on piles having respective lengths of 5.4 and 3.6 m.
force

is

The

coefficients of elastic resistance of the piles to horizontal forces are

They are denned by the relation P x /e!x Figure 1-3 1 furnishes data
on experimentally established frequencies of natural horizontal vibrations
equal.

140

120

100

/
/

so

;6o
40
20
1.0

3.0

2.0

4.0

Pile length I,

5.0

6.0

7.0

8.0

Fig. 1-31. Natural frequency of horizontal vibrations of single reinforced-concrete

piles of

varying lengths.

of single reinforced-concrete piles of different lengths

sections (30

by 30 cm).

The

but of equal cross

distribution of points over the diagram

confirms the conclusion concerning the small influence of the depth of pile

DYNAMICS OF BASES AND FOUNDATIONS

52

driving on the elastic resistance of a pile against horizontal forces.

Thus,
can be held that the employment of long piles will not contribute to the
elastic resistance of pile foundations to horizontal forces.

it

one assumes, as was done on the preceding pages, that the elastic
is determined mostly by its size

If

resistance of a pile to horizontal forces

and the
1.0

rial,

&y

jfL

0.8

-- 9

0.6

cross

pile

elastic properties of its

mate-

then one must conclude that the

on

effect

section

will

have great

its coefficient of elastic resist-

ance to horizontal forces. This is

confirmed by testing piles with different cross sections.
Figure 1-32 gives
graphs of the relationship between

o/

-0.4
0.2

the elastic deformation of a pile and

0.5

1.0

2.0

1.5

Elastic displacements '

x%

2.5

3.0

cm X10.

value of the horizontal force.

the

were performed on wooden

29 to 33 cm diameter. For
piles with a diameter of 29 cm, the
Tests

piles of
Fig. 1-32. Elastic horizontal displace-

ments

of the

heads of two

piles of differ-

coefficient of elastic resistance

ent diameter.

found to be
piles

with a diameter of 23 cm,

it

was

0.41

1.0

was

10 4 kg/cm; for

10 4 kg/cm,

i.e.,

2.5

times smaller.
Accordingly,

it

can be anticipated that for reinforced-concrete piles

much larger than for wooden

the coefficient of elastic resistance will be

piles.

Table

1-17.

Natural Frequencies of Foundations

with and without plles
First

Foundation base

Number

contact

of piles

area,

10.5
8.6
8.3
6.5
10.6
4
4
15

two natural frequencies of

by slid-

vibrations accompanied

16
12
12
9
9

Length

Diameter ing

of piles,

of piles,

5.4

0.29
0.23
0.28
0.23
0.28

&

rotation of the foundation

Type

of

founda-

Lowest

Higher

frequency,

frequency,

cycles/sec

cycles/sec

82
75
86
55
60

266
240
270
270
253

86
92
64

158
222
143

tion

base

Wooden
piles

Natural
soil

base

ELASTIC PROPERTIES

These

OF

SOIL

53

peculiarities in the resistance of piles to horizontal forces

limitations on their

employment

put

for the reinforcement of foundations in

A comorder to increase the lowest natural frequencies of vibrations.

parison of test foundations constructed on wooden piles with those placed
directly

on

soil

indicates that the use of

increase of natural frequencies

when

wooden

piles

has no effect on the

the frequencies depend on the

resistance of the base to vibrations in a horizontal direction.

shown by Table

quencies experimentally established for

for foundations placed directly

Comparison

This

is

two main natural frefoundations placed on piles and

1-17, giving values of the first

on natural

soil.

of the lowest natural frequency of a foundation erected

on

with that of a foundation placed directly on soil reveals that the piles
do not increase the lowest natural frequency; on the contrary, the piles
Thus, pile foundations are characterized
decrease its value somewhat.
piles

by a value

of elastic resistance to shear

tions placed directly

on natural

soil.

which

The

is

not larger than in founda-

fact that the higher fre-

quencies are larger in pile foundations than in foundations placed directly

on natural

soil

shows that the value of the

larger in pile foundations.

elastic resistance to rotation is

EFFECTS

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

GENERAL
Studies and observations of models and existing foundations subjected
to shocks

and vibrations show that they may undergo settlements many

times larger than those imposed by static loads.

happens that
endanger the
produced by
ticular, under
only of the

Therefore

it

often

settlements of foundations undergoing vibrations greatly

safety

of

structures.

An

extremely adverse effect

the vibrations of foundations under machinery

forge

hammers.

hammer

in

is

par-

These vibrations lead to settlements not

foundations, but also of the footings under the

columns or walls of a building where hammers are located. Considerable

deformations of the superstructure result, often endangering the safety

and

stability of the

whole building.

Several cases illustrating the above points are presented in Chap. IX.

Considerable residual settlements caused by a joint action of static

when foundations rest on

loads and vibrations appear particularly often

noncohesive

soils

such as water-saturated sands, even

characterized by low values of porosity.

soils either

by means

if

the sands are

Static investigations of these

of load tests or in a consolidometer

show that they

possess high values of resistance, sometimes larger than corresponding

values of cohesive soils such as plastic clays.

medium and higher-than-medium

Therefore sands having

good natural

densities are considered

bases.
This is true, however, only for static loads. If a sand or other
noncohesive soil is subjected simultaneously to both static loads and
shocks (or vibrations), then, as shown by experience with industrial
This
construction, resistance to external loads decreases considerably.
is

verified experimentally

vibrating

piles,

by plunging

into the soil subsurface vibrators,

pipes, or cylinders with considerable diameters.

54

The

OF VIBRATIONS

EFFECTS

ON

RESIDUAL SOIL SETTLEMENTS

55

settlements of vibrating foundations, as well as the penetration of the

above structural elements into the soil to a considerable depth, may be

observed not only in noncohesive soils, but also in cohesive soils, for
example, in heavy clays. Of course, the quantitative effect of vibrations
and shocks on cohesive soils may be much smaller than on noncohesive
soils,

especially sands.

The

physical processes which cause changes in

soil

properties during

vibrations (such as the decrease in resistance against external loads

leading to the development of residual settlements) are not clarified as yet.

Experiments show that vibrations cause changes in the dissipative

soil, i.e., in the forces of internal dry and viscous friction,
in the forces of cohesion (which determine the initial resistance of soil to
properties of a

shear), in the forces of external friction, in the

hydrodynamic properties

(such as in the coefficient of permeability and in the pore water pressure),

and in elastic and plastic characteristics such as Young's modulus, the

modulus of shear, and the limits of elasticity and of plasticity.
In addition, it was established that if subjected to intensive vibrations,
some soils (especially sandy fills and loose silty and clayey soils) lose their
resistance to shear to such a degree that their mechanical properties are
closer to those of viscous liquids

than of

solids.

It follows that in the process of vibration, all properties of a soil

undergo basic changes, including those properties which govern the

may
soil

resistance to localized loading.

It is

not possible at present to determine with a sufficient degree of

precision the value of the residual settlement of a foundation subjected

to vibrations or shocks.
However, the data available from experimental
and theoretical investigations make it possible to evaluate the influence
of certain factors on the settlement of a foundation and consequently to

take suitable measures to decrease settlement.

11-1.

Effect of Vibrations

on the Dissipative Properties of Soils

a. Internal Friction and Cohesion of Soil during Vibrations.

G. I.
Pokrovsky and associates were the first to investigate experimentally the
influence of vibrations on the coefficient of internal friction.
The
results of these experiments were published in 1934.
They show that
the coefficient of internal friction depends on the kinetic energy of vibra-

energy increases, the coefficient decreases, approaching a

value 25 to 30 per cent smaller than that observed before vibrations.
At the same time, the author investigated experimentally the quantitative
tions; as the

dependence of the
of vibrations.

coefficient of internal friction of

Some

The curves show an

sand on the amplitude

results of these experiments are given in Fig. II-l.

increase in displacement with an increase in magni-

tude of the shearing force.

DYNAMICS OF BASES AND FOUNDATIONS

56

Test 1 was conducted on a freshly filled sand without vibrations.

Test 2 involved vibrations with an amplitude of 0.50
and a frequency
-1
Test 3 was conducted after vibrations ceased. Test 4
of 140 sec
involved vibrations of the same frequency as test 2, but an amplitude of
vibrations of only 0.15 mm.
A comparison of the curves based on data obtained with an without
vibrations permits the conclusion that vibrations have considerable effect

mm

fO

on the resistance of soil to shear.

Without exception, the slope of the

^
3-

shearing - force-shearing - deformation

curve

/>

<

is

smaller with vibrations than

without.
1-

^ rT

the

Since

shear

elastic

is

coefficient

tangent of that slope angle,

2-

of

proportional to the
it is

clear

'

that the elastic resistance of

i3

shear

is

soil

to

lower during vibrations than

0>

in their absence.

The experiments

also

established

that the coefficient of internal friction

is

lower during vibrations than in the

static
4

Displacement,

10

mm

cient
Fig. II-l. Effect of shearing force

on

displacement, with and without vibrations.

state.

With an

increase

in

acceleration, the value of the coeffiof

internal

friction decreases,

approaching asymptotically a limit

which depends on the properties of a
soil.

In addition to experiments with dry soils, analogous experiments were

conducted with sand having a moisture content of 10 to 12 per cent. It

was established that vibrations cause a smaller decrease

of friction in

moist

soils

than in dry

soils.

in the coefficient

Apparently

this

may

be

attributed to the fact that the absolute value of the forces of cohesion

between particles
cent

is

much

of sand having a moisture content up to 10 to 12 per

higher than the corresponding value in air-dried sand. As

the forces of cohesion in a

soil increase,

the influence of vibrations on

changes in physicomechanical properties decreases.

Comprehensive experimental investigations of the effect of vibrations
on the coefficient of internal friction of sand were performed by I. A.
Savchenko under the supervision of the author. Figure II-2 shows the
installation used for these experiments.
It consists of a device for the
determination of the shearing resistance of a soil either subjected or not
subjected to vibrations, a vibrator 1 driven by a direct-current motor,

and a recording instrument

The device

for

2.

measuring the shearing resistance of a

soil consists of

EFFECTS

OF VIBRATIONS

three principal parts: a box

jack

ON

3, in

4, to apply a vertical pressure

RESIDUAL SOIL SETTLEMENTS

which shearing is produced, a spring

on the soil sample, and the setup 5 for

the application of shearing stresses to the

soil,

spanned over a pulley and a loading platform.

consisting of a cable

The shear box

attached to the vibrating platform 6 of the device;

cylinder closed at

its

bottom and two rings inserted into a

4.

The supports

rigidly

is

consists of a lower

it

calibrated spring placed into the guides of the cylinder

part of the spring jack

57

is

socket.

The

the principal

7 of the jack are rigidly attached

to the vibratory platform 6 of the device.

In order to ensure the static nature of the action of the shearing load on
soil, the loading platform 5 is mounted on vibration absorbers.

the

Fig. II-2. Installation used for investigation of the shearing resistance of

soil.

Horizontal vibrations in the direction of shearing are induced by means

of a vibrator of the usual two-shaft type.

the vibratory platform 6

on the

rollers 9

by means

The vibrator

of a turnbuckle 8

mounted on the guide frame

10,

which

is

connected to

and can
is

easily slide

rigidly connected

under the installation.

During tests, the speed of the vibrator could be increased to 3,000 rpm.
The amplitude of vibrations changed up to 1.7 mm, depending on the
magnitude of the moment of the eccentrics of the vibrator.
The influence of vibrations on the coefficient of internal friction was
evaluated by comparing its value determined in the absence of vibrations
to the foundation

with the value obtained during

remaining equal.

The

soil

vibrations,

all

other conditions

was determined as follows: the shear

with sand and was placed on the vibratory platform.
Then the sand was compacted by means of vibrations under a normal
coefficient of internal friction

box was

filled

DYNAMICS OF BASES AND FOUNDATIONS

58

pressure of 0.5
to a certain

kg/cm 2

After that the pressure on the

magnitude and shearing

of the soil

soil was increased

was induced while the box

was vibrated at selected amplitudes and frequencies.

The regime of vibrations was kept constant during a test, but shear was
produced under varying normal pressures. The results of each experiment were presented as graphs of relationship between shearing strength
An example of such a graph is shown in Fig.
s and normal pressure p z
Analogous graphs secured from other experiments show that a
II-3.
linear relationship between the shearing strength of a soil and the normal
pressure on it is observed with and without vibrations.
This relationship
.

is

described as follows:

where tan

<f>

So

=
=

p z tan

<f>

-f-

so

shearing strength of sand, whose value depends

both on forces of cohesion

sand and magnitude of

of

adverse resistance in the device

The value

itself

of the coefficient of internal friction of

was

vibrations
1.0

sand in the absence of

determined in the

same way.

0.8
*-"

,?r
'O

0.6
c

0.4

0.2

sand

coefficient of internal friction of

initial

While
place,

it

shearing of

the

came

soil

took

to the experimenters'

attention that under the steady verti-

cal

pressure

on the plunger

of

the

device there occurred some contrac-

0.4

1.2

0.8
f

1.6

kg/cm 2

Fig. II-3. Relationship between the

shearing strength of a

soil

and normal

pressure.

tion and then some loosening of the

sand in the device. To investigate
this phenomenon, measurements of
the settlement of the plunger were
taken.
These measurements showed

that with an increase in horizontal pressure the

then began a process of expansion of the

was completed.

soil,

soil

at first contracted

which continued

until shear

The

process of initial contraction and subsequent expansion of the

was observed both when the soil was subjected to vibrations and
when it was not. In the latter case this process develops more
intensively, especially the contraction phase.
The tests also showed
that as normal pressure grows, the changes of porosity, produced by
soil

shear, decrease.

The relative change in porosity during shear is small; therefore it does

not substantially affect the shearing strength of a soil as determined by its
coefficient of internal friction.


EFFECTS

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

59

Figure II-4 illustrates the relationship between the coefficient of

and the amplitude of vibrations of a dry medium-

internal friction tan

grained sand while the frequency remains constant.

These graphs show

that the coefficient of internal friction of sand decreases continuously

as the amplitude increases.

The dependence of tan for the same sand on the angular frequency of
As frequency increased up
a> is more complicated (Fig. II-5).

vibrations

1.2
1.0

0.8

Vk
4-4

~0.6

0.4

4-

0.2

0.4

0.8

1.2

1.6

Amplitude

2.0

2.8

2.4

3.2

3.6

A mm
%

Fig. II-4. Relationship between the coefficient of internal friction of dry mediumgrained sand and the amplitude of vibrations: curve 1, = 25 sec -1 curve 2, w =
144 sec -1 curve 3, w = 177 sec -1 curve 4, w = 208 sec -1 (o = angular frequency of
;

vibrations).

40

80

160
120
Frequency of vibrations

200
cj,

sec

240
-1

280

Fig. II-5. Relationship between the coefficient of internal friction of sand and the
frequency of vibrations: curve 1, A = 0.35 mm; curve 2, A = 0.85 mm; curve 3,
A = 1.2 mm; curve 4, A = 1.6 mm.

to 180 sec -1 the coefficient of internal friction slowly decreased; then, as

,

the frequency grew from 180 to 250 sec -1

the coefficient of internal

friction sharply decreased;

subsequent increases in the value of the frequency had almost no influence on the coefficient of internal friction.
These data permit the conclusion that there exists a range of frequencies
which corresponds to small changes in the coefficient of internal friction

of sand.
It follows also that

and the frequency

tan

<j>

for a

of vibrations.

sand depends on both the amplitude

DYNAMICS OF BASES AND FOUNDATIONS

60
1.0

0.8 :
.0.6

zsi^zbzzzzzzz:

0.4
0.2

Acceleration

5
ratio 77

Fig. II-6. Relationship between the coefficient of internal friction of sand and the
acceleration of vibrations.

Figure II-6 illustrates the relationship between the coefficient of

and the acceleration of vibrations, which is
approximately defined by the empirical formula

internal friction of a sand

tan

where tan
tan

<f> st

0oo
f]

=
=
=

(f>

(tan

<f> st

tan

exp

0*,)

(3rj)

tan

(ii-i-D

<f>

value of coefficient of internal friction without vibrations

limit value of coefficient of internal friction
ratio of acceleration of vibrations to

acceleration of

gravity
(3

determining effect of vibrations

coefficient

medium-grained sand,

The

13

dry

(for

0.23)

investigation of the influence of moisture content on the coefficient

was performed on
medium-grained sand, as shown in
Fig. II-7.
It can be seen that the

of internal friction of a soil subjected to vibrations

0.8

*<

0.6

ir^
'

0.4

smallest decrease of the coefficient of

internal friction due to vibrations

"i

when

0.2

8
12
16
20 24
Water content w, percent

28

Fig. II-7. Relationship between the

coefficient of internal friction of

sand

subjected to vibrations and its moisture content: curve 1, co = 144 sec -1

;

-1
2, w = 250 sec
both curves.

curve
for

A =

0.35

mm

was

the moisture content equaled

approximately 13 per cent.

The study of the influence of grain
size on the effect of vibrations was
performed on four varieties of sand
under two regimes of vibrations, as

shown
the
of

in

Fig.

coefficients

these

absence

II-8.

of

The values
internal

of

friction

sands are different in the

of

vibrations;

therefore

it

appeared advisable to compare, not the absolute values of these coefficients, but the values 8 characterizing the effect of vibrations thereon:
5

tan

<}> 8 t

tan

tan

<p st

<f>

EFFECTS

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

61

are coefficients of internal friction respectively

where tan 8t and tan
without and with vibrations.
The value 8 shows the degree of decrease in the value of the coefficient of
internal friction under the influence of vibrations.
The data graphically presented in Fig. II-8 show that the effect of
vibrations increases in proportion to sand-grain diameter.
As indicated above, the effect of
b. Vibroviscous Soil Resistance.
viscous forces of friction is most clearly displayed in noncohesive (freshly
Tests show that in such a
filled) sand undergoing intensive vibrations.
sand the forces of internal dry friction between particles may completely
</>

<f>

1.0

0.8

0.6

.a

|0.4
0.2
0.8
1.2
1.6
Diameter of grains,

0.4

Fig. II-8. Relationship between the effect of vibrations

grains: curve

1,

= 144

sec

-1
;

curve

2,

o>

2.4

2.0

2.6

mm

250 sec -1

and the diameter

0.35

of

sand

mm for both curves.

disappear under the action of vibrations, and the sand may acquire the
mechanical properties of a viscous fluid. Then objects placed on this

sand

will sink into it

with a certain velocity

the unit weight of the sand and will float

if

if

their unit weight exceeds

their unit weight

is

lower than

that of the sand.

property may be characterized by the coefficient of vibroFor the determination of this coefficient and for the investigation of its dependence on the acceleration of vibrations, the moisture
content, and other factors, the method of a falling sphere may be used, as
in the study of viscous fluids.
This method is based on the well-known
Stokes' law, which establishes the dependence of velocity v of the motion
of a sphere in a viscous fluid on the force R acting thereon, the radius r
of the sphere, and the coefficient of viscosity n of the fluid:

This

soil

viscosity.

R =

(II-1-2)

Qirnvr

under the action of weight only, then, designating by 71

(i.e., the weight corresponding to the
unit volume of the sphere) and by y 2 the unit weight of soil, we may
rewrite Eq. (II-1-2) as follows:
If the

sphere

is

the reduced unit weight of the sphere

where

S = nv
2
s = Ar\
yi -

(II-1-3)
72)

(II-1-4)

DYNAMICS OF BASES AND FOUNDATIONS

62

The device used by the author for the investigation of the coefficient of
is shown in Fig. II-9.
A container 1, 30 by 30 cm in cross
section, 40 cm high, and filled with sand to
the height of 30 to 35 cm, was placed on the

vibro viscosity

u u

vibratory platform

which could be sub-

2,

jected to vertical vibrations of a selected

mm

amplitude and frequency. A metallic sphere

3 of 28
diameter was placed on the soil
surface.
A load 4 was imposed on this
diameter)
sphere by means of a thin (2.5

ak

mm

*-5

The sinking of the sphere into the

was recorded by the automatic recorder

steel rod.

soil
5,

which simultaneously registered the sinkand marked the corresponding time

ing

The counterweight 6 balanced

intervals.

pen of the recorder.

Investigations were performed on white
quartz sand with grain sizes from 0.2 to
frame

the
Fig. II-9. Device for investi-

gation of vibroviscous properties of filled

sand.

following

the

mm.

0.5

The

of

method was used

experimental investigation of the

in the

relationship between the coefficient of vibroviscosity

and the acceleration

of vibrations.

In the course of

all

the experiments the moisture content of the sand

22

1-

II

7
/

/k

14

//

*4

5-

6^

//
10

///

It

7=1.6 8;

A/

= 190

~)

mi

V
120

240

600

480

360

720

Time, sec
Fig. 11-10. Relationship between the depth of sinking of sphere into vibrating sand

time: curve
load

1,

1.0 kg;

load

curve

1.9 kg;
5,

load

curve

2,

load

0.6 kg; curve

1.5 kg;
6,

load

curve

3,

load

0.3 kg.

1.2 kg;

and

curve

4,

EFFECTS

was equal

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

63

Prior to testing the sand was compacted by

to zero.

means

of

void ratio was close to its minimum

The unit weight of soil corresponding to this

vibration so that the value of

its

(approximately 0.5).
was approximately 1.77 g/cm 3

porosity

While the value of the acceleration of vibrations remained the same,

the sinking of the sphere was recorded under variable loading.
After the completion of a series of tests with a constant acceleration of
vibrations and selected values of

the

loading,

acceleration

40

was

30

changed and a new series of experiments was conducted, in order

20

to record the sinking of the sphere

10

which

7=1.62
^O

corresponded to various
This procedure

values of loading.

was repeated

successively.

The graphs

0.20

1 0.1

of relationship be-

tween time and the depth

of sinking

sand are
These curves

of the sphere into vibrated

0.4

seen from the

0.2

graphs that the velocity of sinking

varies: it decreases with the depth

0.1

0.6

are plotted for constant acceleration

vibrations

of

loads.

It

of sinking.
sinks,

its

is

first

However, as the sphere

acceleration approaches

0.5

of

inertia

may

be

0.3
0.2
0.1

^S

20 30 40 50 60 70 80 90

S
Fig.

11-11.

Relationship

between the

velocity of sinking of sphere into vibrat-

ing sand and load S.

in Fig. 11-11 present relationships

ss

,y

'

Is

10

neglected.

Graphs

Oi/

= 2.44

0.4

plied

influence

-r

7 =2.02

and its velocity attains a

more or less stable value depending on the magnitude of the load
and the acceleration of vibrations.
Equation (II-1-3) may be aponly for the time interval
corresponding
to
steady sinking velocity, when, consequently,

"

and variable

zero,

the

2"*

<*

^0.05
o

in Fig. 11-10.

7 =1.68

x 0.10

shown

for constant values

r)

of the

acceleration of vibrations between the sinking velocity of the sphere and

S (i.e., the load acting on the sphere). These graphs confirm the linear
relationship, established by Eq. (II-1-3), between the steady sinking
velocity and the load.
It is directly seen from these graphs that the

between the two, i.e., the coefficient

depends essentially on the acceleration of vibrations.

coefficient of proportionality

viscosity,

of vibro-

'

DYNAMICS OF BASES AND FOUNDATIONS

(54

Figure 11-12 presents a graph of relationship between the value of 1/n

is, the reciprocal of the coefficient of vibroviscosity n) and the ratio 77

(that

between the acceleration

of vibrations

and the acceleration

of gravity.

This graph shows that when accelerations of vibrations are lower than

1.5<7, vibrations practically do

not affect the value of the coefficient
of vibroviscosity.
A sharp decrease
in n, and consequently an increase in
1/n, is found only when 77 > 1.5.
77

12

10

/
0.

The

relation between 1/n and 77 may

be approximately described by the
equation

c4
0.5

1.0

1.5

20

2.5

3.0

Acceleration ratio

3.5

Fig. 11-12. Relationship between 1/n,

the reciprocal of the coefficient of vibro-

and

viscosity,

the

4.0

>

a(rj

770)

(II-1-5)

rj

acceleration

of

770 may be designated as the

threshold of the vibroviscose state

where
of

soil.

The fact that

vibrations.

the magnitude of the

forces of cohesion in soils depends

on

makes

their moisture content

possible the assumption that the coeffidepends on moisture content. The method of
problem was analogous to the investigation of the

cient of vibroviscosity also

investigation of this

relationship between the coefficient of vibroviscosity

and the acceleration

the same sphere

The experiments were performed with

of vibrations.

2.5

CM

c)

A\

CD

^1.5' \

\\

,_

0.5

/
V
^o_ -0

Fig. 11-13. Relationship between

viscosity,

and the moisture content

A/
1

6
8
10 12
14
16
Water content w, percent

18

20

22

1/n, the reciprocal of the coefficient of vibroof

sand

fill.

and with the same sand, but with variable moisture content; the acceleration of vibrations and the total load imposed on the sphere (including its
weight) remained the same in all the experiments.
Figure 11-13 shows the relationship between the value 1/n and the
moisture content of a soil. With increase in moisture content to approximately 13 per cent, the coefficient of vibroviscosity increases approximately 220 times. A further increase in moisture content leads to a

EFFECTS

sharp decrease in

ON

OF VIBRATIONS

When

n.

RESIDUAL SOIL SETTLEMENTS

65

moisture content equals some 20 per cent, n

If one is to assume that the

attains about the same value as in dry sand.

coefficient of vibroviscosity of a soil is proportional to the forces of fric-

and cohesion acting between

tion

illustrates

The

how

its particles,

then the graph in Fig. 11-13

these forces change with moisture content.

results of these experiments lead to the

assumption that, other

conditions remaining equal, the residual settlements of foundations (or

the sinking into a

velocity

when

soil

as a result of vibrations) will have the largest

the sand either

is

dry or has a moisture content approaching

the largest possible moisture content for the given

soil.

An

analogous dependence of the coefficient \/n on the moisture content

was obtained for artificially prepared clayey soils.
c.

soil

Damping
by means

The study

Properties of Soils.
of a falling sphere

is

if

the intensity of

so large that the soil loses a considerable part of its shearing

vibrations

is

resistance

and behaves as a viscous body

Bingham.

of the viscous properties of

evidently possible

in the sense of

Newton and

manifested to a
much smaller degree than the elastic and plastic properties, so that the
method of a falling sphere is no longer applicable. However, although
Otherwise, the viscous properties of

the viscous properties of a

have a

soil

significant influence

may

soils are

be manifested very slightly, they

on the following

may

factors: free vibrations,

forced vibrations under conditions of resonance or close to resonance, the

magnitude

of energy spent for the

tion of waves,

and

maintenance of vibrations, the propaga-

others.

Therefore the study of these properties of

literature as

damping

soils,

which are known in the

properties, has a practical interest.

Damping properties of soils cause stresses which depend not only on

deformation but also on rates of deformation. Therefore soil reactions
against foundations depend not only on settlement, but also on rate of
settlement.

To

simplify computations,

generally assumed that

it is

linearly deformable elastic-viscous

body and that

soil

presents a

stresses are linearly

and the rates thereof.

damping properties of materials, including
soils, may be performed by several methods: from observations of the
damping of free vibrations of soil samples, from the amplitude of forced
vibrations under conditions of resonance, from the phase shift between
the existing periodic force and the soil deformation, and, finally, from
related to the deformations

The determination

of the

hysteresis loops.

The

last

method

is

the simplest therefore

;

we

shall dwell

on

it

in

some

detail.

Experiments show that even when stresses are small and are considerably lower than the elastic limit, the relationship between stresses and

DYNAMICS OF BASES AND FOUNDATIONS

66

deformations

is

not linear and single valued: to the same value of stress

there correspond values of deformation which are different under conditions of loading

and

ing and unloading

of unloading.

is

a closed curve which

Therefore, a cycle of successive load-

graphically presented, not by a straight line, but by

is

usually called a loop of elastic hysteresis.

greater degree of nonhomogeneity of a discrete material

is

reflected

by a greater intensity of display of the damping properties of that material

and by a larger area of the hysteresis loop. Since soils (especially sandy
soils) are more micrononhomogeneous and discrete than steel and concrete, for instance, the

much more
The

phenomena

of hysteresis are displayed in soils

intensively than in these other materials.

larger the loop area, the

of the material,

i.e.,

more intensive

the larger

is

are the

damping properties
work

the ability to absorb mechanical

irreversible form by transforming it into heat.

Quantitatively
work AW equals the area of the hysteresis loop. If
is the total
work performed during loading up to the maximum deformation, then the

in

an

this

ratio
,

AW
lv-

the coefficient of absorption which determines the amount of energy

absorbed by the material per unit of energy spent for deformation per
This coefficient may be related to the coefficient of resistance to
cycle.
vibrations, which is the coefficient of proportionality between the soil
is

reaction and the velocity of vibrations.

Let us assume that the diagram of free vibrations of a foundation will

have the shape of a periodic curve, analogous to that shown in Fig. 11-22.
If we take two successive maximum deflections of the foundation from the
equilibrium state with amplitudes of, for example, Ai and A 2, the corresponding energies of the foundation vibrations will equal
k

W, = -AS
where k

is

the coefficient of rigidity of

The energy absorbed by the

= ^A

>
2

soil.

per period of vibration equals

soil

AA
m)

kA
fcAiV
2

(^Y

Hence,

On

the other

hand (with general

reference to Chap. Ill), for free vibra-

tions of a foundation with one degree of freedom,

Ai

= A exp

j(~t)

l"^.(~x)

OF VIBRATIONS

EFFECTS

ON

c damping constant
T = period of vibrations
A = amplitude of vibrations

RESIDUAL SOIL SETTLEMENTS

67

where

without taking into account damping

properties of soilf

Therefore,

4,

exp (-2cT)

hence
Jn

Here

/ is the

Coefficient

is

4tT

natural frequency of vibrations of the foundation.

designated as the coefficient of damping of vibrations.

1.0
>

0.8

<u

0.6
*-*?

*s 0.4

OXJ

H=
H-

0.2

20

10

30
Water content

40
w,

50

60

70

percent

Fig. 11-14. Relationship between the coefficient of energy absorption of a clayey soil

and

its

moisture content.

The values
conditions.

of ^ were determined by I. A. Savchenko under laboratory

His data show that this coefficient is a comparatively stable

soil characteristic.

The experiments

established that the value of the coefficient of absorp-

both for sandy and clayey soils does not depend on the rate of load
application, or on the frequency of changes in the load, or on the
tion

\J/

maximum

stress in the hysteresis loop.

Contrary to existing opinions, it was established experimentally that

in soil undergoing shearing deformations the value of the moisture content
has little influence on the coefficient of energy absorption
of sand.
Investigations of natural clayey soils with some sand and silt content
showed that when the soil undergoes compressive deformations,
\j/

\f/

increases proportionally to the increase in the moisture content (see

Fig. 11-14).

The absolute value

of

\f/

is

smaller for compressive deformations than

for shearing deformations.

The

grain size of sand (Table II-l) has a considerable effect on the

coefficient of absorption: the latter increases

with increase in grain

size.

5=0

t These equations for Ai and A 2 are obtained from Eq. (III-1-18) by setting
(what corresponds to the excitement of free oscillations by a shock) and by setting the
time t equal to one-fourth and five-fourths of the period.

DYNAMICS OF BASES AND FOUNDATIONS

68

Apparently this can be explained by the growing dispersity and micrononhomogeneity of sand with an increase in grain size.
Table

II-l. Relationship between the Coefficient ^ of

Energy Absorption of Sand under Conditions

of Shear and the Grain Size

Grain size of
sand,

mm

\f/

0.10-0.25
0.25-0.50
1.00-2.00

0.64
0.68
0.79

noted that the phenomenon of hysteresis in soil develops

under vibration and under static conditions. This is illustrated by Fig. 11-15, which presents two

It should be

differently
150

hysteresis loops obtained for

A
V

-100

medium-

grained sand undergoing shear deforma-

Loop

tion.

was obtained

sence of vibrations, loop 2

-1

was subjected
zontal

50

co

Ax

155 sec -1

(in

the

mm

1.2

direction of

with frequency

Figure 11-15 shows that

the area of the hysteresis loop

0.2
Displacement, mm
0.1

0.3

smaller

the sand

to vibrations with a hori-

amplitude

shear) of

in the ab-

when

when sand

vibrates.

is

much

This

is

explained by the considerable decrease in

Fig.

11-15.

Hysteresis

loops

in

sand: (1) in the absence of vibrations; (2)

with vibrations.

the shearing resistance of vibrated sand.

For the above

coefficient of

test

conditions,

the

energy absorption of the

sand was decreased 25 per cent by the vibrations. The values of the
coefficient of absorption permit only the estimation of the influence of
certain properties of soil on its coefficient of damping of vibrations .
The true value of depends essentially on the transfer of the energy of
vibrations to the soil.
This transfer depends on the size of the foundation
area in contact with soil, the foundation weight, and the properties of the
Therefore the
soil as a medium in which elastic waves are propagated.
design values of the coefficient of damping of vibrations should be taken
from results of vibration observations on models or existing foundations
(see Chap. III).
11-2.

Effect of Vibrations

on the Porosity and Hydraulic Properties

of Soils

The

by the coefficient
and by pore pressures are basic physicomechanical char-

porosity and the hydraulic properties described

of permeability

acteristics of soils.

Therefore the study of the effect of vibrations on these

EFFECTS

properties

is

and erection

OF VIBRATIONS

RESIDUAL SOIL SETTLEMENTS

69

from the point of view of the design

dynamic loads, but also from the
vibrations for the improvement of engineering

of great interest not only

of foundations subjected to

point of view of the use of

properties of

ON

soils.

Experiments show that the principal vibration parameter which determines the effect of vibrations and shocks on the compaction of soils is the
acceleration, or rather the inertial force, which acts on the soil particles
during vibration.

The

inertial forces are proportional to the density of particles

forming

Therefore, other conditions remaining equal, sys-

a dispersed system.

tems containing particles characterized by high

specific

gravity will

0.9

*0.8
o

p0.7

<y
i

".5

^*~

0.6

0.5

0.4
1

Acceleration ratio

7)

Fig. 11-16. Vibratory consolidation curve of sand

fill.

undergo a more intensive compaction than systems built up of particles

with lower specific gravity. However, there is so little difference between
the values of specific gravity of various noncohesive soils that

considered that the compaction produced by vibrations

is

it

may

be

a function only

of the acceleration of the vibrations transmitted to the soil.

Once the dependence

of the void ratio

on the amplitude and frequency

graphs
showing a relationship between the void ratio and the acceleration of
vibrations.
Figure 11-16 presents such a graph plotted for sand.
On
the abscissa is plotted the ratio
between the acceleration of vibrations
and the acceleration of gravity.
Experimental investigations indicated that the function e = e{r\) is of
the same type for all soils.
In the initial state, all sands are characterized
by a value of the void ratio approaching the greatest limit value; however, when sands are subjected to vibrations, even to those characterized
by low acceleration, they undergo considerable compaction.
As the void ratio decreases, it becomes less susceptible to changes in
vibration acceleration.
For large accelerations, the relative void-ratio
change is small, and its value may be considered to approach the minimum.
of vibration is experimentally established, it is possible to plot

77

DYNAMICS OF BASES AND FOUNDATIONS

70

This kind of dependence of the void ratio on the acceleration of vibrais observed not only in sands, but also in other disperse systems
capable of undergoing vibratory compaction.
By analogy with the static consolidation curves, the graphs establishing

tions

a relationship between the void ratio and the acceleration of vibrations

may

named

be

"vibratory consolidation curves."

the

In highway

construction the dependence of the void ratio on the intensity of impacts

is

The graphs showing

sometimes determined.
'

called the

dependence

this

may

be

'impact consolidation curves."

Experimental investigations indicate the existence of a nonlinear rela-

and the acceleration

tionship between the void ratio

ratio

Therefore the slope of the curve

7]).

e{yi) is

of vibrations (or the

a variable depending on

e m in e min is the minimum limit value

approximation this relationship may be
considered linear; then the differential equation describing the relationship between e and the acceleration of vibrations may be written as follows
the values of

or the difference e

In the

of the void ratio.

first

- a (e -

g=

(II-2-1)

e min )

Here the minus sign means that with an increase

decreases; a

is

vibratory compaction;

if

t\

0,

then

the value de/drj

77,

and

e,

tan u

=
=

where u

in

the coefficient of proportionality, or the coefficient of

slope of curve
initial

value of void ratio

when

77

may be stated as follows:

then the value 1/a equals
the value of the acceleration of vibrations which induces the maximum
possible soil compaction, i.e., a void ratio equal to e m n
By integration Eq. (II-2-1), we obtain
The

if

physical significance of the coefficient 1/a

there

is

a linear relationship between e and

77,

where

is

dition: If

77

e min

+ C exp arj)
(

a constant of integration determined from the following con-

0,

then
e

Cmin

and C =

(6

min)

e min

Consequently

eX P (-<,)

(II-2-2)

Let us assume that prior to compaction by vibration the

loosest possible state, so that the void ratio
e

will

was the

soil

was

in its

largest possible,

e m a X the equation of the vibratory consolidation curve in this case

read as follows:
;

<?min

(e ma x

c min )

exp ( at])

(II-2-3)

If,

ON

OF VIBRATIONS

EFFECTS

RESIDUAL SOIL SETTLEMENTS

as a result of vibrations with the acceleration

will decrease to e,

77

71

the void ratio

770,

then

min

Substituting this value of

exp

em[n )

\yl

"l

(II-2-4)

0:770)

into the right-hand part of Eq. (II-2-2),

we obtain
e

e min

It follows that in the

(<?

m ax

e min )

exp

770

is

above equation

01(77

770)]

the value of the vibration

acceleration needed to bring the soil from the state of least compaction

by the void

(characterized
static

ratio e ma x) to the natural (initial) state of

compaction (characterized by the void ratio

>

e; e max

e).

1.2

1.0

O
0.8,

S0.6 {
c!

0.4

v\

0.2

>yO

uO

K
\

o
V.

10 12 14
w, percent

16

18

20 22

Fig. 11-17. Relationship between the coefficient of vibratory compaction a of sand

its

and

moisture content.

g min and a; while

depend on the soil
moisture content and are affected only by the grain-size distribution and
the size of the largest sand particles, the value of a depends also on the

Equation

(II-2-3) contains three parameters: e max

the parameters

e max

and

by

emin

definition do not

moisture content.
It follows

from the graph

of Fig. 11-17 that the moisture content has

appreciable effect on the coefficient of vibratory compaction a.

the moisture content of a sand

is

low

(less

an

When

than 5 per cent), a sharply

compact under the action

decreases; consequently the capacity of a soil to

of vibrations also decreases.
coefficient

of

When

the moisture content increases, the

vibratory compaction gradually grows,

and when the

moisture content attains some 16 to 17 per cent (corresponding to some

80 per cent of the complete water saturation of voids in sand) it reaches a

maximum value of a = 0.82 to 0.88.

When the moisture content becomes
the coefficient a decreases, and
it

when

equals approximately eight-tenths of

the

optimum moisture content

larger than

the sand
its

of the soil.

is

w =

16 to 17 per cent,

fully saturated

by water

largest value, corresponding to

DYNAMICS OF BASES AND FOUNDATIONS

72

The

existence of an

optimum moisture

at

which the greatest compaction

achieved (other conditions being equal) is characteristic not only for

sands, but also for other soils (for example, for clays with sand and for
clays with some sand and silt).
The foregoing results of investigations concerning the dependence of the
is

coefficient of vibratory

compaction a on the moisture content

of a soil

explain to some degree the considerable settlements of foundations bearing

dynamic loads when erected on water-saturated sands.

When

soil is

subjected to vibrations with a selected acceleration, the

process of compaction of the

soil does not occur instantly, but develops

during a certain period. As the time of action of the vibrations increases,
the void ratio decreases at a diminishing rate and after a certain time
reaches a constant value, regardless of further vibration.

Of great practical importance

the study of the question of the time

is

which is necessary for the

complete compaction of a soil.
Experiments show that dry sands, after having been vibrated for 30 sec,
do not reveal any changes in the void ratio. For cement, the corresponding time interval equals some 120 sec.
Investigation of the effect of vibration time on the compaction of moist
soils shows that complete compaction is not achieved even after the soil is
subjected to vibrations for 3 to 4 min.
If static stresses exceed the elastic limit, then residual deformations
appear.
In the same way, vibrations result in compaction if the acceleration exceeds a certain value which may be termed the "threshold
of vibratory compaction."
The existence of this value is confirmed
interval, corresponding to a given acceleration,

experimentally.
It is also established that

when a

soil is

subjected to vibrations with a

selected acceleration, the threshold of vibratory compaction

And

value equaling this acceleration.

body capable

vice versa,

if

raised to a

is

soil (or

any other

changing its void ratio under the action of

vibrations) has a void ratio e corresponding on the vibratory consolidation curve to the value r/ of the vibration acceleration, then vibrations
with an acceleration lower than 7] will not cause any changes in the void
disperse

of

ratio.

Therefore, the condition of the compaction of

vibrations

may

where

r)

is

soil

under the action of

be written in the form

>

(H-2-5)

Ho

the threshold of vibratory compaction,

i.e.,

the acceleration of

vibrations corresponding to the void ratio prior to vibrations.

is

determined from Eq. (II-2-4).

The concept "threshold of vibratory compaction" makes

it

Its value

possible to

EFFECTS

ON

OF VIBRATIONS

RESIDUAL SOIL SETTLEMENTS

determine the dimensions of the zone of

73

compaction caused by a

soil

selected vibrator.

Laboratory investigations showed that soil compaction under a given

regime of vibrations depends essentially on the magnitude of the load
If a sand already compressed in a consolidometer is
acting on the soil.
subjected to vibrations, the degree of further compaction will be relatively
Moreover, when a static load is applied to a sand, it will be com-

low.

pacted under the action of vibrations only when the acceleration of

vibrations exceeds a certain value, larger than the one needed for compacHence, the threshold of vibratory comtion if no static load is present.
paction depends not only on the initial value of the void ratio, but also
on the static normal stresses acting within the soil. The larger the magnitude of these stresses, the higher the threshold of vibratory compaction of
the

soil.

The

effect of static load

on

soil

compaction by vibration

is

attributed to

the fact that the forces of friction between particles increase with an
increase of pressure on the

Vibrations then cause a smaller change

In other words, under the action of pressure
the sand, so to say, acquires a greater cohesion and consequently does not
respond as much to vibratory compaction.
Static pressure may have an appreciable influence on the vibratory
in the density of the

compaction of

soil.

soil.

soil.

In particular,

may

it

increase in depth the effect of vibrations

When elastic waves are

that the acceleration of

produced in a

soil

soil

vibrations will

the point under consideration.

tions induced, for example,

on

be assumed that with an

soil

compaction

will decrease.

by a local source, it is evident

depend on the coordinates of

Therefore in a

soil

subjected to vibra-

by machinery foundations, the void

ratio will

not remain constant, but will undergo changes which will be governed by
changes in the acceleration within the soil, i.e., by the distance from the
foundation.
(x,y,z) has an acceleration defined by the
whose value depends on these coordinates (77, as before, is the ratio
between the acceleration of vibrations and the acceleration of gravity),
then the void ratio at this point will be determined by Eq. (II-2-2).
It follows from the definition of the threshold of vibratory compaction
that compaction will occur only at those points of the soil where the

If

ratio

a point with the coordinates

7],

inequality (II-2-5)

Beyond the zone

is satisfied.

of vibratory

and at the boundaries

compaction a reverse inequality

v(x,y,z)

Thus the boundary

is

valid,

of this zone

r?

of the zone of vibratory

compaction

of the soil repre-

sents a surface of equal acceleration of vibrations; hence, the determina-

DYNAMICS OF BASES AND FOUNDATIONS

74

tion of the zone of compaction

is

reduced to the computation of the

of accelerations of soil vibrations excited

by an

field

oscillating foundation.
0.40

0.05

12

Acceleration ratio

oo

o
o

Fig. 11-18. Relationship between the coefficient of permeability

and the acceleration

of

vibrations,

Experiments with water-saturated soils show that vibrations lead to

changes in the hydraulic state of a soil as determined by the coefficient of
permeability and the pore pressures in the
ground water.
Figure 11-18 shows two graphs of
changes in the coefficient of permeability
of sands depending on the intensity of
vibrations.
These graphs show that the

CO

24

20

coefficients of permeability increase pro16

portionally to the acceleration.

seen that the effect of vibrations

12

fine-grained

in

than

in

It is also
is

larger

coarse-grained

sands.

The

on the coefficient
be utilized for the
different solutions into a

effect of vibrations

of permeability

injection

soil,

of

may

for example, for soil stabilization

by

chemicals.
40

80

120

160

Time, sec

Fig. 11-19. Changes with time

dynamic hydraulic pressure

in
in

Experimental investigations performed

by N. N. Maslov, V. A. Florin, M. N.
Goldshteyn, and others show that the pore

pressure in water-saturated sands changes

under the influence of vibrations. This is
illustrated by Fig. 11-19, which clearly shows changes with time in the
pore pressure in sand subjected to vibrations with an acceleration of only
15 cm/sec 2
It is seen from this graph that at first the pressure grows,

vibrating sand.

EFFECTS

then

it

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

75

continuously decreases over a long period. When sand is subby shocks, a different pattern of

jected to impulsive vibrations caused

changes in pressure with time is observed; namely, the time interval

corresponding to the increase in pressure is much shorter than in the
The nature of the development of dynamic hydraulic
previous case.
pressures in a

soil

subjected to vibrations or shocks

is

insufficiently under-

There are reasons to assume that the development of

dynamic pressures in ground water leads to a decrease in the shearing
resistance of the water-saturated soil and consequently to the loss of
stood as yet.

stability of earth masses.

11-3.

Elements of the Theory of Residual Settlements of Foundations

Induced by Dynamic Loads

The theory

of the

development

of residual settlements of foundations

subjected not only to static but also to dynamic loads has not been

elaborated so far to a degree sufficient to serve as a basis for the estimation in each particular case of the value of the residual settlement.
is

due not only to the mathematical

This

involved in the theoretical

difficulties

analysis of a foundation subjected to the action of loads which vary with

is no sufficiently verified
have there been established any numerical
values, expressing soil properties, which would permit the determination
of the magnitude of the residual dynamic settlement.
In computations of settlements of structures under the action of static

time, but also to the fact that so far there

rheological soil model, nor

loads, soil

is

usually considered to be a linearly deformable, although

body.
Variations of settlements with time

inelastic,

may be taken into account, for

example, by the introduction of stresses which depend on the rate of
deformation.
Thus in the general case, the computation of settlements
based on the

of foundations subjected to the action of static loads only is

use of a model of a linearly deformable viscous

body whose

elastic limit

equals zero.
It is

case of

ment

hardly

justifiable,

dynamic

however, to use such a rheological model in the

loads, because, as

shown by experiments, the developunder

and dynamic

of residual settlements of foundations occurs intensively

conditions in which the total stress (imposed

by the

static

loads) exceeds the definition of the limit value.

When stresses are lower than the limit value,

settlements of foundations

subjected to vibrations and shocks will be of the same order of magnitude

by static loads. But since the static pressure

by foundations under machinery is comparatively low
(as a rule, it is lower than that imposed by foundations under buildings
and structures), settlements caused by this pressure will also be comas settlements caused

imposed on a

soil

DYNAMICS OF BASES AND FOUNDATIONS

76

paratively low.

may
The

However,

if

the total pressure on the

the process of development of settlements

soil

exceeds a limit

and they
reach considerable values, often up to several tens of centimeters.
settlements are nonuniform and may lead to an impermissible tilting

value

pi,

is

intensified,

of the foundation.

Thus the condition

tions

under machines

for the absence of residual settlements of founda-

may

be written in the form

Pst

Vdy

<

(II-3-1)

Pi

where p st and pd V are the static and dynamic pressures on the

the foundation.

soil

Since vibrations affect the internal friction and cohesion of a

consequently

its

under

soil

(and

resistance to local loads), the value of the limiting

Pdy

7
Pi

1S

Settlement

Fig. 11-20. Required static pressure de-

Fig. 11-21. Simplified settlement diagram

crease in the presence of vibrations.

in the presence of

a limiting pressure

pi.

pressure pi for a given soil will depend on the characteristics of the stress
cycle, i.e., on the relationship between p st and pa y
.

Analogously to the known relationship between the fatigue limit of a

material and the characteristic of the stress cycle, it may be assumed that
the relationship between the limiting values of the static and dynamic
stresses in the cycle will be determined by graphs similar to curve A B in
Fig. 11-20.
Designating by R st the maximum static pressure corresponding to the intensive development of settlements under the action of static
pressure only, and by R dy the maximum dynamic pressure corresponding
to the intensive development of settlements under the action of dynamic
pressure only, and straightening the curve A B, one can establish the
following relationship:

p st

where

<
=

(II-3-2)

qRst
1

- |^

(II-3-3)

tidy

q is the coefficient of required pressure decrease

on

soil in

the presence of

vibrations.

In accordance with the foregoing discussion, let us consider that as long

is lower than its limit value pi, settlement of the foundation will be elastic and, consequently, the foundation
subjected to the action of dynamic loads will undergo only elastic
as the normal pressure on the soil

vibrations.

EFFECTS

When

OF VIBRATIONS

ON

the normal pressure on a

RESIDUAL SOIL SETTLEMENTS

soil

reaches

pi,

77

there appear residual

settlements which increase under the action of this pressure, even

if its

value remains constant. Thus the relationship between the pressure

imposed on a soil by the foundation and the settlement of the foundation

be illustrated by a graph similar to the one presented in Fig. 11-21.

we take into account the soil reactions which depend on the rate of
settlement of the foundation, we arrive at a rheological model of the soil,
will

If

which corresponds to an elastoplastic viscous body.

Let us

first

consider the forced vertical vibrations of a foundation.

The amplitudes

of vibrations of machinery foundations usually lie within

such a small range of values that they undergo vibrations without separat-

ing from the

soil.

Then

the

maximum dynamic pressure on the soil equals

/c

Vdy

where

CuA z

1*,

a 1

l/V,

the amplitude of vertical vibrations of the foundation.

is

Substituting this expression for p dy into inequality (II-3-1) and solving

for the amplitude of forced vibrations, we come to the following condition
for the absence of residual settlements of the foundation:

A <

Pi

Pst

(II-3-4)

Cu

For a great majority of machines whose foundations were designed

more or less correctly, the actual values of vibration amplitudes are much
lower than the limiting values determined by the above inequality. This
explains the fact that foundations

under machines with a steady regime

of operation (turbogenerators, recip-

rocating engines, etc.) do not undergo

even
on water-saturated
fine-grained sands, which are especisignificant residual settlements,

when

erected

changes in properunder the action of vibrations.

ally susceptible to
ties

The determination
tlements

is

of

of residual set-

special

practical

interest in foundations undergoing

Fig. 11-22.

Damped

vibrations due to

sudden impact.

unsteady free vertical vibrations

induced by vertical impact or by waves propagated through soil. Amplitudes of such vibrations may attain a relatively large value and may therefore induce relatively large dynamic stresses.
A diagram of such vibrations will be similar to the curve presented in
Fig. 11-22.
Let us assume that section AB of the vibration diagram

DYNAMICS OF BASES AND FOUNDATIONS

78

corresponds only to elastic settlement of the foundation.

equation of the settlement may be written in the form:

zx

+ jnz H =

2cz,

differential

(II-3-5)

This equation differs from Eq. (III-1-17) only by the presence of a

constant in the right-hand part. This constant is the acceleration of
gravity and takes into account the effect of the weight of the foundation
and machinery. All symbols are the same as those in Art. III-l.
Let us assume that free vertical vibrations of the foundation are caused
by a centered impact. Under the action of this impact, the foundation
acquires an initial velocity in a downward direction equal to vo) its values
are determined by Eq. (V-2-8).
Initial conditions of the foundation movement may be written in the
form: If t = 0, then
z

and

z st

zx

(II-3-6)

where z st is elastic settlement of the foundation under the action of the

weight of foundation and machinery.
The solution of Eq. (II-3-5) with initial conditions expressed by Eqs.
(II-3-6) will be

Z\

where fn d
r
P

=
=
=

z st

pr)
j-5 exp
+ Jnd
(

sinr

(II-3-7)

frequency of natural vertical vibrations of foundation under

consideration of damping: fnd 2

relative time: r = fnd

fn2

C/Jnd

Multiplying both parts of Eq. (II-3-7) by the coefficient c of elastic

uniform compression of the soil under the foundation, we obtain the
following expression for the pressure on soil at the phase of the foundation
lt

movement under

consideration:

Pzi

Vst

exp (- P t) suit

j-

~\

(II-3-8)

foundation and machinery.

Let
As long as p z i > pi,
=
B
point
the
corresponds
which
to
us assume that at a certain time r
n,
on the vibration diagram (Fig. 11-22), the pressure on the soil will reach

p st

is

the pressure imposed

by the weight

of

the foundation settlements remain elastic.

the limiting val^e

pi.

Setting p zl

pi in

Eq. (II-3-8), we obtain the

equation
(Pi
c u Vq

Pst)!m

s/l

_ exp
2

from which we can determine the time

_ pr)

ginr

(II-3-9)

EFFECTS
If

ON

OF VIBRATIONS

the limiting pressure pi

is

RESIDUAL SOIL SETTLEMENTS

reached at the

maximum

79

settlement of the

foundation, then
dzi

dt

from which we obtain

tan

= -

ti

(II-3-10)

The

equation determines the condition of absence of foundation

last

settlements, in the case of foundation vibrations caused

by a

vertically

centered impact.
If the

ments

condition of Eq. (II-3-10)

not

is

satisfied,

of the foundation will develop; section

of Fig. 11-22 will correspond to these settlements.

soil

then residual

settle-

BC in the vibration diagram

Along this section, the
and therefore the

reactions will not depend on the settlement,

differential

equation of the foundation settlement will be

z2

= 0(1 -

2cz 2

where

(II-3-11)

Pi

The
r

must

solution of Eq. (II-3-11)

satisfy the following conditions: If

n, then

Zl

Zi

dzi

dz\

dr

dr

P
c

(II-3-12)

(II-3-13)

The

integral of Eq. (II-3-11) will be

Z2

-BT +

B = Bfl

where

C exp (-2pr)

^+&

+D

(II-3-14)

(II-3-15)

2c u

Using the conditions of Eqs. (II-3-12) and (II-3-13), we obtain two

C and D.

equations for the determination of the arbitrary constants

Solving these equations, we find

r
C =

The development
t

T2,

Dn = BDTl +.B 2+ S +.p

B +S
Q r)
exp (-2
P
.

of residual settlements will proceed

at which time

dz

1
dr

up

to the

moment

DYNAMICS OF BASES AND FOUNDATIONS

80

From

we obtain

this condition

the time r 2

the equation for the determination of

exp [-2p(r 2

= 2-i

n)]

(II-3-16)

Substituting the value r 2 thus obtained into Eq. (II-3-14),

the

maximum

total

elastic

(i.e.,

= - -

22, max

>

and

we obtain

residual) foundation settlement:

B(t 2

+ -

n)

(II-3-17)

move upward under the influence of

undergo an elastic rebound), and the
displacements of the foundation will be elastic only.
Let us consider the motion of the foundation as shown on the third
To simplify the computasection of the vibration diagram (Fig. 11-22).
If r

r2

then the foundation

the elastic reaction of the

will

soil (it will

displacements to the point C; we

time r from this point.
The differential equation of foundation motion corresponding to the

tions, let us transfer the initial point of

shall also read the

elastic

rebound

will

have the form

h +
The

+ fm

2ci 3

z,

g (1

-\

(II-3-18)

solution of this equation satisfying the conditions: If r

Zz

and

0,

then

-r(XT

will

be
z3

When

go
f

[exp (-pr) (cos r

cu

p sin r)

1]

(II-3-19)

the elastic displacement equals

= -

SB,.

^^

Thus the magnitude

(II-3-20)

cu

of the residual settlement only, caused

by one

impact, equals
z

z<i

2 3oo

7T
2p

( T2

(II-3-21)

Tl )

cu

Assuming
(III-3-10),

in Eq.

(II-3-7)

we may determine

= pi/c u and
maximum value

that z x
the

of vibratory foundation motion, at

Vi

it

satisfies

which no residual settlements

occur; the magnitude of this velocity equals

.*

that

_ggj
fg_

Pat

\C

Eq.

of the initial velocity

will

OF VIBRATIONS

EFFECTS

ON

RESIDUAL SOIL SETTLEMENTS

/^

= D.6
-

&-.

81

A..

*2

'/

Xs s

/6 <>

10

11

Fig. 11-23. Relationship between the residual settlement of a foundation

velocity v of

its

and the

initial

motion.

The value of coefficient j depends on p: if p = 0, then j = 1 if p = 0.3,

then; = 1.47; if p = 0.6, then = 1.84.
Using Eqs. (V-2-8) and (V-3-2), we may select the values of foundation
weight and contact area with soil at which no residual settlements of the
;

.7

foundation will occur.

It is not possible to establish directly from Eq. (II-3-21) the effect on
the residual settlements of static pressure
0.20
or the velocity of the initial movement of
0.18

However, by comput-

the foundation.

ing z for various specific conditions,

it is

//

0.16

possible to plot graphs showing the rela-

tionship between residual settlement

it

possible

to

V
0.14

and

-1

These graphs make

certain parameters.

estimate

0.12

the effects of

these parameters.

0.10

Figures 11-23 and 11-24 present such

graphs.
The curves in Fig. 11-23 show
the relationship between residual settle-

ments and initial velocity of foundation

motion for various values of the ratio r
between the static pressure p st and the
limiting pressure

Eq. (II-3-11)].
p
It follows from these curves that after
the initial velocity of the foundation
reaches the value v*, the residual settlements grow approximately in proportion
t

fc|:

0.0

0.06

0.04

0.02

[see

/
0.2

0.4

r=

0.6

0.8

1.0

Pst
Pi

to the initial veloctiy (especially for large

Fig. 11-24. Relationship between

the residual dynamic settlement of
a foundation and the static pres-

values of the coefficient

sure on the

p).

soil.

DYNAMICS OF BASES AND FOUNDATIONS

82

The graph

in Fig. 11-24

shows that qualitatively the

affects residual settlements in a

manner

static

pressure

similar to the initial velocity of

the foundation motion: as the pressure grows, after

some

''threshold"

value the settlements increase proportionally to pressure.

11-4.

Methods

to

Decrease the Residual Dynamic Settlement

of a Foundation

from the foregoing that the residual settlement of a foundaon the dynamic
pressure on the soil (which is proportional to the amplitude of foundation
vibrations), but also on static pressure.
The larger this pressure, the
larger the settlements.
Therefore for foundations undergoing shocks or
vibrations (such as foundations under machinery or under buildings subIt follows

tion undergoing shocks or vibrations depends not only

jected to the action of vibrations), the design should provide a smaller

on the soil than for foundations which support only static

from Eq. (II-3-3) that the coefficient of reduction of
permissible pressure on soil q, in cases where the foundation is subjected
to dynamic loads, depends on the value of the dynamic pressure and,
consequently, on the amplitude of vibrations.
The larger this amplitude,
the smaller should be the static pressure on the soil under the foundation
For example, in the design of foundations under machinery with impact
loads (such as hammers), the coefficient of reduction of permissible
This considerable required
pressures is usually taken as q 0.4 to 0.5.
decrease in the static pressure on the soil is explained by the fact that
the amplitudes of vibrations under hammers may attain relatively large
For foundations undervalues equaling in some cases 1.2 to 1.5 mm.
going vibrations with small amplitudes and not too high frequencies, the
static pressure

loads.

It follows

value of the coefficient of pressure reduction should be correspondingly

For example, under reciprocating engines vibrating at

amplitudes not exceeding 0.2 to 0.3 mm, the coefficient of pressure reduction q may be taken as unity.
Foundations under turbogenerators undergo vibrations at much smaller
amplitudes in the range of several microns or tens of microns. However, taking into account the sensitivity of these machines to nonuniform
settlements and the high frequency of their vibrations, q is assumed to be

increased.

0.8.

For foundations under structures subjected to the action of waves

propagated from various sources (including those under machinery), the
design should provide a reduced static pressure on the soil as compared
to its value

when only

be supported. Disregarding
shops and shops with
lead to considerable nonuniform settlements of

static loads are to

this factor (especially in the construction of forge

crushing equipment)

may

EFFECTS

OF VIBRATIONS

ON

RESIDUAL SOIL SETTLEMENTS

footings under columns or of foundations under

some

83

sections of walls,

may

undergo damage (cf. Art. IX-3).

In order to prevent considerable nonuniform settlements of foundations under structures subjected to the action of vibrations or shocks, it is
necessary (especially in cases where foundations are to be erected on
water-saturated fine-grained sands) to assign varying values of static
pressure on the soil; the pressure values under foundation sections located
close to the source of waves should be lower than those assigned under
as a result of which a structure

The static pressures thus selected

sections located at greater distances.
should be inversely proportional to the amplitude of vertical vibrations
taking place under the action of waves propagated through the soil.

bose^

N?

ID

furoj-.

20
Pile

10

570

>

base

1710

1140

Number

2280

2850

3420

of blows

Fig. 11-25. Relationship between foundation settlement

and number

of impacts.

zone in which dynamic foundation settlements may

Therefore the use of any
measures aimed at an increase in the limiting pressure pi, even at small
depths below the foundation contact area, will contribute to the decrease
of dynamic settlements due to shocks or vibrations.
If the soil under a
foundation consists of sands, then chemical stabilization or cementation
of the soil to a depth of 3 to 4
may be possible.

The

size of the soil

develop

is

not large either in depth or in plan.

may be achieved by any measand consequently the limiting value pi

decrease of the dynamic settlements

ures which increase the soil density

example, by driving short piles).

Figure 11-25 gives a curve of relationship between the settlements of

(for

two identical foundations and the number of impacts. One of these

foundations was erected on natural soil (medium-grained yellow watersaturated sand) and the other on five piles driven to a depth of some 3 m;
four piles were driven at the corners and the fifth in the center.

The

2
foundation contact area in both cases equaled 1.5
the foundation
weight was 5.7 tons. Natural vertical vibrations of the foundation were
;

DYNAMICS OF BASES AND FOUNDATIONS

84

induced by the impacts of a ram having a weight of 160 kg and dropped

from a height of 1 to 1.1 m.
Similar investigations conducted under different soil conditions showed
that a floating pile base had a very favorable effect on dynamic
settlements.

The values
blows on

of foundation settlements (in field experiments) after 1,500

piles,

Table

presented in Table II-2, show that in

II-2.

all

cases the rein-

Data on Decrease of Dynamic Settlements

by the Use of Piles
Foundation area
in contact with

Characteristics of the base

soil,

Loess in natural state

Loess reinforced by seven soil piles 4.5
long.
Loess reinforced by seven wooden piles 4.5
long
Medium-grained water-saturated yellow sand
of medium density
Medium-grained water-saturated yellow sand,

Settlement of the
foundation after
1,500 blows,

mm

2.0
2.0

3.G

2.0

1.0

45.7

1.6

9.0

1.5

19.0

1.5

0.5

8 9

reinforced

by

five

wooden

piles 3

m long.

Medium-grained water-saturated dense gray

sand
Medium-grained water-saturated dense gray
sand reinforced by four wooden piles 3 m long

forcement of the base under the foundation by means of short piles

resulted in a considerable decrease of settlements induced by vibrations.
When the foundation was erected on loess reinforced by wooden piles,
the settlement of the foundation under the action of vibrations was
reduced to one-ninth its value on natural soil. A foundation erected on
short wooden piles driven into dense gray sands had settlements equal to
approximately one-thirtieth of the settlement value on natural sand.
A decrease in residual settlements may also be achieved by decreasing
the amplitude of vertical vibrations by selecting rational dimensions for
the foundation.

THEORY OF VIBRATIONS OF
MASSIVE MACHINE FOUNDATIONS

lll-l.

Vertical Vibrations of Foundations

In general, the investigation of vibrations of a

a. Basic Assumptions.
massive foundation placed on the soil surface can be reduced to the
investigation of vibrations of a solid block resting on a semi-infinite elastic
To date no solution of this problem has been found. Therefore
solid.
several simplifying assumptions concerning vibrations of solid blocks
placed on soil are necessary.
First of all let us assume that there is a linear relation between the soil
reacting on a vibrating foundation and the displacement of this founda-

Then the relation between the displacements and the reactions will
be determined in terms of the coefficients of elastic uniform and nonunition.

form compression, as well as a coefficient of elastic shear. In Chap. I, the

dependence of these coefficients on the elastic properties of the soil and on
the size of the foundation was established; also, the numerical values of the
coefficients for various soils were given.
In addition, it is necessary to
assume that the soil underlying the foundation does not have inertial
properties, but only elastic properties as described by the coefficients.
Thus, the foundation is considered to have only inertial properties and to
lack elastic properties, while the soil is considered to have only elastic
properties and to lack properties of inertia.
These assumptions concerning foundation and soil make it possible, in the general case, to analyze
foundation vibrations as a problem of a solid body resting on weightless
springs, the latter serving as a model for the soil.
Frequently foundations under machinery are embedded into soil to a
In this case, the elastic reactions of the soil act not only
along the horizontal contact surface between soil and foundation, but also
along the side surfaces of the foundation.
These reactions may have
certain depth.

85

DYNAMICS OF BASES AND FOUNDATIONS

86

considerable effect on the frequencies of free vibration of the foundation

and on the

coefficient of

damping.

Therefore, reactions along the side

surfaces of the foundation have considerable effect on amplitudes of free

or forced vibration under conditions close to resonance.
It is difficult to evaluate in each case the effect of side reactions on
foundation vibration. This effect is tentatively taken into account in

by increasing the values of the coefficients of

For example, this method is applied in computafoundations for forge hammers. If a foundation undergoes

design computations
elasticity of the base.

tions of

(as, for example, foundations under reciprocating

machinery), and the design values of frequencies of natural vibrations

only forced vibrations

of this foundation are larger

than the operational frequency

of rotation of the machine, then the effect of the side reac-

tions

is

and can be neglected. In these

above soil reactions is conservative,
a design dynamic stability lower than the

relatively small

cases, disregarding the

since

it

results in

actual stability.
b.

Fig.

III-l.

Vibration of
a

centered

eslng
on a spring.

Vertical Vibrations of Foundations Neglecting the

Damp-

ing Effect of Soil Reactions.

Let us consider vibrations of
the foundation caused by a vertical exciting force P(t) which

changes with time.

We

assume that the center

of

mass

of

foe foundation and machine and the centroid of the area of

foundation in contact with

soil lie

on a

vertical line
,

which

coincides with the direction of action of the exciting force

In this case, the foundation will undergo only vertical vibrations.

Since the foundation is assumed to be an absolutely rigid body, its dis-

P(t).

placement is determined by the displacement of its center of gravity. As

mentioned above, weightless springs serve as a model for the soil. Thus,
the problem of vertical vibrations of a foundation is reduced to the
investigation of vibrations of a centered mass resting on a spring (Fig.
III-l).
Let us denote by z the vertical displacement of the foundation
computed with respect to the equilibrium position. We shall consider z
to be positive in a downward direction.
If the displacement of the center
of gravity of the foundation equals z, then the reaction of the spring (i.e.,
the foundation base) will equal

R= W +
where

W
cr

=
=

=
A =

(III-l-l)

weight of foundation and machine

coefficient of rigidity of the base
cr

cu

c rz

cu

uniform compression of soil

horizontal contact area of foundation with soil
coefficient of elastic

(III-l -2)

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

87

has the dimensions tons per cubic meter and A is expressed in

c r will evidently have the dimensions tons per meter.
Using d'Alembert's principle, we may obtain the differential equation of
According to this principle, the equation
vertical foundation vibrations.
of motion may be written in the same way as the equation of statics if one
adds the inertial force to the external forces acting on a moving body.
Then the equation of motion for the foundation will be
If c u

square meters, then

W + P(t)

-mi +
or,

using Eq. (III-l-l),

we obtain

mi
where

m=
g

mass

- R =

of foundation

c rz

P{t)

and machine:

(III-1-3)

m = W/g

acceleration of gravity

Dividing both parts of Eq. (III-1-3) by the mass m, we rewrite this

equation as follows:
z

+Uz =
2

where

pa)

/n2

(III-1-4)

p(t)

mmjA
&m
2l

(III-1-5)

Equation (III-1-4) describes the vertical vibrations of a foundation

under the action of an exciting force.
Let us consider the case in which no exciting force acts on the foundation, but the motion results from an impact or from an initial displacement of the foundation. Setting p(t) = in Eq. (TII-1-4), we obtain
i

+ frJz

(III-1-6)

This equation corresponds to the case in which the motion occurs only
under the action of the inertial forces of the foundation and the elastic
reaction of the base.
Such vibrations are called natural or free vibrations.
For example, foundations under forge hammers may be subjected to such
vibrations.

The

general solution of the

homogeneous

differential

Eq. (III-1-6)

may

be written as follows:
z

= A

sin/n2 Z

+B

cos fnz t

(III-1-7)

Hence it is seen that free vibration under the action of elastic reactions
and inertia forces is a harmonic motion with frequency fnz called the
,

"natural circular frequency of vertical vibrations of the foundation."

According to Eq. (III-1-5), this frequency is determined only by the

foundation mass and the elasticity of the base and does not depend on the
nature or condition of the exciting force.

DYNAMICS OF BASES AND FOUNDATIONS

88

Since the frequency of vibration

is

number

the

second, the period of the natural vibration,

tion, is related to the circular

i.e.,

of oscillations per

the time for one oscilla-

frequency/^ by the following equation:

T nz J-

2t
,

J nz

The numbers of oscillations per minute and per second

circular frequency of vibrations

are related to the

by the following simple formulas:

One oscillation per second is called a hertz.

The coefficients A and B in Eq. (III-1-7) represent the amplitudes
natural vibrations of the foundation.
initial

conditions of motion,

i.e.,

of

Their values depend only on the

on the magnitudes

displacement) of the foundation at a certain

of the velocity (or the

moment

of time taken as the

Natural vibrations of foundations under machines are
usually caused by an impact, i.e., the foundations experience a certain

initial

moment.

initial velocity.

Therefore

Let us assume that at

z

let

us consider only this particular case.

0,

and

(III-1-8)

Differentiating both parts of the solution (III-1-7) with respect to time,

we obtain

the following expression for the velocity of the foundation:

2

Setting

Afnz cos fnz t

and
and B:

in Eqs. (III-1-7)

expressions for constants

A=^

Bfnz

sin

(III-1-9),

fnz t

we obtain

(III-1-9)

the following

B =

Jnz

Thus, when vertical natural vibrations of the foundation are caused by

an impact, the displacement is determined by the equation
a

mnfj
r Jnz

(III-1-10)

While the frequency of natural vibrations of a foundation depends only

on the inertia and the elastic properties, the amplitude, i.e., the maximum
deflection from the equilibrium position, depends also on the initial conditions of the motion, being proportional to initial velocity.

Returning to Eq. (III-1-4) for forced vertical vibrations of foundations,

us consider the case in which the exciting force P(t) is a harmonic

let

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

function of the time, for example, P(t)

= p

sin at

(a> is

89

exciter frequency

and
p

where

An
is

is

P
=

the exciting force).

exciting force which changes with time according to sin ut or cos

cot

of special interest in the study of forced vibrations of foundations, since

in design work, exciting loads

functions of time.

imposed by machines are usually harmonic

Substituting into the right-hand part of Eq. (III-1-4)

the expression
p(t)

we obtain

sin

cot

the equation for forced vertical vibrations of foundations:

z

The

+ fm z =
2

sin

(III-l-ll)

oot

general solution of this differential equation presents the

sum

of

two solutions, corresponding to free and to forced vibrations caused by a

Due to the action of damping soil reactions, free

damped out a short time after the beginning of the forced
The
foundations, and there remain only forced vibrations.

given exciting force.

vibrations are

motion of

solution of Eq.
vibrations,

is

corresponding only to these steady-state

(III-l-ll),

as follows:
z

We

= A

sin

obtain the expression for the amplitude

substituting the formula for z

(III-l-ll); then

[Eq.

of forced vibrations

(III-1-12)]

by

into differential Eq.

we have

A = -jj-^m(fnz
s

The

(III-1-12)

a>t

=r
2

(III-1-13)

to )

solution (III-1-12) shows that the frequency of forced vibrations

is

Thus, unlike the frequency

of natural vibrations, the frequency of forced vibrations does not depend
on the inertial and elastic properties of the foundation and its base.
Since the exciting loads of machines are usually repeated periodically with
every revolution of the machine, in many cases the frequency of the
exciting force equals that of rotation of the machine.
In general, this conclusion is valid for all linear mechanical systems not
capable of producing self-excited vibrations. Therefore, the identity of
the frequency of forced vibrations and the frequency of exciting loads acting on the foundation holds so long as the relationship between elastic
foundation displacements and soil reactions is linear. Numerous comparisons of the frequency of forced vibrations of machine foundations
with the frequency of exciting forces developed by these machines confirm
equal to the frequency of the exciting force.

DYNAMICS OF BASES AND FOUNDATIONS

90

Thus

the coincidence of these frequencies.

it

is

clear that in forced

vibrations of a foundation, there really exists a linear relationship between

the foundation displacement

and the

soil reaction.

Figure III-2 presents graphs of the effect of the magnitude of the

on the amplitude of forced vertical vibrations of a test

These graphs substantiate the linear character of the
relationship established by Eq. (III-1-13).
It follows from the same equation that the amplitude of forced vibrations depends also on the mass of the foundation and the difference
between the frequencies of free and forced vibrations.
In order to better understand the influence of the mass and the natural
frequency of the foundation, we transform expression (III-1-13) into
exciting force

foundation.

A =
3

Since

mfm
where

st

is

mfn S

(III-1-14)
a,

//-

st

the displacement of the foundation under the action of force

0.30

the latter were applied stati-

if

cally, expression (III-1-14)

0.25

i/

0.20

be

rewritten as

0/

A* ,y

may

A = vA
z

st

(III-1-15)

0.15

\*

where

77

is

a dynamic modulus (or

magnification factor)

0.10
1

(III-1-16)
0.05

7?

and
200 400 600 800 10001200
Pz

co//nz is

The value

the frequency ratio.

of the

dynamic modu-

depends only on the interrelationship between the frequency of

lus
Fig. III-2. Relationship between the
amplitude of vertical vibrat ions A z and the
exciting force

the exciting force and the natural

frequency of the foundation.

frequency of the excited vibrations is small in comparison with
the natural frequency of the foundation, then the value of the dynamic
modulus is close to unity and the amplitude of forced vibrations of the
foundation does not differ much from A st the latter represents a static
displacement of the foundation under the action of the exciting force P.
With an increase in the frequency of the exciting force, also increases;
thus the denominator in expression (III-1-16) decreases, leading to an
increase in the value of the dynamic modulus.
When = 1, i.e., when
the frequency of the exciting force equals the natural frequency of the
If the

',

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

91

foundation, the amplitude of vibrations of the foundation theoretically

equals infinity.

This corresponds to resonance.

With further

increase

becomes larger than 1; the

dynamic modulus continuously decreases with an increase of , and when
For ranges of frequency
= y/2, the dynamic modulus again equals 1.
of exciting force corresponding to > \/2, the dynamic modulus uniformly decreases, asymptotically approaching zero. Hence it follows
that an exciting force, the frequency of which is large in comparison with
the natural frequency, may induce forced vibrations with an amplitude of
in the frequency of the exciting force,

infinitesimal value.

This conclusion

is

used as the guiding principle for

the design of various devices for insulation from vibrations, particularly

for insulating

machines and engines.

When

given, the design of a device for insulating a

be

made

device

is

in such a

way

the exciting frequency

is

machine or an engine should

that the frequency of natural vibrations of the

as small as possible

compared

to the exciting frequency (for

example, the frequency of oscillations caused by traffic).

Foundations under reciprocating machinery are usually designed in
such a way that the natural frequency of the foundation is higher than
If one increases the
the operating frequency of the machine; i.e., < 1.
foundation mass without changing the foundation area, the frequency of
It follows that
natural vibrations decreases and the value of increases.
the denominator in expression (III-1-16) decreases, causing an increase

dynamic modulus.

Thus, other conditions being equal, an increase

accompanied by an increase in the amplitude
This is the reason why modern foundations for
of forced vibrations.
machines (especially for reciprocating machinery) are designed as blocks
with large bases and minimum height.
Vertical Vibrations of Foundations Considering the Damping Effect of
c.
As mentioned in the foregoing discussion, under condiSoil Reactions.
tions of resonance the amplitude of forced vibrations theoretically
approaches infinity. However, this contradicts experimental data
which show that under conditions of resonance, the amplitude of vibrations still remains finite.
This contradiction between experience and
theory is explained by the fact that amplitudes of vibrations with a
frequency close to the natural frequency of the foundation are affected by
deviations of the mechanical properties of the soil from those of an ideal
elastic body.
Like any real body, soil deviates from an idealized model
represented by an ideally elastic solid.
In reality, irreversible processes,
characterized by energy dissipation, occur in soil.
This deviation of soil
properties from those of an ideally elastic solid may be taken into account
if one assumes that the reaction of the soil depends not only on the displacement of the foundation, but also upon its velocity. Since the
velocities of foundation vibrations are rather low, it can be taken as a
in the

in the foundation height

is

DYNAMICS OF BASES AND FOUNDATIONS

92
first

the

approximation that the damping reactions of

power

first

vibrations of foundations

may
z

soil

are proportional to

Then the equation

of the velocity of vibration.

of free

be written as follows:

+ f Jz =

2cz

(III-1-17)

This expression differs from Eq. (III-1-6) by the presence of the term
2cz.

Here
a
(in-i-i7o)

2m
is

damping constant

called the

double value equals the coefficient of

Usually c < /,. The solution

its

resistance a per unit of foundation mass.

of Eq. (III-1-17) corresponding to this case

= exp (~ct)(A

sin

fnd

is

as follows:

+B

cos fnl t)

(III-1-18)

where fn d is the natural frequency of vertical vibrations of foundations in

cases where the reaction of soil depends not only on the displacement, but
also on the velocity.
Substituting
solution

hand

(III-1-18)

part

we

(III-1-17),
will

Fig. III-3. Effect of

Hence

damping on vibration

curve 1, damping coefficient c is small; curve 2, c approximately

equals natural frequency / z curve 3, c is
larger than fnz
z

A and B

it

Eq.

any

if

fm 2

C5

follows that the

ing properties of a

left-

find that the solution

/ind

amplitude

the

into

differential

satisfy this equation for

values of

(time)

of

soil

damp-

decrease the

natural frequency of vibration of

the

foundation.

If

the damping

constant

is

small in comparison to

the natural frequency of the foundation, then the influence of the

properties of the soil on the natural frequency

damping

may

be neglected.
However, the effect of the damping reactions of soil on the amplitudes

of free vibrations of a foundation

small values of

c.

is

It follows directly

rather considerable, even in cases of

from Eq. (HI-1-18) that amplitudes

This is illustrated by

of vibrations decrease exponentially with time.

curve

of Fig. III-3.

== fm, then the character of free vibrations of foundations will

correspond to curve 2.
For large values of damping constant, when
c > fm, free vibrations are not possible, and, under the action of an
impact or an initial displacement, the foundation will undergo nonIf c

periodic motion, as

Thus damping

shown by curve

reactions of the

3 of Fig. III-3.
soil

have considerable

effect

on

free

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

93

vibrations of foundations and on amplitudes of forced vibrations, especially

We

under conditions close to resonance.

obtain an equation for forced vibrations of foundations, including

the effect of damping reactions of the soil, if we insert into the right-hand
part of Eq. (III-1-17) the value of the exciting force; as before, we assume

= (P/m)

the latter to equal p sin wt

z

The

2cz

sin ut;

+ frJz =

then we have
(III-1-19)

sin at

solution of this equation, corresponding only to steady forced

vibrations of foundations, will be

= A*

Here A*

is

sin (ut

(III-1-20)

y)

the amplitude of forced vibrations:

A* =

m Vifnz 2

CO )

(III-1-21)

4C 2 C0 2

The phase shift between the exciting force and the displacement
induced by this force equals
tan y

=
Jnz

Similarly, Eq. (III-1-21)

may

rj*

(III-1-22)

W2

be reduced to the form

A* = v*A
The dynamic modulus

st

in this case will equal

77*

(III-1-23)

A = ^~

where

Jnz

Figure III-4 presents curves of interrelationship between

77*

and

the

frequency of forced vibrations to the natural

frequency of the foundation. These graphs are plotted for different
magnitudes of A, proportional to the damping constant.
latter is the ratio of the

The

particular case

A =

corresponds to the previously discussed case

where the damping effects of soil reactions were

not considered. Only here will the amplitude at resonance increase
without limit. At all other times, when A^O, the amplitude remains
of foundation vibrations

finite.

The

larger the value of A, the smaller the amplitude.

conditions of damping, the

to

maximum

Under

value of the amplitude corresponds

DYNAMICS OF BASES AND FOUNDATIONS

94

Thus if damping occurs, the resonance frequency decreases somewhat,

and the dynamic modulus at resonance equals

2A

(III-1-24)

VI -

Hence it follows that the larger the damping constant, the smaller the
dynamic modulus at resonance.
5.0

A
A
A

4.5

= 0.0 5 C
= 0.07L

4.0

A=0
A =01150
A =01175

1L- A=0.i'0

3.5

A=0./5

3.0

if

7*2.5

-A

=0.20

-A=<0.25

2.0

1.5

1.0

0.5

0.5

1.5

1.0

2.0

Fig. III-4. Relationship between the dynamic modulus of

damped

of

damping

coefficient A.

increases,

rj*

and

The graphs presented in Fig. III-4 also show that

damping reactions of soil is observed in the zone of
is

vibrations

actual to natural frequencies <a/fng for varying values of the reduced

the ratio

value of

2.5

approximately equal to unity.

the influence of

forced vibrations decreases;

damping
when is

soil

When

the greatest effect of

resonance,

when

the difference

the

properties on amplitudes of

large in comparison to unity, this

may

be neglected. Since foundations under machines with a

steady regime of work are usually designed in such a way that there is a
significant difference between and unity, the effect of soil damping may

effect

be neglected in

many computations

of amplitudes of forced vibrations.

on Forced Vertical Vibrations of Foundations.
The foregoing theory of vertical vibrations of foundations is based on the
assumption that soil reactions may be represented by weightless springs
d.

The

Effect of Soil Inertia

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

characterized by the coefficient c u

the real properties of

This model

95

differs considerably

from

Therefore, the results obtained should be

soil.

considered as a first approximation only. As stated previously, an

accurate solution of the problem of vertical vibrations of foundations

on

resting

soil

necessitates a consideration of the problem of vibrations

of a solid resting

on an

which in the simplest case presents a

Limiting our analysis by considering only

elastic base,

semi-infinite elastic solid.

vertical forced vibrations

changes with time, we

+R

ml
where

R =
e

induced by the force P sin at which harmonically

write the equation for this case as follows:

may

exp

(tat)

= P

exp

[i(at

magnitude

of soil reaction against

phase

between exciting force and

shift

(III-1-25)

e)]

foundation
soil

reaction

In order to solve Eq. (III-1-25) it is necessary to determine the dependence of the value R upon the displacement, the characteristics of the
foundation, and the

soil

properties.

E. Reissner 37 gave an approximate solution of the problem of vibrations of a solid body resting on an elastic semi-infinite mass; for the
computation of R, he used the magnitude of the settlement of the soil
under the center of a uniformly loaded absolutely flexible circular area.
0. Ya. Shekhter 41 showed a mistake involved in Reissner's solutionf and
solved the same problem taking the magnitude of settlement of soil
(needed for the computation of the value R) as an arithmetic mean
between the magnitudes of settlement under the center of a flexible
circular area and under its edge.
As a result of rather complicated
computations which are omitted here, the following simple relationship
between R and z was established:
z

where

= ~

(/i

exp

if2)

(III-1-26)

(iat)

G = modulus of elasticity in shear of soil

= radius of a circle = \/A/ir
A = contact area between foundation and soil
/ = functions depending on ratio between radius
r

/1,

of circle

and

propagated by foundation under

machine, and also depending on Poisson's ratio of soil
The following formulas give values of /i and / 2 corresponding to a
Poisson ratio of v = 0.5, with a degree of precision sufficient for practical
computations:
length

of

shear waves

= -0.130 + 0.0536* - 0.0078* +

h = 0.0545Ji(1.047*)[l + Jo(1.047x)] +
/i

0.0474*

0.0065* 3

f This mistake was also noted in the paper by R. N. Arnold, G. N. Bycroft, and
G. B. Worburton in /. Appl. Mech., vol. 22, no. 3, pp. 391-400, 1955.

DYNAMICS OF BASES AND FOUNDATIONS

96

where %
L =
J Jo =

2irr

/L

length of shear waves propagated from foundation

Bessel's functions of first kind, of order one, zero

1}

The length

of the shear

waves

is

L =

Z7T

03

where v s is the velocity of shear waves in soil. For conventional designs

machinery foundations the value x is considerably smaller than unity.

of

^<3

0.28
0.24
0.20

~*il

jAy = o\

^^r^l
0.16

U[v =
0.12

0.08

0.04

\jV =

^
^
6
>P*
0.2

0.6

0.4

0,8

1.0

1.4

1.2

/^)

1.6

1.8

X
Fig. III-5. Auxiliary diagram for the solution of Eq. (III-1-26).

Figure III-5 presents graphs of }i and /2 depending on magnitudes of

the independent variable x; these graphs are plotted for Poisson's ratio
equal to

0,

0.25,

and

when determining
of x

and

of

0.5

and make

of Eq. (III-1-25),

(/i

From

this

possible to avoid computations

v.

Substituting the value of

hand part

it

values of /i and / 2 corresponding to selected values

if2)

exp

determined by Eq. (III-1-26) into the

we

(tut)

obtain:

+R

we obtain two equations

Gr

(t? +

exp

(iwt)

= P

for determining

R = P

sin

R = P

cos

exp

[i(o>t

R and

e)]

left-

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

we

Solving this system of equations,

tan

find

(mw 2 /Gro)f2

(raco

fi

x
/TTT (III-1-27)

TinTi
/0 )/i

V[l

(mco /(?r

97

)/i]

(III-1-28)
[(mco 2 /6V )/2 ] 2

expression thus found for R is substituted into the right-hand part

Eq. (III-1-26) by neglecting the imaginary part, we obtain the following formula for the amplitude a of forced vibrations of the foundation:

The

of

Gr Mil
\[1

(mco

/GV

+f:
2

)/i]

{{m^/Gn)}^

(III-1-29)

between the exciting force P and the displacement z

where tan a = /1//2 (phase shift between displacement and reaction of soil). Using Eq. (III-1-27), we obtain

The phase

equals

<p

shift

e,

tan

<p

+ Oco T77T-\m
/Oo)(/i +^
J.

/1

Let us introduce a dimensionless value

6

is

the

soil

density.

(III-1-30)

2\
2
)

-2L.

(III-1-31)

7r

where 7

4-

Then
WCO 2

Gr Q

and formulas

for the

amplitude and phase of vibrations

will

be rewritten

as follows:

aGr

-4
\

tan ^

= -

Ji

+/
+ W2)
2

(1

* 2 &/i) 2

jfi

^2
6% 2 (/i

+;

2
2

(IH-1-32)
(III-1-33)

Figure III-6 presents graphs of changing aGr /P depending on

changes of x, or, what is the same thing, changes in the frequency of
excitement.

These graphs have much in

common

with resonance

curves for a system with one degree of freedom subjected to damping.

Thus, although the

damping properties

initial

Eq. (III-1-25) does not take into account

of soil, amplitudes of vibrations never reach infinity

with changes in frequencies of excitement as is the case in an ideally

elastic system with one degree of freedom.
This means that even an
ideally elastic soil has a damping effect on the amplitude of foundation

DYNAMICS OF BASES AND FOUNDATIONS

vibrations.

This

foundation, due to

is

explained by the fact that the energy of a vibrating

its

propagation into the

soil, is

continuously dissipated

therefore the vibrations of a foundation, even of one resting on an ideal

elastic solid representing a semi-infinite elastic

mass, are

damped with

0.8

V- =
1

0.7

\^ b = 20
0.6

0.5

&Ko.4

I/)
0.3

--10

\^b = 5

h- P

>

6=

U^'C,

0.2

0.1

0.6

V- -M
|

0.5

b=20

^' >?-10 }

0.4

<~>

--5

^0.

^b

2
-b

0.2

0.1

0.3

^b = 20
^b=I0
!

v = V2

^b = 5
y

b =2

<^K = 0_

0.1

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

Fig. III-6. Auxiliary diagram for the solution of Eq. (III-1-32).

time.

Since,

other conditions being equal, amplitudes of maximum

it follows that the

vibrations depend upon the value of Poisson's ratio,

damping

of

foundation vibrations also depends thereon.

comparison
shows

of graphs in Fig. III-6, plotted for several values of Poisson's ratio,

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

99

an increase in dissipation of energy

that an
into
the
soil,
and hence to a result equivalent
foundation
from a vibrating
properties
of the soil.
damping
to an increase in the
of
vibrations
depend considerably on the
The fact that amplitudes
properties
of soil are determined not
damping
that
the
value of b shows
(inertia
elastic
and
properties)
characteristics
but also by the
only by its
the
foundation.
mass
of
size and
The interrelationship between resonance values aGr /P and the value
b is illustrated in Fig. III-7 by
0.8
dashed lines. It is seen that this
A
relationship is close to a linear one
increase in this ratio leads to

of the type

/J

On
P
where a r

is

The value

S^/r

pb

(III-1-34)

S/
//

the amplitude under

's

^,

^=-

0.2

b is determined by Eq.
Taking into account

fi^

15

&

^ J>^ -j^5g

that

m =

tv

conditions of resonance.
(III-1-31).

y^

0.6

20

[A
Fig. III-7.

Dependence

of

peak

(reso-

nance) values of Fig. III-6 on Poisson's

where
is the weight of the foundaand machine and A is the
contact area of the foundation with

ratio

v.

tion

soil,

Eq

(III-1-31)

may

be rewritten

as follows:

b.=
where y
p st

=
=

unit weight of

normal

7 J^L
y/A
Vst

(III-1-35)

soil

static pressure

Substituting this value of 6 into Eq. (III-1-34), we obtain for the

amplitude of vibrations under conditions of resonance (for P = 1)

ar

lor*

Ijl
GA* + Gy A

Pst

(III-1-36)

Since the unit weight of soils varies within a comparatively narrow

range, the value of y in Eq. (III-1-36)
equaling, for example, 1.70 tons/m 3

may

that the inertial characteristics of

soil

be considered to be constant,

Hence the conclusion is possible

have small effect on amplitudes

under conditions of resonance.

Equation (III-1-36) also shows that the resonance amplitude (reduced
to unit of exciting force) increases with an increase in normal static

DYNAMICS OF BASES AND FOUNDATIONS

100

pressure on the foundation base and decreases with an increase in the

foundation contact area. Thus, under otherwise equal conditions, the
damping of foundation vibrations by the soil decreases with an increase
With an increase in the
of the static pressure beneath the foundation.
foundation contact area, resonance amplitudes decrease, i.e., damping
Consequently, damping properties of soil depend not only on
increases.
its physicomechanical properties, but also on the foundation size and
Hence it is understandable why a damping constant established
mass.
on the basis of the resonance curve of forced vibrations or on the basis of

observation of natural vibrations of a foundation depends on the charac-

foundation itself.
In order to compare formulas for computations (III-1-32) and (III-1-33)
with the results of the theory of vibrations, let us rewrite expression
(III- 1-21) for the amplitude of vibrations of a foundation on a base of
zero inertia, considering the base to consist of weightless springs.
teristics of the

A* =

V(c m
r

where

t co

(III-1-37)

(aco) 2

denotes the total mass of foundation and

m =m+m
t

soil,

Assuming that the foundation contact area is a circular area with radius
we transform Eq. (III-1-37) by introducing x and b as variables.
Multiplying (III-1-37) by Gr Q we obtain

*Gro

1^

[(cr/Gu)

According to (III-1-5),

(m*coVGV

m= m

if

Cr

we

Wltjnz

CU

ff

"

Cs

(1

-y/A

(III-1-38)

(au/Gro)'

Further, using (1-1-9), (1-2-12), and (1-2-13),

Cr

have

will

)]

we obtain

V* Gr

c s is a coefficient which depends on the geometric form

foundation contact area; for a circular area it equals 1.083 if one
considers settlement as the arithmetic mean of settlements under the

Here, as in (1-2-14),
of the

center of the circular area and

its

edge.

s-V
2m,fm

We
i

denote
(III-1-39)

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

where a

refers to

Then

Eq. (III-l-17a).

aw V _ 4m
~
GrJ
GVo
(
2

Let m = 13m, where /?

due to participation of
t

is

101

/rtg

a>

8 Vtt

co

l-v

Gr

the coefficient of increase in foundation mass

soil.

~G^ =

Khf>

Consequently,

A*

~P =

V[3.84/(l

Taking the derivative

equating

it

to zero,

we

v)

6/3]

15.35x 2 fr 2 /V(l

of the right-hand side of Eq.

(III-1-40)
v)

(III-1-40)

and

find that resonance corresponds to the following

value of the independent variable x:

3.84(1

6(1

2 2 //3)
v)0

Let us assume that for selected values v and b, the value of the amplitude of vibration computed from Eq. (III-1-32) will attain its maximum
Assuming that the maximums of amplitudes computed from
at x = x
Eqs. (III-1-40) and (III-1-32) correspond to the same value of the
independent variable x, we obtain
.

3. 84(1

9(1

gjVg)

(HI-1-41)

y)0

Let us assume also that the maximum values of amplitudes of vibracomputed from Eqs. (III-1-40) and (III-1-32) coincide with each

tions

other.

Then we

V[3.84/(l

obtain:

v)

6/5]

15.35xo 2 6^/(l

Equations (III-1-40) and (III-1-42)

ar

(III-1-42)

v)

be transformed as follows:

~ (ki)
L2P/f
= 3.84(1
OXq (1 v)
1

may

(III-1-43)

Figure III-8 presents graphs of

different values of Poisson's ratio

and

(3

as functions of b for soils with

v.

If we take from these graphs values of the coefficients and jS corresponding to various values of b and v, and then, using Eq. (III-1-40),

DYNAMICS OF BASES AND FOUNDATIONS

102

plot resonance curves of forced vibrations,

we

shall see that these curves

coincide fairly well with graphs plotted on the basis of Eq. (III-1-32).

Such comparisons were made for b = 2.5, 10, and 20 and v = 0, 0.25,
and 0.50. The results of computations from Eq. (III-1-40) using values
of coefficients and /3 taken from Fig. III-8a and b coincide so well with
the results of computations from Eq. (III-1-32) that the two curves
merge completely.

Thus it is established that if numerical values of the reduced coefficient

damping and the coefficient of increase in mass are taken from the
graphs of Fig. III-8a and b, then the results of computations of amplitudes
of

0.70

1.70
\

0.60

1.60

\
\

0.50

1.50

0.40

1.40

>sa
^5_

0.30

1.30

0.20

1.20

0.10

1.10

10

\\

20

15

10

(a)

ib)

Fig. III-8. Auxiliary diagrams for the determination of the reduced

and the

coefficient of

mass increase

15

damping

20

coefficient

/3.

from the approximate formula (III-1-40) (the latter

obtained on the basis of an assumption that soil is represented by
weightless springs) coincide fairly well with the results of a more precise
of forced vibrations

theory which takes into account inertial properties of

An

analysis of Fig. III-8a

respect to the effect of

and

soil inertia

leads to

soil.

some general conclusions in

on foundation vibrations.

It follows

with a decrease of b the

directly from
coefficient of increase of mass (3 will increase hence the effect of soil inertia
on forced vertical vibrations of the foundation will grow.
According to Eq. (III-1-35) the value of b is proportional to the pressure
p 8t on the foundation contact area and inversely proportional to the
square root of the area.
Small values of b correspond to foundations with
Fig. 111-86 that, for

all values of

v,

small height and large contact area.

more strongly by the

siderable height and

Such foundations are influenced

than are foundations with con-

effect of soil inertia

relatively small contact area.

having the shape of thick slabs cause larger masses of

do foundations having the shape of high blocks.

Thus foundations
soil to

vibrate than

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

Numerical values

103

machine foundations attain 7 to 15. For

mass (3, even for v = 0.5, does
Hence, the soil mass participating in founda-

of b for

these values, the coefficient of increase of

not surpass the value 1.23.

tion vibrations does not exceed 23 per cent of the foundation mass.
Since
the period of natural vertical vibrations is inversely proportional to the
square root of mass, it follows that the results of calculations of the period
of natural vibrations of a foundation which include soil inertial effects are

not more than 10 per cent smaller than those which neglect soil inertial
Thus a correction accounting for the influence of soil inertia
properties.
not be significant. Errors in computations of natural frequencies or
amplitudes of forced foundation vibrations are usually not less than
10 to 15 per cent; therefore the effect of soil inertia on machinery foundawill

tions

may

be considered to be so slight that

it

may

be neglected in

many

engineering calculations.

The foregoing
tion resting

the

discussion on the effect of soil inertia refers to a founda-

on the

soil surface.

reacts not only

soil

Where

the foundation

on the horizontal foundation base

is

embedded and
but also on

area,

the foundation sides, there can be a considerable soil-inertia effect.

1 1

Rocking Vibrations of Foundations

1-2.

Let us consider vibrations of foundations due to the action of a rocking

moment which changes with time according to the function
sin ut and which lies in one of

external

the principal vertical planes of the

foundation

(Fig.

III-9).

It is

as-

sumed that the center of inertia of

the mass of the foundation and the
centroid of
lie

on a

its

horizontal base area

vertical line located in the

plane of the rocking moment.

Let us further assume that the

elastic resistance of the soil against

stmzsmzsi

^mm^^m.

y/Aww/i

sliding of the foundation is so large

in

it

dR=-cf l<pdA

comparison to the resistance of

the foundation against rocking that

may

be considered to be infinitely

Fig. III-9. Analysis of rocking motion

of a foundation.

great.

In this case, the motion induced by an external moment

sin ut will
be a rocking around the axis passing through the centroid of the area of
foundation in contact with soil, perpendicular to the plane of vibrations.

The

position of the foundation

the angle of rotation

the point

cp

is

determined by one independent variable

around the axis passing through

of the foundation

(Fig. III-9).

DYNAMICS OF BASES AND FOUNDATIONS

104

Let us assume that at a certain instant the foundation has rotated a

small angle

tp

around

The equation

this axis.

motion

of its

be

will

-W^ + ZMi =
W

where

= moment

mass

of inertia of

of

(III-2-1)

foundation and machine with

respect to axis of rotation

XMi = sum of all external moments with respect to same axis

In this case, the foundation weight and the soil reaction are external
forces.

The moment

Foundation Weight.

a.

of this force

W in respect to the

axis of rotation is

LW<p

where

is

the distance between the axis of rotation and the center of

gravity of the vibrating mass.

b.

Soil Reaction.

with

soil,

by the

An

soil

dA

element

located at a distance

c v is

is

acted upon

dA

of the elementary force

soil

dR with

compression.

respect to the axis of

is

dM
If it is

Cpl<p

the coefficient of elastic nonuniform

The moment
rotation

axis of rotation,

reaction

dR =
where

of the foundation area in contact

from the

==

ldR= cJLy dA

assumed that the foundation does not

then the total reactive

with soil is

moment

lose contact

with the

soil,

against the foundation area in contact

Mr = -c v <pj*dA = -c 9I<p

(III-2-2)

where / is the moment of inertia of the foundation area in contact with

with respect to the axis of rotation of the foundation.
By adding the exciting moment
sin cot to the two other moments, we

soil

obtain the equation of forced vibrations of the foundation:

W^lp +

WL<p

or

c^Iip

W,ip

By

equating

M to zero,

( C<P I

M sin
-

cot

WL)co

=
=

we obtain the equation

M sin

JTT 9

^
}

cot

of free rocking vibra-

tions with respect to the y axis:

W,lp

(cJ

WL)cp

(III-2-4)

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

The

solution of this equation

if

where

fn<p

is

= C
=

105

sin (fnv t

WL

r T

w)

(III-2-5)

-^r^
W

(III-2-6)

and jn

=
=

<p

c, cpo

natural frequency of rocking vibrations of foundation

arbitrary

motion

The

determined from

constants

initial

conditions

form similar to that for forced

but instead of P, m, and fnz the

solution of Eq. (III-2-3) will have a

vertical vibrations [Eq.

(III-1-13)],

Wo, and fnv should be inserted.

In the case under consideration, the following expression
obtained for the amplitude of vibrations:
values of

A =
Since the product

term

of

of foundation

may

WL

is

Wo(fnS

will

be

(III " 2 " 7)

0>)

usually small in comparison with c^I, that

be neglected in Eq. (III-2-6)

we then obtain

for the frequency

of natural rocking vibrations

/,

If the

and a

is

(HI-2-8)

foundation base area has a rectangular form with sides a and

the side perpendicular to the axis of rotation,

I
Jfnip

6a 3

6a 3 Cp

b,

we have

12

It follows from the latter formula that the length of the side of the
foundation area in contact with soil and perpendicular to the axis of rotation has considerable effect on the natural frequency of rocking vibrations
of the foundation.
Depending on the selected length, the natural fre-

quency

may

change considerably; hence the amplitude of forced vibraThe length of the other side of the foundation
area, i.e., the one parallel to the axis of rotation, does not much influence
the values of fnv and A v
This length is usually selected on the basis of
tions will also change.

design considerations.

The amplitude

of the vertical

component

of vibrations of the edge b of

DYNAMICS OF BASES AND FOUNDATIONS

106

the foundation area in contact with

soil is

A = -A

Ma
2W (U ~

A =

or

"2)

Rocking vibrations occur mostly in high foundations under machines

having unbalanced horizontal components of exciting forces and exciting
moments. For example, such vibrations
o
occur in foundations under sawmill log-sawing
These foundations are usually high
frames.
6.0
and project above the first floor of the sawmill.
o /

Figure 111-10 illustrates the distribution of

amplitudes of horizontal forced vibrations
along the height of the foundation under a log
frame.
It is seen from the graph that the

5.0

E
.4.0

o
o

centroid of the foundation area in contact with

-o

3.0

may be used for the

computation of the amplitude of forced vibrations of foundations under log frames induced
by the horizontal component of exciting forces
developed in these frames.
Therefore Eq. (III-2-7)

*-

o
^_

JZ

5
X

subjected to forced rocking vibrations.

soil is

2.0

1.0

Vibrations of Pure Shear

111-3.

lilt

0.2 0.4 0.6

Amplitude,

0.8

soil to compression is
comparison with the resistance to
shear, then displacement of the foundation
under the action of horizontal forces will occur
mainly in the direction of the action of hori-

the resistance of

If

1.0

large

mm

Fig. 111-10. Variation of the

horizontal amplitude of rock-

in

ing forced vibrations along

zontal exciting forces.

the height of a foundation.

Pt

force

sin ut acts

Let us assume that a horizontal exciting

on the foundation. The equations of forced and free

vibrations will be analogous to the equations of vertical vibrations of a

e.g., Eq. (III-l-ll), in which c T should be inserted instead
thus the equation of forced horizontal vibrations will be

foundation,
of c u

',

fnx

= Pt

sin

cot

(III-3-1)

where x is the horizontal displacement of the center of gravity of the

foundation and
e

Jnx

jnx

is

lA

the natural frequency of vibrations in shear

(III-3-2)

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

The

solution of Eq. (III-3-1)

is

(m -3 -3)

A* '
m(/J - g)
conclusions

All

107

and formulas obtained while considering

vertical

apply also to vibrations in shear.

In addition to the foregoing type of vibrations of foundations, characterized by horizontal displacement of the center of gravity of the
vibrations of foundations

vibrating mass, vibrations in shear

may have

a form of rotational vibra-

tions with respect to the vertical axis passing through the center of

gravity of the foundation and the centroid of the area of

Jz

\J/

sin

cot

Cp

its

base.

Letting

= moment of inertia of vibrating mass with respect to above axis

= polar moment of foundation base area
= angle of torsion of foundation
= exciting moment acting in horizontal plane
= coefficient of elastic nonuniform shear

we obtain the following equation of forced vibrations of a foundation

induced by an exciting moment:

Wrf

CfJ gf

= Mz

particular solution of this equation

may

where Jn ^

c^J z

/W

is

under consideration.

and Vertical Displacement

Equations of Vibration.

The foregoing

of vibrations of massive foundations in

by

(III-3-5)
CO*)

Vibrations of Foundations Accompanied by Simultaneous

Rotation, Sliding,

a.

be presented in the form

the square of the natural frequency of vibrations

of a foundation for the vibrations

111-4.

(III-3-4)

cot

Wz(U 2

sin

discussion concerned cases

which the

soil

was characterized

with respect either to compression or to shear. Let us

consider now the simplest case of vibrations of foundations, where the soil
is able to offer elastic resistance both to compression and to shear.
As
before, it is assumed that the center of gravity of the foundation and
machine and the centroid of the foundation base area are located on a
vertical line which lies in the main central plane of the foundation.
External exciting forces induced by the machine also lie in this plane if
(Fig. III-ll) is the center of gravity of the vibrating mass, then these
forces may be reduced to the exciting force P(t) applied at
and a couple
with moment M(t). Let us match the origin of a coordinate system with
infinite rigidity

DYNAMICS OF BASES AND FOUNDATIONS

108

the center of gravity of the foundation and machine mass at an instant

when the foundation is motionless; the direction of coordinate axes is

shown in Fig. III-ll.

Under the action of loads

P(t)

and M(t),

the foundation will undergo a two-dimen-

motion determined by the values of

sional

three independent parameters: the projec-

and z of displacement of the foundaon the coordinate

axes and the angle <p of rotation of the
foundation with respect to the y axis which
tions x

tion center of gravity

passes through the center of gravity of the

foundation and machine, perpendicular to

the plane of vibrations.
Fig.

Analysis

III-ll.

of

By

com-

projecting

all

foundation at time

bined types of foundation motion.

on the
on the x and z

forces acting
t

them the projections

we obtain, according to d'Alembert's

axes and adding to

on the same axes of the

inertia forces,

principle,

-mx +
-mz +
where

m =

ZXi =
2Zi

(III-4-1)

foundation mass

projections on x, z axes of all external forces acting on

foundation
The equation of the moments with respect to the y axis should also be
added to Eqs. (III-4-1).
Xi, Zi

-M m $ + XMi =
where

The
1.

Mm

the

is

moment

of inertia of the

mass with respect

following forces are acting on the foundation at time

Weight

on the x

W of foundation and machine.

axis equals zero, that

on the
Zi

2.

(III-4-2)

Soil reaction

of weight.

z axis

to the y axis.
t:

A projection of the force

equals

-W

caused by settlement of the foundation under the action

projection of this force on the x axis also equals zero;

the projection of this force on the z axis equals

Z 2

where

cu =
A =
=
z
8t

coefficient of elastic

c u Az s t

uniform compression of

area of foundation in contact with

elastic settlement

and machine

soil

soil

caused by action of weight of foundation

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

The

soil

tion

and

reaction
It

soil.

is

109

applied to the centroid of the contact area of founda-

produces a

moment with

respect to the y axis:

Mi = WL<p
where

the distance from the center of gravity of the mass to the

is

foundation base.
a given instant of time

If at

measured from the equilibrium

this

displacement

t,

the foundation has a displacement z

position, then the soil reaction induced

Z =

c u Az

3.

Horizontal reaction of elastic resistance of

the x axis

=
=

cT

Its projection

on

coefficient of elastic

is

Axo

uniform shear of

soil

displacement centroid of contact area of foundation

Xo

where x

soil.

is

Xi =
where

by

is

L<p

horizontal displacement of the

common

center of gravity of

foundation and machine.

Substituting this value of x Q into the expression for Xi,

Xi = c T A(x

The moment

we obtain

Lip)

of this force with respect to the y axis is

c T AL(x

Lip)

Reactive resistance of soil induced by rotation of foundation base

In order to compute the moment caused by this resistance, let us
single out an infinitely small element dA of the foundation area in con4.

area.

tact with

soil.

The

reaction

dR

of soil

dR =

c,pl<p

on

this

element

is

dA

= coefficient of elastic nonuniform compression of soil

= distance between area element dA and axis of rotation
The moment of this elementary reaction with respect to the y axis
where

c^

dM =
z

is

cJP<pdA

By integrating over the whole foundation area in contact with soil, we

obtain the total reactive moment of soil developed when the foundation
base contact area turns an angle <p.
This moment is

M
where I

is

the

moment

= -c v Iip

of inertia of

foundation contact area with respect

HO

DYNAMICS OF BASES AND FOUNDATIONS

to the axis passing

through the centroid of

this area, perpendicular to the

plane of vibrations.

To

(t)

these four forces there should be added the projections

of the external exciting force P(t)

moment

P x (t)

and

on the coordinate axes and the

M(t).

Substituting into Eqs. (III-4-1) and (III-4-2) the established values of

the projections of forces on the x and z axes, as well as the values of

moments with

respect to the y axis, we obtain, after some elementary

transformations, a system of three differential equations of forced vibrations of a foundation:

mz

mx

Mm

lp

CrALx

N. P. Pavliuk 32 was the

(cj

first

cT

Ax

- WL +

cu

Az = PJt)

c T AL<p

c T AL*)<p

= P x (t)
= M(t)

(III-4-3)
,

TTT
111

4"4

'

to give these equations of foundation

vibration.

Equations (III-4-4) are interdependent because each of them includes

Equation (III-4-3) in no way depends upon Eqs. (III-4-4).
<p.
Hence it follows that vertical vibrations of foundation occur independently of vibrations associated with the other two coordinates. If a
foundation is acted upon by exciting loads having no vertical components,
then no vertical vibrations of the foundation develop. In this case the
foundation will undergo rotation around the y axis and horizontal displacement in the direction of the x axis. If a foundation is acted upon by
an exciting load producing only a vertically centered force, then the
foundation will undergo only vertical vibrations.
In the same way, if the equilibrium of a foundation at a certain instant
of time is disturbed only by a vertical displacement of its center of gravity,
or if at this moment the foundation is given a velocity in the vertical
direction, then the foundation will undergo only vertical natural vibrations.
If the equilibrium of a foundation is disturbed by a displacement
x and

of its center of gravity in the horizontal direction or

if it is

given a velocity

no vertical vibrations appear. In this

case the foundation motion will be characterized by changes in two
coordinates: x and <p.
The same coordinates characterize the foundation
motion if its equilibrium is disturbed by changes in either x or <p.
The fact that vertical vibrations of foundations are independent of
vibrations in the directions x and <p gives us a chance to consider each
type of vibration separately. An investigation of vertical vibrations has
already been made in Art. III-l. Therefore it remains to investigate
vibrations corresponding to the system of Eqs. (III-4-4).
b. Free Vibrations.
If the equilibrium of a foundation is disturbed by
subjecting it at the initial instant of time to certain changes in the

in the horizontal direction, then

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

coordinates x and

and the

<p

and

velocities x

1 1

then during the time which

<p,

follows the foundation will be subjected to elastic soil reactions

and

inertial forces

undergo

will

The equations

free vibrations.

and

of these

vibrations are as follows:

mx

Mm -

cr

ip

ALx +

+ c Ax
T

- WL +

(c/

Particular solutions of these equations

where

= Aa

A a B a and

a)

<p

may

mT
^L-l-o)
.

be written in the form

= Ba

sin (fn t

a)

a are arbitrary constants.

sin (fn t

=
=

CjALip

c T AL*)<p

Substituting these solutions into (III-4-5) and reducing

eliminating sin (fn t

a),

we obtain two homogeneous

(c T

-c ALA a
T

(c v I

A - mfn

WL +

)A a

crAL*

cT

terms by

all

equations:

ALB a =

M m f *)B
n

mT
"^
^ ii D

The constants A a B a and

f should satisfy these equations if the particular solutions are to satisfy the system of differential equations of free
,

vibrations of the foundation.

System

(III-4-6) does not permit the determination of values of all

three constants

A a B a and/n
,

In order to do

the initial conditions of the foundation.

this, it is necessar}^ to

However,

we

if

A a and B a (that is, only the amplitudes

unknown, then we obtain from the first equation

Eqs. (III-4-6) only

are

A ~
Aa
Substituting this expression for

B a [-cM
If
it is

Ba

L*

(c/

cT

of vibrations)

CtAL
B
A - mfj a

Aa

- WL +

know

consider that in

into the second equation,

cr

AU

M m fJ)(c A

- mf n

we obtain
2

)]

does not equal zero, then, in order to satisfy the above equation,

necessary to assume that the factor in brackets equals zero.

we obtain
Mfn) 2 =

Then

the frequency equation

~CM L* +
2

(C VI

- WL +

Cr

AL* -

M m fn )(C A
2

- mfj) =
(III-4-7)

This equation contains only one unknown constant

quency

/,

the natural fre-

of vibrations of the foundation.

Let us transform Eq. (III-4-7) by opening brackets and grouping

members containing the same powers of /. Then we obtain a seconddegree equation for f n 2

DYNAMICS OF BASES AND FOUNDATIONS

112

After dividing

members

all

of the

mM m

new equation by

may

it

be

rewritten as follows:

Jn

f
V

- WL

cT I

cjL
"*"

rn

+ Mm\

L*m

Mm

'

Jn

c,I

T'

- WL

c^A

Mm

_
~

m0 the moment of inertia of the total vibrating mass

and machine) with respect to the axis passing through the
the base contact area and perpendicular to the plane of vibra-

Let us denote by
(the foundation

centroid of
tions; this

moment

equals

Mm0
Let

where

Mm + ml?

M^o
1

>

>

Substituting
it

0.

Mm =

yM m

/ cZ

- WL

into the equation for frequencies,

we

rewrite

as follows:
Jn

Mm

But according

to (III-2-6)
C *T

CrA\ fj
) t

cvI

^ m
and

~ WL ~

Mmo
is

CrA

(III-3-2) the expressions

Uf *

c
_iA

and

very large in comparison to

its

f
m ~ Ux

represent limiting frequencies of the foundation

to shear

- WL

yMm

when

the resistance of

soil

resistance to rotational vibra-

tions or vice versa.

Using these two expressions we obtain the

final

equation of frequencies

in the following form:

A(fn 2 )

This equation

will

fn"

f
Jn<P

4. UX
/ 2

fn

f 2f
J%
* J% *

(III-4-8)

have two positive roots fnl and / 2 corresponding to

the two principal natural frequencies of the foundation.

It can be proved that the natural frequencies which are the roots of
Eq. (III-4-8) have the following interrelationship with the limiting fre-

quencies fn(p and fnx'. the smaller of the two natural frequencies (for
example, /n2 ) is smaller than the smallest of the two limiting frequencies;
the larger natural frequency is always larger than fnv and fnx
In the case under consideration, involving a foundation with two
degrees of freedom, specific forms of vibrations correspond to the fre.

quencies fn i and/ 2 of the foundation; these vibrations are characterized

certain interrelationship between the amplitudes A a and B a which
depends on the foundation size and the soil properties, but does not

by a

depend on the

initial

conditions of foundation motion.

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

Let us determine from the

A a/B a

first

113

equation of system (III-4-6) the ratio

A-

Jnx

(III-4-9)

fn

If the

-LJ

foundation vibrates at the lower frequency fn2 then, according to

,

the above statement,

and p also is larger than zero; consequently, the amplitudes A a and B a

have the same sign. It means that during vibration at frequency
fn2, when the center of gravity deviates from the equilibrium position,
z
1
1
/

rj
i

-tr

-~4-

fVb

WMiW/

/-,

L
////\\

Pi

^-A

^^A^y///\V\V///W
VW//\\V^AW//W^
//A

y/Av

(A)

{a)

Fig. 111-12.

Two

types of foundation vibrations which correspond to Eq. (III-4-9).

for example, in the positive direction of the x axis, the rotation of the

foundation will be also positive, and changes of amplitudes A a and B a will

be in phase. The form of vibrations in this case will be analogous to that
shown in Fig. III-12a; i.e., the foundation will undergo rocking vibrations
with respect to a point situated at a distance p 2 from the center of gravity
of the foundation.

The value

of p 2 is

determined by the absolute value of

However, if a foundation
substituted for fn
vibrates at the higher frequency fnl then, since fnx 2 fm 2 < 0, p will be

expression (III-4-9)

if

f n2 is

and A a and B a will be 180 out of phase. Figure III-126 illustrates the form of vibrations corresponding to this case.
Here the
foundation vibrates around a point which lies higher than the center of
gravity and at a distance pi determined from expression (III-4-9) if /i is
negative,

substituted for fn
There is a simple relationship between p 2 and
.

P1P2

where
i is

V2

pi:

Mrr

the radius of gyration of the mass of foundation and machine.

DYNAMICS OF BASES AND FOUNDATIONS

114

main dimensions of a foundation which determine its mass, base

and moments of inertia are selected, then the limiting natural frequencies fnip and Jnx will depend only on the coefficients of elastic nonuniform compression and shear c^ and c T of the soil.
Often the exact values of these coefficients are not known and only the
range of the most probable values of c v and c T may be assumed. Then
the computation of the natural frequencies jn \ and/n2 should be performed
If

the

area,

for the

whole range of values of these

The natural

frequencies fn \

and/w2

coefficients.

of the foundation are

determined as

the roots of Eq. (III-4-8)

/m,2

= Zy

[fn/+U
2

V(W +U

47/V/n*

2
]

necessary to compute a range of possible values of frequencies

corresponding to the most probable values of elastic coefficients of the soil,
If it is

then

it is

more convenient
2

Z1.2

where

/z

is

to transform the latter expression as follows:

1

27

[1

+v

\/(l

m)

4 7 m]

the ratio of the squares of the limiting natural frequencies:

f
J nx

Jrup

If the dimensions of a foundation are selected, then the value f3 depends

only on the assumed values of the coefficients of elastic compression and

shear.

After the selection of a range of values for these coefficients,

easy to calculate

all

it is

possible values of frequency.

Figures III-13a and b present graphs of

as functions of

(3i t2

/x.

The

and /3 2 are plotted along the y axis, the values of ju along the
x axis.
Curves are plotted for different values of 7, from 7 = 0.4 (high
foundations) up to 7 = 0.9 (low foundations).
With these graphs it is easy to determine a possible range of changes
in fn i and /n2 using a given range of values of c? and c T
The frequencies
are determined from the formula
values of

/3i

/nl,2

/nv> /3l,2

depend not only on (3 lt2 but also onf n<P in order

to determine the smallest and the largest value of f n i and /n2 it is necessary to compute the minimum and maximum values of the right-hand part
of the above expression.
c. Forced Vibrations.
Returning to Eqs. (III-4-4), describing forced
Since the values of

ni

vibrations of foundations, let us consider separately several particular

cases of the action of exciting loads.

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

Assume that a horizontal

force of

magnitude

115

sin wt is applied at the

center of gravity of the foundation and machine.

This case is of the

Equations (III-4-4) (forced
vibrations of a foundation accompanied by changes with time in the
greatest interest in engineering practice.

ID.U

= o.4

y^0.5
7 = 0.6

10.0

y = 0.7
y = 0.8
y = 0.9

fiy1

5.0

3.0

2.0

1.0

4.0

5.0

(a)

y=0.9

1.00

0.80

0.60

0.40

0.20

Fig. 111-13. Variation of the coefficients

/8i

frequencies fn i and /re2 of the foundation

natural horizontal and rocking frequencies

and /3 2 (which govern the two main natural

shown in Fig. 111-12) with the ratio n of
and the ratio y of moments of inertia which

inversely reflects the height of a foundation.

<p and the horizontal component x of the coordinates of

the center of gravity) will be rewritten as follows:

angle of rotation

mx

Mjp -

CrALx

(cj

+ c Ax
T

- WL +

cT AL(p
CrAL^ip

= PT
=

sin ut

(III-4-10)

DYNAMICS OF BASES AND FOUNDATIONS

116

We shall

seek particular solutions of this system, corresponding only to

the forced vibrations of a foundation, in the form

x
<p

A x sin cot
= A v sin cot

Substituting these expressions for x and

<p

into Eqs. (III-4-10),

that selected particular solutions will satisfy this system

Ax

and

Av

if

we

find

the coefficients

are to be the roots of the following system of equations:

(c A mco )A x CrALAp = P T
- WL + cTAL - M m )A^ =
2

-CrALAx

Solving this system,

(c 9 I

we

co

find the following expressions for the amplitudes

of forced vibrations:

A(co 2 )

where
If

(III-4-11)

AL

cT

mMm (fm

<o

)(/n2

co )

moment

vibrations are caused by an exciting

Mi, the equations of

forced vibrations of the foundation will be

+ c A c AL<p =
- WL + c AL
=

mx

Mm

lp

CrALx

For the amplitudes

(c^I

of forced vibrations

AL

)<p

sin

cot

we obtain

,,

c
A x = -y^- Mi
T

A(

cr

A;

mco 2

A(co 2 )

(in-4-12)

MU

Changes in the frequency of exciting forces lead to changes in the

amplitudes of vibrations, even when magnitudes of exciting forces remain
the same. The phenomenon of resonance is observed when one of the
natural frequencies coincides with the frequency of exciting forces.
Since in the case under consideration, the foundation has two natural
frequencies, two resonances are possible when amplitudes grow rapidly.
Figure 111-14 illustrates the general character of resonance curves for
the forced vibrations of foundations under discussion; here an experimental resonance curve is plotted on the basis of data obtained by the

author during investigations of a test foundation with a 4-m 2 base area

in contact with

soil.

To every frequency

of forced vibrations of a foundation there corresponds a particular form of vibrations which is characterized by the
magnitude and sign of the radius vector p connecting the center of

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

gravity of the foundation and the point

which the foundation rotates.

mined from the equation
p

ence of the form of vibrations upon

changes in the frequency of the excit-

t0.3

rfe^ L

Jnx

i.e.,

\)
V

i}

then according to (III-4-12),

p

>

0.1

(in-4-14)
*

CO

4
co

is

It

subjected only to

moment

component

0.2
is

the action of the exciting

fnx,

\
o

ing force.

When

\
Horizontal

0.4

a foundation

around

b),

of p are deter-

The character of the exciting loads

causing forced vibrations of a foundation also has an effect on the depend-

If

and

(Fig. III-12a

The magnitude and sign

(III-4-13)

117

small in comparison with

Number

P does not differ much from L,

at low exciting frequencies, the

16
second (hertz)

12

of oscillations per

Fig. 111-14. General character of resonance curves corresponding to Eq.

foundation will vibrate with respect to

(III-4-12).
the axis passing through the center of
gravity of the base contact area, perpendicular to the plane of vibrations.
With an increase in exciting frequency, the denominator of (III-4-14) will
decrease rapidly and p will grow,

F"
L
i

0)

l^-r

co

fnx,

large.
will

!/

then p will be indefinitely

In this case the foundation

undergo

only

vibrations

shear (sliding) with a certain

!/

With a further

plitude.

of

am-

increase

in the exciting frequency, p changes

Fig. 111-15. General character of changes

in the radius vector p with the frequency

of vibrations

the foundation vibrations will

be accompanied not only by changes
in p, but also by changes in x^; in
other words, the foundation contact
If
sliding.
undergo
area
will
i.e.,

l\

its

sign with an increase in

;

tinuously

co.

zero as

co

decreases,

becomes

co,

p con-

approaching

infinitely large.

This means that at exciting frequencies considerably larger than the

limiting frequency fnx a foundation will principally undergo rotation
around the axis passing through its center of gravity. Figure 111-15
,

illustrates the general character of

changes in

p,

depending on

co.

<

DYNAMICS OF BASES AND FOUNDATIONS

118

main vertical axis versus frequency

computed from Eq. (III-4-14) on the
Circles
basis of an experimental value of fnx and a selected value of w.
plotted on the same figure show amplitudes of the horizontal component
of vibrations corresponding to some magnitudes of exciting frequency.
These amplitudes were measured at various foundation heights. The
Figure

o),

1 1 1- 16

gives position of the

plotted for a test foundation and

experimental points agree well with values established on the basis of the
foregoing theory.

The Effect on the Natural Frequencies of Eccentric Distribution of the

Foundation and Machine Mass. An eccentric distribution of the machine
d.

->

T"

"
i

/
\ /

10

\i

\/

15.7

23

/
\

\
\f

\\

25

35

'

/
/

108.7

120

\
\

130

1 \
1

135

1 \\

160

170

main

\
\

\ 1

Fig. 111-16. Variations in the position of the

changes in the frequency

\l\

\ 1

108.5

90

\
I

78

50.7

/
/

1?

(?

rP-rQ-r-r

-o
1

180

vertical axis (Fig. 111-12) with

to.

mass may occur when a machine and a generator or a motor are coupled
on the same shaft. Sometimes an eccentricity in the mass distribution
is caused by asymmetry of the foundation resulting from various cavities,
channels, etc.
The asymmetry can often be eliminated by moving the
Sometimes this
centroid of the foundation area in contact with the soil.
cannot be done; then foundation vibrations should be computed with
the asymmetric distribution of mass taken into account.
Let us consider the simplest case of asymmetry of foundation mass,
that in which the center of gravity of the foundation and machine mass
and the centroid of the foundation contact area lie in one of the main
foundation planes, but not on the same vertical line. We shall investigate
foundation vibrations in the main plane, in which both the centers of
gravity

lie.

The foundation motion

projections x and

again determined by three parameters: the

of the displacement of the foundation center of
is

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

gravity on the corresponding coordinate axes and angle

ip

119

of rotation of

the foundation around the y axis, passing through the center of gravity
and perpendicular to the plane of vibrations.

Let us assume that foundation motion has been caused by an initial

disturbance (for example, an impact) and examine the forces acting on
the foundation during its motion. Then we shall set up the differential
equations for this motion taking into consideration the soil reactions
produced by foundation displacements only.

As before, we match the origin of the coordinate system used for the
study of the foundation with the center of gravity, when the foundation
is

at rest.

We assume that at a certain instant of time, the projections of the

displacement of the center of gravity of the foundation will equal x and
z, and the projection of the rotation vector on the y axis will equal <p.
We measure these values from the equilibrium position of the foundation
when it is subjected to the action of weight and to static soil reactions.
The

horizontal displacement of the centroid of the foundation area in

contact with

soil

equals

xq

where L, as before,

is

L(p

a distance between the center of gravity of the

foundation and the foundation base; hence, the soil reaction caused by
horizontal displacement of the centroid of the contact area equals

R x = c A(x

Lip)

The

vertical

component

displacement of the centroid of the

of the

contact area equals

Zq

where

The

tip

the eccentricity of the foundation and machine mass.

e is

soil

reaction caused

by

this displacement of the centroid is

R = c A(z
3

Let us compute the moments of

The moment
The moment

all

e<p)

forces with respect to the y axis.

of the force of gravity equals zero.

of the soil reaction

caused by the static action of weight

equals

M
where

The

is

= WLip

the total weight of the foundation and machine.

horizontal

component

M
The moment due

of the soil reaction

c T A(x

to the vertical

moment

L<p)L

component

c u A(z

produces the

eip)e

of the soil reaction is

DYNAMICS OF BASES AND FOUNDATIONS

120

The

reactive

by the angle

<p

moment produced by

M
where I

is

the

due to rotation

soil

of the

foundation

equals

moment

= -c v I<p

of inertia of the foundation contact area with

respect to the axis passing through

centroid, perpendicular to the plane

its

of vibrations.

moments into Eqs.

we obtain the following differential equations for
distribution of masses now under consideration:

Substituting the above expressions for forces and

(III-4-1)

and

the eccentric

(III-4-2),

mx
m'z

Mm

lp

cT

ALx

WL +

{c^I

c u Ae

+ c Ax c AL(p
+ c Az c u Ae<p =
T

+ c AL

)<p

c u Aez

=
(III-4-15)

unlike the previous

vertical
here the

Equations (III-4-15) show that

case, in

which the

center of inertia of mass and the center of gravity of the foundation

contact area lay on the same

line

equations of motion are interrelated.

Therefore,

if

three differential

at the initial

moment

a change took place in only one parameter related to the motion, then
as a consequence there would be changes in all three parameters deter-

mining the position of the foundation. Thus, if at the initial instant the
foundation is subjected to the action of a disturbance inducing only a
horizontal displacement of its center of gravity, then it will move not
only in this direction, but also in the vertical direction, and will undergo
rotational vibrations around the y axis as well.
If the asymmetry in the distribution of masses is very small (e
0),
then Eqs. (III-4-15) at this limit value of e are transformed into the
systems (III-4-3) and (III-4-4).
We shall proceed as before in order to obtain the frequency equation for
the asymmetrical case under consideration.
First of all, let us simplify the differential equations of motion by denoting

A
c uA =
L )A =
cT

c<J

- WL +

{c u e

cr

Inserting these notations into Eqs. (III-4-15),

m'z

We

will

(III-4-16)

z e<p

then seek a particular solution of Eqs. (III-4-16) in the form

x

where

<p

we obtain

+ c x c L<p =
+ cz c =
c Lx + C(p c ez =
mx

Mm

cx

cz

= Aa

sin

A a B a and C a
,

jn t

= Ba

sin/n

are arbitrary constants.

<p

= Ca

sin

/J

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

Substituting these solutions into Eqs.

121

we obtain

(III-4-16),

three

homogeneous equations:

- mfn *)A a ~ Cx LC a =
- mfn *)Ba - c eC a =
- M mfJ)C a - c x LA a - c eB a =
(C X

(cx

(c

The

A a B a Ca

constants

(III-4-17)

and

/ should satisfy these equations

selected particular solutions are to satisfy the system

our

if

(III-4-16)

of

Equations (III-4-17) are linear and homogeneous.

In order for these equations to have solutions other than zero for A a B a
and C a their determinant should be identically reduced to zero
differential equations.

(c x

- mfn *)(c - m/n

M mf

)(c

n *)

If

(c x

(c z

mfj)c s 2 e*

- mfn *)c x

L*

(III-4-18)

the eccentricity in distribution of mass equals zero, then this equation

reduced to the equation

of frequencies is

fe

Thus, when

mfj)[(c x

- m/n

M m/

)(ci

2
c x IJ]

0,

f 2
Jn

_
~

A(/n 2 )

where

C\

_ 2l
- m

f 2
Jnz

- m/n )( Cl - M m fJ) c^I WL + c L A

2

(c x

cJTJ

(III-4-19)

Equation (III-4-19)

is

identical with Eq.

Denoting the

(III-4-8).

natural frequencies of vibrations of the foundation which correspond to

the limiting case

may

by f ni and /n2 and assuming that jn \

,

>

fn2,

we

rewrite Eq. (III-4-19) in the form

A(/; 2 )

On

(fn

fnl )(fn

/n 2

2
)

(III-4-20)

the basis of the general characteristics of the interrelationships

systems with a limited number of degrees of

dependence exists between the
and / 8 corresponding to e ^ 0, and the frequencies

between frequencies
freedom, we

may

when

of

state that the following

frequencies fnh fn2

fm, fni

mM m

0.

fnZ

<

fn2

<

fn2

<

We rewrite Eq. (III-4-18) in the form

(cx - mfn *)(c - m/n )(ci + c,e - M mfn
2

< fnl

Jul

(c x

- ?n/n )c e
- fe - mfJ)c x *L =
2

Dividing by m, we obtain

t(U 2

/ )(/nl

/n )(/n2

~ fn*) ~

(/na:

/, )/n

/n ,

DYNAMICS OF BASES AND FOUNDATIONS

122

from which we

find
e

/n

where

:(y-/n

-/n 2 )(/n2 -/n 2

2W
(f 2_f
\Jnx
J n JJnz
2

)(/nl

(III-4-21)

The right-hand part of Eq. (III-4-21) does not depend on e. Let us

draw a graph of this member, taking fn 2 as an independent variable
(Fig. 111-17).
This graph will be formed by two separate curves with an
asymptote

corresponding

value fn 2 = fv
Branch
the x axis at point fn 2 = /n2 2

B
fn

the

meets

Branch

the same axis at point

crosses
2

to

= fS and

111-17

that

is

point fn 2 = fnz 2 (Fig.

plotted on the assumption

fm >

Straight line

fni)-

responds to the

left

The

(III-4-21).

member

cor-

of

Eq.

abscissas of points

at which the curves cross this straight

line give the

unknown

roots

n3 ,/2

2
,

and fm of Eq. (III-4-21). The

Fig. 111-17. Graph illustrating Eq.
graph clearly illustrates the effect of
(III-4-21).
eccentricity e on the natural frequencies of foundation vibration.
It is seen that in the case of an
eccentric distribution of mass, the two smaller frequencies /n3 and fn i
become somewhat lower, while the largest frequency becomes higher.
If the eccentricity e is small, the frequencies Jn3 fn2 and fn \ do not differ
much from n i,/,l2 &ndfnz and at the limit, when e = 0, they become equal.
For foundations having a relatively small eccentricity, say 5 per cent of
2

the length of a side of the foundation contact area,

neglected and computations
111-5.

a.

may

its effect

may be
e = 0.

be based on formulas derived for

Experimental Investigations of Vibrations of Massive Foundations

Verification of the

Theory of Vertical Vibrations.

The

investigations of vibrations of foundations were performed

first

field

by the author

together with A. Mikhalchuk 3 on a porous water-saturated silty clay with

some sand. On the site of investigations the clay had a thickness of

m and

was underlaid by a sand bed having a thickness of about 4 m.

rested on a thick layer of clay.
The ground- water level was
20 to 30 cm higher than the base of the test foundations, which were all
placed in the same excavation at a depth of about 2.5 m. For dynamic
investigations three test foundations were employed with areas in contact
with soil of 2.0, 4.0, and 8 m 2 and weights up to 30 tons.
In addition to dynamic investigations of forced and free vibrations,
4.5

The sand

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

123

were also performed in order to determine the

static investigations

cient of elastic uniform compression of

coeffi-

soil.

Figure 111-18 presents resonance curves of vertical vibrations of a

2
for different eccenfoundation with an area in contact with soil of 8
tricities e of unbalanced mass of a vibromachine.
Analogous curves were
obtained for other foundations. Table III-l presents data on computed

0.9

0.8

0.7

0.7

*-

=z 15 rnm

0.6

0.6

0.5

/T0.5

E 0.4

<

w V v>

s>

0.3

5rr

0.3

>.5n im

0.2

0.2

,
I
[

*
7

0.1

0.1

^ 4r
4

Number

12

16

of oscillations per sec

20

Number

(hertz)

12

16

of oscillations per sec (hertz)

Fig. 111-18. Resonance curves of vertical

Fig. 111-19. Resonance curve of 1.5-m 2

vibrations for three different eccentricities

test of

of

Table

III-2.

unbalanced mass of a vibromachine.

and experimentally established values of natural frequencies of foundation

vibrations.
The computation was performed on the basis of values of
c u established by static investigations.
The author and Ya. N. Smolikov performed analogous investigations
on a water-saturated soft gray clay with an admixture of organic silt.
Investigations were performed on test foundations with base areas of
0.5, 1.0, and 1.5 m 2
.

Figure 111-19 presents one of the resonance curves of forced vibrations

2
obtained for a foundation with a contact base area of 1.5
In addition

DYNAMICS OF BASES AND FOUNDATIONS

124

Table III-l. Comparison of Natural Frequencies Computed and

Observed during Tests on Water-saturated Clay with a
4-m-thick Sand Layer at 4.5 m Depth
Frequency

of natural vertical

vibrations, sec -1

Mass

Foundation
contact

area,

of

cu

from

static

system,
tons X sec 2 /m

investigations,

1.66
1.92
3.05

4.40
2.45
2.05

kg/cm 3

Computed

Established from
observations of
forced vibrations

72.8
71.4
73.2

88.0
60.0
69.0

Table III-2. Comparison of Natural Frequencies Computed and

Observed during Tests on Soft Saturated Silty Clay
Frequency of natural vertical
vibrations, sec

Mass

Foundation

cu

system,

contact area,

of

tons

from

investigations,

sec 2 /m

-1

static

Established from

kg /cm 3

Computed

observations of
forced vibrations

0.332
0.520
0.685

0.5
1.0
1.5

Table

III-3.

3.5
2.52

72.5
69.5
67.8

2.1

72.8
69.0
70.2

Comparison of Natural Frequencies Computed and

Observed during Tests on Loess
Frequency

of natural vertical

vibrations, sec"
cu

Foundation
contact
area,

Mass

of

static

system,
tons

sec

investi2

/m

gations,

kg/cm 3

0.81
1.40
2.00
4.00

0.44
1.08
1.10
1.74

_1

from

14.2
10.8
10.3
8.2

Computed

Established

Established

from obser-

from obser-

vations of

vations of

natural

forced

vibrations

vibrations

162

158

159

118

113

107

137
137

117
118

117
121

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

to (fynamic investigations, the coefficient c u

was

determined by the

also

Results are presented in Table III-2.

method.

static

125

Similar investigations were performed by the author, Ya. N. Smolikov,

and P. A. Saichev on

loams and on loess.

from static investigations and
vertical vibrations established by two different

loessial

Table III-3 gives values

natural frequencies of

of c u secured

dynamic methods.
Table III-4 gives results

of similar investigations

performed on water-

saturated gray fine dense sands containing, in places, peat and organic
The first two foundations listed in Table III-4 were placed on sand
silt.

with an admixture of peat and organic

on pure sand.

Table

silt;

the remaining three rested

Comparison of Natural Frequencies Computed and

Observed during Tests on Fine Saturated Sands

III-4.

Frequency

of natural vertical

vibrations, sec
cu

Foundation

area,

Mass

of the

system,

contact
2

tons

sec

static

investi2

/m

gations,

kg/cm 3

0.38
0.84
0.84
2.76
3.60

1.0
4.0
4.0
8.0
15.0

from

3.96
4.45
7.54
5.55
4.00

Computed

102

145
189

126
127

Established

Established

from obser-

from obser-

vations of
natural
vibrations

vations of
forced

103
136
155
126
121

vibrations

95
143
181

130
124

were placed either directly on the soil

Therefore the soil reacted
only along the foundation base area. The free vertical vibrations were
excited by impacts; the forced vibrations by special vibromachines.
Frequencies of natural vertical vibrations of test foundations were
computed from Eq. (III-1-5), which does not take into account the
In

all

tests the foundations

surface, or at the

bottoms

of excavations.

inertial properties of soil.

Errors in evaluation of results of experiments are about 10 per cent.

The

analysis of Tables III-2 to II 1-4 leads to the conclusion that there is

only a small difference between the values of natural frequencies of
vertical vibrations as computed by the two methods, i.e., from the data

and from those established on the basis of observaand forced vibrations. Hence, Eq. (III-1-5) is in rather good

of static investigations

tion of free

agreement with experimental data.

The

conclusion

is

also possible that in all foundations studied, the soil

DYNAMICS OF BASES AND FOUNDATIONS

126

had small

effect on the natural frequencies of vertical vibrations

Apparently this is explained by the fact that for these
foundations the value b, found from Eq. (III-1-35), was comparatively

inertia

of foundations.

large, reaching

20 in some cases.

Therefore the effect of

soil

inertia

on

the natural frequency of vertical foundation vibrations was comparatively

small, being within the range of errors involved in the experiments and

the evaluation of results.

Experimental Investigations of

b.

III-5

summarizes the

damping

of vibrations

measured amplitudes

Coefficient

Damping.

of

Table

The values

were determined from the

This was done
and Eq. (III-1-21) becomes:
of

of forced vibrations at resonance.

as follows: at resonance

co

fnz

A*
since

the

results of the determination of the coefficient of

**- rfis

from Eq. (III-l-17a),

2mUnz

m2cfns

a/2m and from Eq.

(III-1-39).

{2m fnz) ~ (2m/)

t

if

one sets

tion

m = m
t

(i.e., if

one considers only the mass

and neglects the mass

of the

founda-

Then:

of the soil).

p
A Tes 2mfnz

The values of b were established from Eq. (III-1-35).

The analysis of Table III-5 leads to the conclusion that the coefficient
of damping is much smaller for soft gray silty clays with some sand than
Table

Summary of Test Results for the Determination

of the Reduced Coefficient of Damping

III-5.

Foundation
contact

Soil description

area,

Water-saturated brown
clay with some sand

silty

4
8

Water-saturated soft gray clay

with sand and organic silt

0.5
1.0

1.5

Water-saturated
sand

fine

dense gray

1
1

1
1

Weight

of

system,

16.5
6.7
4.0

0.145
0.133
0.181

3.25
5.10
6.72

26.0
14.5
9.0

0.071
058
0.051

6.76
6.76
6.76
16.4

19.0
20
20
55

0.132
0.190
0.175
0.083

tons

16.3
18.8
30.0

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

127

with some sand and for fine gray sands. In foundaon sands, much higher values of b were observed than in
foundations placed on brown silty clay with some sand. Hence it follows
for

brown

silty clay

tions placed

that for similar values of b, the coefficient of damping will be larger in

sands than in brown clays.
The value of the coefficient of damping is strongly influenced by several
factors very difficult to take into account (for example, the backfilling of

Figure 111-20 presents two resonance curves

foundation excavations).

of vertical forced vibrations of a founda-

tion with a contact area of 1.0

1

2
;

1.0

curve
0.9

corresponds to the situation in which

is exposed along all its

the foundation

height; curve 2 characterizes the

0.8

same

The depth
was around 2 m. The

backfilling

of

0.7

foundation, but back-filled.

0.6

foundation was placed on silty clays

interbedded with sands; the groundwater level was considerably below the

0.5
0.4

foundation base.
It follows from the comparison of
graphs 1 and 2 of Fig. 111-20 that
amplitudes of vibrations of a backfilled
foundation at resonance are about 3.5
times smaller than those of an exposed

foundation.

damping

is

amplitude
it

Since

the

coefficient

0.1

-2

&J /
5

12
hertz

of

of vibrations at resonance,

follows that in the case under con-

damping

ad

0.2

16

20

24

inversely proportional to the

sideration, the value of the coefficient

of

0.3

Fig. 111-20. Resonance curves of a

1.0-m 2 test foundation: curve 1, sides

footing free; curve 2, sides of

of

foundation backfilled.

for a backfilled foundation

be approximately 3.5 times larger than that for an exposed foundation.

considerable effect of backfilling on the value of was also observed
in investigations of foundations placed on gray sands.
For example,
increased from 0.19 to 0.32 when a foundation was backfilled to the
will

height of

m.

Even when foundation

sides are not subjected to the influence of soil

by ground water, the value grows. For

example, investigations on the same fine gray sands established that submerging of the foundation to a height of 1 to 1.5
is accompanied by
reactions, but are flooded

an increase in the

damping by 1.5 to 2 times its value.

The increase in the coefficient of damping when foundation sides are
not free is explained by the increase in the total foundation surface dissicoefficient of

pating energy into the

soil.

There

is

an increase

in the dissipation of

DYNAMICS OF BASES AND FOUNDATIONS

128

energy of foundation vibrations, and therefore an increase in the coeffiBesides, the value of is affected by the forces of
cient of damping.
friction, whose magnitude increases with an increase of the area of backfilling of

the foundation.

Experimentally established values of were in most cases smaller than

computed values taken from graphs of Fig. III-8a, corresponding to
For foundations characterized by
selected values of the coefficient b.
data presented in Table III-5, no experimental value of was greater
than 0.20. However, according to the graphs of Fig. III-8a, the values
of this coefficient (corresponding to the value of b for the test foundaApparently the absolute
tions) should almost always be larger than 0.20.
values of the computed coefficients are close to the values typical for
backfilled foundations.

Table

III-6.

Data on the Vibration of Pile Foundations

Data on

Frequency

forced vibrations,

foundation

K
Mass

tion

contact
area,

of

sec 2 /m

sec
z

-1

from

static

Dis-

system,
tons

of natural

pile

Number of

tance

investi-

Length

gations,

of piles,

kg/cm

be-

tween

piles

Established

Com- from
puted tion

observaof forced

vibrations

piles,

m
10.8
8.6
8.3
6.5

If

3.0
2.3

16

1.7

12

2.0

12

0.81
0.81
0.81
0.81

5.4
5.6
5.4
5.6

153

104
105
55

X
X
X
X

10 4
10 4
10 4
10 4

227
215
247
166

201

186
235
138

one plots a resonance curve of forced vertical vibrations of a founda-

on the basis of computed values of the coefficient at resonance, it

turn out that the computed amplitudes of forced vibrations are not
This
in full agreement with experimentally established amplitudes.
divergence is partly explained by errors involved in the operation of
tion

will

frequency-measuring devices. In addition, values of apparently depend

on the frequency of vibrations and, possibly, on the amplitude; therefore
the values of are different for different sections of the resonance curve.
Consequently, amplitudes computed on the basis of an assumption
that

remains constant only approximately correspond to the true values.

Available experimental data permit the assertion that if the coefficients

of elasticity of the soil are correctly selected, then the divergence between

romputed and

test values of

not exceed 10 to 20 per cent.

amplitude outside the resonance zone

will

THEORY OF VIBRATIONS OF MASSIVE MACHINE FOUNDATIONS

c.

Investigations of Vibrations of Pile Foundations.

129

Investigations of

were performed on test pile foundations. Results

are presented in Table III-6.
The natural
Piles were driven into water-saturated fine dense sands.
frequency of vertical vibrations of the foundation was computed from
Eq. (III-1-5) the foundation mass was considered to be the only vibratvertical vibrations

It is natural, therefore, that the natural frequencies of vertical

ing mass.

vibrations established

by

static investigations

turned out to be somewhat

higher than the frequencies obtained

from investigations of forced vibrations.

For all foundations investigated, a
magnitude of vibrating mass was computed on the basis of values of the
coefficient of elastic

uniform compres-

by

sion,

established

tions,

and on the basis

static investigaof the resonance

frequency of vertical forced vibrations.

This vibrating mass was, on the average, 30 per cent larger than the mass
restricted to the foundation only.

1.2

E
6

1.0

0.8
I
JO

\
\

o
to
0.6
o

I
V

fo.4
E
J
r
0.2

V*

^<\J

s*
Investigations
Experimental
of
c
Complicated Forms of Foundation VibraY
4
8.
12
16
Other forms of vibrations of
tions.
Number of oscillations per sec (hertz)
foundations were also investigated.
For example, forced vibrations in one Fig. 111-21. Resonance curves of a
1.5-m 2 test foundation of silty clay
of the principal planes, induced by a
with sand subjected to forced horihorizontal force, were studied.
Figure
zontal vibrations.
III- 14 gives a resonance curve of forced
2
horizontal vibrations of a foundation with a contact area of 4
placed

d.

on brown

with sand; Fig. 111-21 shows a similar curve for a

2
foundation with a contact area of 1.5
placed on water-saturated soft
silty clays with sand.
Similar investigations were performed on loess and
on gray sands. In the first two cases, for all foundations investigated,
two frequencies fn i and fn i were obtained which correspond to a sharp
rise in the amplitudes of vibrations.
Thus, there is experimental corroboration of a theoretical conclusion
concerning two maximums in the resonance curves of forced vibrations
of foundations, corresponding to the two frequencies fn \ and / 2
No
static tests for determining c v were performed during the investigations
of foundations on water-saturated silty clays with some sand.
Therefore,
silty clay

no way to verify directly how far the frequencies

computed from values of c? and cT (established by means of

in this case there is

fni

and

/ 2

static investigations), coincide

with those which were established experi-

DYNAMICS OF BASES AND FOUNDATIONS

130

mentally. Static tests for determining c v and c T were performed on

foundations resting on loess. Using the established values, it is possible
to compute the frequencies /i and /n2 and to verify how close the computed values of these frequencies are to those established experimentally.
Results of this analysis are presented in Table III-7.

Table III-7. Data on the Two Fundamental Frequencies of a

Foundation Subjected to Forced Horizontal Vibration
Frequencies computed

from

Founda-

Experimentally
established

static

values of / 2

investigations

tion

contact
area,

fni, sec

-1

174.0
181.0
140.0
167.0

0.81
1.40
2.60
4.00

/2, sec

From

58.4
73.5
65.2
89.2

free

From

forced

vibrations

vibrations

65.3
73.5
69.1
77.8

48.3
64.0
50.2
54.0

and c T were performed on gray

with some sand; also /n i and fn2 were experimentally deterSince these frequencies are connected with the limiting fre-

Static investigations for determining c u

silty clays

mined.

fn<p and fnx it was possible to establish the latter analytically.

Then, on the basis of the values of fn<p and/nx the values of cT were determined. Results of this processing of data from dynamic investigations
on gray silty clay with some sand are presented in Table III-8.

quencies

Table

Data on the Coefficient

III-8.

ct

of Elastic Uniform Shear of Soil

cT ,

Foundation

kg/cm 2

contact
area,

0.5
1.0
1.5

It is seen

From experimental

From

static

frequencies fn \ and fn2

investigations

1.88
1.64
1.27

1.90
1.58
1.40

from the tables that the values of c T computed on the basis

and those obtained as a result of the investigations

of static investigations

with a satisfactory degree of accuracy.

Available data, including those cited above, lead to the conclusion

of vibrations coincide

that the theory of vibrations of foundations for several-degrees-of-freedom systems, as presented in this chapter, is supported by experiments.

IV
FOUNDATIONS UNDER RECIPROCATING
ENGINES

IV-1.

General Directives for the Design of Foundations

Design Values of Permissible Amplitudes of Foundation Vibrations.

types of reciprocating engines belong to the group of unbalanced
machines which are dangerous in respect to vibrations. The fact that
these engines usually operate at comparatively low speed increases the
probability that vibrations may develop in adjoining buildings or structures.
Therefore a thorough analysis of vibrations in such foundations
a.

Many

is

of the

The

utmost importance.

greater the amplitude of vibrations of the foundation, the

more

danger there is to adjoining structures. In addition, if the amplitude

of foundation vibrations is large, the foundation may lose its stability
and undergo a nonuniform settlement endangering the normal work of
the engine.
Finally, vibrations of large amplitude may lead to the
destruction of the foundation and to damage of the engine for example,
it has been observed that sometimes crankshafts of log-sawing frames
fail as a result of vibrations of their foundations and frames, to which the
crankshafts are rigidly tied. Similarly, foundation vibrations often cause
dangerous vibrations of machine connections.
It is extremely difficult to establish a limit for the permissible value
of amplitude of foundation vibrations on the basis of general principles.
There are some cases in which vibrations with an amplitude up to 0.4
to 0.5
did not have any harmful effects.
However, many cases
have been observed in which foundations under engines with low-frequency vibrations underwent vibrations at smaller amplitudes than those
cited above, but induced strong vibrations of structures located at a
distance of several tens of meters.
It may happen that even when the
amplitude of vibration of a machine foundation is smaller than an
;

mm

131

DYNAMICS OF BASES AND FOUNDATIONS

132

accepted permissible

limit, the adjoining structure will vibrate

due to

resonance.

On

the strength of data gained by experience

it is

possible to state

no resonance is to occur in adjoining buildings and structures,

then the amplitude of vibrations of a foundation should not exceed 0.20
This range of amplitude values may serve as a basis for
to 0.25 mm.
evaluation of the adequacy of foundation-design computations.
b. Machine Data Required for Foundation Design.
Data supplied by
the maker of the engine or motor is the basic information for a foundation designer, together with data on soil conditions and on the exciting
loads imposed by the engine.
The following information should be given
that

1.

2.

if

The normal speed and power of the engine

The character, magnitude, and point of

application of dynamic

loads which will develop in the process of operation of the engine.

If

data cannot be supplied, the designer of the machine foundation

should be given all the data needed for the computation of exciting forces
3. The distribution of static loads imposed by the engine over the
foundation surface
this

4.
5.

The
The

size

and shape

of the engine supporting plate

location of openings

and grooves

in the foundation provided

for anchor bolts, pipe lines, the flywheel, etc.

c.

Foundation Material.

The following loads are imposed on

a machine

foundation:
1.

2.

The weight of
The dynamic

the machine and equipment

loads which develop in the process of machine

operation

For diesel engines, the total load is such that the reduced pressure
on the upper surface of the foundation usually does not exceed 3 to
5

kg/cm 2

In any
still smaller.
and masonry are considerably higher.
It is natural that since pressures imposed on foundations are small, the stress analysis should be performed only for cross
sections weakened by large openings or grooves.

For horizontal piston engines

this

value

is

case, the permissible bearing values of concrete

Thus the question

masonry is

crete or

material on the

site.

of selection of a material for the

first of all

foundation

con-

and of availability of
usually employed for founda-

a question of cost

Concrete type 100|

is

tions under reciprocating machinery.

f Translation Editor's Note: The figure 100 indicates the 28-day compressive
cube strength in kilograms per square centimeter of the concrete mixture used (100

kg/cm 2 equals

1,420 psi).

FOUNDATIONS UNDER RECIPROCATING ENGINES

d.

Comments in Regard

to

133

Foundations under reciprocating

Design.

engines are usually built as massive blocks provided with grooves and

channels for machine details and openings for anchor bolts.

Due to the massive shapes of such foundations, when studying their
vibrations

it is

possible to consider

them

of

and
and amplitudes the theory

as absolutely rigid bodies

to use in the computations of frequencies

vibrations of a solid resting on an elastic base, as presented in

Chap.

III.

to be observed when designing a machine foundaminimum dimensions of the foundation should be

way that the amplitudes of its forced vibrations will

The main condition

tion

is

as follows: the

selected in such a

not exceed the permissible value.

If a foundation is erected on a natural
be

less

among

than that of frost penetration:

soil

base, its depth should not

There

is

an established opinion

practicing engineers that in order to decrease the transmission of

vibrations the depth of a machine foundation should be no less than the

depth of the footings of adjoining walls and columns. Theoretical and

experimental data on wave propagation in soils, presented in Arts. VII-2
and VII-5, lead to the conclusion that provision for machines of foundations deeper than footings for walls has

vibrations to walls.

no

effect

on the transmission

Therefore the depth of a machine foundation

of

may

be selected without taking into account the transmission of vibrations.

In order to obtain uniform settlement of the foundation, t is recom-

mended

to place the

common

center of gravity of the system

the foundation and machine) on the

same

(i.e.,

of

vertical line with the centroid

foundation area in contact with the soil. In any case the eccenmasses should not exceed 5 per cent of the
length of the side of the contact area.
Satisfying this condition makes
it possible to simplify the computation of foundation vibrations.
When
the common center of gravity does not lie on the same vertical line as
the centroid of the foundation contact area, it is necessary, as stated
above, to solve at least three interrelated differential equations of vibraIn order to simplify the computations, an attempt should be
tions.
made to achieve a condition such that the plane of action of the exciting
forces imposed by the machine coincides with one of the principal planes
of the

tricity in the distribution of

of inertia of the foundation.

In order to decrease the transmission of vibrations to adjoining parts

gap between the foundation of an
unbalanced machine and the adjoining structures (footings, walls, floors,
and so on). As a rule, the machine foundation is not allowed to serve as a
support for other parts of the building or for mechanisms not related to a
given machine. If it is not possible to avoid placing unimportant parts
of a building on the machine foundation, measures should be taken to
of buildings, it is necessary to leave a

DYNAMICS OF BASES AND FOUNDATIONS

134

soften the connection

by providing gaskets made

of rubber, cork, felt,

or other insulating materials.

If several machines are to be installed in the same shop, and if the
distances between these machines are comparable to the foundation

dimensions, then, in soft

soils, it is

recommended

to place the foundations

under similar machines on one common mat of sufficient thickness. The

rigidity of this mat should be selected so that its possible deformations
remain small in comparison with the amplitudes of vibrations. Only
then may a group of machine foundations installed on the same mat be
regarded as an absolutely solid block resting on an elastic base. The
computation of vibrations of such groups of foundations is very difficult.
If an asymmetry and an alternating phase shift are observed, then computations are practically impossible.

Therefore

if

several foundations are erected on the

conditionally broken
tions

up

same mat,

it

is

into sections corresponding to separate founda-

the computations of vibrations proceed as

if

each foundation were

Then the design value for the permissible amplimay be increased somewhat (by 25 to 30 per cent).

installed separately.

tude of vibrations
To avoid a distorted tilt of the master shaft of the machine, its external bearing should be placed on the same machine foundation.
This
directive refers also to the installation of motors coupled directly to
reciprocating engines such as electromotors and generators.
In any case, the foundation area in contact with soil should be selected
in such a way that pressure on the soil does not exceed permissible
values.

The

larger the foundation contact area, the smaller the reduced pres-

and the higher the natural frequencies of the foundation.

importance for low-frequency machines, including
The easiest way to change the natural
the reciprocating engines.

sure on the

This

is

most

of

soil

of considerable

frequencies of a foundation

is

to increase or decrease the dimensions of

the foundation contact area and change

its

configuration in plan.

fore a final selection of the contact area should be based

There-

on requirements

obtained as a result of design computations regarding vibrational loads

on the foundation.
Foundations under low-frequency machines should be designed so
that their natural frequencies are much higher than the operational
frequencies of the machines.

The natural frequencies

of

foundations are affected by the absolute

value of the foundation mass and by

its

distribution in space.

The

designer should try to distribute the mass so that the smallest possible

value of

its

moment

of inertia is

obtained with respect to the principal

axis passing through the centroid of the foundation contact area.

meet

this requirement, the

minimum foundation height should

To

be selected.

FOUNDATIONS UNDER RECIPROCATING ENGINES

IV-2.

Unbalanced

Inertial Forces in

135

Reciprocating Engines

Forced vibrations in foundations under piston engines are largely

caused by the unbalanced inertia forces in the moving parts of crank
mechanisms.
Figure IV- 1 illustrates the main features of a
a. Single-line Machines.

The piston A and the piston rod B execute

an alternating motion; the connection rod C executes a complicated
periodic motion; all points of the crank D execute a rotational motion
around the main axis 0. Any of these parts may have unbalanced
inertial forces which independently may cause foundation vibrations.
reciprocating mechanism.

Fig. IV-1.

The main parts

According to the laws of

acting at point

statics,

and place

of a reciprocating engine.

we may

replace

all

these forces

by forces

at this point the origin of a coordinate system

x, y, z.

Let us place the z axis in the direction of piston movement, the x axis
perpendicular to this direction, and the y axis perpendicular to the plane
of the drawing.
Since all points of the crank mechanism move in the
plane xz, the y ordinates will remain constant for all these points. As a

unbalanced inertia forces by forces acting at point 0,

and one couple. Resolving these into their components along the coordinate axes, we obtain the force components P x
and P z and the moment
y
The position of any point of the crank mechanism is determined by
result of replacing all

we obtain one

force

one independent variable: the angle of rotation <p of the crankshaft.

Therefore the coordinates xi} y i} and Zi of any point i of the moving
machine parts are functions of the angle <p, the latter being a function
of time.

Let us denote by Xi and z\ the projections of the acceleration of an

element i of the crank mechanism on the x and z axes; then the projections of the inertial forces acting on this element will be as follows:
X xi

fiv%JLt%

zi

nt%K"i

DYNAMICS OF BASES AND FOUNDATIONS

136

The

projections of the inertial forces acting on

crank mechanism

will

Px =
Considering the

Xi

all

elements of the

be

and

Zi

P = SmA

Vrmxi

(IV-2-1)

coordinates as functions of
dxi

t,

we obtain

dxi dip
''

It

dip dt

dip

but

'

CO

dt

where

co

is

constant.

the angular velocity of the machine rotation, hereafter assumed

Therefore,
dxj _

dxi

dt

dip

Taking the derivative with respect

Xi

similarly,

z{

Substituting the values for

i\

P
The
will

to

t,

=
co

and

find

d 2 Xi

d 2 z-

d?

<*

we

z\

into Eqs. (IV-2-1),

,v

d2-

dip-

we obtain

(IV " 2 ' 2)

mechanism evidently
moving parts: the crank,
Consequently the component of the

total resultant inertial force of the crank

equal the

sum

of the inertial forces of its

the piston, and the connecting rod.

resultant of the inertial forces acting in the direction of the piston motion

can be described by the formula

P =
3

where

P zX
=
Pz?> =
Pzi

P,l

+P +
22

P*Z

projection of inertial forces of crank on the z axis

projection of inertial forces of rod, crosshead, and piston
projection of inertial forces of connecting rod

FOUNDATIONS UNDER RECIPROCATING ENGINES

137

Further,

Pz2

The

first

moment,

ity,

of the

d 2 z 2i

03-

>

m 'ii-f^:

w2

for example, of the

to the rotation axis will equal

mass

d *z *

crank and Xi and

d<p 2

mass

On

Xmi&u.

z\

d 2 z zi
3i

of the

crank with respect

the other hand,

if

is

the

are the coordinates of its center of grav-

then

XmuZu = MiZi
Differentiating this equation twice with respect to

d 2 Zi

,,

<p,

we obtain

d 2 zi

I
Using

we

this relationship,

find

^g

Pel

Pz2

"*-&

Pit

=^4

similarly,

The

JY

P.l

^^l

Px2

-^S?

P.I

^S

expressions for the projections of the resultant inertial force will be

* -*(" + *? + * 8?)
(IV-2-3)
2

co

/ ,,

d 2 zi

(Mi -j-i

-h M
.

T.

d 2z 2

3-f

+
.

T
M-

'

Without limiting the general validity of the solution, and without involving any significant error, it is possible to concentrate the entire mass of

DYNAMICS OF BASES AND FOUNDATIONS

138

the crank mechanism not at three points, as has been done above, but
This will simplify the expressions obtained for P x and P e
at two points.
.

It
rate.

has been assumed that the crank executes its motion at a uniform
Therefore the magnitude of its inertial force will be the magnitude
of the centrifugal force;

i.e.,

P = RiM
c

ICO'

where Ri is the distance between the center of

gravity of the crank and the axis of rotation.
This force

will be directed along the radius of

Let us consider that it is applied not
to the center of gravity of the crank, but to point
a, i.e., to the crankpin (Fig. IV-2).
In order to

rotation.

obtain a force equal to

it is

and applied at point a,

u is concen-

necessary to assume that mass

trated at this point and

same proportion

as

Ri

is

is

smaller than

Mi

in the

smaller than R; thus

we

should set

Ri

In

Mi

Since point h (the crosshead) executes a reciprocating motion which does not differ from the
Fig.

IV-2.

motion

Reciprocat-

may

ing engine, illustrating

derivation of

the

Eq.

Finally the mass

M
M

is

mass

2,

we

concentrated at

this will not change the magnitude of its

induced inertial force.
of the connecting rod may be replaced by masses

point b

(IV-2-4).

of the center of gravity of

consider that mass

;

However, this distribution

32 concentrated at points a and b.
so
as
not
change
the magnitude of the
made
to
should
be
of
3
inertial force of the connecting rod computed on the assumption that
its mass is concentrated at the center of gravity (3,23).
3i and
mass

Denoting the coordinates

(xb,Zb),

P,3

On

of points a

and

b respectively bjr (x a ,z a )

we obtain
d 2z

(,

d<p*

dW
+ M.32^

the other hand,

l

M,

dhz
d<p*

Equating the right-hand parts

of these expressions,

31za

Miozi,

zz

we obtain

and

FOUNDATIONS UNDER RECIPROCATING ENGINES

139

Similarly, deriving the expression for the projection of the inertial force

on the x

of the connecting rod

we obtain the second equation

axis,

M nx + M
a

= MiX%

d2 Xb

from the equations obtained for

3i and
32 that the values
masses should be selected so that their center of gravity lies at
point (xzjZz), i.e., the masses should be distributed in an inverse relationDenoting these distances by L\
ship to their distances from mass
3
It follows

of these

and L 2 we obtain
,

Mn = L

Mm
On

Li

the other hand,

ikf 32 ,

we

Mn + M
Solving these equations for M and
Z\

Af ,1

where

is

j?

M^

32

j-M

find

L = L

the length of the crank.

Thus the three masses, concentrated

at the centers of gravity corre-

sponding to the parts of the crank mechanism, may be replaced, without

changing the magnitudes of the inertial forces of the mechanism, by

two masses:
1.

Mass

a,

concentrated at the crankpin:

M
2.

Mass Mb, concentrated

M
The
will

M n = ~ M, + h- M*

M +M
2

inertial force of

this will

a o)

sin

-f

mass

mass Mb

P za =

will

tLi1 ^3

(IV-2-5)

M on the coordinate axes

M ^R cos
a

<p

only have a projection on the

z axis;

be equal to
zb

The

<p

P =
will

32

projections of the inertial force of

(IV-2-4)

at the crosshead:

be

P xa +
The

In +

-M

h U3

Zh

projections of the resultant inertial force of the whole

mechanism

equal

Px =
P =
z

M R sin
M ^R cos
a o>
a

<p

<p

Mw
b

Zb

(IV-2-6)

DYNAMICS OF BASES AND FOUNDATIONS

140

from

It follows directly

zb

We

= R

cos

+L

ip

cos

<p

have, from the triangle Oab,

from which

Expanding

cos

we

IV-2 that

Fig.

jS

cos

/?

= y sm

sin

*P

a2

sin 2

</?

<p

Newton's binomial theorem,

into a series according to

(3

sin

obtain

cos

(3

(1

sin 2 <p)*

a2

Ya

sin 2

<p

sin 4

y&a 4

<p

He"

sm<5

<P

'
'

'

Using the formula converting an even exponential trigonometric

function into a linear one,

2-l(_

gm n V =

l)/2

CO g U(p

+
.

n C0S
1)

n(n
j

cos (n

2)(f
.

4)<p

(n/2)n(n -

I)

(n/2 -h

we

replace powers of sines

cos

where Aq,
teristic

2,

4,

number a

of the

by the cosines

+A

= A

(3

" V

cos

2<p

of multiples of

At

cos

4<^

n/2

Then

2<p.

are constants depending only on the characcrank mechanism

A = 1 - M - %4 - %56 A = ^ + K6 + 1 %12 +
A = ~(M4 + %56 +
)
4

'

'

'

Substituting the expression established for cos

into Eq. (IV-2-6),

fi

we

obtain
zh
2

rf

26
2

= R

cos

<p

= R(cos

+
<p

- (A

-\-

+A

cos

2<p

cos

-f-

2(p

+A

Ba cos 4^

dip

where

4A
B =
2

2? 4

'

"

I6A4

Substituting into these formulas the expressions for

in

cos 4<p

view of the small value of

a,

disregarding

all

and

4,

and,

terms containing

its

FOUNDATIONS UNDER RECIPROCATING ENGINES

fourth or higher powers,

141

we obtain

*.--$

B.-..(l+Sf)
Therefore,

-~ = R\

cos ^

afl

+x) cos ^

T cos ^V

Substituting the expression established for d 2 Zb/d<p 2 into the formula for

zh ,

we

find that

P =

MbRo) 2

zb

cos

<p

all +-T-)

Substituting this expression for

zb

cos

^ ~ ~T cos ^

into Eq.

(IV-2-6) describing the

projection on the z axis of the resultant inertial force of the machine

and replacing

we

<p

by

cot,

where

cos

co

is

the angular velocity of machine rotation,

finally obtain

Px =

Rco 2

P =

Rco

sin

(M a

cot

b)

cot

+ aM

( 1

+~

cos

2cot

M a* COS
b

Thus the formula describing the

(IV-2-7)

J
.

4:C0t

exciting loads causing the forced

vibrations of a foundation contains terms depending not only on the

machine rotation, but also on the double, quadruple, etc.

However, the coefficients preceding cos 2cot and cos
decrease very quickly, and these terms may be disregarded in engi-

frequency

co

of

of this frequency.
4tot

neering calculations.

The terms containing cos cot are called primary inertial forces (the first
harmonics); those containing cos 2cct, secondary forces (secondary
harmonics) and so on.
;

The foregoing

discussion leads to the conclusion that rotating machin-

ery masses produce primary inertial forces; reciprocating masses produce

and forces of even higher orders.

mass M'a on a shaft, it is possible to
balance inertial forces induced by mass
If M'a is fixed on the shaft so
a
that the angle between the radius vectors of masses
a and M'a equals w,
then, in order to have the inertial forces of the rotating parts balanced,
one of the two following conditions should be satisfied: Either

both primary

By

inertial forces

installing counterweights of

M. - g M =
'.

or

V Ml + ^M -M'
3

DYNAMICS OF BASES AND FOUNDATIONS

142

In order to balance fully the projections of primary inertial forces in the

direction of piston motion it is necessary to select mass M'a and distance I
(between its center of gravity and the axis of rotation) so that either

M
^M +M +M
a

or

2o)t

and cos

4co,

Usually

Mb

is

larger than

M' =

(IV-2-8)

one of the above equations,

remain terms depending only on

satisfy

will

while the expression for

Px =

mass

- I M'a =

If the selected values of I and

then in the expression for P z there

cos

R^M

Px

will

be as follows:

sin ut

therefore the selection of a counterweight

satisfying Eq. (IV-2-8) leads to

an enlargement

of inertial forces in

the direction perpendicular to the

sliding of the piston.
1st cylinder

/f-th

cylinder

b.

Multicylinder

method

The

Engines.

of determination of exciting

loads in multicylinder engines

principle

the

same as

in

is

in

single-

cylinder engines.

Consider

vertical

engine

in

which the cylinders are situated in

the same plane, parallel to each
other (the so-called linear arrange-

ment of cylinders). Usually the

number n of cylinders does not

*/

exceed

10.

Unbalanced

inertial

forces are calculated similarly for

vertical

and horizontal reciprocat-

ing engines.
Fig.

IV-3.

Multicylinder engine,

illus-

trating the derivation of Eq. (IV-2-9).

Let us direct the y axis (Fig.

IV-3) along the crankshaft of the

and horizontal, the z axis

upward, along the axis of sliding of the pistons. Let us place the origin
at the mass center of the foundation and engine and let us assume that
the yz plane passes through the principal axis of engine rotation.
We confine ourselves to the case in which the engine has only main
We
cylinders (no auxiliaries such as compressor and exhaust cylinder).
first
denote by (3 k the angle between the crank of the kth cylinder and the

engine, the x axis perpendicular to the shaft

FOUNDATIONS UNDER RECIPROCATING ENGINES

By

crank (the wedging angle).

143

the reasoning of Art. IV-2-a

we obtain

the following expressions for the component exciting force along the

x and z axes for the

/cth cylinder:

Pxk = R k u*Mak sin (cct +

P = R k [(M ak + Mbk )
cc

2k

cos

ak

The terms

fa)

3
Mbkak
7

fa)

-j- cos 2 (cct

bk

and

(cct

bk

a k cos

(cct

fa)]

cos

...

4(co

fa)
.

/3ft)

have been neglected.

In order to obtain the resultant exciting force transmitted to the
foundation from all engine cylinders, it suffices to sum the above expressions for

all

Px =

co

RkMah

co

Then we have

cylinders.

sin

(cct

fa)

Pz

2
ft

X=

ftbtW*

Af) cos

(art

/J*)

+ M^a* cos 2(co +

/3ft)]

(IV-2-9)

In addition to exciting forces, there are exciting moments; their magni-

tudes equal
n

= X
=
ft

If

PJyk

PdAib

k=1

= X ^*A*
=
1

lb

the crank mechanisms are identical in

all

cylinders, then the equations

for the exciting force will be simplified:

Px =

Rcc 2

a
ft

X=

sin

-f

(cct

/5ft)

??

P = Ru*[(M a +
z

M X

cos

b)

(co*

fa)

it

cos 2 (cct

follows that to balance the

harmonics

first

n
(cct

The second harmonics

fa)

will

X
ft=i

be satisfied

/3ft)]

of the exciting forces,

the following equations should be satisfied

X cos
ft=i

ft=i

ft=i

Hence

+Ma X

cos2(co*

sin

if

fa)

(cct

fa)

DYNAMICS OF BASES AND FOUNDATIONS

144

To balance

moments

the exciting

of the first harmonics, the following

equations should be satisfied:

X=

I* cos

(cot

ft)

/:a

sin

{col

&) =

lyk

sin

(co

ft)

i
1

=1
fc=i

Similar conditions hold for the second harmonics.

Let us consider several particular computations of exciting loads

imposed by multicylinder engines, assuming all cylinders are identical and
neglecting all higher harmonics of exciting loads.

PzZ

Pi\
i

J~

I,
1

>/\w%

Wffl.

?\w///,

0*

Fig. IV-4. Derivation of equations for a vertical two-cylinder engine.

c.

Let us assume that the engine

Vertical Two-cylinder Engines.

mounted asymmetrically on the foundation

(Fig. IV-4).

are identical.

Case

Cranks

1.

in

Same Direction.

ft

Assuming

in

Eq. (IV-2-9) that k

Pxi
Pzl

The

resultant

=
=

The components

we obtain

1.2,
2

Pzl

COS

Lot

of the exciting forces will

Px =
P =
z

2tt

= Rw M a sin cot
= RcC (Ma + M b

Px2

components

ft

Here,

2Rco 2

sin

2Rco 2 (M a

of the exciting

M
M
M

cot

moment

= P sl (l

h)

cos

equal

2k)

2l x )

2r x iL

= P xl (l

cot

be

is

Both cylinders

FOUNDATIONS UNDER RECIPROCATING ENGINES

The values of lx and l z are shown in Fig. IV-4.

The engine under consideration belongs to the

145

I,

class of highly unbal-

anced engines, dangerous with respect to vibrations.

Case

Two-cylinder Engine with 90 Crank Angle.

we have

2.

On

the

basis of Eqs. (IV-2-9),

M
Rco M

Pxl = Rw

P x2 =

sin

sin (at

cot

P rl = Ra*(M a +

P z2 =
The

P x = Rw

P =

Ro)

Ro> 2 (M a

resultant

(M a

Af6) cos

+
6

+ 1) =

o>*

= \/2 Ru
sin

it

j|f 6 )

sin

sin ( ut

|j

&>*)

=
Hence

-Rco 2 (M a

of exciting forces are

cos ui)

M )(cos

l!

co

cos (at

6)

components

(sin at

=
g (* + 1)

follows that the resultant

V2 P<o

(M a

components

6)

cos (at

+ I)

of exciting forces are

1.41 times the resultant forces in each cylinder.

Let us determine the components of the exciting moment:

M
M
M

= Pel (l + y + PJy
= (Pxl + P x2 )l
= Pxl (l + y ) + Px2 ly
l

Analogously,

This case, the 90 crank angle,

is

the most characteristic for two-

cylinder engines.

Case
01

0,

3.
(3 2

Two-cylinder Engine with 180 Crank Angle. Here

According to Eq. (IV-2-9), the resultant components
it.

of the exciting forces will equal

Px =

The components

P =
z

of the exciting

moment

= P sl

equal

= Pxl

d. Reciprocating Horizontal Compressors.

These engines usually have
two cylinders with 90 crank angles. The expressions for the exciting
forces here are the same as for a vertical engine the only difference is that
in the equations x should be changed to z and vice versa.
;

DYNAMICS OF BASES AND FOUNDATIONS

146

The

exciting

moments

will

equal (Fig. IV-5)

= P gl yl - P z2
l

My = P

M
crank angles;

i.e.,

ft

0,

sin
first

harmonics

Pxllyl

120,

l y2

PJ.
PxllyZ

These engines usually have 120

and /3 3 = 240. Since

+ cos 120 + cos 240

+ sin 120 + sin 240

cos

the

Three-cylinder Engine.

Vertical

e.

Zlx

=
=

of the exciting forces are balanced:

Px =

P =
z

spaced

If all three cylinders are

alike,

then the exciting moments of the

engine are

M
M
M

= P el (2l

ly)

ly)

+ PJy

= Pxl (2l

ly)

+ Px2 +

ly)

+ PxZ

P,*{1

f.

0i

Vertical
0,

(l

This engine

Four-cylinder Engine.

exciting forces

180,

and

180,

(3 A

is

360.

and moments are balanced as a

ly

so

All

designed that

components

of

result of this arrangement.

S'lyZ

lyX

*~P.u
x\*

P.
r
xZ

WR

W7R.
1

Fig. IV-5. Derivation of equations for horizontal piston compressors.

g.

Vertical

Six-cylinder

In this engine

Engine.

crank angles are

usually as follows:

2t

4tt

For such crank

positions, the first

forces are balanced.

ft

The

exciting

V3 PJ

4tt
181

and second harmonics

of disturbing

moments equal
y

V3 PJ

Their absolute values are comparatively small and they cannot cause
Therefore

vibrations with an amplitude exceeding the permissible value.

FOUNDATIONS UNDER RECIPROCATING ENGINES

147

computations of foundations under six-cylinder engines there is

no need to compute forced vibrations.
If the engines have auxiliary cylinders (a compressor and an exhaust
cylinder in addition to the main cylinders), then in the computations of
exciting forces a load imposed by the auxiliaries should be added to the
However, the exciting loads
loads produced by the main cylinders.
caused by auxiliary cylinders are small in comparison to loads caused
by the main cylinders, and they may often be neglected in computations
of foundation vibrations.
in design

IV-3. Stresses

In

many

Imposed by

Belt Pull

some types
by means of a belt

cases reciprocating engines set into rotary motion

of operating machines, usually electric generators,

Driven
pulley

Driven belt

Drivin>

pulley

-Driving belt

Fig. IV-6.

drive.

On

Diagram

the other hand,

motion by means

of forces transmitted to pulleys

by

belt.

some reciprocating engines are given rotary

from an electromotor. Saw frames and

of a belt drive

compressors belong to this group.

When a belt drive is in operation, the force of the belt pull acts on the
engine bearings and consequently on the foundation.

Let us consider a reciprocating engine set in rotation by means of a

and examine the stresses transmitted to the foundation.
If Ti is the magnitude of belt tension in the slack side of a belt (Fig.

belt drive

IV-6) and T 2 is tension in the driving side, the resultant force of pull
transmitted to the bearings of the engine, and consequently to the
foundation, equals

T
The

T2 = P

(IV-3-1)

peripheral tension transmitted by the belt to the driven pulley

is

the difference between tensions in the driving and driven belts:

Pr

= T ~
l

T,

(IV-3-2)

DYNAMICS OF BASES AND FOUNDATIONS

148

If
it is

is

the engine power and

known

the peripheral speed of the belt, then

v is

that

W
V

Since

where

=
R =

speed of engine,

The

= Rw =
rpm

radius of driven pulley

then

belts

NR
60
2tt

60

^2~,NR = g 55 WR
-

<

interrelationship between the pull values in the driven

and driving

approximately expressed by the formula

is

T 2 = T&
where

IV-3 "3 >

<p
ix

=
=

(IV-3-4)

smallest angle of arc of belt contact

coefficient of friction

between belt and pulley

The magnitude of depends on the type of flexible connection

From Eqs. (IV-3-2) and (IV-3-3) we have:
ll

7W

used.

9.55^

Substituting here the expression for

T from Eq.
2

(IV-3-4),

we obtain

RN 1 -

e^

1
T
1

e*

and consequently,
T,

9.55

4kr
RN

Substituting these expressions for 7\ and

(IV-3-5)
v

into the right-hand part

Eq. (IV-3-1), we obtain the following expression for the force transmitted to the foundation by the belt pull
of

p-

The

^mr^

(IV - 3 " 6)

direction of this force depends on the respective locations of the

axes of the driving and driven pulleys.

FOUNDATIONS UNDER RECIPROCATING ENGINES

149

the straight line passing through the axes of rotation of the driving

If

and driven pulleys forms an angle y with the horizontal

the horizontal component of belt pull equals

P =
The

vertical

component

of pull tension usually

may

be neglected since
In cases in which the

small in comparison with the engine weight.

engine

is

then

Pt cos 7

tx

it is

(Fig. IV-6),

driving, the expression for

remains the same but the direction

sign changes.

Therefore

if

mounted on the same

the driven and driving pulleys are

foundation, the forces imposed by a driving gear represent internal forces

and do not influence the displacement

of the foundation.

Examples of Dynamic Analyses of Foundations for Reciprocat-

IV-4.

ing Engines

Example

Dynamic computations

1.

shaft with

Design Data.

1.

crank angles,

/3i

an

0,

of foundation for a vertical compressor coupled on a

electromotor

two-cylinder compressor has the following characteristics:

= tt/2; compressor weight, 12 tons; electromotor weight,

j3 2

4 tons; operational speed, 480 rpm.

The

harmonics of the exciting forces equal

first

(in

tons)

in the direction of sliding

of the piston,

P =
zi

and

3.0 cos

Pz2 =

cot

-3.0

sin

cot

in the horizontal direction perpendicular to the shaft axis,

Px =
\

where w

is

0.4 sin

PX =

cot

0.4 cos

cot

the angular velocity of rotation of the compressor, equaling

co

The base

0.104

480

50 sec -1

of the foundation consists of silty clays with

some sand characterized by

the following design elastic coefficients:

cu

2.

5.0

10 3 tons/m 3

cv

10.0

Design Diagram of Foundation.

10 3 tons/m 3

To

cT

2.5

simplify computations,

10 3 tons/m 3

it is

advisable

to shape the foundation in plan as simply as possible, avoiding all small grooves, projections,

asymmetry, and

so on.

Figure IV-7 gives a design diagram for the founda-

tion under consideration, selected on the basis of the foregoing reasoning.

larger dimensions were selected for the projection of the foundation slab

hand

Somewhat
on the

left-

due to the eccentric distribution of the equipment on the foundation.

3. Centering of the Foundation Area in Contact with Soil and Determination of Static Pressure on Soil. Let us determine the coordinates xo, yo, and zo of
side,

DYNAMICS OF BASES AND FOUNDATIONS

150

the common center of gravity of the system (the foundation and compressor with the
electromotor) with respect to the axes shown in Fig. IV-7:
XrriiXi

Xo

where

3/o

= ^mnji

ZrriiZi

= masses of single elements of system

= coordinates of centers of gravity of
m = mass of system

ra

single elements with respect to axes

Xt, yi, z%

We will consider

that the masses of the compressor and electromotor are concentrated

at the height of the level of the master-shaft axis (at a distance of 0.8

from the

foundation surface).

,/
4 tons

12tons

-2.2

2.3

-5

4-1

1
z i_

00

x -2

o
6.0-

>

3.0

H-0.5

4.8

0.7

-1
*

Dimensions,

X4 tons

12 tons

X-+

Fig. IV-7. Design diagram of foundation, example 1:

(1)

foundation slab;

(2)

upper

part of foundation.

The results of computations of static moments of single elements of the system are
given in Table IV- 1. Using those data, we obtain the coordinates of the common
center of gravity of the system
10.35
1.5

3-0

6.91

The

?/o

20.55

-6^r

7.52

ftQ
98m

1.09
6.91

relative values of the eccentricity in the directions of the x

and y axes equal,

in

per cent:
3.0

2.98

3.0

100

0.7

These values of eccentricity in the distribution of the masses are so small that they
be neglected in further computations of the foundation. Thus we obtain: the
weight of the whole system,

may

= mg =

6.91

9.81

67.5 tons

FOUNDATIONS UNDER RECIPROCATING ENGINES

the foundation area in contact with

A =
and the

static pressure

on

6.0

3.0

18

soil,

W = ?1
67 5
=
= JL

p Bt

151

soil,

lo

A.

tons/m 2 = 0.38 kg/cm 2

3.8

4. Possible Forms of Foundation Vibrations and Design Values of Exciting

Loads. The foregoing data lead to the conclusion that horizontal components of the

Table

Elements
system

Summary of Data for the Solution of Example

IV-1.

Dimensions

Coordinates of

of

center of

Mass

elements,

of

<h

az

Compressor
Electromotor
Foundation

slab

Upper part

element,

sec 2 /m

tons

Static
of

gravity of

element,

m
ax

of

Vi

1.23
0.41

1.5
1.5

2.2
4.5

moments

mass

of

elements,
tons X sec 2

m
z%

Xi

IV-4-1.

rrnxi

mnji

1.85
0.62

2.70 2.82
1.84 0.90
6.06 0.55

2.3
2.3

0.5

2.02

1.5

3.0 0.25 3.03

4.8

1.0

3.25

1.5

3.0

7YliZ %

of

foundation

4.85 9.75 3.25

1.0

10.35 20.55 7.52

6.91

Total

disturbing forces of the compressor are small in comparison with vertical components.

Therefore the dynamic analysis of the foundation

ing the amplitudes of forced vibrations caused
forces

The

and

of their

moments.
component

resultant vertical

Pz

The design value

Pz\ cos

4.2 cos (tot

cot

of the A^ertical

sin

-i

to

will

to determin-

of the exciting

cot

3.0 (cos

cot

sin

cot)

component

of the exciting loads will be

4.2 tons

Due

components

P =
This load

vertical

of the disturbing forces equals (see Art. IV-2)

may be confined merely

by

induce vertical forced vibrations of the foundation.

asymmetric position of the compressor, the foundation will be subjected

the action of the disturbing moment
The magnix with respect to the x axis.
to the

tude of this

moment

is

= P tl (l

y)

+ Pz2 v
l

DYNAMICS OF BASES AND FOUNDATIONS

152

where

I
lv

=
=

distance between cylinder axes; in the case under consideration

1.3

distance between second cylinder and vertical axis passing through center
of gravity of complete system in the case under consideration l v = 0.2

Thus

The design value

3.0(1.3

4.6 cos (

u>t

3.0

0.2 sin ut

|J

moment

of the disturbing

0.2) cos ut

should equal

its

greatest magnitude:

Xm

4.6 tons

Under the action of this moment, vibrations will develop in the plane parallel to yz;
they will be accompanied by a simultaneous sliding of the foundation in the direction
of the y axis and a rotation of the foundation with respect to an axis parallel to the
x axis and passing through the common center of gravity.
5. Computations of the Amplitude of Forced Vertical Vibrations of the
Foundation. From Eq. (III-1-5) we determine the frequency of vertical natural
vibrations of the foundation:

/=
The amplitude

'

Hence

it

ox
';;

xi8 =

i3.o

-Is) X

6.91(13.0

103

"

io3 sec

-,

found from Eq. (III-1-13):

of forced vertical vibrations is

A -

much

X 10- -0.06 mm

0-058

follows that the amplitude of vertical vibrations of the foundation will be

mm).
Determination of the Moments of Inertia of the Foundation Area
Contact with Soil and of the Mass of the Whole System. The moment

in

through

its

smaller than permissible (0.15

6.

inertia / of the foundation contact area with respect to the axis passing

center of gravity perpendicular to the plane of vibrations

Ir a

The moments of
same

respect to the

inertia of the

Xyy6

54.0

of

is

masses of separate elements of the system with

whose mass is considered to be con-

axis are: for the compressor,

centrated at the height of the shaft axis,

/oi

2.3 2 )

1.23

5.94

7.3 tons

X m X

sec 2

= W2 (1.5 +

2.3 2 )

0.41

7.55

3.1 tons

Xm X

sec 2

wi(0.8 2

for the electromotor,

J 02

for the foundation slab,

U, =

(a 3 , 2

a ?J)

4-

3 /i3

^~

(6.0 2

0.5 2 )

4- 2.02

0.25 2

6.1 tons

Xm X

sec 2

FOUNDATIONS UNDER RECIPROCATING ENGINES

(h 3

is

the distance between the center of gravity of the

153

mat and the foundation con-

tact area).

For the upper part of the foundation, located above the mat, the moment
from an analogous formula, is
OK

Q
/04

The
axis

total

= -j~

moment

(4.8 2

1.0 2 )

of inertia of the

3.25

X LO =

mass

of the

Xm X

9.8 tons

of inertia,

sec 2

whole system with respect to

this

is

Wo =
The moment

hi =

7.3

+3.1 +6.1

9.8

Xm X

26.3 tons

sec 2

=l

of inertia of the

whole system with respect to the axis passing through

the center of gravity of the whole system perpendicular to the plane of vibrations
I
since h

The

ratio

_ mh* =

1.09

26.3

6.91

1.09 2

18.2 tons

Xm X

is

sec 2

m.

between the moments of

inertia

is

18.2/26.3

0.69

7. Computation of Amplitudes of Forced Vibrations of a Foundation

Accompanied Simultaneously by Sliding and Rocking. The limiting natural

frequency of rocking vibrations of the foundation, according to Eq. (III-2-6),

w
The

10

103

Jn

2.5

.
2

The frequency equation

10 3

(20.5

6.5)

10 3

0.69

solving this equation

/m 2 =

18

c _
6.5

g-jn:

fn *

compute the

^
X

1Aa3 sec
10

_2

we
33.4

39.2

20.5
,

i~

10 3/n 2

_
"

6.5

0.69

193.0

10 6

find the natural frequencies of vibrations of the

10 3 sec" 2

/n2 2

5.8

10 3 sec" 2

coefficient A(co 2 ):

A(C0 2 )

is

for the foundation [Eq. (III-4-8)] is

or

We

X LOO _

67.5

limiting frequency of vibrations in shear, from Eq. (III-3-2),

fnx

By

54

is

= mlUnl* - " )(/n2 - CO

= 6.91 X 18.2(33.4 - 2.5) (5.8 = 13.8 X 10
2

2.5)

10 6

system

DYNAMICS OF BASES AND FOUNDATIONS

154

From

Eqs. (III-4-12) we determine the amplitudes of sliding shear and rotation of

The amplitude of sliding shear of the center of gravity of the whole

the foundation.

system

is

2.5

A, =
;v

The amplitude
A<p

10 3

of rotation

2.5

The maximum

10 3

18

1.09

= nniN/in-
0.016 X 10

4.6

m = nm
0.016 mm

is

18

n8

6.91

2.5

10 3

_ nnn ^
= 0009
X

10 9

..
_.
in
10~ 3 radians

horizontal displacement of the foundation surface in the plane yz

A = A y + hiA 9 = (0.016 +
= 0.020 X 10~ m ^ 0.02
3

0.41

0.009)

is

10~ 3

mm

The foregoing computations show that the amplitude

of horizontal vibrations, as

well as the amplitude of vertical vibrations, lies within the range of permissible values.

Hence the conclusion is possible that the dimensions of the foundation for the machine
under consideration were selected properly.
It is clear that in the case under review an increase in foundation height would lead
to greater amplitudes of vibrations; hence an increase in height would not only raise
the cost of the construction, but would also have a negative effect on the dynamic
condition of the foundation.

This conclusion holds for all cases in which the natural frequencies of a foundation
supported on soil are higher than the operational frequency of the engine mounted on
the foundation. This occurs in the overwhelming majority of reciprocating engines.

Example

Design Data.

1.

The
is

Dynamic

2.

analysis of a foundation for a reciprocating horizontal compressor

The

reciprocating horizontal compressor has two cylinders.

distance between the axis of the engine master shaft and the foundation surface

0.9

m.

The

Maximum

operational speed is 167 rpm.

values of unbalanced inertia forces of the engine are: horizontal com-

ponent in the direction of piston motion

Px

12.8 tons; vertical

component

P =
z

0.73 tons.

The foundation
kg/cm 2

value of 2

on a soil of medium strength having a permissible bearing

The design values of the coefficients of elasticity of the soil may

rests
.

be selected according to Table 1-8 as follows:

Coefficient of elastic uniform compression:
cu

Coefficient of elastic

4.0

10 3

tons/m 3

nonuniform compression:

Coefficient of elastic uniform shear:

cT

The dimensions

2.0

10 3 tons/m 3

of the foundation are not limited

or plant equipment.

by

structures,

communication

lines,

FOUNDATIONS UNDER RECIPROCATING ENGINES

2.

Selection of a Design Diagram for the Foundation.

155

The dimensions

of

the foundation are to be selected according to design considerations based on the

requirements of the plant management. Figure IV-8 gives the design diagram
Concrete type 100 f is to be employed
selected on the basis of these considerations.
for the foundation.
3. Centering of the Foundation Area in Contact with Soil and Determination of Pressure on the Soil. Let us determine the coordinates Xo, yo, and z of
-2210-

-960->

Section along longitudinal axis of the foundation

-1950-

1950

*-750>H-750

0.00
_5Z

-1.20

V.

_1
-2.00

'///////////'/s///

'///////St
-3000-

3000

6000Section along axis of main shaft

Fig. IV-8. Design diagram of foundation, example

2.

the center of gravity of the whole system (the foundation and engine) with respect to
the axes shown in Fig. IV-8:

Xnnyi

"SrriiXi

liiriiZi

yo

where m = masses
Xij y%, Zi

of separate elements of system

coordinates of centers of gravity of these elements with respect to
x, y, z

m =

mass

axes

complete system
Separate elements of the foundation are marked in Fig. IV-8 by the numbers
of

1, 2,

and

so on.

The foundation should be divided

magnitudes of masses and coordinates
f

into elements of such shape that the data on

of centers of gravity of separate elements

See footnote in Art. IV-l-c, p. 132.

may

DYNAMICS OF BASES AND FOUNDATIONS

156

moment of inertia of the mass of the whole system will be

be used
computed.
The data for the computation of coordinates of the center of gravity of the system
are given in Table IV-2.
Masses corresponding to cavities in the foundation are
shown with minus signs. From the data of Table IV-2 we find the coordinates of the
center of gravity of the system
later,

Xo

when

89.14

oi g

the

4- 16

64.52

==

01 E

3 -05

28.55

1.33

The displacement of the center of gravity of the mass of the system with respect to
the center of the foundation area in contact with soil is:
In the direction of the x axis:
4.16

4.00

0.16

3.05

3.00

0.05

In the direction of the y axis:

The

relative

magnitude

of the eccentricity in the direction of the x axis

i (\

--

which

is less

than 5 per cent.

100

The

is

2 per cent

eccentricity in the direction of the y axis

is

even

smaller.

Since the eccentricity in the mass distribution is small, its influence on the amplitudes of forced vibrations will be insignificant. Therefore we neglect hereafter the
eccentricity and consider that the center of gravity of the mass of the system and the
centroid of the foundation contact area are located on the same vertical line.
The pressure on the soil imposed by the static load is assumed to be uniformly distributed over the foundation contact area and equals

p Bt

Thus the

mg =
=
p

static pressure

21.5

on the

9.81

soil is

4.4

.
.. .
,
2
tons/m 2 = n
0.44 kg/cm

considerably smaller than the permissible

pressure.
4.

Computation of Amplitudes of Forced Vibrations of the Foundation.

Since the horizontal component of the unbalanced inertial forces of the engine in the
direction perpendicular to the motion of the piston is zero, and since the vertical

component

of the

above forces

is

insignificant,

we compute the amplitudes

of forced

by the horizontal component of

the direction of piston motion (the system will be sub
plane).
We also neglect the action of exciting moments

vibrations only for foundation vibrations caused

unbalanced

inertial forces in

jected to vibrations in the xz

tending to produce rocking vibrations of the foundation.

We begin by establishing the data needed for the computation of amplitudes of
foundation vibrations. The frequency of machine rotation equals

to

w2

= 0.105V =
= 300 see -2

0.105

167

17.3 sec" 1

^
=
N

N
o CO
o o
H N O CN

00 to CO * CO
OS t^.
CO

O O

W(D

*.

OO

CT>

11
|

i
=3

>>

H 00 C N 00
NN
IO 00 O 00 o o

and
gravit

IfS

center

o o o

betwee

nt

SogoJ
* a
m

a r
o o
d
o o
o

es

d
"
i

r]

OS
CO

tJH

a>

Q
<m

Sb

s?

-2

IS

.-i

>

.-I
ity,

CD
CD

CD
IO

i-h

o o o

i-i

cn

O
O kO
00 o o

00 00
>o iO

*}<

-s .S
'S "8

4)

8Q

ffi

<*-

o d

rj

.25

el
to

through
gravity

'.

of
Moment

3 o

CD

* h d||d 00
Nh
oo

iTa

tons

respect

mass

O o
03

49

iO O)
!D t(i O)
CN CO iO * CO i-h

iO
io

^tH (ON H H
^

CN

WN^hOhN
<0 H
M O

CN
tO

<t<

n
s

*>

(O

g S i

00

ci

g -x

-i-3

3)

% *

as
3

<S

48

1
"5

WO
O O

center

element

to

ij

"? ja

"g

08

.13

00 CONN *
N
0 O rH CN CO
H M
IO >o H
N >C
|| 00
"J

00

CM CO CO CO CO CO

<*

CO

sS
.

i-i

rt<

tO Tf

Tf
CO

t 00
00

axes,

of

HM

O bO O
h o
o O
o o o

spect

es

N CN
II

>0 >C OS O >0

O
Ol IN N H Tf N
O
N NO HN H H

*4

of

"^

O)

ID O) CO

co io

Na

CD
O
o O
o to
* ^ ^

T*<

Tj<

<*

<*

>o

IO
OS

03
oa

03

CO
CN

N 00N
O N
CO CN rH

|x

CN

-H

43

tJH

CT>

IO

C CO rH

CN

1
1

la

5a

_,

4)

i-O
"O
Oi IO CN

'(0

Q^

o
>0
N o

10HHH

O
O O
CO

CD
tCO "* CN

oo r- CO CN CN

l-l

a
" a
g
d * o
4)

O
"#
O

2 a
a

bCO

o
o
UJ UJ

'ohOhh

03
03
43

sys
(engin

Elem

of

'3

a-

d
3

id

Z>

CN

157

'
1

DYNAMICS OF BASES AND FOUNDATIONS

158

The
is

common center of
moment of the engine

distance from the axis of the master shaft of the engine to the

gravity of the mass of the system

hi

1.53

m.

12.8

1.53

is

The

exciting

then

= PJbx =

X m

19.6 tons

The moment of inertia of the foundation contact area with

through

its

respect to the axis passingcenter of gravity perpendicular to the plane of vibrations equals
-

The weight

whole system

of the

W
The moment

X
8.00 =
3

6.00

._.
256

,
4

is

= mg -

21.5

9.81

211 tons

mass of the whole system with respect to the axis

center of gravity perpendicular to the plane of vibra-

of inertia of the

common

passing through the

tions equals

Wm
The moment

=
=

z 0i 2 )
K22rrii(a X i* 4- a zi *) 4- m2(z 2
12.09 = 101.78
102 tons X
89.69

mass

of inertia of the

m X

sec 2

whole system with respect to the axis

of the

passing through the centroid of the foundation contact area perpendicular to the

plane of vibrations equals

Wo + Im = mh =
2

The

ratio

102

between the moments

7

21.5

1.33 2

of inertia of the

102

i4o

Xm X

140 tons

masses

sec 2

is

0.73

The limit value of natural frequency of rocking vibrations of the foundation

determined from Eq. (III-2-6):

M
The

x 2 se_ L3 3 x

103

21.1

^6 x

is

10S sec _ 2

limit value of the natural frequency of sliding shear vibrations,

from Eq.

(III-3-2), is

10'

48

4 4(

21.5

We

set

up the frequency equation

fn

14.6

10

4.46

_2

1Q 3 gec

of the foundation according to

10

/n

0.73

fn

14.6

26.0

10

4.46

10

/,,

89.0

/m.2 2

[13.0

V(13.0 2 -

(13.0

89.0)]10 3

Hence

Thus the natural frequencies

fnl = 21.9

8.9)10 3

of the foundation will

10 3 sec" 2

/ 2 2

10
10

0.73

Solving this equation,

/nl.2

Eq. (III-4-8):
3

be
4.1

10 3 sec" 2

FOUNDATIONS UNDER RECIPROCATING ENGINES

We

compute the

coefficient:

=mWm (fnl

A(co 2 )

159

21.5

We

" 2 )(/ 2 2

102(21.9

0.30) (4.1

0.30)

10 6

18.0

10 10

then compute the amplitudes of forced vibrations induced by the horizontal force
= PJir, according to Eqs. (III-4-11) and (III-4-12) the
the moment
horizontal displacement of the common center of gravity of the foundation and the

P x and by
engine

Ax

is

(8

10 3

256

211

From

the same equations

10 3

we

X 10 X 48 X
18.0 X 10 10
X 48 X 1.33 X 19.6
18.0 X 10 10

1.33

1.33

ni

102

vin _

17.3 2 )12.8

n ,=0.17

mm

find the amplitude of rocking vibrations of the founda-

tion about the horizontal axis passing through the center of gravity of the foundation

perpendicular to the plane of vibrations:

(2

~
=

0.019

10 3

0.17

The design value

1.33

(2

18.0

10 10

12.8

10 3

48

21.5

17.3 2 )19.6

10- 3 radians

Thus the amplitude

A =

48

of forced vibrations of the

10~ 3

of the

1.04

0.019

upper edge of the foundation equals

10~ 3

0.19 10" 3

m <

0.2

mm

amplitude of vibrations does not exceed the permissible

value; hence the dimensions of the foundation are selected correctly.

The foregoing computations show that vibrations

mainly by

component

its

of the

foundation are produced

horizontal displacement in the direction of the action of the horizontal

of the disturbing force of the compressor.

This

is

explained by the fact

that the dimensions of the foundation in the direction of the action of this force
large in comparison with the height of the foundation.

is

Therefore rocking results only

dynamic displacements.
Hence the rocking vibrations of a foundation may be neglected when computing the

in small

amplitude of forced vibrations

if

the foundation

is

elongated in the direction of action

of the horizontal exciting force; in this case the vibrations of the foundation

considered to be vibrations of sliding shear.

dynamic computations.

may be

This assumption greatly simplifies

However, this simplification of computations should be very cautiously applied.

For example, if such a simplification were made in the case of the foundation under
consideration, then from Eq. (III-3-3) we would obtain for the amplitude of horizontal
displacements of the foundation

21.5(4.46

The computed amplitude

0.30)

10 3

0.14

10- 3

m =

0.14

mm

26 per cent smaller than that obtained by means of the

This cannot be admitted as a good approximation.
The results of computations will be more accurate if we add to the amplitude of
is

foregoing computations (0.19

mm).

DYNAMICS OF BASES AND FOUNDATIONS

160

vibrations of sliding shear a displacement produced

by rocking vibrations

of the

foundation, computed from the formula

where A? is the amplitude of rocking vibrations when no shear is present, determined

from Eq. (III-2-7), and h is the full height of the foundation.
Assuming that the horizontal exciting force acts at height hi from the base of the
P x h we obtain
foundation, and assuming that, in Eq. (III-2-7),

Pxhhx
Wo{jn 2 ~

Ax

For the foundation under consideration, h

X
A" "
14oi? -^W) X10 =
2

Thus the

2.0

CO

x ,

2
)

and

- 38

hi

X1 "'

2.9

m "

m; consequently,

- 4

mm

total amplitude of the horizontal displacement of the foundation will equal

A = A x + A x<p =

0.14

0.04

0.18

mm

An amplitude computed by means of the above approximate method will not differ
much from the value obtained as a result of computations taking into account vibrations of the foundation

Methods

IV-5.
a.

accompanied by simultaneous

sliding shear

and rocking.

for Decreasing Vibrations of Existing Foundations

Counterbalancing of Exciting Loads Imposed by Engines. As stated

methods of balancing primary inertial

in Art. IV-2, there are different

forces

by means

of counterweights.

component in the direcmotion and partly a component in the direction of piston motion.
Or, the dimensions of counterweights and their
distances from axes of rotation may be selected to counterbalance completely the first harmonic of the component exciting forces in the direction
of piston motion.
Then the component in the perpendicular direction
It

is

possible to counterbalance completely a

tion perpendicular to piston

will increase.

Usually the

first

method

is

employed

engines because stresses in the engine

for the

counterbalancing of

itself are smaller than those occur-

ring when the other method is used.

Another advantage of the first
method is that it requires a smaller counterweight mass.
The efficiency of a certain method of counterbalancing the exciting
forces induced by an engine for the purpose of decreasing foundation
vibrations depends on the type of engine and on special features of the

foundation.

For a horizontal reciprocating engine, the most dangerous foundation

vibrations are those which are accompanied simultaneously

and

may

by rocking

In this case, a decrease in the vibrations of the foundation

be achieved by counterbalancing the inertial forces of the engine

sliding.

FOUNDATIONS UNDER RECIPROCATING ENGINES

by the second method, even

if

this leads to

161

some increase

in vertical

such an engine was counterbalanced by means

of the first method but impermissible horizontal vibrations were observed
after the construction of the foundation, then counterbalancing by
means of the second method (i.e., by changing the character of counterTherefore,

vibrations.

may

balancing)

if

be recommended as one of the simplest measures to

decrease these vibrations.

In cases in which vertical vibrations of an impermissible amplitude are

present in systems with horizontal motors, the second method
able,

and the

first

method should be

Similarly, for a vertical motor, the

is

unsuit-

applied.

method

of counterbalancing selected

depends on the type of foundation vibrations

vertical, horizontal, or

rocking.

The

installation of counterweights for balancing a

motor does not

require dismantling or prolonged interruption of operation.

ruption

is

The

inter-

only for the time needed to attach the counterweight to the

sides of the crank.

b.

Chemical Stabilization of

Soils.

If

a foundation rests on sandy

soil,

then, in order to decrease vibrations, chemical or cement stabilization

of the soil

under the foundation may be used. Such soil stabilization

an increase in the rigidity of the base and consequently in

will result in

an increase

in the

natural frequencies of the foundation.

Therefore

very effective when natural frequencies of the foundation

on a nonstabilized soil are higher than the operational frequency of the
engine which is usually the case. An increase in rigidity will increase
still further the difference between the frequency of natural vibrations
and the frequency of the engine; consequently the amplitudes of foundathis

method

is

When a foundation resting on a natural

has natural frequencies smaller than the operational frequency of the
engine, then soil stabilization may cause an increase in the amplitudes
This may be undesirable if a soil is stabilized to such a
of vibrations.
degree that frequencies of natural vibrations of the foundation merely
tion vibrations will decrease.
soil

approach the operational frequency. But if a soil is thoroughly stabilized

and natural frequencies of the foundation became much higher than the
operational frequency of the engine, then such soil stabilization may
result in a considerable decrease in amplitudes of vibrations.
Chemical and cement stabilization of soils is economically advantageous, since its costs are low in comparison, for example, with structural
measures. The principal advantage of this method lies in the fact that
it can be applied without a prolonged interruption in the work of the
engine.

The

nected with

interruption

is

soil stabilization

over-all result

is

only for the period of direct work conThus the

for 2 to 3 days more.

and then

that the engine will be inactive only for a few days.

DYNAMICS OF BASES AND FOUNDATIONS

162

The limits of the stabilized zones of soil and their shape are determined
by the character of the vibrations. If, for example, a foundation is subjected mainly to rocking vibrations about an axis passing through the
centroid of the base contact area, then it suffices to stabilize the soil
near the foundation edges, perpendicular to the plane of vibrations, and
it is not necessary to stabilize the soil under the entire foundation.
The
depth of the stabilized zone should be no less than 1 to 2 m.

This method of decreasing vibrations was applied at one of the Soviet

when it became necessary to decrease the amplitudes of vibrations
of an operating horizontal compressor without a long interruption in its
work. Soil was stabilized to a depth of about 1.0 m; the zone extended

plants

cm beyond the foundation edges. The results of foundameasurements before and after stabilization showed that
the amplitudes of vibrations, on the average, decreased by 50 per cent.
The work of the compressor was stopped only for the period of injection
of the silicates; the engine was set in motion immediately after silicatization was completed.
It can therefore be assumed that when the compressor renewed its motion, the stabilized soil had not as yet formed a
sufficiently rigid base, and it is possible that foundation vibrations acted
unfavorably on the stabilized zone of soil, which had not fully hardened.
The use of structural measures for decreasing
c. Structural Measures.
horizontally 30
tion vibration

foundation vibrations often requires a long interruption in the engine's

operation and considerable expense of funds and materials. Therefore
the use of this method may be suggested only in cases in which for some

may be applied. At the same time, it should

be noted that the correct change in foundation design may prove very
effective in decreasing the amplitude of vibrations.
Structural measures are applied with the purpose of changing the
reason no other methods

natural frequencies of a foundation in such a

way

as to achieve the largest

between them and the operational frequency of the

engine.
The choice of structural measure depends on the nature of the
vibrations and the interrelationships between the frequencies of natural
and forced vibrations. The operational frequencies of reciprocating
engines are usually lower than the fundamental frequencies of foundations; therefore most of the structural measures are directed towards
possible difference

increasing

still

further the natural frequencies of the foundation.

This

achieved by increasing the foundation contact area and its moments

of inertia, as well as by increasing the rigidity of its base by means of piles.
In addition, it is possible to increase the foundation mass without
inducing changes in the frequency of foundation vibrations. This results

is

amplitudes of vertical vibrations.

check calculations of the natural frequencies of a vibrating
foundation show that they are lower than the operational frequencies of
in a decrease in the

When

FOUNDATIONS UNDER RECIPROCATING ENGINES

163

the engine, an enlargement of the foundation contact area or an increase

may not decrease the amplitudes of vibrations,

in the soil rigidity not only

but

may

even increase them.

In this case,

it is

better to decrease

still

more the natural frequency of the foundation. This may be achieved

by enlarging the foundation mass without an increase in its area in contact with

The

soil.

depends on local conFor example, if a vibrating foundation lies close to another

foundation, it can be attached to the latter. As an illustration we will
selection of particular structural measures

ditions.

Fig.

IV-9.

Machine foundation which developed horizontal vibrations

of

high

amplitude.

describe here the structural measures which were applied to a foundation

under a horizontal compressor in order to decrease

The

horizontal

system was 30 tons.

with clay laminae.

The foundation
area 7 by 8

about 8 m.

2
,

its

vibrations.

induced by the whole

The foundation rested on a medium-grained sand

component

of the exciting forces

consisted of a block about 4.6

placed on 55 situ-cast

piles.

high, with a base

The length

of the piles

was

Figure IV-9 shows a cross section of the foundation.

Horizontal vibrations of extremely large amplitude (around 0.9 mm)

were observed while the engine was in operation. At the same time,
there occurred settlement of the basement of an adjoining structure, under
which no piles were provided. On the side nearest the foundation under
discussion, settlement of the basement reached 70 mm.
It appears that
this considerable settlement of the basement was caused by vibrations
transmitted from the foundation. These vibrations furthered the loosen-

DYNAMICS OF BASES AND FOUNDATIONS

164

ing of the soil and the carrying away of soil particles from beneath the
foundation and basement by ground water. When reinforcement of the
foundation was started, it was found that soil under it was washed out
or had subsided to a depth of about 0.5 m.
No damage was found in
the foundation block.

The ground-water

level

was approximately

1.5

above the

level of

the foundation base.

Reinforcement
decreasing

its

of the

foundation was undertaken with the purpose of

After the ground-water level had been arti-

vibrations.

New

piles

Old piles

Fig. IV-10. Reinforcement of foundation

shown

in Fig. IV-9.

soil was removed from beneath the foundation to a depth

Thus an excavation of a total depth of about 1.25 m was
formed. This excavation was filled with concrete. The foundation
area in contact with the soil was extended on all sides and 33 new situ-

ficially

lowered,

of 0.75

m.

cast piles were installed.

mended

for reinforcement.

Figure IV-10 illustrates the measures recomReinforcement of the new part of the founda-

tion provided a good connection with the old part.

To avoid

settlement

under the building walls due to excavation of the soil,

sheetpiling and chemical stabilization of soil were used beneath the footings under the walls.
This reinforcement of the foundation was very effective. The amplitudes of foundation vibrations decreased from 0.9 to 0.05 mm, i.e.,
of the footings

18 times.

This decrease in amplitudes was caused by considerable increase in the

natural frequencies of foundation vibrations due to an increase in the

FOUNDATIONS UNDER RECIPROCATING ENGINES

165

foundation area in contact with the soil, as well as to an increase in the

moment of inertia of the contact area. In addition, a considerable effect
was produced by the extra foundation mass and the increase in the
rigidity of the base due to the installation of supplementary piles.
To facilitate the application of various structural measures for decreas-

recommended

in doubtful cases to leave

be used, if necessary, for the attachment of an additional mass to the foundation or for the extension of its
These measures, of course, should be applied
area in contact with soil.

ing foundation vibrations,

it is

projecting reinforcement which

only

after

recognizing

that the foundation

vibrations

of

an

is

the

may

fact

undergoing

px

s'mfl *

impermissible
]

^A

magnitude.

The use

By means

tions.

of special slabs, first pro-

posed by Professor N. P. Pavliuk

and Engineer A. D. Kondin^ ma y
be considered as a structural measure to decrease foundation vibra-

iyn

Analysis of the effect of

taching a slab

of these slabs, it is possible in

to

an engine foundation,

some cases

to decrease

the amplitudes of rocking and horizontal vibrations of foundations.

Let us assume (Fig. IV-11) that slab A, resting on soil, is attached to a

foundation undergoing rocking vibrations around the axis passing through
the centroid of the foundation area in contact with soil.
Let us set up
the equation of forced rocking vibrations of the foundation.

The

follow-

ing symbols will be used:

Wo = moment

PT

sin ut

of inertia of foundation mass and of mass of engine

with respect to axis of vibrations
moment of inertia of foundation area in contact with soil,
with respect to same axis
magnitude of horizontal exciting force induced by engine
and transmitted to foundation, where co = frequency of

engine rotation
distance between line of action of exciting force and foundation contact area

distance between center of mass of foundation and engine,

and foundation contact area

mi = mass of attached slab

A =
i

contact area of attached slab

Hi = distance between place of connection of foundation with

attached slab, and foundation contact area

W = foundation weight
cv

cr

coefficients of elastic

nonuniform compression, shear

of soil.


DYNAMICS OF BASES AND FOUNDATIONS

166

The

equation

differential

forced

of

vibrations

of

the

foundation

together with the attached slab will be as follows:

(Wo

miflV)^

- Wh + H^CrAJv = P T H

(c+I

(IV-5-1)

sin ut

From this we obtain in the usual way the expression for the natural
frequency of rocking vibrations of the foundation with attached slab
:

J n <pl

The amplitudes

Cyl

- Wh +
~

H^CrAx
ff

(IV -0-4/

of rocking vibrations of the

foundation will be found

from the equation

A<pl

P TH

m^XW -

(Wo

(IV " 5 ' 3)

co*)

Equation (IV-5-2) shows that under certain conditions an attached

may have no effect on the natural frequency jntp of rocking vibrations

slab

of the foundation.

To determine

lowing expression:

these conditions, let us use the fol-

U1 > U
2

or since

fnip

(IV-5-4)

cj - Wh

we

substitute into the left-hand part of (IV-5-4) the expression for fntp i 2
because of
from Eq. (IV-5-2); then, neglecting the term containing

Wh

its

smallness,

we obtain
IFo

m^
CrA_i

Hence

mi

or

fnxl *

Thus

if

- Wl
>
>

(1V ' 5_5)

c^I

W
/n ,

one selects the attached slab so that the frequency of

(IV-5-6)

its

natural

vibrations of pure shear equals the frequency of rocking vibrations of the

foundation, then the attached slab will have no effect on the magnitude

frequency of natural vibrations of the foundation. Besides, the

amplitude of forced vibrations of the foundation will decrease according
of the

to the ratio

A 9l
Av
where

^i#i
= -^=
Wo

1
1
2

(IV-5-7)
5

FOUNDATIONS UNDER RECIPROCATING ENGINES

co;

Usually in foundations under engines, fn<p i

therefore, approximately,

A*i

is

167

considerably larger than

PtH
cJ

+ HSc A
T

To determine

the amplitude of foundation vibrations before the slab

co and the value of
in comparison with c^I is
fn<p
small, we have from Eq. (IV-5-3)

is

attached,

if

WH

PtH

A "~

c vI

cJ

Hence
c^I

A.M

Hi 2 c T Ai

or
1

(IV-5-8)

HftrAx/cJ

It follows from Eq. (IV-5-8) that the effect of the attached slab on
the decrease in foundation vibrations will be proportional to the frequency

fnx of natural vibrations of shear of the slab and proportional to the height
Hi of the slab above the foundation
^ sIn 0Jt
base.
The foundation area in contact with soil should always be as
large as possible; the contact area of

the slab

ditions

and

tions.

is

Since

limited

local con-

economic considerathe

value

depends also on the

the elastic shear of

foundation

by

may

of

fnxl

TTii

C
C\

coefficient of
soil c T ,

sl

Fig. IV-12. Analysis of the effect of attaching dampers

2 /2 to an engine foun-

pile

be installed under

dation

Wli.

the slab to increase fnx \ as much

as possible.
Short frictional reinforced-concrete piles should be used.

Professor N. P. Pavliuk and A. D.

reinforced-concrete slab

Kondin describe a

case in which a

was used

to decrease the vibrations of a foun-

By

installation of the slab, foundation

dation under a compressor.

vibrations were practically reduced to zero.

d.

Dynamic Vibration Dampers.

Consider a foundation with mass mi

induced

(Fig. IV-12) subjected to the action of external exciting forces

by an engine and producing only vertical vibrations. Let us assume that

two masses, each equaling m 2 /2, are attached to this foundation by
means of elastic ties (elastic rods, springs, etc.). Then the foundation
with attached masses will have not one but two degrees of freedom.
Formulas from the theory of vibrations of a system with two degrees of

DYNAMICS OF BASES AND FOUNDATIONS

168

freedom

may

be directly applied when considering the forced vibrations

of the system.

For example, for the amplitudes A\ and A 2 of forced vibrations of the

foundation and the attached masses, respectively, we have:

/22

/r-

fna *

where fna
Co

=
=

S&)
Also,

CO

IV " 5 " 9 )

= SL

natural frequency of vibrations of attached masses

coefficient of elastic rigidity of tie

between these masses and

foundation

= magnitude

of exciting force per unit of foundation

A(co)

where

/i

Unl

C0

)(/n2

(IV-5-10)

CO )

and fn % are natural fundamental frequencies

mass

of the foundation

with dampers.
It follows

from the first of Eqs. (IV-5-9) that the amplitude

becomes zero when

of

founda-

tion vibrations

U2
i.e.,

when

=fn J

(IV-5-11)

the frequency of natural vibrations of the attached masses

equals the frequency of the exciting force.

In order to determine the

amplitude of vibrations of the damper masses, we substitute / 22 into the
right-hand part of Eq. (IV-5-10) in place of co; then

mi
Substituting this expression for A(co 2 ) into the second of Eqs. (IV-5-9),

we obtain

the following expression for the amplitude of vibrations of

the damper:

Since

it

follows that

v =

A =
2

^
c2

(IV-5-12)

FOUNDATIONS UNDER RECIPROCATING ENGINES

169

Thus, the amplitude of vibrations of the damper equals its static

produced by a force of magnitude equal to the maximum
value of the exciting force Pit).
Equations (IV-5-11) and (IV-5-12) determine the selection of the
damper. It should be noted that neither the frequencies of the damper
nor the amplitudes of its vibrations depend on the properties of the soil
deflection as

base or the mass of the foundation.

It follows

from Eq. (IV-5-9) that theoretically

it is

possible to

damp

vibrations of infinitely large foundations subjected to the action of

periodic exciting forces by attaching dampers to these foundations, even
dampers with small masses. However, the smaller the mass of the
damper, the smaller should be the rigidity c 2 of its elastic tie with the
foundation, and consequently the larger will be the amplitude of its
vibrations.

At values of c 2 smaller than a certain limit, amplitudes of vibrations

damper may attain magnitudes endangering its strength. Therefore the minimum value of the damper mass is limited by permissible
values of stresses in the elastic tie between the damper and the foundation.
It has already been mentioned that when dampers are used, the founof the

dation has not one, but two natural frequencies of vibrations, determined
as roots of the equation

/n

where

The

[fm

/ z

(1

mi

rifnaVn

d =

+ fntfna

cu

(IV-5-13)

roots of Eq. (IV-5-13) are:

/nl,2

where fnz
fna

2
liifnS +/na (l

/i)

Vlfnz +/n
2

(l

m)

2 2
]

4/n2 2/na

2
)

= frequency of natural vertical vibrations of foundation

= = average operational machine rotation
= m 2 /nii = ratio between masses of dampers and foundation
co

mass

When

damper

is installed, fni will be larger than both fnz and co, and
be smaller than these frequencies. Besides, either /ni or fn2 will
lie close to u, and the other will be close to fnz
Let us assume that fnz > co. Then the lower fundamental frequency
If the engine
/ 2 will be close to co; the higher one, fnh will be close to/n *.

fn2 will

has varying angular frequency, then, with the installation of the damper,
the danger arises that one of the values of co will coincide with fn2 i.e.,
that resonance will occur with the lower frequency of the system "founda,

tion

and damper."

In this case, the amplitudes of foundation vibrations

DYNAMICS OF BASES AND FOUNDATIONS

170

be sharply increased and the masses attached to the foundation

as dampers, but as intensifiers of the vibrations.
In order to avoid such an intensification of vibrations, ra 2 should be

will

work not

will

maximum

selected so that the

decrease in operational frequency of the

compared with the average value of this frequency, is smaller

than the difference between the average operational frequency of the
engine co av (equaling the natural frequency /2 2 of the damper) and the
engine, as

lower frequency fn2 of the foundation. Hence, the following condition

should be satisfied when the damper sizes are selected

<

Jn2

Or

y (,U
2

+fnJ(l

W m in"

- VlfnS + fna

/i)

Solving this inequality for

(l

M)P

fn Z %a'

<

0> mi n

and noting that

/jl

- M

we obtain
V-

>

2
2
/n 2 (w a v

W min 2 )

OJ min

Let us assume that the nonuniformity in engine speed

is

as follows:

CO a v

Hence

= w av (l

o> min

Substituting this expression for

inequality obtained for

co min

e)

into the right-hand part of the

ju,

where

^-

^>

(i

*y

e)

and we obtain:

f9

Wav

<

If

/3

<

1,

then

co

max

>

co av

(1

(l

~
t
(1 f
+
e)

The

e(2

e)

(IV-5-14)

show that the selection of a proper interbetween the damper mass and the foundation mass depends

inequalities obtained

relationship

not only on the values of the irregularity in the engine speed, but also
on the interrelationship between the natural frequency of the foundation

and the average operational speed

If

of the engine.
is higher than the operational
then the damper mass should be

the frequency of natural vibrations

frequency of the engine

(i.e.,

/?

>

1),

FOUNDATIONS UNDER RECIPROCATING ENGINES

selected in proportion to the value of .

with an increase in

If

/?.

The nonuniformity

/?

1,

fi

If

(3

<

should have

1,

171

the value of

its

y.

decreases

lowest value.

in the speed of reciprocating engines lies in the

range from 0.01 to 0.10. The most uniform speed is observed in multicylinder diesels with flywheel, where e is about 0.01 to 0.02; in saw frames e
is

0.05 to 0.10.

The frequencies

of natural vertical vibrations of foundations are usually

than the operational frequencies of low-speed

higher
engines,

i.e.,

usually

/?

>

1.

With

reciprocating

variations in the angular speed of

may develop resonance with the lower fre"foundation-damper" system. If one assumes that the
average smallest irregularity in engine speed is around 0.075, then in
order to avoid resonance, it is necessary that the lower frequency of the
system differ by at least 3 per cent from the average operational speed
Thus, in calculations of the smallest value of n, the design
of the engine.
value of e should be taken not less than 0.03. For this value of e, and
with |8 = 1.3, the value y = 0.05; if ft = 1.6, then the damper mass
should be about one-tenth the foundation mass. If the natural frequency
of vertical vibrations is two times larger than the operational frequency
of the engine, the damper mass should not be less than 20 per cent of the
foundation mass.
The weights of foundations under reciprocating engines may reach
For the previously mentioned values of e and (3
several hundred tons.
In practice it is
the damper weight will equal several tens of tons.
difficult to attach a mass of this size elastically to the foundation so that
the frequency of natural vibrations of this mass corresponds exactly to
For example, for a foundathe average operational speed of the engine.
tion weighing 300 tons, for e = 0.03 and ^ = 1.5, the damper should
weigh not less than 27 tons; for e = 0.05 it should weigh 45 tons. The
difficulty in attaching such blocks to the foundation limits the use of
dynamic dampers even for machines with uniform speed. It is out of
the question for such machines as saw frames, in which the irregularity
the engine, the foundation

quency

of the

in speed attains 0.1.

For high-frequency engines, such as turbogenerators and electromotors

with small irregularity in speed, the employment of dynamic dampers
may be effective because of the low value of e and because usually /? < 1
for their foundations.

By

introducing damping into the system of the damper,

it is

possible

and decrease the amplitude of foundation vibrations when go approaches /n2

However, these effects will take
place only for some optimum values of damping.
In order to achieve in
practice these optimum values, the damper should be thoroughly tuned
up.
The maintenance of constant damping is especially difficult under

to increase the difference

co av

/2

DYNAMICS OF BASES AND FOUNDATIONS

172

working conditions. Temperature, moisture, and contamination may

a damper's natural period and its damping, thus upsetting its

affect

tuning.

When

foundations undergo vibrations close to those of pure sliding

the above interrelationships will be valid, except that fnx
should be inserted everywhere in the equations instead of fnz
They are
shear,

all

valid also

when foundation

vibrations are close to being pure rotational

vibrations around an axis passing through the centroid of the foundation

area in contact with

soil.

In this case, ^ designates the ratio between the

4
6
Number of

10

12

14

16

vibrations per second (hertz)

Fig. IV-13. Experimentally determined shift of resonance peak as a result of the use
of a vibration damper.

moment

of inertia of the damper masses with respect to the axis passing

through the center of gravity of the foundation perpendicular to the plane
of vibrations and the moment of inertia of both the foundation and engine
mass with respect to the same axis.
Let us note in conclusion that the author and his associates investigated experimentally the effects of dampers on model foundations
These experiments confirmed the fundamental theoretical conclusions.
For example, the experiments verified that after the damper is installed,
one of the resonances of the newly formed "foundation-damper" system
(Fig. IV-13) appears close to the operational frequency of the engine.
This resonance is dangerous even at negligible changes in frequency of

engine rotation.
IV-6. Analysis

and Design of Foundations with Vibration Absorbers

In some cases much lower than usual permissible amplitude values

machine foundation vibrations are necessary. It is very difficult

of

to

FOUNDATIONS UNDER RECIPROCATING ENGINES

means

decrease these amplitudes by

173

of proper selection of the

mass or the

foundation contact area or by increasing the rigidity of the base.

However, the amplitudes of foundation vibrations under reciprocating
engines may be considerably decreased by means of special spring
absorbers.

These absorbers are comparatively inexpensive, reliable in operation,

and effective in decreasing the amplitudes of forced vibrations of foundaAbsorbers considerably decrease vibrations produced not only by
tions.
the main (first) harmonics, but also by higher harmonics of exciting loads,
as well as by loads developing as a result of various factors not taken into
account by design computations.
Therefore spring absorbers are sometimes used to decrease vibrations of
4 ~^-\
XL
y
machine foundations having unbalsecond
harmonics.
This
anced
is
2>,
done in order to eliminate completely
i

the transmission of the inevitable

7^

vibrations to adjacent structures and

especially to

measuring

6^^|

equipment and precision

instruments.

Human

/OL

y/A v/////:

beings feel vibrations of even very

small amplitudes.
vibrations interfere

Fig. IV-14. Small one-spring vibration

Sometimes small
with the work of

absorber.

precise devices or are the reason for undesirable distortions in various

technological processes (for example, in the operation of precision devices,

in molding, etc.).

There are various types of absorbers employed, depending on the type

machine to which they will be attached, on the static load transmitted
to the absorber, and on special requirements in regard to assembling and

of

adjusting.

Figure IV-14 shows a sketch of a small one-spring absorber used for

small engines producing no

This absorber consists of a

The

coil

considerable

spring

which

unbalanced exciting loads.

fits

into the adjusting slab 2.

regulating bolt 3 rests on this slab.

The frame 4

5 of the absorber.

The

is

regulating bolt 3.

To

position of the frame

is

placed on the

lid

adjusted by turning the

eliminate harmful external effects on the spring,

6, having insulating pads 7 (made

rubber or cork) which protect the spring from water and dirt.
Such light absorbers are used for vibroisolation of small diesel engines,
ventilation units, presses, pumps, and other machines.
For the vibroisolation of reciprocating engines of medium and high
capacities, absorbers containing several springs are used.
An absorber of
The
this type with four springs is shown schematically in Fig. IV-15.

the latter
of

is

enclosed in the housing

DYNAMICS OF BASES AND FOUNDATIONS

174

housing and parts of the absorber are

made

of steel plates

and several

other metals.

The main

characteristics of the springs

(i.e.,

the diameters and

number

dynamic computations.
In addition to spring absorbers, rubber absorbers may also be employed
for vibroisolation of light engines and devices.
In comparison with
spring absorbers, rubber absorbers are simpler and less expensive.
Besides, they are characterized by a larger coefficient of resistance to
of coils) are selected according to the results of

vibrations,

useful

when they

are used for vibroisolating machines of

Fig. IV- 15. Four-spring vibration absorber.

irregular performance.

tion in their

modulus

disadvantage of rubber absorbers

of elasticity,

which depends on the

load.

is

the varia-

Computa-

tions related to vibroisolation always involve relatively large errors

rubber

is

if

used.

Depending on the balance

of the engine

and

its

operational speed,

arrangements may be used for the vibroisolation of foundations

by spring absorbers. Fundamentally these arrangements can be reduced
to two types: supporting and suspension springs.
When designing the vibroisolation of a foundation for a high-speed
engine (more than 300 rpm) which is relatively well balanced, i.e., no
first harmonics of exciting loads are present, it is not necessary to provide
It may be designed as a reina heavy foundation above the springs.
In this case, a
forced-concrete slab of comparatively small thickness.
in
which
the absorbers
"supported" type of vibroisolation is employed,
IV- 16
Figures
are installed directly under the mass above the springs.
IV-16
shows
Figure
and IV-17 illustrate vibroisolations of this type.
the arrangement of absorbers employed for the vibroisolation of a sixHere
cylinder diesel engine operating on one shaft with a generator.
rolled
of
made
frame
metal
under
a
the absorbers are installed directly
steel shapes, used instead of the cast-iron frame of the diesel and gendifferent

The mass above the springs consists here of the mass of the
motor with the generator and the supporting frame; no portion of the
foundation lies above the springs. This arrangement of absorbers can
erator.

FOUNDATIONS UNDER RECIPROCATING ENGINES

175

be used only for high-speed engines. The supporting frame should be

very rigid to avoid the harmful effects of its deformations on the connector
of the shaft.

nnnnrn
annnnn
t

x
^

.gf^i.

Fig. IV-16. Supporting-spring type of vibroisolation of a six-cylinder diesel with

generator on the same shaft.

Figure IV-17 shows the "supported" vibroisolation of a high-speed twocylinder diesel engine having unbalanced

first

harmonics

of exciting loads.

In this case, the mass above the springs was increased by means of a
special thick reinforced-concrete slab under which the absorbers were

They rest on supporting slabs which also support rolled steel

beams embedded in the lower part of the foundation above the springs.
placed.

4
r^f.

-^v-

Fig. IV-17. Vibroisolation of high-speed two-cylinder diesel with generator on the

same

shaft.

System has unbalanced

first

harmonics of exciting loads.

DYNAMICS OF BASES AND FOUNDATIONS

176

rigid

coupling between the absorbers and the beams

bolting a

beam

in the concrete

The

to the cover of each absorber.

formed

by-

above the springs to permit access to the absorbers.

installation

of

successively as follows:
creted.

is

Special cavities are left

Usually

it

vibroisolation
first

of

the supported type proceeds

the foundation beneath the springs

is

con-

consists of a reinforced-concrete slab with a thickness

m, depending on the type and size of the engine and on soil

After the concrete has hardened, the surface of the slab is
covered with Ruberoid, tar paper, or plywood, on which the lower slabs of
the absorbers are placed in the proper order. Above these slabs a
of 0.20 to 1.00

properties.

beams is installed. Then the form work

prepared for the concrete of the foundation above the springs; cavities
should be left for each absorber. Then the concrete is poured. Due to
the presence of the layer of Ruberoid (or tar paper or plywood), the
concrete of the upper part of the foundation will not bind to the concrete
beneath the springs.
After the concrete of the upper part of the foundation has hardened,
prefabricated frame of rolled steel
is

the absorbers are mounted.

The

springs are placed on the lower slabs of

the absorbers and are covered by the upper supporting slabs, which are

bolted to girders.

permit

Finally, the restraining anchor bolt

lifting of the

regulated by

means

mass above the

of a level.

If

springs.

The

is

installed to

lifting is carefully

the absorbers are installed correctly,

the lifting and regulation of the mass above the springs do not take

much

time and do not involve any difficulties.

The vibroisolation of low-frequency engines by means of absorbers
leads to the necessity for providing a heavy foundation above the springs.
Therefore, if the foundation is placed directly on the absorbers, the latter
should be installed at a level considerably lower than the floor of the shop.
This hinders access to the absorbers and their mounting, regulation, and

maintenance.
In such cases, absorbers of the "suspended" type are used. It is seen
from the sketch of an absorber of this type in Fig. IV- 18 that it differs
from the previously described 'supported" type only by the considerable
length of the restraining anchor bolt passing through the absorber.
Projections cantilevered from the body of the foundation above the
springs are attached by girders to the lower end of the restraining anchor
The absorbers are placed on the upper edges of the foundation
bolt.
mass below the springs. This mass is designed in the shape of a box in
'

which the mass located above the springs is inserted.

The procedure for mounting and regulating absorbers of the suspended
type does not differ much from that used for the supported type.
Generally the dynamic computations of a foundation with absorbers
are reduced to an investigation of vibrations of a system having up to

FOUNDATIONS UNDER RECIPROCATING ENGINES

However,

12 degrees of freedom.

177

since absorbers are mostly used for the

vibroisolation of engines with vertical cylinders, the analysis

many

cases be limited to

an investigation

Then
of a

the problem of foundation vibrations

system with 2 degrees of freedom.

"3ZZZZ

is

may

in

of vertical vibrations only.

reduced to an investigation

rzzzzz>

s
Fig. IV-18. Suspended-type absorber.

Let us assume that the masses of the foundation above and beneath the
springs are concentrated in their centers of gravity, located on the

same

Let us further assume that an exciting force P(t) sin oot acts
on the mass m 2 above the springs (Fig. IV-19). The differential equations of forced vertical vibrations of the system under consideration will
be as follows:

vertical line.

m{i\

where P(t)
co

Z\,Z2

Ci

c\Z\

c 2 (z2

tn^2

c 2 (z2

zi)
zi)

=
=

(IV-6-1)
P(t) sin ut

= magnitude of exciting force

= frequency of exciting force
= vertical displacements of centers of gravity

and above springs mi,

ra 2

coefficient

rigidity

of

elastic

of

of

masses below

base under foundation

beneath springs
Cl

cu

(IV-6-2)

DYNAMICS OF BASES AND FOUNDATIONS

178

A
c =
cu

coefficient of elastic uniform compression of base

area of foundation beneath springs, in contact with

n n 2d

C;

=
=
=
=
=
=

Wi

n2
n
er

D
G

soil

total coefficient of rigidity of all springs

number
number
number

(IV-6-3)

of springs in each absorber

of absorbers

each

of coils in

spring-

diameter of spring
diameter of coil

modulus

of elasticity of material of springs

Each spring should be designed

so that stresses developed therein under

the action of static and dynamic loads will not exceed

a permissible value.

P sin cot

Limiting our discussion to forced vibrations only,

we take the

solution of the system (IV-6-1) in the

form

A\

Zx

sin wt

z2

= A

sin

cot

where the amplitudes A\ and A 2 of forced vibrations

of the foundation beneath and above the springs are

f i2
Jnl
x

At

miA(co 2 )

(1

+ #.)/*' + Mf -

Fig.

IV-19.

Deri-

(IV-6-1)

concern-

the

vibration

ing

of foundation
masses above and

where fni

rt

'

m2A(co *)

777777777?

vation of Eq.

(IV-6-4)

p(t)

*>

(Iv.g. 5)

the limiting frequency of natural vibra-

is

above the springs, computed

on the basis of the assumption that the foundation
beneath the spring is infinitely large; the value of
jn is determined by the equation
tions of the foundation

below absorber
springs.

f ;
Jnl

(IV-6-6)

fniz is

the limiting frequency of natural vibrations of the complete system

when

it is

assumed that no absorbers are used

fnu
Finally,

C]

mi

= m

mi

+m

(IV-6-7)
2

FOUNDATIONS UNDER RECIPROCATING ENGINES

The

determined by the expression

coefficient A(co 2 ) is

A(>)

CO

179

(1

+ f 'W +

H)(jnl *

(1

nlz

(IV-6-8)

/O/.tf-b*

Returning to Eq. (IV-6-4), let us investigate the dependence of the

amplitude of forced vibrations of the foundation beneath the springs on
The value
fni, which is proportional to the rigidity c 2 of the absorbers.
of the exciting force is proportional to the square of the frequency of
engine rotation; therefore

P(t)

where 7

is

a coefficient which depends on parameters of the engine.

Substituting this expression for

(IV-6-4)

yo> 2

into the right-hand part of Eq.

and dividing the numerator and denominator by

co

2
,

we obtain

the following expression for the vibration amplitude of the foundation

beneath the springs:

Al =
where

& =

&

(u

CO

If

+),

(1

fc

(IV-6-9)

&*&*)

03

no absorbers are used and the upper and lower parts

of the

founda-

tion are rigidly connected, then according to Eq. (III-1-13) the amplitude
of vertical forced vibrations will equal

A =
*

mi n

(1

The degree

of absorption of vibrations will

Ai

(1

+m)*

(1

(IV-6-10)

- n
1)

wt 2
m)(^

be

+^ 2

+m)(^ - 1H
2

&&)

(IV-6-11)

Let us investigate the effect of changes in the value of 1 on the value

We assume that fni is very small in comparison with co; i.e., the
7],
value & is also very small.
It follows directly from Eq. (IV-6-11) that
of

If

& *

0,

then

7]

00

hence it follows that if the natural frequency of foundation vibrations

above the springs is small in comparison with the frequency of engine
rotation, then the amplitude Ai of the foundation with absorbers is small
in comparison with the amplitude of vibrations of the same foundation
without absorbers.
Suppose & > 00 this corresponds to a very large value of /?. One
;

DYNAMICS OF BASES AND FOUNDATIONS

180

can see from Eq. (IV-6-11) that in this case > 1 i.e., absorbers will not
have any influence on the amplitudes of foundation vibrations.
Figure IV-20 gives a graph of changes in
depending on changes in
rj

rj

It is evident that the absorbers cause a decrease in the

fi.

of vibrations only

+1.0

when

77

amplitudes

>

It is

seen from the graph that the zone of

usefulness of absorbers is limited to
a very narrow range of values of
lying between

(1

& =

and

~ \2(I

&

+ m)W - 1
+ /*)(&.* - 1)
(IV-6-12)

When

>

1,

the following inequality

exists

Diagram illustrating Eq.

and the limits of usefulness

Fig. IV-20.

(IV-6-12)

M
^

of vibration absorbers.

"^
r

Consequently, the use of absorbers

does not bring any advantage, but on the contrary is harmful, because
when 1/| < 1 the amplitude of vibrations of the foundation with absorbers
is larger than that of the same foundation without absorbers.
The foregoing theory of vibrations of foundations with absorbers leads
1

have a favorable effect

foundation vibrations, the following condition is

to the conclusion that in order for absorbers to

on the amplitudes

of

necessary: the frequency of natural vibrations of the mass above the

springs should be as small as possible in comparison with the frequency of

A decrease in the frequency of natural vibrations of the

foundation above the springs may be achieved by the use of absorbers
engine rotation.
of

stiffness and by an increase in the mass above the

For high-frequency engines the required relationship between
can be easily achieved without a considerable increase of the
the foundation above the springs.
For low-frequency engines

suitable

springs.

and fn
weight of

co

is usually difficult to achieve by just decreasing the

absorber because, due to strength requirements, this
decrease cannot extend below a certain limit determined by the strength
of the springs.
In such cases a decrease infn is achieved by providing a

the relationship
rigidity

of the

massive foundation above the springs.

If the degree of absorption of foundation vibrations
from Eq. (IV-6-11) we obtain for

77

is

specified,

then

z :

(1

(!+)(*-

ju)W

l)(fc,-

(IV-6-13)
1)

FOUNDATIONS UNDER RECIPROCATING ENGINES

Example.

Design computations for a foundation with absorbers under a

181

vertical

compressor

A foundation for a 120-k\v vertical compressor is to be designed. Foun1. Data.

dation vibrations are objectionable, since a precision apparatus is adjacent to the
foundation.
The base of the foundation is formed by a soil characterized by a coefficient of elastic uniform compression c u of 2 X 10 3 tons/m 3
In order to avoid harmful influences of the foundation on the apparatus, the amplitude of foundation vibrations should not exceed 0.03 mm.
The main exciting force imposed by the compressor is the vertical exciting force
P z = 2.6 tons; the compressor speed is 480 rpm.
In order to have an amplitude of foundation vibrations under
2. Computations.
the compressor smaller than 0.03 mm, the foundation should be very heavy and
should have a large area in contact with the soil.
For example, assuming that the ratio between the frequency of natural vertical
vibrations of the foundation and the engine frequency equals 2, for A z = 0.03 X 10~ 3
and P z = 2.6 tons, we obtain from Eq. (III-1-13) the following foundation weight:
.

w
W

2.6

_
~

0.03

Then the foundation area

10- 3

in contact

X 9.81
X 3 X 2.5 X
with

soil

_ U0
~

10 3

t0nS

should equal

_/w ,TT_ 4X2.5X10'X115 _ 5/5

575m2
m
A ~
"
"
c ug
2 X 10 X 9.81
3

For such a low-power engine as the compressor under consideration it would be

unreasonable to construct a foundation with weight and dimensions as large as
those obtained above. In order to meet the requirements in regard to the amplitude
of foundation vibrations, it is better to use spring absorbers.
The selected dimensions
of the foundation with absorbers are shown in Fig. IV-21.
The data for the computations are as follows: the foundation area in contact with
2
soil A = 12.5
the weight of the foundation beneath the springs is 21.0 tons;
the weight of the foundation above the springs (together with the engine) is 35.0
tons.
The coefficient of rigidity of the base is

ci

The mass

cu

A =

of the foundation

mi

The mass

21.0/9.81

12.5

35.0/9.81

25.0

10 3 tons/m

is

2.15 tons

above the springs

m =
The

10 3

beneath the springs

of the foundation

sec 2 /m

sec 2 /m

is

3.56 tons

limit natural frequency of vertical vibrations of the

whole system (assuming

that no absorbers are used) equals

u>
coefficient

&z

is

25.0

cx

mi

+m

computed

&*

2.15

to

10 3
4.38

+3.56

be

/ z

4.38
2.5

X 10 =
X 10
3

1.75

10 3

DYNAMICS OF BASES AND FOUNDATIONS

182
Absorbers-^

>;[

1
10
CM

"
/

\\

II

Si

^>CN14
CNIO^x Layer of Ruberoid or
\ impregnated cardboard-.
^

/I

1
:r_zz::jr:: -z^zzzdczzzzziuciiz

mm

C\J

*
\-

Section along

BB

'

F^l

11

^M#^#===#^4
ii

#^#^4^
ikJbmXimijJ
-100-

-100

/-H

100

400
J

Axes

Fig. IV-21.

The

ratio

Example

absorbers

of

Section along

AA-AA

of design computations for foundation with absorbers.

between values

masses

of the

iu

is

= 3.56
2.15

ra 2

mi

1.65

If no absorbers were present, then for the selected dimensions of the foundation
the amplitude of vertical vibrations would equal

2.6
(2.15

+3.56) (4.38

=
2.5)10 3

0.25

10~ 3

0.25

mm

In order that the permissible amplitude of vibrations of the foundation with

is not exceeded, the degree of absorption of vibrations should equal

absorbers

0.25

8.4

0.03

Let us assume that the design value of rj equals

determine the required value of the coefficient &:

tf

=
(1

1.65)

(1

10.

4- 1.65)1.75

(-10 -

1)(1.75

=
1)

From Eq.

0.165

(IV-6-13)

we

We

FOUNDATIONS UNDER RECIPROCATING ENGINES

183

determine the required value of the limiting frequency fn

of natural vertical

iz

vibrations of the mass of the foundation above the springs:

Uz 2 = &
The required

co

0.165

2.5

10 3

fniz

m =

414

3.56

414 sec" 2

be

total rigidity of all the absorbers will

ci

If c sp is

1,480

tons/m

the rigidity of one spring, then

Ci

Cup

7ll7i2

where n\ = number of springs in each absorber

n 2 = number of absorbers
Assume n = 2 and n 2 = 8; then the required

rigidity of

On

1,480

c*p

2X8

the other hand, using Eq. (IV-6-3),

no

one spring

will

be

tons/m

we obtain

.-%\

dV-6-14)

where G is the modulus of elasticity in shear of the spring material; its value may be
assumed to be 7.5 X 10 6 tons/m 2
We assume there are five coils in a spring. Substituting values of n and G into
.

formula (IV-6-14) we obtain

92.0

Hence

^f^

w> -

raffia

Let us assume the diameter of the spring d

Dna _
~
D =
3.

^
4.9

9.3

cm

10- 4

_
_

39.0
4.9

value

Rp
is

is

40

XUH*

1-88

2.5

10

"4

10 -2

X 1Q-* _
X 10-< "

8 '

m; then

10

permissible load on the spring equals

z
Rv
8D

where

4 9

The

Stress Analysis of the Spring.

ird

(IV-6-15)

the permissible torsional stress for the spring material;

10 3 tons/m 2

we assume

Assuming in accordance with the foregoing computations that d = 2.5

and D = 9.3 X 10 -2 m, we obtain for the permissible load on the spring

PD =

its

3.14

15.6

10- 6

9.3

X 40
10- 2

10 3

aA 4
2 64 t0nS
-

10~ 2

DYNAMICS OF BASES AND FOUNDATIONS

184

In order to find the actual load on each spring, it is necessary to determine the
amplitude of forced vibrations of the foundation above the spring. From Eq.
(IV-6-8), we determine the value of A(co 2 ):
A(co 2 )

X 10 - (1 + 1.65) (0.414 X 10 4- 4.38 X 10

X 2.5 X 10 + (1 + 1.65) X 0.414 X 10 X 4.38 X
6

6.25

10 3

10 3

21.0

10 6

According to (IV-6-5),
(1

1.65)

4.38

10 3
3.56

0.34

The dynamic

10- 3

+ 1.65 X 0.414 X
X 21.0 X 10

W-2

2.5

2.64

imposed on the springs as a

force

10 3

0.34

10~

1.48

result of vibrations

10

3.56

is

0.5 tons

Thus, the actual load on each spring equals

-Pact

which

is

0.5

35.0

^-x

16

00 tons
2.22
,

smaller than the permissible load.

Fig. IV-22. Detail of absorbers with suspension system.

A schematic diagram of the main part of the arrangement of the foundation with
suspended absorbers is shown in Fig. IV-22.
The construction of a foundation with absorbers proceeds analogously to the
procedure described in Art. V-7.

V
FOUNDATIONS FOR MACHINES
PRODUCING IMPACT LOADS

V-l.

General Directives for the Design of Foundations for

Forge Hammers

Forge hammers are divided into

a. Classification of Forge Hammers.
two groups: drop hammers for die stamping and forge hammers proper.
The side frame of a drop hammer is
mounted on an anvil (Fig. V-l), thus
giving rigidity to the system.

The

side

frame, together with guides for the ram,

precision of blows
This peculiarity in the
design of drop hammers predetermines to
a certain degree the design of their foundations, since the foundation block under the
anvil serves as a support for the whole

contributes

to

the

required in forging.

Frame

Ram

hammer.
Free forging operations are usually per-

formed by forge hammers proper. The

anvil and the side frame, as a rule, are
mounted separately. Forge hammers are

Anvil

built as single-support frames (Fig. V-2)

and as double-support frames

The

(Fig. V-3).

can be of the arch or bridge

Pneumatic hammers with air com-

latter

types.

hammers.
a Hammer Founda-

pressors are single-frame

b.

tion.

Design Data for

The

hammer

Fig. V-l.

Drop hammer with

frame mounted on

anvil.

following technological data are required for the design of a

foundation
185

DYNAMICS OF BASES AND FOUNDATIONS

186
1.

The nominal weight

hammer.

of

dropping parts, which usually characterizes

In drop hammers the real weight of dropping

parts, in addition to the weight of the ram, piston, and rod, includes also
the upper half of the die.
Therefore in these hammers the real weight of
the power of the

dropping parts

is

greater than the nominal weight or that

shown

in

catalogues.

The

design of a

hammer

Fig. V-2.

Hammer

foundation

The

weight of dropping parts.

made on

is

the basis of the real

total average height of the

with single support

frame.

Fig. V-3.

Hammer

upper and

with double support

frame.

lower halves of the die lies within the range 250 to 500 mm. It should be
noted that some dies for long units (axes, shafts, etc.) are provided with

heavy upper

parts, whose weight reaches 100 per cent of the nominal

weight of dropping parts. Such cases should be mentioned in the technological assignment.
2. The weight of the anvil and the dimensions of its base area.

3. The weight of the hammer, including side frames, ram cylinders

with the anchor plate, etc., but excluding the anvil weight. When no
data are available on the weight of the anvil and frames, it is permissible

to

assume the anvil weight

weight of the
weight of the ram.
total

4.

to be 15 to 20 times the weight of the

hammer and

The maximum height

anvil

of the

is

ram drop

(or the

stroke).
5.
6.

The upper piston area.

The average working pressure on

ram; the

taken to be 25 to 30 times the

the piston.

maximum

piston

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

187

For the design and analysis of the foundation, either the machine
assembly drawing or the following data should also be made available:
7. Dimensions in plan, the thickness and elevation of the top of the
pad under the frame and anvil of the forge hammer.
8. The position, diameter, and length of anchor bolts.
9. The elevation of the anvil base with respect to the floor level of the
shop.
10.

Dimensions

in plan

and the thickness

of the

wooden pad under the

anvil.
11. The location of the hammer in the shop with respect to adjacent
foundations under engines, machinery, and supporting structures of the
building; the dimensions, elevations, and depths of these foundations.

Foundations
c. Material of the Foundation and Pad under the Anvil.
under hammers with a weight of dropping parts in excess of 1 ton are
made of concrete type 150,f with coarse aggregate of hard rocks with a
Normal portland
compressive strength not less than 250 kg/cm 2
cement, of a type not below 300, is used for concrete. The latter is
reinforced according to design data or according to instructions given on
.

the job.

The pad under the anvil is usually made of oak. Experience in the
hammers under war conditions showed that for hammers
with a weight of dropping parts up to 2 tons, pine and larch timber may be
operation of

used as material for under-anvil pads. Timbers of the best quality,

having a moisture content below 15 to 18 per cent, should be used.
d. Directives for Design.
Until recently there was a tendency to design
foundations under hammers as massive blocks embedded to a considerable
depth in the soil. 44 The purpose was to provide such dimensions of the
foundation that its static elastic settlement would be larger than the
amplitude of its vertical vibrations. Since design values of amplitudes
of foundation vibrations under hammers were selected within the range
2.0 to 2.5 mm, the height of the foundation had to be increased considerably to obtain the desired static settlement.
Figure V-4a shows a
typical design of a foundation under a hammer, as used until recently.
Here for each 1 ton of dropping parts weight, there correspond 80 to 100
and often 120 tons of foundation weight.
The discussion which follows will show that the use of such heavy
foundations is not necessary, especially since they involve a considerable
increase in the cost of construction.

Foundations under hammers should be designed as blocks or slabs

loaded from above by backfill.
Figure V-46 is a typical design of such a
block foundation. The ratio between the weight of the foundation and
that of the dropping parts is about 40.
f

See footnote, Art. IV-l-c, p. 132.

DYNAMICS OF BASES AND FOUNDATIONS

188

In block-type foundations the thickness of the part below the anvil

should be selected as follows: for a ram weight up to 0.75 tons, the thickness should be not less than 0.75 m; for a weight of 0.75 to 2.5 tons, the
thickness should be 1.5 m; for a weight of 2.5 tons and more, the thickness

must be

1.25 to 2.50

m, depending on the power

Previously, foundations under forge

of the

hammer.

hammers were

anvils separated from the footings under the frames.

built under the

This decreased the

hammer frames during eccentric forging. However, the

separation of the foundation elements results in their tilting with respect
stresses in the

to each other

and

nonuniform settlement of the foundaIn recent years foundations under forge hammers

in considerable

tion under the anvil.

have been designed by the method shown

in Fig. V-4c;

i.e.,

the footings

Fig. V-4. Types of hammer foundations: (a) deep block foundation; (6) slab block
foundation weighted by backfill; (c) anvil block foundation 2 supports frame 1;
(d) single block foundation for anvil and frame.

under the frame were not placed directly on the soil, but on the block
under the anvil, and boards 2 to 3 cm thick (or several layers of Ruberoid)
were placed between these two footings. Thus tilting of the anvil with
respect to the frame was prevented.
However, due to insufficient elasticity of the pad, this foundation design was not very effective in decreasing the stresses which developed as a result of nonuniform forging.
These stresses may be decreased much more efficiently by means of
spring washers in foundation bolts and by oak timbers installed under the
In such cases the foundation
anvil and under the frame of the hammer.
under the forge hammer should be designed as a single block (Fig. V-4cT).
Reinforcement is placed as directed on the job. The reinforcement
used for the foundation under the anvil consists of 2 to 4 horizontal steel
grillages formed by 8- to 12-mm bars spaced at 10 to 20 cm; the upper
grillage is placed at a distance of 2 to 3 cm from the foundation surface.

Near the foundation

consists of

surface in contact with

soil,

the reinforcement

or 2 horizontal grillages formed by 12- to

20-mm

bars and

spaced 15 to 30 cm apart. Distances between the grillages are 10 to 15

cm in the part of the foundation under the anvil and 15 to 30 cm near the
foundation contact surface.

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

The number

of grillages is

189

determined by the power of the hammer.

It

should be kept in mind that double-acting hammers belong to the group

of heaviest hammers with respect to their impact effect on the foundation.

Pads under the anvil are made of square timbers from 10 by 10 cm to

20 by 20 cm in cross section. Timbers are laid flat in one or several rows,
one over the other. Each row is braced by transverse bolts every 0.5 to

m and forms a

separate mat.
rows of timbers are employed, then in order to decrease wear
and tear and make the pad more rigid, the timbers are placed in the form
The upper row of timbers is laid along the short side of the
of grillages.
1.0

If several

Table

V-l.

Thickness of Timber Pads under the Anvil

Thickness of pad, m, if
weight of dropping parts of hammer

Type

No.

of

hammer
hammer

Double-action drop

Single-action drop

Forge

anvil base.

is:

hammer

hammer

The mats must be

Up

to

Up
Up
Up

to

ton

1-3 tons

Over 3 tons

20

0.20-0.60
0.10-0.40
0.20-0.60

0.60-1.20
0.40-0.90
0.60-1.00

0.10
to 0.20
to

strictly horizontal,

easily fitted into the excavation.

smoothly planed, and

They should be checked by means

of a

water level. To prevent decay resulting from moisture, it is advisable to

impregnate timbers with wood-preserving solutions.
The space between the pad and the excavation walls may be filled with
petroleum asphalt. In order to prevent a horizontal displacement of the
anvil along the pad, four timbers are placed around it near the base.
The pad thickness is shown on the assembly drawings of the hammer or
indicated in the technical assignment.
Tentative values of pad thickness
under the anvil are given in Table V-l.
The thickness of the pad should be selected so that the stresses therein
do not exceed permissible values, which are as follows:

Oak:
Pine:

Larch:

300 to 350 kg/cm 2

200 to 250 kg/cm 2
150 to 200 kg/cm 2

e. Remarks on
Concrete for the footings
Construction Procedures.
should be placed using vibrators. In the presence of ground water
containing chemicals which may produce deterioration of concrete,
pozzolan cement should be used or special measures should be provided

to protect concrete

from the action

of water; the velocity of

water flow

DYNAMICS OF BASES AND FOUNDATIONS

190

and

possible fluctuations of ground-water level should be taken into

account.

In the process of foundation construction, special care should be taken

(if the latter are
foreseen by the design) and the excavation for the anvil or frame.
The lower part of the excavation for the anvil should be strictly
to provide accurate location of holes for anchor bolts

horizontal; no additional corrective pouring of concrete

is

permitted.

supplementary cement grout has to be placed under the frame of the

forge hammer, then the foundation surface in contact with the cement
grout should be roughened, cleaned, and washed. The underlying soil
should be compacted by tamping in of broken stone. In moist soils a
working mat of concrete type 50 f is placed under the foundation.
Concrete should be placed in horizontal layers and, as a rule, without
interruption in the work.
In case of an emergency interruption, the
following measures should be taken to secure the monolithic character of
If

the foundation:
1.

Dowels

of 12 to 16

of the joint to a
2.

mm diameter should be embedded on both sides

depth of 30

Prior to placing a

new

cm

at a distance of 60

cm from

each other.

layer of concrete, the previously laid surface

should be roughened, thoroughly cleaned, washed by a jet of water, and

mm

covered by a layer of a rich 1 2 cement grout, 20

thick.
The
placement of concrete should be started not later than 2 hr after this
mixture is laid on the surface.
:

The

may

be mounted on the foundation only after hardening of

not less than a week after its placement. The foundabe put in operation as soon as the concrete attains the design

anvil

the concrete,
tion

may

i.e.,

strength value.

may result in some soil falling

between the anvil base and the upper row of the timber pad.

Vibrations of the forge-hammer anvil

into the space

This

may

this, after

lead to tilting of the anvil or

the anvil

is

damage

built a protective wall

is

to the pad.

To avoid

usually installed around

it,

extending from the top of the foundation to floor level. Such a protective
box permits an easy and rapid inspection of the anvil and pad; it also
simplifies the mounting and dismounting of the anvil.
V-2. Initial Conditions of Foundation

Motion under Impact Action

The Velocity of Dropping Parts at the Beginning of Impact. Large

into two groups: those with an unrestricted
drop of the ram, and those with a restricted ram movement.
The first group includes hammers with frictional hoisting of the ram
and hammers in which the ram, rigidly tied to the piston, is lifted by
a.

hammers may be divided

See footnote, Art. IV-l-c, p. 132.

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

191

steam pressure from underneath. In frictional hammers, the ram is

connected to a plate which moves between two frictional disks pressed
When the ram is lifted to the height desired, one of
against this plate.
the disks is moved away from the plate and the ram falls, moving along
guides.

its

In single-acting hammers, the ram, which

is

rigidly tied to the piston by

by the pressure of steam released through a valve

located under the piston and opened when the latter is in its extreme low

means

of a rod, is lifted

After the piston

position.

raised to the height desired, the access of

is

steam into the cylinder under the piston is stopped, the valve opens, and
The piston, together with the ram,
the steam or compressed air escapes.
drops at increasing speed.
After the access of steam is discontinued and the exhaust valve opens,
steam cannot escape at once from the space in the cylinder under the
Therefore a counterpressure against the ram drop is created,
piston.
resulting in a loss both in the ram's velocity

and

in the kinetic energy of

drop.

its

The

velocity

the

v of

ram drop under

the condition of unrestricted

motion equals
v

where

h
77

=
=
=

\/%gh

(V-2-1)

acceleration of gravity

height of

ram drop

coefficient

which takes into account counterpressure and

frictional forces

The numerical value of depends on the design of the hammer, its

working order, the regulation of valves, etc.
Modern forging practice mostly employs the large double-acting
hammers. In these hammers, steam or compressed air acts on the ram
not only while it is being lifted, but also during its drop; therefore the
velocity and kinetic energy are considerably larger at the moment of
impact of the ram against the workpiece.
Changes in steam pressure during the drop of the ram, both under the
piston (counterpressure) and above the piston (active pressure), depend
on many varying factors such as the proper operation of the valves, the
tightness of the piston in the cylinder, and the working order of the
stuffing box.
It is impossible to take into account all these factors with
sufficient accuracy.
Therefore in computations of ram velocity, one
usually works with the average pressure of steam and air in the feed pipe.
Then the velocity of the forced motion of the ram under the action of its
own weight and the steady pressure will equal
rj

Mw+AM

(v-2-2)

DYNAMICS OF BASES AND FOUNDATIONS

192

where

A =

piston area

total pressure

=*

W=
t]

b.

on piston
dropping parts

total weight of

correction coefficient

Experimental Determination of

the

Correction

Coefficient

rj.

values of the correction coefficient in Eqs. (V-2-1) and (V-2-2)

The

may

be

determined only empirically by comparing the velocities obtained from

the equations with the corresponding values obtained experimentally.
Table

V-2.

Results of Hammer Efficiency Measurements

Velocity at beginning
of impact,

Nominal

Type

of

weight of
dropping

hammer

parts, tons

Measured

m/sec

or (V-2-1) for
v

Double-acting

Hammer

hammer

with unre-

stricted action

Ratio

77

between
Computed from measured and
computed
Eq. (V-2-2)

velocities

5.4
3.6
2.25
1.8
1.125
1.0
1.0
0.635

6.2
6.0
5.4
4.5
6.2
6.8
5.8
5.5

9.0
8.4
8.6
8.1
8.6
8.5
9.8
9.0

0.69
0.71
0.63
0.56
0.72
0.80
0.59
0.61

0.54
1.125

3.3
3.5

3.56
3.93

0.96
0.89

The author performed such measurements for ten drop hammers of

powers and makes. The measurements were made under

different

working conditions

in shops,

without any special adjustment. Therefore

may be considered to be characteristic

the results of these measurements

for

working conditions.

The
also
V

results of these

gives

velocities

measurements are presented in Table V-2, which

computed from Eqs. (V-2-1) and (V-2-2) for

I-

This table shows that the measured velocity of dropping parts at the
of impact is much lower in double-acting hammers than the
values computed from Eq. (V-2-2).
For these hammers, the ratio
between the values of measured and computed velocities lies within the
range 0.45 to 0.80; the average value of 77 in Eq. (V-2-2) may be taken to

moment

equal 0.65.

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

In addition,

not depend

by the

it

193

follows from Table V-2 that the absolute velocity does

the power of the hammer.
This is explained in part

much on

ram and the steam or air

hammers only within comparatively narrow

fact that usually the height of drop of the

pressure vary for different

Therefore in many cases the design velocity value may be taken

equal approximately the average value of velocities measured in
hammers with different powers. This value for double-acting dieranges.
to

stamping hammers equals 6.0 to 6.5 m/sec.

In hammers with unrestricted drop, especially frictional hammers, no
counterpressure of steam is encountered, and a decrease in the velocity
of dropping parts is mostly caused by friction in the guides.
When a

hammer

is

properly adjusted, the

effect of friction will

be negligible

therefore the correction coefficient

in Eq. (V-2-1) will be close to unity.

This conclusion

confirmed by

is

data for hammers with unrestricted

drop, presented in Table V-2.
c.

tion

Initial

Velocities

Motion.

Let

of

Founda-

us investigate

Fig. V-5. Derivation of Eq. (V-2-4).

the values of velocity characteristic

end of a vertical eccentric impact. Such an

impact occurs in drop hammers when edge grooves are stamped.
We assume that no pad is present under the anvil and that the anvil
and foundation form one single body whose elasticity may be neglected
in comparison with the elasticity of the soil.
Let us also assume that the
foundation can be represented by the body shown in Fig. V-5, that the
eccentric impact produced by the falling mass occurs in a plane which we
shall consider to be one of the principal planes of the foundation, and
that the center of gravity of the foundation and the centroid of its
contact area with the soil lie on the same vertical line.
Due to the
impact, the foundation undergoes vibrations which occur in the aforementioned principal plane.
If the ram and foundation are considered as one closed system, it
may be assumed that linear momentum is conserved during the impact.
for the foundation at the

The foundation

momentum

is

motionless before the impact; at that time linear

equals the

velocity of the falling

momentum

of the

ram of mass m Q at

the

ram,

i.e.,

moV, where v

is

the

moment it touches the founda-

tion (the beginning of impact).

After impact,

ram detaches
foundation

i.e.,

itself

during the period following the instant when the

momentum of the ram and

from the foundation, the

is

Q Vi

nwo

DYNAMICS OF BASES AND FOUNDATIONS

194

where

V\

velocity at which

m =

foundation mass

initial velocity of

Vq

ram rebounds from foundation

forward motion of center of mass of founda-

tion

The momentum

of the

system before the impact equals the

momentum

after the impact; therefore

m
In addition to progressive

moVx

mv Q

(V-2-3)

downward motion under the action of an

movement around

eccentric impact, the foundation undergoes a rotational

the axis passing through

in

center of gravity, perpendicular to the plane

its

which impact occurs.

The moment

of

m vr = m Vir +
where / = moment

mass

of inertia of

momentum

will

be
(V-2-4)

I<Po

of foundation

and hammer

in

regard to axis of rotation

r

eccentricity of impact

rotation of foundation
Vo
Equations (V-2-3) and (V-2-4) include three unknown values. In
order to derive a third equation, let us use Newton's hypothesis concerning the restitution of impact. According to this hypothesis, if
there occurs an impact between two bodies moving in relation to each
other, the relative velocity after the impact is proportional to the relative
The ratio between these two depends only
Velocity before the impact.
on the material of the bodies which underwent the impact. The foundation was motionless before impact; therefore the relative velocity of the
ram equals v. After impact, the absolute velocity of ram motion equals
Vi, but the point of the foundation which was subjected to impact acquired
a velocity whose vertical component equals v + r^ it follows that the
relative velocity of the ram after the impact equals v + r<p vi.
According to Newton's hypothesis,
initial velocity of

"

rv

Vl

(V-2-5)

where

e is

From
of the

the coefficient of restitution.

Eqs. (V-2-3) to (V-2-5)

(1

u
where

we obtain

expressions for initial velocities

foundation motion:

+
=

fji)(r

I
ra

+
u

m
=

Wo

r2

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

If

the impact was at the center of the foundation, then

vo

\-~
1

When

<p

195

and

(V-2-8)

the elasticity of the pad cannot be neglected,

sidered that the initial velocity of motion

is

it

should be con-

acquired only by the anvil

hammers) or by the anvil and frame (for drop hammers).

(V-2-6) and (V-2-7) remain the same, but the symbols m and
/ denote the mass and the moment of inertia either of the anvil or of
both the anvil and the frame, without taking into account the foundation
(for forge

Then Eqs.

mass.
d.

Coefficient of Restitution

e.

It follows

from the foregoing formulas

that the initial velocities of motion of the foundation or anvil depend

If the impact was pere.
and consequently e 1. For the impact of a
body against a plastic one, vi = 0, and consequently e = 0. For

considerably on the coefficient of restitution

fectly elastic,
rigid

then

vh

numerical values of e lie within the range

< e < 1.
In forge hammers, e depends on many factors, the most important of
which are: the temperature of a forged piece, the dimensions and forms of
grooves (in stamping hammers), and the elastic properties of materials
of the ram, head, and anvil.
real bodies, the

Since the design values of the amplitude of

tions

depend on the selected values

hammer

foundation vibra-

of the coefficient of restitution

e,

the

designer of a foundation naturally has a practical interest in knowing

its

However, the answer to this question is poorly elucidated

in special publications on heat treatment of metals.
In this connection
the author carried out special measurements to determine numerical
values of the coefficient of restitution of hammers. 11 Measurements were
performed under working conditions with both single- and double-acting
hammers. The computation of e was made from measured values of the
heights of fall and rebound of the ram after impact or from the interval
of time between two rebounds of the ram.
The results showed that the values of the coefficient of restitution
depend to a great extent on the state of a forged piece. Figure V-6
presents a graph of changes in e as a function of the number of blows on a
forged piece under a hammer having a weight of dropping parts equaling
5.3 tons.
Analogous graphs were obtained for other hammers. It
follows from these plots that during the first blows against the forged
piece, when its temperature is high and it is in a plastic state, the coefficient of impact velocity restitution is very small, equaling approximately
real values.

0.10.

As the number

piece decreases,

of blows increases, the temperature of the forged

the impact rigidity increases, and consequently the

DYNAMICS OF BASES AND FOUNDATIONS

196

For the last blows, when a comparatively cooler

e increases.
being
forged,
the
coefficient of restitution approaches 0.5.
piece is
Measurements of this coefficient during idle blows and under conditions of cold
value of

forging showed that

its

value does not exceed

Since computations of

hammer

0.5.

foundation vibrations should be per-

formed for the most unfavorable conditions of operation, the design

value of the coefficient of restitution for

should be taken as

hammers

forging steel parts

0.5.

0.50
0.40
0.30

0.20
0.10

10

Number

of

12
blows

14

16

Fig. V-6. Variation of the coefficient of impact velocity restitution

of

hammer blows on

20

18

with the number

the forge piece.

Values of e for forge hammers proper are much smaller than those for
stamping hammers, and corresponding design values may be taken to
equal 0.25.
Finally, for

hammers

forging nonferrous metals, this coefficient

considerably smaller than for

hammers working on

steel parts

is

and may be

considered to equal zero.

V-3. Natural Vibrations of a

Hammer and

its

Foundation as a

Result of a Centered Impact

The
a. The Main Assumptions Involved in Design Computations.
foundation and hammer present a system which includes at least seven
bodies: the frame, the dropping parts, the forged piece, the anvil, the
elastic

From

pad under the

anvil, the foundation block, and, finally, the soil.

phenomena which develop as a

impact of the ram against a forged piece lying on the anvil
are extremely complicated and may be analyzed only with a high degree
the point of view of mechanics, the

result of the

of approximation.

The main problems in computations for a hammer foundation are to

determine the amplitude of foundation vibrations and to establish the
values of stresses in the pad under the anvil.

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

The
the

solution of these problems

hammer

anvil,

is

197

usually based on the assumption that

frame, the forged piece, the anvil, the elastic pad under the

and the foundation block form one

body. Such an assumpand the frame is justified by

solid

tion in regard to the foundation, the anvil,

the fact that the deformation (due to impact) of each of these bodies

small in comparison with

fore

may

soil

is

settlement under the foundation and there-

be neglected.

However, deformation of the pad under the anvil may be much larger
than soil settlement. Therefore the assumption that the pad has an
infinitely large rigidity may lead in some cases to large errors in computaThis assumption is permissible only when the masses of both the
tion.
anvil and the frame, if the latter is placed directly on the anvil, are
comparatively small in relation to the foundation mass. Only in this
case will the pad have no considerable effect on the amplitude of foundaOtherwise, the elasticity of the pad cannot be neglected.
tion vibrations.
In the case under consideration, the computation setup will be reduced
to a system of three bodies: the ram, which is the striking body; the
anvil, which is separated from the foundation by an elastic connection;
and the foundation on an elastic base. The anvil and the foundation are
the impact-receiving bodies.

In determining the amplitudes of foundation vibrations, it is possible

assume that the time of actual impact is small in comparison with the
period of natural vibrations of the system; therefore, during the impact,
there is no time for the foundation and anvil to undergo displacements
comparable to their displacements during the vibrations which follow
the impact.
Since the reactions of the pad and the soil depend only on
to

damping
assume that during impact no additional
reactions occur from the pad and soil.
Thus only static reactions develop,
imposed by the weight of the foundation, hammer, and anvil. These
reactions existed before the impact and balanced the weight of the

the displacements of the anvil and the foundation (we neglect

reactions), it is possible to

installation.

Therefore during impact, the foundation (with anvil and frame) and
first approximation, may be considered to be
free bodies.
Then an analysis of the impact of the system may be

the dropping ram, in the

reduced to the analysis of a free impact of two or more absolutely solid

bodies moving with given initial velocities.
The striking body (the ram) in all computations is assumed to be
absolutely rigid.
b.

Equations of

the Vibrations of

Foundation and Anvil.

We

begin by

considering the simplest conditions: those in which pad elasticity

may

be neglected and the vibrations of the foundation, anvil, and frame occur
as vibrations of a body with only one degree of freedom.

DYNAMICS OF BASES AND FOUNDATIONS

198

In this case the equation of vertical free vibrations of the foundation

be (Art. III-l)

will

z+U
where

(V-3-1)

mass
measured from equilibrium position

vertical displacement of center of

anvil,

fm 2 = square

of

foundation and

of frequency of natural vibrations of foundation:

nz

A =

foundation area in contact with

m =

total vibrating

soil

mass

c u = coefficient of elastic uniform compression of soil

Equation (V-3-1) is the equation of free vibrations of the foundation
without damping. The general solution of this equation is

= A

The constants A and B,

ditions of motion.

the impact of the

fnz t

+B

cos fns t

ram

(V-3-2)

as usual, are determined

Taking as the beginning

initial

we

obtain, for

con-

when

0,

Vu

we obtain

initial conditions,

from the

of readings the instant

against the anvil ends,

z

Using these

sin

- h.

B =

f
Jnz

Equation (V-3-2)

will

take the form

z

sin fmt

(V-3-3)

J nz

The maximum

deflection of the foundation

IT

h =

2fn~z

Its value will

be

(V-3-4)
Jnz

If

one

is

to take into account the soil reactions

which are proportional

to the velocity of foundation displacement, then the amplitude of real

computed without considering the

a difficult task to evaluate the influence

vibrations will be smaller than the one

damping

forces.

However,

it is

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

of

damping

by means

forces

these forces depend on

many

As

of computations.

199

stated in Chap. Ill,

factors (for example, the foundation area in

contact with soil, the foundation mass and the period of its free vibrations,
and the foundation depth).
The pad under the anvil is fairly elastic in comparison with the anvil
and foundation; therefore the anvil and frame (if the latter rests on the
anvil) will not only participate in vibrations of the foundation on soil,
but will undergo some vibration with respect to the foundation.
In order to evaluate the amplitude of vibrations of the anvil in relation
to the foundation, it is necessary to consider vibrations of a system with
two degrees of freedom. Free vibrations of such a system are determined
by the following differential equations:
ntiZi

+ CiZi
mz +
i

m =

where mi,

c 2 (z 2

=
=

zi)
\

21)

n7
(V-3-5)

masses of foundation, anvil (with frame,

mounted on

d =

c 2 (z 2

cu

if

latter

is

anvil)

coefficient of rigidity of soil base

under foundation

= (E/b)A = coefficient of rigidity of pad under anvil

A = base area of pad
b = thickness of pad
E = Young's modulus of material of pad
= displacements of foundation, anvil measured from
zi, z
c2

equilibrium position

We

denote by /ni and fa2 the natural frequencies of the system whose
motion is determined by Eqs. (V-3-5) by
;

f 2 _
Ua
" S
m
2

we denote the frequency of natural vibrations of the anvil with the frame
(or for forge hammers proper that of the anvil on a motionless foundation) then we obtain a general solution of the system of Eqs. (V-3-5)
;

zx

z2

= difna )
= CJna sin Umt
2

Setting

sin (fnl t

+ a J + C fna
C^ = d cos ai
CM = C

aJ

cos a 2

(fna

sin (fn2

fn 2

sin (fn2 t

C = d
C^ = C

a2)
(V-3-6)

a.)

sin ai
sin

a2

we obtain
21

Z2

2
2
/nl *) Sijlfmt + C^(fna - fnl ) COS fnl t
2
2
+ C^iU - /n2 ) Sin fn2 t - C^(fna 2 ~ fn2 2 ) COS fn2 t
C^fna* Smfnl t + C^fn J COS fnl t + C^fna * sill fn2 t
+ C< 4 V\ cos fn2t

C<-(fna *

(V-3-7)

DYNAMICS OF BASES AND FOUNDATIONS

200

The

natural frequencies fn \ and / 2 are determined as roots of the

equation
/*

(J?

+ /

where

Ml)/*

\x

)(l

Ji

)fl%a 2

(V-3-8)

hammer

the limiting frequency of the foundation together with the

is

placed on

The
(at

+m

mi

nil

/*

(1

initial

the condition that the pad

soil (for

infinitely rigid).

conditions of motion in the case under consideration

0) are as follows:
zi

where

is

v a is

z2

ii

i2

va

the initial velocity of motion of the anvil,

+e

= y-y

Va

and

Ha

Particular solutions of system

which correspond

(V-3-5)

to

these

conditions are as follows:

initial

Una 2
1

-C

Jv2
9/
2

fna (fnl
Zi

-T-o
Jul

-r~2

Jnl

f Slli fnl t

fnl*)

SID. f,

9\

J"

Jn2

2
/f
Jna

V
^a

)UnJ

-T

_ Jfnl

7""^
Jul

/nl

sm/

fl

i/

/n2

f
J na

_ Jfnl

t
Jn2

/
\

q Q\
v -5-;

,\

sin/n2 ^

With these expressions it is possible to compute stresses which develop

pad as a result of combined vibrations of the anvil and foundation.
The maximum stress a in the pad evidently will equal

in the

^(z -

(V-3-10)

Zl )

V-4. Experimental Studies of Vibrations of Foundations under

Forge Hammers
of vertical vibrations of hammer foundabased on some assumptions which may be
verified only by comparing the results of computations with experimental
This refers primarily to the negligibility of the mass and damping
data.
a.

Introduction.

tions,

The theory

presented in Art. V-3,

properties of

soil.

is

As stated

in

Chap. IV, due to the fact that forced

vibrations of foundations under reciprocating engines are usually characterized

by a frequency

different

vibrations, the influence of

and

may

be ignored.

from the natural frequency

damping

soil

of

foundation

reactions in such cases

is

small

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

201

Under an impact, the foundation below a hammer undergoes free

Therefore the damping soil reactions will have conThe
siderable influence on the amplitudes of foundation vibrations.
vertical vibrations.

introduction of damping-reaction values into the computations of these

somewhat complicate the formula used, but the calculabe practicable from a mathematical point of view. However, in order that this complication, caused by the introduction of
damping reactions, should be of some practical value, it is necessary to
vibrations will
tions will

know

still

the constants characterizing the dissipative properties of the

Great

difficulties are

soil.

involved in establishing these constants, because

their values depend not only on the soil, but also on the design of the
foundation (in particular, on the depth of the foundation, the ratio
between the length and width of the foundation, the foundation height,
It is very difficult to take
and the material and density of the backfill)
.

into account the influence of

constant of a

all

these factors on the value of the

damping

soil.

The inertial properties of the soil, which were not considered by the
theory of vertical vibrations presented in Art. V-3, also may have great
In addition, the results of computations may be influenced by
effect.
values of the coefficient of elastic rigidity

c 2 of

the pad under the anvil.

This coefficient depends not only on the properties of the material of the
pad under the anvil, but also on its design and on other special features
which cannot always be taken into account by computations.
The pad under the anvil in hammer foundations of conventional
design usually consists of several shields
together.

The

made

of

timber beams bolted

horizontal surfaces of these shields, just as the base of

the anvil and the surface of the foundation under the anvil, are not

smooth surfaces and consequently do not come into contact with

Because of this, some sections of the surfaces
all points.
the pad, the anvil, and the foundation are subjected to considerable

ideally

each other at
of

stresses while others are not loaded

at

all.

properties of the whole pad depend not only on

the conditions of

its

Only by means
ciently large

number
It is

its

result,

the elastic

material, but also on

contact with the surfaces of the foundation and anvil.

of

measurements
of operating

elucidate the influence of

vibrations.

As a

all

of vibrations occurring in a suffi-

hammer

foundations

is it

possible to

the above factors on vertical foundation

obvious that measurements do not give us an oppor-

tunity to establish separately the influence of each of these factors; for

example, that of damping and inertial properties of soil. However, the

measurement data do make it possible to introduce corrective coefficients
into the formulas of Art. V-3, which then permit the adjustment of the
results of computations performed on the basis of these formulas to the
results of vibration

measurements.

DYNAMICS OF BASES AND FOUNDATIONS

202

As

early as 1939, the author carried out a large-scale investigation of

foundation vibrations. 8

He

studied 47 foundations under

hammers loca-

ted at six different plants.

That study had a threefold purpose: the verification of formulas for

computations of hammer foundation vibrations, the collection of data
on the design of normally operating foundations, and the determination
of values of vibration amplitudes which could be accepted by designers
as permissible and on which the main dimensions of the foundation would
depend.
Table

V-3.

Data on Plant Sites where Hammer-foundation

Measurements Were Made

Plant
Geological

no.
1

A brown sandy

and hydrogeological description

of site

clay with yellow inclusions comes to the surface everywhere

site of the plant.

This clay has a thickness of 0.1-2.5
and is
underlaid by a fine quartz sand alternating with lenses of clayey sand and
of clay with some sand and silt.
The thickness of these layers is not
uniform over the area of the plant, varying from 0.3-6 m. With increasing

on the

depths, sands free of clay admixture predominate.

is

at a depth of 7

m,

i.e.,

below

all

the

hammer

Ground-water

Yellow medium-grained dense sands at a natural moisture content

Yellow medium-grained dense sands at a natural moisture content

Fine dense sand, ground-water level at a depth of 2.2-2.8 m,

the base of the hammer foundations

Medium-grained sands

Heavy brown

of

clays with

medium

level

foundations.

i.e.,

above

density reaching to a depth of 9.4

some sand and

silt

Foundations of various designs were studied. Slab-shaped foundapredominated at one plant only. At other plants, only deeply
embedded block-type foundations were present. This design of foundations was very popular at the time of the investigation (1939); slabshaped foundations embedded to a small depth were seldom used then.
b. Description of Bases and Foundations.
The greatest part of the
forge hammers mounted on the foundations studied (35 out of 47) were
double-acting steam or air stamping hammers. Only 6 foundations were
under drop hammers of unrestricted action. The remaining 6 foundations were under forge hammers proper.
The foundations investigated
were located at six different plants. The geological conditions for each
tions

plant are given in Table V-3.

Prec. Results of Measurements of Foundation and Anvil Vibrations.
liminary measurements of foundation vibrations showed that, in addition

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

203

hammer foundations also undergo rocking vibraHowever, the latter are less important because their amplitudes
are much smaller than those of vertical vibrations.
The vibration amplitudes of the foundation, the anvil, and the frame
For example,
are strongly affected by the state of the forged piece.
to vertical vibrations,

tions.

during the

impact

is

first

impacts of the

largely

consumed

hammer

in plastic

against the piece, the energy of

deformation of the metal, and the

The
coefficient of restitution is small, as are the vibration amplitudes.
amplitudes of vibrations of the anvil and foundation grow with each
subsequent impact. The largest amplitudes come with the last few
impacts, when the forged piece is already deformed to such a degree
that the greater part of the impact is taken not by the forged piece, but
by the lower die, which transfers the impact energy to the anvil and
Since the last few impacts induce the most unfavorable
dynamic conditions for the foundation and anvil, their vibrations were
measured during these impacts.
Figure V-7 shows samples of vibrograms obtained for some of the
hammers investigated. It is seen from these vibrograms that vibrations
of the hammer foundation and anvil, in most cases, differ considerably
from damped sinusoids, which could be assumed on the basis of theoretical
This shows that the foundation together with the anvil
considerations.
presents a much more complicated vibrating system than was assumed

foundation.

in Art. V-3, wherein vibration equations

were derived.
In addition, as was to be expected, the vibrograms reveal considerable
influence of the damping reactions.
In some cases this influence is so
large that the motion is almost aperiodic.
Finally it was found that identical foundations built under the same
geologic conditions and subjected to the action of identical impacts
underwent vibrations of varying amplitudes, sometimes sharply differing
from one another. For example, two identical foundations under 3.6-ton

hammers were

mm

them had 0.48

amplitude of
In the same way, two identical foundations under 2.25-ton hammers had 0.80- and 1.80-mm amplitudes of
investigated; one of

vibrations, the other 0.78

mm.

vibrations.

These data on vibrations

of existing foundations

permit the assump-

by theory.
Thus the observed differences in the amplitudes of foundation vibrations
under hammers operating under the same conditions are apparently
explained by the influence of the following factors: (1) the state of the
timber pad under the anvil; (2) the contacts between this pad and both
the anvil and the foundation; (3) the backfill of the foundation; there
tion that they are greatly affected

may

by

also be influences of other factors

design computations.

factors not considered

which are

difficult to include in

DYNAMICS OF BASES AND FOUNDATIONS

204

Vibrograms

Nominal
weight

of

vibrations

(amplitudes are given

of falling
parts, tons

in

mm)

Foundation

Anvil

0.2 sec

0.2 sec

0.1 sec

0.2 sec

ou0.2 sec

0.2 sec

L_
p.2 sec

0.2 sec
|

Fig. V-7. Typical vibrograms of operating

hammer

foundations.

Therefore the results of computations carried out on the basis of the

formulas of Art. V-3 should be considered as tentative values only, showing
the order of magnitude of vibration amplitudes, but not their absolute
values.

comparison of computed and measured vibration amplitudes of

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

hammer

foundations leads to the conclusion that

if

205

data required for

calculation are selected correctly, then an average error in computation

be around 30 per cent.

Amplitudes of Foundation Vibrations. On the basis of investigations of 47 foundations, points are plotted in Fig. V-8 giving measured
vibration amplitudes versus actual weights of dropping parts of hammers. It is seen that the amplitudes of hammer vibrations never attain
the value of 2.0 mm, i.e., the value which in the past was taken as permissible in computations. 44 With a decrease in power of the hammer, a
will

d.

2.0 -

o
<s

-21.0
<D

CPO
o

^oo
r>9Sb Op

o
o

0123456789
ft

Weight of striking parts, tons

Fig. V-8. Measured vibration amplitudes of 47

the weight of the striking parts.

hammer

foundations plotted against

decrease in the amplitudes of foundation vibrations

is

observed.

The

overwhelming majority of foundations studied had vibration amplitudes

of about 1.0 to 1.3
or less.
Foundation vibrations characterized by
these amplitudes did not exercise any noticeable harmful influence on the
structures of forge shops.
Similarly, no considerable settlements of
foundations were observed where amplitudes of vibrations were of the

mm

order of 1.0 to 1.3

mm.

However, foundations having vibration amplitudes greatly exceeding

1.0 mm underwent considerable settlements.
For example, a foundation
with an amplitude of around 1.8
underwent a settlement reaching
0.3 m.
The above discussion leads to the conclusion that a design value of
vertical vibrations of hammer foundations may be taken in the range 1 to

mm

1.2

mm.

Amplitudes of Anvil Vibrations. Amplitudes of vibrations of anvils

of stamping hammers are much larger than amplitudes of
foundation vibrations. For powerful hammers having thick pads under
e.

and frames

the anvil, absolute values of anvil vibrations reach 5 mm, although most of
the hammers studied had amplitudes in the range 2 to 4 mm.
With an

DYNAMICS OF BASES AND FOUNDATIONS

206

power of the hammer, the amplitude of anvil vibrations

For hammers with a weight of dropping parts up to 1 ton, a
typical amplitude reaches 1 mm; for 2-ton hammers this amplitude
for hammers in which the weight of dropping parts exceeds
reaches 2
3 tons, the amplitude of anvil vibrations is usually 3 to 4 mm.
These
values of amplitudes of anvil and hammer vibrations may be taken as
increase in
increases.

mm

permissible.

The above values show that when stamping hammers

are in operation,

the anvil rebounds on the pad.

Shocks of the anvil and frame (if the

latter is attached to an anvil whose amplitude of vibrations reaches considerable values) have a harmful effect on the condition of the hammer.
In addition, the larger the amplitude of vibrations of the anvil, the more
kinetic energy of impact is consumed by these vibrations and consequently the smaller the hammer's efficiency.
Loss of impact energy
due to vibrations reaches 10 per cent of the work of the hammer's dropping parts.

The

by insufficient
pad under the anvil, which in some hammers had a thickThere is no reason to use such thick pads either from the
ness of 1.5 m.
point of view of the forging process or from a structural point of view.
The pad thickness is usually assigned by the hammer supplier on the
basis of traditional recommendations of the manufacturer and is not
substantiated by any design data. Therefore different plants producing
hammers of the same power recommend pads of different thicknesses.
The thickness of the pad should be selected so that the vibration
amplitudes of the anvil do not exceed a particular value; in addition,
Table V-l
stresses in the pad should not be greater than is permissible.
was compiled on the basis of these considerations. Thicknesses of pads,
as recommended in that table, are somewhat smaller than those which
The decrease in pad
usually have been employed up to the present time.
thickness as compared with usually accepted thicknesses is based on
large vibration amplitudes of anvils are explained

rigidity of the

considerations concerning the harmful effects of anvil vibrations of large

amplitude.

The Determination of Elastic Constants of the "Anvil-Foundation"

If one is to consider a foundation together with the anvil
mounted thereon as a system with two degrees of freedom, as was done in
Art. V-3, then vibrograms of the anvil and foundation vibrations will
show two sinusoids of different periods superimposed on each other. It
follows from theory that the amplitudes of these sinusoids will be inversely
/.

System.

proportional to their frequencies.

measured foundation vibrograms in

many cases approach aperiodic curves. In no case was it possible to
determine from vibrograms both natural frequencies of the combined

As stated

before, the shapes of the

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

207

Vibrograms usually reveal only

vibrations of the anvil and foundation.
Therefore, it is possible
the vibrations at the lower principal frequency.
to consider (with a precision sufficient for practical purposes) that in

Eqs. (V-3-9) the amplitude of vibrations for sin fni t (where fnl > fn2 )
Then approximate expressions for dynamic displacement

equals zero.

of the foundation

and anvil

fnl
fn2 )(fna
2/f 2
f 2\f 2
Jna \Jnl
Jn2 )Jn2

_ (/n
/

Si

77

f
Jna

Z2

be as follows:

will

{Jnl

-fnl

(V-4-1)

'ft
J n2t

S111

V(1

2\ f

"abUl J n 2b

Jn2 )Jn2
Zj.

Hence
Z\

where

fn2

-7

(V-4-2)

(V-4-3)

7^
Jna

Thus, having found, from vibrograms obtained for the anvil and foun/3 and the lower natural frequency of vibrations, one
can establish from Eq. (V-4-3) the limiting frequency fna of vibrations of
From the formula
the anvil on the pad.
dation, the value

Jna

~ bm

one can establish the value of the modulus of elasticity

E of the

pad under

the anvil.

Then no

difficulties are

involved in establishing the value of the second

higher frequency fnh as well as the limiting frequency fi of the natural

vertical vibrations of the entire installation on the soil.
After some trans-

formations (not shown here), we obtain

/nl

_
~

+
1

Here, as before, n
the frame mass
2

fi

is

is

Q _

^2\^2

w2

>

the ratio between the anvil mass (in drop

also included)

and the foundation mass.

hammers
Knowing

from the formula

Ji

mi

+m

one can establish the real value of the coefficient of elastic uniform compression c u of the base under the hammer foundation.

DYNAMICS OF BASES AND FOUNDATIONS

208

The

results of

computations of moduli of elasticity for pads and

coeffi-

cients of uniform compression c u for bases, performed in accordance with

the above methods for the several foundations studied, lead to the following conclusions:

As was to be expected, the computed moduli of elasticity E for pads

under the anvils of different hammers vary within a comparatively wide
range of values. The probable reasons for this have already been disThe moisture content and working life of the pad should also
cussed.
be mentioned, as their influence may be very noticeable.
The average value of E was established from computations to equal
4.7 X 10 4 tons/m 2 i.e., approximately two times smaller than the value
customarily used in stress analyses of oak beams when the latter are
compressed across their fibers. On the basis of these results, it is recommended that a design value of 5 X 10 4 tons/m 2 be used for the modulus
of elasticity of timber pads.
In most hammers the dynamic stress in the pad under the anvil does
not exceed 200 tons/m 2 i.e., it is much lower than the permissible value
of about 300 to 350 tons/m 2 for oak timbers compressed across their
This attests to the fact that pads which were employed up to this
fibers.
time have had a considerable safety factor. As stated before, this is
because the thickness of the pad has usually been taken much larger than
was necessary from the point of view of dynamic computations.
In low-power hammers, dynamic stresses in pads do not exceed 100
Therefore it is possible to employ in these hammers pads
tons/m 2
,

made

of pine or larch instead of oak.

Special investigations

showed that the design value

of the coefficient

uniform compression of the soil base of hammer foundations

However, the average value of this coefficient
was about 4.0 kg/cm 3
obtained from measurements of hammer vibrations was around 25 kg/cm 3
Such a large divergence between
i.e., approximately six times larger.
these values attests to the fact that the amplitudes of foundation vibrations under hammers are greatly affected by factors not considered by
In particular, this theory, as stated
the theory presented in Art. V-3.
above, does not consider the influence of the damping and inertial properties of the soil, but this influence may be considerable, just as in the
In addition, the value of c u may
case of natural foundation vibrations.
c u of elastic

be affected by the backfill of the foundation.

Investigations of a test foundation showed that with backfilling the
value of the coefficient c u of elastic uniform compression increases

approximately two times as compared with the value established from

Consequently, in computations of
tests on an exposed foundation.
natural vertical vibrations of foundations under hammers, the value of
the coefficient of elastic uniform compression c u should not be taken equal

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

209

to that used in computations of vibrations of other machines whose

foundations do not undergo natural vertical vibrations. According to
the above data, in design computations of hammer foundations the coefficient c u

should be used instead, where

c'u

With allowance

for

(V-4-5)

kc u

some safety reserve

for

dynamic

stability of the

may

be taken as equal

foundation, the value of the correction coefficient k

to 3.

Comparison of Different Formulas for

g.

To

Computation of Amplitudes.

the

simplify the practical computations of vibrations of the foundation

and
This

independent of each other.

anvil, the vibrations are considered to be

is

equivalent to the assumptions that the presence of the pad under

the anvil has no influence on the amplitudes of vibrations of the foundation

and that the vibrations

of the anvil are not affected

mass

by

elastic

This assumption leads to considerable simplification of formulas for the computation

According to
of vibration amplitudes of the foundation and anvil.
Eqs. (V-3-4), the amplitude of vibrations of the foundation can be
properties of the soil base or the

of the foundation.

established from

Az

and the amplitude

(1

(W

+ e)W v
+ W )fnz
Q

of vibrations of the anvil

Aa =

fl

(V 4 b)

+eW*>

from

(y. 4 .7)

rV 2jna

It is interesting to establish and compare the computational errors

involved in these equations and the more accurate Eqs. (V-3-9). Taking

into account the fact that in Eqs. (V-3-9) the amplitudes for sin/^ are

much

smaller than the amplitudes for sin fn2 t, we may neglect the terms
containing sin fn \t in these formulas. Then the vibration amplitudes of
the anvil and foundation will be determined by Eqs. (V-4-1).

Table V-4 gives the results of computations of vibration amplitudes of

the foundation and anvil, performed for several foundations studied both
;

computation were employed. The same values of the

of elasticity of the pad under the anvil and the coefficient of
elastic uniform compression of the base under the foundation were
taken for all hammers (respectively, 5 X 10 4 tons/m 2 and20 X 10 3 tons/m 3 ).
The same velocity of dropping parts (6.5 m/sec) was used for all hammers.
It is seen from Table V-4 that there is considerable difference between
the amplitude values of foundation vibrations as computed by Eqs.
(V-4-1) and (V-4-6).
Hence it follows that a control computation of

methods
modulus

of

t N ^ N^ N

lO iC 'O
b- 00 >0 <M
-tf

'

CM

tJh

tf

CO

rj(

CO

to

H
N lO h
Oi CO o CO
O H O rH

JO l>
rH

o o o o o o
X X X X X X
CM CO rH C5 O CO
Ol CO H (N M
C
CO CN CM CM CM
1

"tf<

>>

o o o o o o
X X X X X X
CM GO ^ o o ^
o

th o
t- n<
CO i lO OS <# 00

cr

<^

H W >

a
2 r^

as ii
?! S W

I
<D

"5

JJ

Cft

CO

-c

*a
cm

^^
e

-O.

S'S
o
"43
o3

T3

en

fi

-"S

.o

>

'-

>

.2

x>

b- o

coo

>

<y

+3

'S
a
o o a

-4J

S 2 3
d -

11

-*

<fl

CO

M h

II

et

^
210

&&
S s
Oj
H

nj

II

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

vibrations of a foundation under a

hammer

211

should be performed using

Eqs. (V-4-1), considering the effect of the pad under the anvil on these
vibrations.

However, Table V-4 shows that there is only a small difference between
computations of anvil vibration amplitudes by Eq. (V-4-7) and by the
Therefore it is permissible to use the simpliprecise Eqs. (V-4-1).
formulas in computations of anvil vibration amplitudes and in stress
analysis of the pad under the anvil.

more
fied

V-5. Selection of the

Weight and Base Area of a Hammer Foundation

It was formerly held that the weight of the foundation for a hammer
and the size of its area in contact with the soil should be selected in such
a way as to meet the following requirements: the total pressure on the
soil should not exceed the bearing capacity of this soil and the foundation
should not bounce on the soil. These conditions may be written as
;

follows

where p st

pay

Pst

Pst

>

Pdy

< ap

(V-5-1)
(V-5-2)

Pdy

=
=

permissible bearing value under condition that only

coefficient of required reduction

static,

dynamic pressures on

soil

static load is acting

The condition expressed by Eq. (V-5-1) is based on an assumption that

and dynamic pressures are equivalent or that the "coeffi-

either the static

not a constant for a given soil and foundadynamic pressure transmitted to soils
(especially to granular soils) may induce settlements and deformations
which are tens and hundreds of times larger than those caused by static
Therefore if one of the items of the
pressure of the same magnitude.
left-hand part of expression (V-5-1) changes, but the sum of these items
remains constant, the total settlements and deformations will change.
Hence it follows that under the assumption that a has a constant value
for a given soil and foundation, the condition expressed by Eq. (V-5-1)
cannot be accepted, because it is contrary to the physical nature of the
cient of required reduction"
tion.

As stated

in

Chap.

is

II,

phenomenon.

The condition expressed by Eq. (V-5-2) has no practical significance,

because the bouncing of a foundation is of no essential importance and
cannot be observed under working conditions. However, an observance
of this condition led to carrying foundations

depth.

mm

down

to a considerable

Since the design value of vibration amplitude was taken to equal

and more, in order to obtain a static settlement of this value

necessary to increase considerably the heights of foundations.

it

was
The

DYNAMICS OF BASES AND FOUNDATIONS

212

was that all hammer foundations erected with designs complying

with the conditions of Eqs. (V-5-1) and (V-5-2) represented massive
Figure V-4a shows a
blocks carried down to a considerable depth.
typical foundation of this kind.
Limitation of the vibration amplitude of a hammer foundation is the
result

most important condition which must be

foundation.

The

smaller

the

by the design

satisfied

vibration

amplitude,

the

of such a

smaller

the

influence of vibrations on adjacent structures

and buildings, on the

foundation, on the soil, and on the hammer. Another significant condition is the limiting of the value of static pressure on soil; as shown in Art.
1 1-3,

the smaller the static pressure, the smaller the settlement of the

foundation (other conditions remaining equal).

Thus instead of the conditions of Eqs. (V-5-1) and (V-5-2), the foundation design should satisfy the following two conditions:

A < A
< apo

As indicated

hammer

in Art. V-4,

(V-5-3)

Pst

(V-5-4)

an average value

of vibration amplitudes of

foundations, obtained from the results of numerous investiga-

tions of operating

hammers,

is

approximately

as a design value for the permissible amplitude.

of

Eq. (V-5-3)

may

mm.

This may be taken

Therefore the condition

be rewritten as follows:

A <
z

10" 3

(V-5-5)

In the simplest case, under the assumption that the foundation together
with the anvil presents a system with one degree of freedom, the value of
the vibration amplitude is determined by Eq. (V-4-6), and the condition
of Eq. (V-5-5) may be written in the form
(1

+ eWoV <

Vkcu WAg

10-

(V-5-6)

dimensions are in tons, meters, and seconds.

Eqs. (V-5-4) and (V-5-6) values of foundation contact area and
weight can be found for which the amplitude of foundation vibrations
will not exceed 1
and the static pressure on the soil will not exceed

where

all

From

mm

the value ap

thus we obtain

(1

W, =

(1

M
+ >^y^
10'

meters

(V-5-7)

\/kc u ap g
e

lo-i

- Wa

tons

(V-5-8)

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

213

Wf =
a =

weight of foundation together with backfill, if present

weight of anvil and frame
Dividing both parts of Eq. (V-5-8) by TFo, we obtain a formula for
determining the reduced foundation weight corresponding to a unit of
actual weight of dropping parts of the hammer:

where

n,

+ t^ aP% X

(1

10'

(V-5-9)

Ua

\/kc u g

n'

where

W
w

Wa

In accordance with the values of p and

c u for different soils

given in

Art. 1-2, the following approximate relationship can be used:

82

The corrective coefficient

we set equal to 3.0; the
we set equal to 0.4.
V-4,

Table
Type

of

V-5.

k,

0.07

in accordance with data presented in Art.

coefficient of reduction of bearing capacity ,

Values of Some Hammer Coefficients

hammer

m/sec

na

nf

Stamping hammers:
Double-acting

hammers (stamping

of

steel pieces)

Unrestricted

Stamping
Stamping

6.5

0.5

30

48

4.5
4.5

0.5
0.0

20

34

6.5
4.5

0.25
0.25

30
20

hammers:
of steel pieces

of nonf errous metals

Forge hammers proper:

Double-acting
Unrestricted

16

Substituting the values of these coefficients into (V-5-9),

35
25

we obtain a

simple formula for the tentative determination of the foundation weight

depending on the velocity

of

dropping parts of the

hammer and

the

coefficient of restitution:

n,

8.0(1

e)v

na

(V-5-10)

According to data of machine-building plants, one can take approximately

:

For double-acting hammers:

For unrestricted hammers
Numerical values

of

nf

n f = 30
n a = 20

for different

hammers

are given in Table V-5.

DYNAMICS OF BASES AND FOUNDATIONS

214

In order to compare the computed values of n f with data secured from

some of the drop-hammer slab
foundations investigated at one of the plants.
experience, Table V-6 shows values of n f for

The average experimental value

of

nf

for the

hammers

of

Table V-6

is

should be mentioned, however, that the actual weight of dropping

parts of the hammers studied is much larger than their nominal weight.
In most other cases, the operating dies are lighter and consequently the

46.

It

difference

parts

is

between the actual and the nominal weights of the dropping

Therefore such hammers have a larger value of n f

smaller.

Table V-6. Comparison of Actual and Computed Reduced

Foundation Weights n f [Eq. (V-5-10)] and Reduced Foundation
Areas a f [Eq. (V-5-11)] of Several Hammer Slab Foundations
Nominal

weight of
dropping
parts, tons

tons

1.0

1.2

1.25
5.4
2.25
1.35
0.54
2.25

2.10
7.5
4.10
2.0
0.75
4.10

Wb = weight

f,

tons

49.7
92.0
502
217
73.5
28.4
168

A,

Actual
value

Actual

of nf

of aj

41.1
43.8
67.0
53.0
36.8
38.0
41.1

12.2

14.7

20.0
65.6
38.2
16.5
9.9
40

of concrete foundation

value

Com-

Com-

puted

value of

Wo

value of
af

32.6
31.3
30.8
32.4
36.8
31.0
34.0

12.2

h]

nf
48.0
48.0
48.0
48.0
48.0
48.0
48.0

9.6
8.8
9.4
8.2
13.0
10.0

without consideration of the

puted

9.5
8.8
9.3
8.2
13.2
9.8

backfill.

reaching 50 to oo for the same foundation weights.

tions,

which

ratio

is

For massive foundawere accepted by all design organizations, this

lying in the range 70 to 80 and often reaching

until recently

much

larger,

100 to 120.
If in slab

foundations one takes into account only the weight of conweight of backfill above the slab, then the value of

crete, neglecting the

nf

will

be around 30 to 35.

In massive foundations no such backfill

is

present; therefore the value of n f of 70 to 120 represents the ratio between

the weight of the foundation and the actual weight of dropping parts.

Thus the expenditure

of material for slab foundations is

than for massive foundations.

instrumental investigations of
amplitudes of their vibrations

mm

And, as stated

hammer
lie

two times smaller

in Art. V-4, the results of

slab foundations

show that the

within the range of permissible values

1
and less).
Equation (V-5-7) for the selection of the foundation area in contact
with soil may be simplified on the basis of the following considerations:
According to available data on the values of c u and p for different soils,

(around

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

it

215

can be stated that

^ 0.5
=

Hence

10- 2 Po

lO" 3

all dimensions are tons and meters.

Dividing both parts of Eq. (V-5-7) by Wo, we obtain an expression for
the reduced contact area of foundation per unit of the actual weight of
dropping parts:
A_
e)v X 10 3
(1

where

= nr =
Wo

af

Setting as before k

3.0

Po

and a

y/akg

we obtain a simple formula

0.4,

10~ 2
for

the tentative determination of af and consequently of the entire contact

area of the foundation:

af

20(1

e)v

(V-5-11)

Po

Equation (V-5-11) establishes the dependence of the foundation conon the hammer characteristics, but also on soil proper-

tact area not only

ties;

the required dimensions of the foundation contact area increase in

an inverse proportion to the bearing capacity of soil.

values of af computed for different types of soils.
Table

V-7.

Table V-7 presents

Values of Reduced Foundation Contact Areas

Required for Different Soils

a/

Values of a/ for following groups of

Type

of

hammer

Weak
Po
1.5

soils,

<

kg/cm 2

Soils of

medium

strength, p =
1.5-3.5 kg/cm 2

soil:

Soils of

high

strength, p

3.5-6

kg/cm 2

Stamping hammers:
Double-acting

hammers (stamp-

ing of steel pieces)

Unrestricted

13

13-5.5

5.5-3.3

9-4
6-2.5

4-2.5
2.5-1.5

hammers

Stamping of steel pieces

Stamping of nonferrous metals
Forge hammers Proper
Double-acting
Unrestricted

11

7.5

5-3
3-2

11-5
7.5-3

For example, it is seen from this table that double-acting drop hammers
used for stamping steel pieces, i.e., the type most frequently employed

by the following
on the average about 9 m 2

in forge shops, are characterized

of

medium

strength,

ratio:

when

the

soil is

of foundation contact

area will be required per unit of actual weight of dropping parts.

DYNAMICS OF BASES AND FOUNDATIONS

216

Table V-6 gave values of a f which were used for hammer slab foundawhen p = 2.5 kg/cm 2 It follows from the data of this table that
values of a f established from Eq. (V-5-11) are close to those accepted for
the design of hammer slab foundations.
Massive hammer foundations
designed as blocks are characterized by values of a f of 7 to 8, i.e., by
somewhat smaller values than those used for slab foundations.
tions

V-6. Design of a
Example.
1.

Hammer Foundation

Dynamic computations for

Design Data.

the design of a

hammer foundation

hammer

double-acting stamping

has the following specifica-

tions:

Nominal weight

of

dropping parts:
3.0 tons

Actual weight of dropping parts:

Wo =
Height

3.5 tons

of drop:

1.0

Piston area from above:

A =

0.15

Steam pressure:
p

Weight

of the anvil

90 tons

atm

and frame:

Base area of the anvil:

A =

4.75

0.60

Thickness of pad under anvil:

h

Soils on the site of the foundation consist of brown clays with some sand and silt,
with a permissible pressure p = 2 kg/cm 2 if only static pressure is acting.
2. Velocity of Dropping Parts at the Beginning of Impact.
From Eq.

(V-2-2),

0.65

/2 X 9.81 X
\j ^
'-

1.0(3.5

80

X
J
^-0.15]
=
-

6.1

m/sec

3. Preliminary Computation of the Required Values of Foundation Weight

and Soil Contact Area. Design computations for determining the required weight
of the foundation corresponding to a unit weight of dropping parts are made from
Eq. (V-5-10). The coefficient of restitution is taken as c = 0.5. The weight of the
anvil and frame corresponding to a unit weight of dropping parts will be:

na

90

or;
7
25.7

According to Eq. (V-5-10), the weight of the foundation corresponding to a unit

weight of dropping parts equals
nf

The required weight

8.0(1

0.5)6.1

25.7

of the foundation (together

3.5

47.3

47.3

with backfill) equals

166 tons

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

H-95-H40H

H40h-95

380

v0- 00
'

12mmd
@200-

217

rx:lOmmd @100
18mmd

(a

200

7*
650

A-

Section

B^i
.

m\

oI

in

A
t

T"
\

o
C\J

T
o
vr

.
m
1

B^
Plan

240-

H-115-H40
w/AW/$j

40 h-115-

$mm, w

550Section

3-8

Fig. V-9. Design of foundation for example of Art. V-G.

The foundation contact area corresponding to a unit weight of dropping parts

determined from Eq. (V-5-11):
a/

20(1

+ 0.5)6.1
20

The required foundation area

in contact

A =

9.2

with

3.5

9.2

soil

32.2

is

cm 2

equals

4. Design of Foundation.
On the basis of the values determined above for the
required foundation weight and area in contact with soil, we design the foundation in
the form of a slab.
The outline of the foundation design is shown in Fig. V-9.

DYNAMICS OF BASES AND FOUNDATIONS

218

The
V,

actual

volume

6.5

The volume

5.5

of backfill

V =
The

total

1.5

3.2

5.5

0.40

1.0

4.60

0.95

weight of the foundation and backfill

3.8

1.15

0.40

58.0

1.6

35.7

Amplitude of Foundation Vibrations.

pad under the anvil to equal

We

58.0

2.2

with

in contact

A =

20.9

20.9

is

The foundation area

is

Wx =

5.

be as follows:

of concrete will

161.4 tons

soil is

6.50

5.5

take the modulus of elasticity

of the

E =
We

10 3 tons/m 2

take the thickness of the pad under the anvil from the design data:

The

50

coefficient of rigidity of the

50

pad under the anvil

10

4.75

of the

hammer

9.18 tons

10 4 tons/m

90/9.81

sec 2 /m

limiting frequency of natural vibrations of the anvil on the oak timber

We

'

12

The square

10 3

of the foundation together

mi =

is

10 3 sec" 2

10 3 tons/m 3

10 3

A;

=
12

3.

Then
10 3 tons/m 3

base under the foundation equals

coefficient of rigidity of the

d =

43

pad

uniform compression to equal

of the correction coefficient

c'u

The mass

f8

set the coefficient of elastic

and the value

104

39 5

cu

The

equal

is

m =
The

will

39.5

0.60

The mass

0.60

35.7

43.8

10 4

with the backfill

=
o'i

16 -5 tons

tons/m

is

sec 2 /m

of the limiting frequency of natural vibrations of the

whole system

is

then

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

The

ratio

between the mass

with the backfill

hammer and

of the

219

the mass of the foundation together

is

9 18
'

Using Eq. (V-3-8), we

up the equation

set

for determining the frequencies of

natural vibrations of the foundation-hammer system:

/n

(1

0.557) (43.0

10 3

16.7

+
-

Or

fn

Solving this equation,

we obtain

/ni.2

92.5

10 3 )/n 2

+ 0.0557) X 43.0 X
+ 1115 X 10 =

(1

10 3/ 2

V(46.25) 2 -

[46.25

10 3

16.7

10 3

1115]10 3

(46.25

32.2) 10 3

Hence we have the frequencies:

fm 2 =

78.5

14.1

/n2

We

determine the

X
X

10 3 sec" 2

initial velocity of

=
=

fnl
2
fn

10 3 sec" 2

288 sec -1
199 sec"

the motion of the anvil together with the

frame:

+0.5)3.5 X6.1

(1

3.5

From Eqs. (V-4-1) we

The amplitude

establish the amplitudes of vibration of the foundation

_ (43.0

The amplitude

10 3

43.0
1.07

0.342 m/sec

90.0

of vibration of the foundation

anvil.

14.1

10 3 )(43.0

10 3 (78.5

10 3

and

is

10 3

14.1

78.5

10 8 )

10 3 )119

mm

of vibrations of the anvil together with the

(43.0

(78.5

10 3

10 3

78.5

14.1

10 3 )V342

10 3 )119

1.6

frame
10- 3

is

m =

1.6

mm

Thus the results of computations show that the amplitude of vibrations of the
foundation will not exceed the permissible value of 1.0 to 1.2 mm.
The dynamic stresses in the pad under the anvil approximately equal

o-

which
6.

is

- A

ci{Aa
-.

much

z)

39.5

10 4 (1.6

X IP" +
3

1.07

10~ 3 )

4.75

The foundation

V-9 according to practical requirements pointed out

type 150f is used for the foundation.

in Fig.

Standard

OOQ
222 ,tons/m 2

smaller than the permissible value of 300 to 350 tons/m 2

Reinforcement of the Foundation.

and design

Illustrative

reinforced as

in Art. V-l.

shown

Concrete

Hammer Foundations. Computation

stamping hammers of different powers, as

Designs of

of foundations for

is

See footnote, Art. IV-I-c, p. 132.

DYNAMICS OF BASES AND FOUNDATIONS

220

well as foundation design for forge

manner

hammers

may

proper,

be performed in

similar to the preceding numerical example.

Figures V-10 to V-17 show several standard illustrative foundation

hammers and

designs for stamping

forge

hammers

of several systems.

In the preparation of these examples, actual data on hammers were

borrowed from instructive design manuals which had been compiled with
the author's participation as a consultant. 18
500

100

1100

^800-

H 500

500 *~ 750- *-75C )-* 500

:0.00>

800

500> 500*

4
m^W^H-^E
t:10mmd

16mmd
@2Q0-

mm d

3200-

<

Section

8 i
i

ir>

CO

1
I

o
o
m

500 ^

1100

IbUU

* 1100

'

'

> *

~T

>

Actual weight of falling parts,

tons
Velocity at beginning of
impact, m/sec
Weight of anvil and frame,
tons
Thickness of pad under
anvil,

Volume

500

lbUU

@100

Concrete type 110

o
o

ni

2500

A -A

>

II

@ 200-

V_ 10 mmd

j|

16

A 7

|!

100

Section

crete,

Weight

B+-

26.6

20

8.7

of foundation

and
22.6

backfill, tons

Fig. V-10. Foundation for 750-kg frictional

5.5

of foundation con-

Foundation contact area,

Plan

0.75

hammer

of

"KLF"

plant.

Design computations for these foundations were performed for a soil of

strength with the coefficient of elastic uniform compression of
soil c u equal to approximately 4 kg/cm 3

medium

V-7.

Computation and Design of Hammer Foundations with

Vibration Absorbers

a.

General Directives

on Computation and Design. Sometimes the

hammer foundations is of great

decrease in vibration amplitudes of

practical importance.

It follows from the approximate Eqs. (V-4-6) and (V-4-7) that vibration
amplitudes of the foundation and anvil are inversely proportional to the
square roots of the products of the base rigidity and mass. Conse-

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

-600^

221

-900 -f*600->

0.00

14

mm d

14

mmd s:

@25-^

@ 25

"'

1I
vx -

1
^

10mmd @1Q0

<

2100-

Section

B-B

Concrete type 110

Actual weight of falling parts,

tons
Velocity at beginning of
impact, m/sec
Weight of anvil and frame, tons
Volume of foundation concrete,

1.0

4.2
6.0

5.0

Weight of foundation and

11.0
5.5

backfill, tons

Foundation contact area,

Plan

Fig. V-ll. Foundation for 1.0-ton free-falling cable

-575

2650-

400

H400]*575>

^-Pour asphalt-^

16mm d@20CK'' 10mm d@

hammer

H 400H

650|400H1400

<650*

"

20

*S10 mmd

@100

16mmd @ 200
-3500

4600
Section

m\wA

10mmd@100
-12mmd@200

0.00

A-A

Section

B-B

B-

qConcrete type 110

S A

Actual weight of falling

parts, tons
Velocity at beginning of
impact, m/sec
Weight of anvil and frame,
tons
Thickness of pad under
anvil,

Volume
575'

400k
fc1325
2300

1325

400^575^

2300-

crete,

Weight

6 5
3

of foundation con-

of foundation

backfill,

Plan

17 7

and
50

tons

Foundation contact area,

Fig. V-12. Foundation for 1.2-ton steam or compressed-air stamping

16

hammer.


DYNAMICS OF BASES AND FOUNDATIONS

222

quently,
tion
to

if

one

is

by increasing

make

to try to decrease the vibration amplitude of a foundaits

mass,

design value of 0.3 to 0.4

to increase the

that this

difficulties will arise; for

example,

the amplitude of vibrations three times smaller

mm instead of 1.0 to 1.2 mm)

it

if

to use a
be necessary

will

weight of the foundation at least nine times.

Similarly, it is very
is impracticable.

It is clear

method

50

^7

77

difficult

-200-

one wishes

(i.e.,

TJ

to

50

@ 20

;ia|E||pgglp;__
""^"lOrnmd

16mmd @

10

25^

350
Section

B-B

Concrete type 110

Actual weight of falling

parts, tons
Velocity at beginning

of impact, m/sec
Thickness of pad under
anvil,

Volume

40

of foundation

concrete,

34

Weight of foundation
and backfill, tons
Foundation contact
area,

Plan

80
15 8

Fig. V-13. Foundation for 2-ton steam or compressed-air forging

hammer.

decrease the amplitude of foundation vibrations by increasing the foundation contact area or the rigidity of the base.

they are considered together with

vibrations of the anvil, depend not only on the parameters of the foundation c and mi (i.e., on the rigidity of the base and the mass of the foundation beneath the springs), but also on the parameters c 2 and ra 2 (i.e., on
the rigidity of the absorbers and the mass of the foundation above the
springs), theoretically it is possible to decrease the amplitudes of foundaSince foundation vibrations,

tion vibrations

by

if

selecting suitable values of c 2

and

2.

These parame-

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

223

be selected so as not to increase sharply the vibration amplitude of the anvil in comparison with values customary in forging practice.
Thus the problem is reduced to the following: values of c 2 and m 2
should be found for which the corresponding vibration amplitudes of the
ters should

foundation and anvil do not exceed selected values. Hammer characteristics (the weight of dropping parts, their velocity at the beginning of
*72.5

40

-325-

40

72.5

0.00

40

-77.5^

40

-165

77.5-:

_2_

Pour asphalt-

m WftM^

mMwm

12mmd
@ 200
16mm

&
^lOmmd

lOmmd @ 100
10mmd O 200-

@100

16mmd

@ 200 -v

@200

400-

550Sectior

A-A

Section

B-B

Concrete type 110

\

ID
s

o
o

C\J

"->

ft

Weight of anvil and

E5

frame, tons
Thickness of pad

Actual weight of falling parts, tons

Velocity at beginning
of impact, m/sec

o
*

ID

under

Volume

anvil,

area,

Plan

6.5

0.40

of foundation

concrete,

B+-

5.5

33

Weight of foundation
and backfill, tons
Foundation contact

*72.5> 40

Fig. V-14. Foundation for 2-ton steam or compressed-air stamping

111

22

hammer.

impact, and the coefficient of restitution) are considered to be assigned

and

fixed.

Let us consider vibrations of the foundation above the springs (the

anvil), as a first approximation, to be a system with one degree of freedom.
Then, according to Eq. (V-4-7), the amplitude of vibrations of this portion
of the foundation equals

Aa

(V-7-1)

VczWi
Equation (V-7-1) was derived from Eq. (V-4-7) as follows, assuming
the foundation does not move the natural frequency fna of the hammer on
:

DYNAMICS OF BASES AND FOUNDATIONS

224

1000 -*k- 1000^

-^-1000

\\Wf^

14mm

@ 200

Section

B-B

Concrete type 110

Actual weight of falling parts,

5

tons
Velocity at beginning of im-

pact, m/sec
Weight of anvil and frame,

4 7

tons
Thickness of pad under anvil,

Volume
crete,

Weight

of foundation con-

16 5

of foundation

and
36 3

backfill, tons

Foundation contact area,

Fig. V-15. Foundation for 1.5-ton cable

the pad under the anvil

=
<J

c2

=
=

hammer,

hammer

weight of

Substituting the above value of fna into Eq. (V-4-7),

where a

is

(1

e)W

V'c2g/W 2

+ e)W v
Vg VC2W2
(1

VC2W2

equaling
(1

denote by

z2

+ e)W v
Vg
Q

(V-7-2)

the static settlement of the foundation above the

springs on absorbers; then

W
z-

c2

We

we obtain

a coefficient depending only on the characteristics of the

hammer and

We

pad

rigidity of

Aa =

is

/..

where

consider z 2 to be fixed.

(V-7-3)

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

40

97.5

225

40 102.5

14mmd

lOmmd
100

@ 200

18mmd
@ 200

18mmd

lOmmd
@ 200

@ 200

Section

B-B

Concrete type 110

Actual weight of falling

ID
CO

in

A
t

ID

Velocity at beginning of

impact, m/sec

Weight

ID

anvil,

i
97.5

40

162.5

162.5

7*

<

-3

DO

Volume

97.5

crete,

40

>*

30

of anvil

6.5

and frame,
92.5

tons
Thickness of pad under

in

3.0

parts, tons

0.60

of foundation con-

52.5

Weight of foundation and

backfill,

261

tons

Foundation contact area,

Plan

Fig. V-16. Foundation for 3-ton steam or compressed-air stamping

30

hammer.

14mmd

@200
18mmd

EC

@ 200-^

*^H0mmd
@ 200
Sectioi i

A-A

Section

B-B

t
m
to

t
in

Concrete type 110

1
o
ID

-)

ft
o

o
id

->

in
to
I

-130-* 50 *-200^ ^-200-^ 50

38C

B+Plan

Fig. V-17.

H30-*

Actual weight of falling

parts, tons
Velocity at beginning of
impact, m/sec
Weight of anvil and frame,
tons
Thickness of pad under
anvil,

Volume
crete,

4 2
6 5

107 5

60

of foundation con-

107

Foundation contact area,

50

Foundation for 4.2-ton steam or compressed-air stamping hammer.

DYNAMICS OF BASES AND FOUNDATIONS

226

From Eqs. (V-7-1) and (V-7-3) we determine approximate values for

the weight of the foundation above the springs and the total coefficient
of rigidity of all absorbers:
2

~ VS

(V-7-4)

c2

(V-7-5)

The mass and

with the

From

soil

the area of the foundation under the springs in contact

on the basis of design considerations.

are selected

the selected parameters of the system, the amplitudes of

its

computed taking into account the fact that the system has
not one but two degrees of freedom. The design values of vibration
amplitudes computed from Eqs. (V-3-9) should not exceed permissible

vibrations are

values.

The installation
hammer, which is

of absorbers should not decrease the efficiency of the

where

is

the

Wi

Kin

w~~

work done by the dropping parts

of the

hammer, equaling

2g

W\

is

the energy lost on the rebound of dropping parts:

W^

Mil

where, approximately,

Vi

in is

the

maximum

on the timber pad or

value of the kinetic energy of vibrations of the anvil

of the

foundation above the springs on the absorbers:

Kv

in

The values
and

its

AJc~
-

W and W\ do not depend on the characteristics of the anvil

base; therefore the following condition

the efficiency of the

hammer without

hammer

with absorbers

subscript

is

must be
not

less

absorbers:

Aa0 2 C20

The

ev

refers to the design

> Aa

C2

without absorbers.

satisfied so that

than that of the

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

227

Since the amplitudes of foundation vibrations above the springs are

about the same whether or not absorbers are used, the last condition may

be rewritten as follows:

<

C2

C2

Computations for a stamping-hammer foundation with absorbers

Example.

The following specifications are given:

1. Data.
Weight of dropping parts of the hammer:

Wo =
Weight

2.0 tons

of anvil together with frame:

33.7 tons

Coefficient of restitution:

0.5

Velocity of dropping parts:

6.0

m/sec

Design values of permissible amplitudes are as follows:

Aa =
A

For the anvil:

For the foundation:

mm
mm

0.2

2. Computations.
The soil is of medium strength, with a coefficient of elastic
uniform compression c u equaling 3.3 kg/cm 3
According to data of Art. V-4, the
value of the coefficient of rigidity of the base under the hammer foundation will be
.

cu

kcu

3.3

10

kg/cm 3

Let us assume that the static settlement of the mass above the springs equals 0.01 m.
We determine the hammer coefficient from Eq. (V-7-2):
(1

a =

+ 0.5)2 X

6.0

_
7_
5.75

V9.81

From Eq.

(V-7-4)
the springs:

we determine

Wl =

the tentative value of the weight of the mass above

10-'

^=

Vl

191 t0nS

The weight
to the

of the concrete block of the foundation above the springs, which

weight of the hammer, is:

W/2 = 191

From Eq.

(V-7-5)

we determine
c2

The foundation above the

5.0

in plan.

It

33.7

is

added

157.3 tons

the required rigidity of the absorbers:

191/10- 2

19,100 tons/m

and 6.0 by
springs is designed as a block of height 2.3
has a depression for the anvil, which is placed not on timber beams

but on a pad made from steel wool.

The foundation under the springs is designed in the shape of a box. The thickness
of the protecting walls will be 0.3 m.
The cross-sectional dimensions of the columns

DYNAMICS OF BASES AND FOUNDATIONS

228

which support the absorbers will be 0.45 m; the thickness of the supporting slab
be 1.0 m. Thus the area of the foundation beneath the springs in contact with

A =

7.76

6.80

52.7

soil

Figure V-18 gives a sketch of the foundation with absorbers.

mass under the springs equals W\ = 180 tons.

The weight

100

of the

100

*-750-

3030

h-750->

Pipe for

will

MM

Depression
drainage

for

of fluids

7760-

Fig. V-18. Design of foundation for example of Art. V-7.

The foundation both above and beneath the

springs

is

built of properly reinforced

concrete type 110. f

Vibrations are computed from Eqs. (V-3-9).

The

coefficient of rigidity of absorbers

is

Co

The mass above the

springs

191/9.81

The mass

c'u

A =

of the foundation

10

10 3

The square

19.5 tons

52.7

beneath the springs

mi = 189/9.81 =

springs

10 4 tons/m

sec 2 /m

under the foundation equals

coefficient of rigidity of the base

ci

1.9

is

m =
The

52.7

10 4 tons/m

is

18.4 tons

sec 2 /m

of the frequency of natural vertical vibrations of the foundation

is

c2

fSee footnote, Art. IV-l-c,

1.9

10 4

19.5
p. 132.

0.975

10 3 sec" 2

above the

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

The square

of the frequency of vibrations of the foundation

/>

X
mi+m

^^Tr
+

18.4

The

between masses

ratio

><

103 SeC

"2

" 106

then set up the frequency Eq. (V-3-8):

(0.975

10 3

10 3 )(1

29.7

/n

we

Solving this equation,

/ 2

then determine the

10 3

0.975

=
=

+ e)W ov
+W

(1

X 10 sec"
X 10 see"
3

29.7

motion

of the foundation

+ 0.5) X 2.0 X
(2.0 + 191)

6.0

above the springs:

0.093 m/sec

The displacements of separate parts of the foundation are found from Eqs.
The displacement of the foundation beneath the springs is

(0.975

0.946

0.093

The displacement
22

10 3

0.975

29.7

0.975

10 3

fe^r

10 3

X
- 29.7 X

^-g

10 3

i! = 0.101

- 29.7 X
X 10

10

10 3 )

= -- 0199

sin

0.946

M+

'

we obtain

mm

'

109 sin

mm

is

10 3

172

\
smfnit)

(V-3-9).

above the springs

/ 0.975
10 3 \

0.946

10 3

0.942

Hmt)

Neglecting terms containing sin fn \t,

Thus the

10 3

of the foundation

10 3 ) (0.975

10 3 (29.7

0.093

0.946

initial velocity of

Wo

21

10 3

find the natural frequencies of the system:

/ni

(1

+ 1.06)/
+ (1 + 1.06) X 13.4 X
X 10 /n + 27.0 X 10 =

13.4

or

We

'

is

is

We

13 4

19.5

229

beneath the springs

_..,_

nn/m*

0.0007 sm/i<t

+
.

no sin/ <
3.03
2
.

mm

for the amplitudes of vibrations

A =
2

selected dimensions of the foundation

3.03

mm

mass above the springs and the

selected value of the coefficient of rigidity of the absorbers lead to an amplitude of

vibrations of the foundation above the springs approximately one-half the design value

mm). The foundation will be practically motionless and no harmful influence

be exercised by the vibrating foundation on structures or on technological processes.
Absorbers are made of cylindrical standard springs used in railway rolling stock.
The dimensions of the springs are chosen as follows:

(0.2
will

Diameter of the coil D:

Diameter of the spring d:

80
30

Number

5.5

of coils n:

mm
mm

DYNAMICS OF BASES AND FOUNDATIONS

230

If the number of absorbers is n\ and the number of springs in each absorber

then the required rigidity of each spring will be

is

n 2}

---

niri2

On

the other hand, as mentioned in Art. IV-6,

d4
940 r- 10 3

c sp

Equating the left-hand parts of the two

tons/m

we obtain

latter expressions,
c2

nm2 =

940^

10

Substituting here the corresponding numerical values as given above,

If

will

each absorber
be

is

1.91

nim =

X 8 X 1Q- X
X 3 X 10-

10 4

made

940

5.5 .. in _ 3

10

of four springs, then the required

ni

70

number

of absorbers

70
70
1Q

= = 18
n
4
2

Thus the actual number of springs will be 72.

The permissible torsional stress for the spring

To

Then the

rigidity of

_
"

Trd*T

8D

3.14

10 3
~

_
~

K
b
6 tonS
-

one spring equals

1.91

10 4

=2

permissible deflection of one spring

20

= ORA
264 ,tons/m
)

is

=
^~
264

c sp

The

X 10~ X 40 X
X 8 X 10-2
6

33
8.8

c sp

The

10 3 tons/m 2

permissible load on each spring will be

^p _
~
The

40

is

0.020

20

mm

actual deflection will be smaller, namely,

z

z2

+A =
2

10

3.03

13.03

We design the absorber to be of the suspension type

mm

Each absorber,
(see Art. IV-6)
placed in a case made of steel channels welded together.
A general view of the absorber as designed is shown in Fig. V-19. The inside dimenThe
sions of the case 5, within which the springs are placed, are 248 by 248 mm.
The lower guide disks 4,
case is fastened by bolts 14 to the lower supporting plate 2.
for springs 10, are fastened by screws 11 to the same plate.
The upper pressure plate

consisting of four springs,

1 is

is

placed on the springs and

is

also

provided with guide disks for the springs.

The

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

231

mass below the springs may be

lifted by tightening the regulating bolt 6 by means of

nuts 9; the anchor cap of this bolt fits between two edges of girders built into the
lower part of the foundation above the springs.
The walls of case 5 containing the absorbers are fastened on shelves formed by
girder pieces embedded into projections of the walls of the foundation below the

springs.

The

cantilevers of girders

embedded

springs should be designed in such a

in the lower part of the

way

foundation above the

that bending stresses produced in

them by

the action of the weight and inertia forces of the foundation above the springs do not
exceed a maximum value.

Construction Procedure for a Foundation with Absorbers.

b.

founda-

tion with absorbers should be constructed as follows:

1.

Place the concrete of the foun-

dation under the springs, walls, and

projections.
2. Place two or three layers of
Ruberoid or tar paper on the sur-

>-4*

^-<

face of the foundation slab.

Install girders on the

Ruberoid

or tar paper, thus forming a lower

frame; projecting sills of this frame

serve as a support for the anchor
plates of regulating bolts of the
absorbers.

Thoroughly check

all

required dimensions and positions

of girders, then weld the frame.
Install the frame in a position corresponding to the design location of

absorbers.
4.

Install

absorbers

without

tightening the springs; insert caps

of

regulating bolts between each

pair of girders forming the frame.

5.
6.

Fig. V-19. Design of vibration absorber

for

example of Art. V-7.

Place the concrete of the foundation above the springs.

After the required period of time has elapsed (not less than 10 days),

erect the

hammer.

When the hammer is mounted,

lift the foundation above the springs.

done by gradually tightening the regulating bolts so that the mass
above the springs is lifted 1 to 1.5 cm without tilting. A level is used
to check that no tilting has occurred.
8. Cover the absorbers and the foundation above the springs with a

7.

This

is

demountable metal

When

plate.

the foregoing procedure has been completed, the

ready for operation.

hammer

is

'

DYNAMICS OF BASES AND FOUNDATIONS

232

V-8. Pressures

in

the Base under a Foundation Subjected to

Horizontal Impacts
If horizontal impacts are transferred from an operating engine to a
massive foundation, natural vibrations of this foundation will develop.
When a horizontal impact occurs in one of the vertical principal planes
of inertia of the foundation, the equations of this vibration do not differ

from Eqs.

(III-4-5)

mx

Mm
=

where x

cT

ALx +

(c v I

c T A(x

- WL +

L<p)

c T AL*)<p

=0

(Van
^v-a-i;

projection on a horizontal axis of displacement of center of

mass

(p

ip

of foundation

angle of rotation of foundation with respect to axis passing

through foundation mass center perpendicular to axis

of

vibrations

m =

mass

of foundation

and engine

W
cT

= weight of foundation and engine

= coefficients of elastic shear, elastic nonuniform compression
A = foundation area in contact with soil
/ = moment of inertia of foundation area in contact with soil,

c,,,

with respect to axis passing through

centroid and per-

its

pendicular to plane of vibrations

Mm

= moment

mass

of inertia of

and engine with

mass

of foundation

respect to axis passing through center of

L =

distance between center of mass and foundation base

Solutions of Eq. (V-8-1) should satisfy the initial conditions;

t

when

0,

where Xq and

<po

(p

(p

(pQ

are respectively the initial velocities of forward motion in

the horizontal direction and of rotation around a horizontal axis passing

They

through the center of mass of the system.

(V-2-6)

and

are established from Eqs.

(V-2-7).

Solutions of Eq. (V-8-1) which correspond to these initial conditions

are as follows:
1

fnl

~-

W
fn2

( frJt
V
V

sin

fnit

fn
2

fnx X0c

fnAU ~ ML

(f?ix

fnl )(.fnx XQc

fnl

i'0

sin/n2

M
'

fn2

&())

sin/nit

fnl

Unx 2

fn2 )(fn X

X Gc

2
fnl X

sin j n ot

(V-8-2)

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

233

where fnx

/ni; fn2

frequency of natural vibrations in shear, accompanied by

sliding of foundation
frequencies of foundation determined from solution of Eq.

Xoc

initial

(III-4-8)

with

velocity of centroid of foundation area in contact

soil

Xq c

From
stresses

Xo

Lip

these solutions for x and <p it is possible to find the dynamic

which develop in the base of a foundation as a result of an

impact.

Dynamic compressive

stress near the

edge of the foundation

contact area will be

o>

where 26

is

a.

-cjoip

(V-8-3)

the length of the foundation contact area.

V-9. Foundations (Bases) for

Location of Drop

hammers

Drop Hammers

Hammer

within

the

works

to

Break Scrap

Iron

Special

Steelworks.

drop

up scrap iron,
pigs, and large blocks.
These hammers are distinguished by the great
kinetic energy of ram impact required to break the scrap.
While in
are installed at metallurgical

for breaking

hammers the kinetic energy at the moment of

impact against the forged piece does not exceed 10 to 12 tons X m, in
modern powerful drop hammers used in scrap yards the kinetic energy of
Therefore these hammers may
the dropping ram attains 150 tons X m.
become a powerful source of elastic waves spreading through soil; sometimes they may also have a harmful influence on various technological
double-acting 5-ton forge

processes.

scrapyards with drop hammers should be located as

from other structures. It will be shown in Chap. VIII
that the propagation of waves through soil is greatly affected by soil
properties; therefore the minimum permissible distance from a shop with
drop hammers will depend on soil conditions. In addition, it is clear

Because of

this,

far as possible

that the greater the kinetic energy of the

hammer, the

greater the energy

waves propagated through

soil, and consequently, other conditions

being equal, the larger should be the values of minimum permissible

of the

distances between the drop

The

hammer and

other structures.

location of a scrapyard with drop

hammers

also

depends on the

character of technological operations in certain structures and on the

vibration amplitudes which are permissible in connection with these

between a scrapyard with drop

smaller than that between a
scrapyard and a laboratory with precision instruments or a shop where
precision machines are operating.
operations.

It is clear that the distance

hammers and a warehouse can be much

DYNAMICS OF BASES AND FOUNDATIONS

234

Generally it is not possible to establish by means of computations the

dependence of the minimum safe distance from a scrapyard with drop
hammers on the three factors indicated above. In each case this problem
should be solved on the basis of the following data: (1) results of experimental investigations of wave propagation at the construction site under
study; (2) values of permissible vibration amplitudes for local technological processes; (3) data on construction characteristics of structures.
Table V-8 gives data on tentative values of minimum distances
depending on power of drop hammers and soil conditions.

Table

V-8.

Data on Minimum Distances between Drop-hammer Installations

Used to Break Up Scrap Iron and Other Structures

Minimum

distances to the drop

hammers,

ram weights

of:

3-7 tons

Over 7 tons

30

50

Over 70

30
50

50
80

Over 70
Over 100

30
20

40
30

Over 60
Over 50

m,

for

Soil conditions

Up
Plastic clays, clays with
silt,

to 3 tons

some sand and

moist sands

Sands, clays, clays with some sand and

silt

below ground-water

Swamp

level

soils

Dry sandy
sand and

soils,
silt,

hard clays, clays with

loess, loessial soils

Rocks

b.

Design of Crushing Platforms under Drop Hammers.

Up

to the

present time, bases under crushing platforms have been designed accord-

methods which have much in common with methods of design and

computation of forge-hammer foundations. However, the energy of
the dropping part (ram) of a crushing drop hammer is many times greater
than the energy of extremely powerful forge hammers; therefore if forgeing to

hammer

foundation requirements are applied to foundations for crushing

platforms, the latter turn out to be extremely heavy blocks sometimes

weighing more than a thousand tons.

Figure V-20 shows a massive foundation designed for a breaking
hammer with a 10-ton ram weight and a drop height of 30 m. The
foundation for the crushing platform was designed similarly to foundations and anvils under forge hammers; the only difference is that sand
and crushed rock were used as a pad under the metal anvil instead of the
oak timber girders generally used in hammer foundations. To reduce the
cost of construction, the lower part was made of Cyclopean concrete;
the upper part is of heavily reinforced concrete type 130. f The total
t

See footnote, Art. IV-l-c,

p. 131

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

235

weight of the whole structure reaches 900 tons, the depth of the foundam, the foundation area in contact with soil 85 2
Technically and economically, such massive foundations cannot be

tion 9

considered rational.

moment

The ram

velocity of the

hammer

just described at

impact is 24.2 m/sec, and the kinetic energy is 295

ton X m. If one considers that the impact occurs not against the scrap
lying on the anvil, but directly against the anvil, so that the coefficient
of impact velocity restitution is of the order of 0.5, then the foundation
should undergo vibrations of an amplitude within the range 5 to 15 mm,
the

of

VW/AVWAWAWAt
Anvil,

-2.25

^ = 110

tons

Sand with crushed stone

Concrete type 130

Foundation under
anvil

-6.0
o".

".'

'".^

Cyclopean concrete

'.':;."&".*. '.-..'4

^-9.o[v:^s^;;tv:;\;:V;-:-;.:v.^;v;-::--:A>e.v>:..:

Z?=8.25

Fig. V-20.

heavy foundation

for a scrap-crushing

hammer

installation.

depending on soil conditions. For a soil with a coefficient of elastic

uniform compression equaling 5 kg/cm 3 the amplitude of foundation
vibrations will be 7.5 mm, and the dynamic pressure on soil will be of the
order of 4 kg/cm 2
The impact of a ram weighing several tons dropping
from a height of 20 to 39
will induce large stresses in the anvil and
foundation. Therefore the foundation under the anvil should be made of
concrete of better quality and should be thoroughly reinforced.
In spite
of this, cases have been recorded in which the anvil and the portion of the
foundation under the anvil were destroyed in the operation of drop
hammers breaking up scrap.
Foundations for crushing platforms can also be designed as hollow
cylinders made of reinforced concrete and filled with sand and small scrap.
Figure V-21 shows a sketch of this type of base. In order to increase
the efficiency of the whole installation, the largest possible degree of
compaction should be achieved in filling up the cylinder. Sand may be
used for filling; compaction can be accomplished by means of vibrations
,

DYNAMICS OF BASES AND FOUNDATIONS

236

applied to successive layers, each layer about 0.5

thick.
covered by a layer of broken scrap to a thickness of 1.5 to 2
with the sand which has been subjected to vibration.

The sand

is

m and mixed
To

protect

against flying chips, joists are suspended on hinges from a metallic ring
installed
If

above the cylinder and are tied to each other by a rope.

the walls of the cylinder are sufficiently high, they

propagated inside of the cylinder and thus

Fig. V-21. Design of foundation for

of

waves

beyond

in the soil

be especially useful

if

it.

may

hammer

may

screen waves

prevent the propagation

to break

up

pig-iron scrap.

Therefore a cylindrical foundation

may

the soils are dangerous in regard to the spreading of

vibrations and settlement under the action of vibratory loads.

The

larger the depth of the cylinder, the larger its effect

ing of waves.

The

frequency of waves induced by the

smaller in loose soils than in dense

soils;

on the screen-

ram impact

is

therefore, other conditions being

equal, cylinder depth should be larger in poor soils than in strong soils.

Waves propagated

in soil

under the action of an impact

may

be classified

as of high frequency, since the number of such waves is of the order of

-1
1,000 min
and more. As we shall see in Chap. VIII, the dimensions of

a screening device should be selected according to the frequency of the

propagating waves. For waves of 1,000 cycles/min and more, the depth

medium strength should not be less than 5 to 6 m.

no known accurate stress analysis of a hollow cylinder, filled
with a material whose strength properties are other than its own, subjected to the action of elastic waves propagated inside of the cylinder.
of the screen in soils of

There

is

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

237

Therefore the stress analysis of the cylinder is necessarily limited by a

very rough approximation of the real distribution of stresses in its walls.
Let us determine approximately the amplitude of vibrations of the
ram which is dropped on the broken scrap the equation of its vibrations
will be
;

where

fng

2cz

(V-9-1)

= damping constant; its value for the case under consideration

may be approximately taken to equal 0.5 to 0.7 fnz
= frequency of natural vertical vibrations of ram, equaling
f
Jnz

cu

+ fH

coefficient of elastic

impacts;

its

cu

Ag

uniform compression of base subjected to

be approximately taken to be of the

may

value

order of 3 to 5

~_

10 3

kg/cm 3

A =
W=

base area of ram

weight of ram
After impact, for a certain time (equal to one-fourth the period of its
natural vibrations) the ram will be pressed into the scrap.
Its velocity
at the beginning of impact will be
v

where h

= y/2gh

(V-9-2)

drop height.
Let us take as the start of readings the instant at which the ram touches
the scrap. The solution of Eq. (V-9-1) for the time
< t < T/4 (where
T is the period of natural vibrations of the ram on the scrap) will be
is

where

is

the

maximum
A,

= A

sin

penetration of the

ram

first

approximation this wave

three-dimensional wave.
will exert

(V-9-4)

C2

=* 0.7/.

wave spreadvolume contained by the cylinder.

against the scrap will induce an elastic

ing from the point of impact over the

In the

into the scrap.

vgw

Ul = VfnS of the

ram

-^exp(^)
Jnz

Impact

(V-9-3)

fnzl t

After the

may

be considered to be a spherical

wave reaches the

cylinder walls,

it

a pressure on them, inducing stresses therein.

Neglecting the absorption of wave energy by the

medium

filling

the

can be approximately estimated that the wave amplitudes

decrease in an inverse proportion to distance from their source.
cylinder,

it

DYNAMICS OF BASES AND FOUNDATIONS

238

If

is

of radius

may

the amplitude of the source acting on the surface in an area

then the amplitude A r of soil at a distance r from the source
,

be approximately taken as

A = A
r

The amplitude
wall

of the

(V-9-5)

wave component perpendicular

to the cylinder

is

A = A

R
r

(V-9-6)

VR2 +
2

R =

where

inside radius of hollow cylinder

= depth

of element of cylinder wall under consideration

Using Eqs. (V-9-4) and (V-9-5) and substituting \/R 2 2 + 2 for

r,

we obtain

When computing A we
mass

neglected the absorption of vibrations by the

Besides, it was assumed that all the impact

consumed only by the formation of elastic waves. As a matter
of fact, a considerable part of the energy is spent in breaking up the
iron blocks and on vibrations of the cylinder, together with the mass it
contains, acting as a solid body on an elastic base.
Therefore the
assumption is possible that the values of A established by Eq. (V-9-7)
are larger than actual.
However, taking into account that the dynamic
wave propagating in the mass contained by the cylinder exerts a dynamic
inside of the cylinder.

energy

is

pressure on the cylinder walls,

it is

possible, in static

strength, to evaluate stresses with sufficient accuracy

computations of

by using the ampli-

tudes established from Eq. (V-9-7)

It follows

from the condition stipulating continuous contact between

the mass included in the cylinder and the cylinder walls that the ampli-

A computed

tudes

for the soil

may

be taken as equaling the amplitudes

of elastic expansion of the cylinder walls.

Stresses in the material of

the cylinder should be established from these amplitudes.

The

elastic

impact wave

is

not propagated instantaneously in the mass

inside the cylinder, but with a certain finite velocity of the order of

2,000 to 2,500 m/sec.

First

it

exerts a pressure on the elements of the

wall situated at the level of the impact.

ward

As the wave

travels in a

down-

it exerts a pressure on
Consequently, under the action of the moving wave, the cylinder is
subjected to a nonuniform pressure along its height which results in a

direction,

the lower elements of the cylinder.

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

239

bending of the cylinder walls as shown by the dashed line in Fig. V-22.
It is very difficult to take into account stresses which develop in the
cylinder walls as a result of this bending, but it is necessary to provide
longitudinal reinforcement along the inner
(for

example, rods of 16

mm

and outer faces

of the cylinder

diameter spaced every 0.30 to 0.40 m)

Fig. V-22. Estimation of stresses in cylindrical walls of foundation in Fig. V-21.

For computation of radial and tangential stresses in the cylinder walls,

may be used.f For radial stresses:

formulas for thick-walled cylinders

R2 2

For tangential

is

R1 2

RA q
P

(V-9-8)

R:

1+ f)

(V-9-9)

the outside radius of the cylinder and

Ri

__

stresses:
<j^

where Ri

RS-RS\

<7 r

<

<

R\

The magnitude of internal pressure acting on a cylinder ring at a depth

below the level of impact is determined by the formula

EA
R
where

and

[(Ri*

R2 2 )/(Ri 2

are the Poisson ratio

R2 2 )

and the modulus

(V-9-10)
v]

of elasticity.

f Cf., for example, S. P. Timoshenko and J. M. Lessels, Applied Elasticity, Westinghouse Technical Night School Press, East Pittsburgh, Pa., 1923. (Translated in

German and

Russian.)

DYNAMICS OF BASES AND FOUNDATIONS

240

Example.
1.

Dynamic computations

for a cylindrical base under a crushing platform

Given Data. A scrap hammer has the

of ram breaking up scrap:

following dimensions:

Weight

5 tons

Area of the ram base:

A = lm 2
Reduced radius

ram base

of

area:

R =

0.56

Radii of the cylinder: internal:

2.

=
=

Ri
Ri

external:

4.5

4.8

m
m

Assumed Data

Coefficient of elastic uniform compression of scrap:

cu

Modulus

kg/cm 3 =

10 3

10 6 tons/m 3

of elasticity of concrete:

E =
The Poisson

10 6 tons/m 2

ratio for concrete:

Damping constant

0.35

for vibrations of the

We

Calculations.

3.

ram on

0.7/

determine the frequency of natural vibrations of the ram:

J3X10 X1X
=
6

fm =

From Eq.

(V-9-2)

scrap:

we determine

= nAA
2.44
X

9.81

the velocity of

in 3, sec" 1
10

ram motion

at the instant of

impact

against scrap:

9.81

15

17.1

m/sec

From Eq (V-9-7) we find the amplitude of the normal component of displacement

caused by the wave propagating in scrap as a result of impact. The computation is
performed for the highest stressed upper zone of the cylinder, i.e., where = 0.
17.1

A "
2.44X10'
From Eq. (V-9-10)
zone of the cylinder:

we

-3 X

10 6

4.5(4.8-

Assuming

in

0.56

/
0.77r\
(' ~2-) =

CIp

find the value of

0.29

X
=

R2,

__

we

on

- 29

10

of the

wave on the upper

find the

"

> 2

4 82

12 9
-

4 52

"

maximum

172 tons/m2

"

value of the tangential

stresses in the cylinder wall:

**

mm

" 10n+
12 9 tons/m = - 11Q1
8 kg/cm
,

0.35

ir._,

dynamic pressure

10- 3

+4.5)/(4.8-4.5')+

Eq. (V-9-9) that p

- 29

17 2 k s/ cm2
'

FOUNDATIONS FOR MACHINES PRODUCING IMPACT LOADS

241

An adequate reinforcement should be installed to resist these tensile stresses.

Since the latter decrease along the depth of the cylinder, the lower zones must be
From computations performed on
reinforced less intensively than the upper zones.
the basis of the above formulas, it is easy to see that in a cylinder having, for example, a depth of 5 m, the lower zone will be under the action of tensile stresses which
are approximately two times smaller than those acting on the upper zone.

Large iron pigs or Martin blocks

Pad made

Fig. V-23. Crushing platform on good

If soils are relatively

of

compacted sand

soil at

a considerable distance from building.

strong and no buildings or shops with technological processes

may be affected by vibrations are located

platform may be made "without a foundation."

which

and scrap placed

Fig. V-23.

directly

on

soil

nearby, the base of the crushing

It may be formed by iron blocks
or on a layer of a compacted sand, as shown in

VI
FRAME FOUNDATIONS FOR MACHINERY

Vl-l. Instructions for the

Design and Construction of Frame Foundations

a. Field of Application of Frame Foundations.

Frame foundations do
not limit a designer in the location of the engine and its auxiliary equipment as do massive foundations. For example, condensers, pipelines,

air vents, and electric wiring for turbodynamos and electromotors can
be arranged much more conveniently if the machines are mounted on
frame foundations.
The use of frame foundations facilitates considerably the inspection of
and access to all parts of the machine. Therefore frame foundations
are often employed for turbodynamos (turboblowers, turbocompressors,
and turbogenerators) of varying power. In the course of recent years
a tendency has appeared, in the practice of foundation design for these

engines, to limit the use of frame foundations to low-power

turbodynamos

only (up to 10 to 12,000 kw), and to use massive foundations for turbodynamos of higher power. However, this tendency is not at all justified,
since observations of frame foundations under high-power
to 100,000

kw) show that these foundations are

in

dynamos (up
cases more

many

economical than massive foundations and, as has been indicated, they

are advantageous in many respects in regard to the mounting and maintenance of the engine. In addition, investigations established that very
often cracks are formed in massive foundations under turbodynamos
due to the stresses induced by settlement or by temperature changes,
while no cracks due to these causes are observed in frame foundations.
Frame foundations can also be successfully used for various electrical
machines, such as motor generators, synchronous compensators, highpower dynamos, and electromotors, in which no sudden changes in load
occur.

Lately there have been cases in industrial design practice where frame
foundations were used for reciprocating engines, in particular for com242

FRAME FOUNDATIONS FOR MACHINERY

243

This use of frame foundations is most rational in cases where

some reason a foundation should have a considerable height; this

pressors.
for

may

happen, for example, if it cuts through a basement.

We shall not consider frame foundations under reciprocating engines,
because these foundations are of such high rigidity that they should be

computed

as rigid bodies resting

on

elastic bases.

Design Assignment. In addition to data on soil conditions, the

following information is required for the design of a foundation:
b.

Foundation diagrams showing dimensions, distribution and required

and openings in the foundation, and distribution and sizes of
foundation bolts and pads under
1.

sizes of pipelines, tunnels, channels, grooves,

bolts

Design Assignment for the

of the condensation
floor within the limits of the edge
of the lower slab of the foundation
3. A Design Assignment for the
installation of a platform around
the turbosystem at floor level of
2.

installation

room
Data concerning the layout

the machine
4.

of

auxiliary equipment, in particular

chambers of the air-cooling appa-

ratus and the generator outlets

5.

diagram

of

static

loads
Fig. VI- 1.

acting on the foundation, imposed

by both stationary and rotating parts

(the

Frame foundation.

magnitudes

of loads

and the

points of their application should be indicated)

6.

7.

Power of the engine in

The distribution of hot

kilowatts and speed

pipelines

and the temperatures at the outer

insulation surfaces

frame foundation (Fig. VI-1) is

more transverse frames embedded
At the top these frames are tied
in a sufficiently thick foundation slab.
together by longitudinal girders and an upper (erection) platform having
openings necessary for stationary machine parts. Often a layout of the
frame foundation is more complicated. Transverse walls are inserted
between columns of the transverse frames, or two-story frames are used.
Sometimes the rigidity of transverse frames is increased by structural
measures to such a degree that the foundation cannot be considered an
elastic frame system, but should be treated as an absolutely rigid body.
c.

Instructions for the Design.

usually designed to be built of three or

DYNAMICS OF BASES AND FOUNDATIONS

244

Figure VI-2 shows an isometric projection of a frame foundation with

transverse walls, designed for a 100,000-kw turbogenerator.
The computations for such a foundation, particularly dynamic computations, are
very complicated. Therefore the foundation should be designed so that
the diagram of stresses transferred from the machine to the base

is

as

simple as possible and secures the most efficient distribution of internal

stresses in the foundation, as well as the simplest forms of foundation

Fig. VI-2. Isometric view of a frame foundation for a 100,000-kw turbogenerator.

vibrations.

In this respect, the foundation design should satisfy the

following conditions:

The geometric layout

sections,

and

of the foundation, the shapes of girder cross

their reinforcement should be basically

symmetric with

respect to a vertical plane passing through the rotation axis of the engine.

The frame beams should be placed

directly under bearings, so that cenwhich develop during engine operation are transmitted
directly to the transverse frames.
Axes of columns and transverse frame
beams should lie in the same vertical plane perpendicular to the rotation
axis of the motor.
To prevent the appearance of torsional stress in
trifugal forces

transverse girders, eccentric loading of the latter should be avoided as

much as possible. The direction of the load should, if possible, pass

through the center of gravity of the beam cross section.

The beams and

FRAME FOUNDATIONS FOR MACHINERY

girders should be designed of rectangular or

accordance with the

minimum
15 cm for

official

T-shaped cross

245

sections.

In

Technical Rules and Construction Code, the

cross-sectional dimensions of unloaded elements should be

slabs

The upper

and 25 cm

for girders.

erection platform of the foundation should be as rigid as

possible in its plane.

One method of achieving this is to extend the

beams towards the outer faces of the

longitudinal and edge transverse

If one attempts to increase the rigidity of the upper platform by extending the dimensions of the horizontal elements of the foundation in the direction of surfaces which limit the space assigned for the
installation of machine parts, this change in dimensions should be coordinated with the machine manufacturer.
In order to increase the general rigidity of the frame foundation,
haunches should be provided at the intersections of beams and columns.
Turbodynamos and electrical machinery are relatively safe in regard
to the transmission of vibrations to buildings.
No cases are on record
of vibrations of entire buildings induced by these machines.
However,
occasionally it happens that turbodynamos cause objectionable local
vibrations in columns, isolated wall sections, and especially floors and

foundation.

other building elements.

An

extensive instrumental investigation of

foundations under turbogenerators was conducted by the author. 4

In

the course of this investigation, considerable vertical floor vibrations

were found in places where the foundation was rigidly connected with
the floor of the machine room.
These vibrations, especially when caused
by high-frequency machines with speeds of, for example, 3,000 rpm, produce a very adverse effect on people standing on the vibrating sections,
The vibraas they cause an unpleasant feeling in the soles of the feet.
tions also result in the displacement of pieces of equipment not tied to
the floor. These phenomena are observed during floor vibrations with
an amplitude of 0.02 mm. For this amplitude and a frequency of 3,000
oscillations per minute the vibration acceleration is about 0.2g.
In order to decrease the transfer of vibrations from the upper erection
platform of the foundation under the turbogenerator to the building, and
particularly to the floor of the machine room, it is recommended that a
gap be provided around the entire contour of the upper foundation platform. The floor beams should be placed on separate columns supported
by footings independent of the machine foundation.
Foundations under low-frequency electrical machines cannot produce
the floor vibrations described above, since the frequency of natural vibrations of the foundations is considerably higher than the operational frequencies of the machines. Therefore in the design of these foundations
there is no necessity to provide a gap between the foundation and the
floor of the machine room.
Bearing floor elements may be supported

DYNAMICS OF BASES AND FOUNDATIONS

246

directly by the frame beams and the columns of the foundation under
low-frequency electrical machines. In some cases such a support may be
effective in decreasing the amplitudes of machine foundation vibrations.
Maintenance records of foundations under high-frequency turbo-

systems indicate cases of relatively large vibrations of cantilevered parts

of the erection platform of the foundation.

Figure VI-3 gives graphs of the distribution of amplitudes of vertical

vibrations of cantilevered elements along one of the foundations investi-

These graphs show that in some places the amplitudes of vibramm, which corresponds to an acceleration of vibrations
equaling some 0.5*7. Vibrations with such high acceleration resulted in
Figure VI-3 indicates
the formation of cracks in the erection platform.
gated.

tions reached 0.06

C^r

0.02

0.013

H-115

f-

0.003
(115 -*!*-

0.017

0.014

0.010

0.023

0.032

155 -*}*- 155 -*)"- 155 -*}155 ->f0.003

0.021

155-4*-

155

0.01

0.004

0.01
0.053
0.033
0.009
155 ->+- 155 -*f+- 155 -*{*- 155 -*|*- 155

0.005

0.006

Fig. VI-3. Recorded vibrations

around a machine foundation.

(in

0.002

0.003

*+- 155 ->}*- 155 *+* 155 *(*-155 -*j*-155 -*f-

155

-"4^155

0.032

0.014
> <

180

0.002

*\

0.002

H* 180

millimeters) of a floor platform cantilevered

(1) a crack in the cantilevered slab and in the edge girder; (2) the zone of
assumed deformation in the platform; (3) a crack in the edge girder.
Such vibrations occur only when the frequency of natural vibrations

of the erection platform, acting as a cantilever of variable cross section,

close to the frequency of machine rotation.
Therefore the cantilever
elements of the foundation erection platform should be designed to be
much more rigid than is required by static computations their frequencies
of natural vibrations should be much higher than the frequency of
is

machine rotation.

The

cantilevered elements of the erection platform usually are

T beams

computation of the frequencies

of natural vibrations of these elements involves some difficulties and is
extremely laborious. The design of the erection platform should ensure
sufficient rigidity of such cantilever elements.
This may be achieved
by the installation of a rigid circumferential edge beam resting directly
on the cantilevers; another method consists in the installation of special
of variable cross sections; therefore the

rigid stiffeners.
The cross-sectional height of the cantilever at the
embedment point should be no less than 60 to 75 per cent of its span.
Turbodynamo and electrical-machine bearings should be thoroughly

adjusted, and the shafts should be in strictly horizontal position.

fore designs of foundations under these

There-

machines should ensure proper

FRAME FOUNDATIONS FOR MACHINERY

For

centering of the masses.

in

mass distribution

is

many machine

permissible

foundation area in contact with

ment
and

up

soil,

of the center of gravity occurs.

electrical

machines should,

case, its value should not

if

exceed

247

foundations an eccentricity

to 5 per cent of the side of the

in the direction in

The

which displaceturbodynamos
to zero; in any

eccentricity in

come

possible,

close

to 2 per cent.

The author studied 36 foundations under turbogenerators and found

that only in 2 foundations the amplitudes of vertical vibrations of the

lower slabs were 0.002 to 0.003 mm; in 3 foundations the amplitudes of

vertical vibrations of the slabs were on the order of 0.001 mm; the amplitudes of vibrations of the lower slabs of the remaining foundations were

The vibration amplitudes of the

1 micron (0.001 mm).
foundation slabs were much smaller than the amplitudes of vertical
vibrations of the upper parts of the foundations.
The results of instrumental measurements of foundation vibrations
smaller than

lead to the conclusion that in practice the lower slabs of foundations

under turbogenerators are not subjected to vibrations and consequently

do not transmit any dynamic pressure to the base. Therefore the pressure on the soil under turbogenerator foundations is determined only by
static loads, i.e., by the weight of the foundation and equipment thereon.

Hence

clear that

it is

it is

not necessary to follow the traditions of recent

practice in assigning design pressures under turbogenerators not to

exceed 0.5 to 0.6 of the permissible pressure on

soils

determined with

respect to static loading only.

The introduction

of a coefficient equaling 0.5 to 0.6

of permissible pressure
piles,

and consequently

on the

soil led to

and the reduction

the necessity for employing

to considerable rise in construction cost.

It

should be noted that the above-mentioned extensive investigation of

machine foundations established that the use

of pile foundations did

not

safeguard against considerable settlements and tilting of foundations

under turbogenerators.

The lower foundation

vibrate at

all;

sure on soil

may

The depth

slabs

under turbogenerators practically do not

therefore the coefficient of reduction of permissible pres-

be taken to equal 0.8 to

1.0.

foundation under turbodynamos and electrical machines

has no effect on the transmission of vibrations to adjacent structures.
Therefore, when necessary because of design considerations or other
of

reasons, the depth of foundation under the

machine

may

smaller than the depth of footings under walls or columns.

be

made even

a foundaunder a turbodynamo is to be erected close to footings under walls,

columns, and other machines, then special care should be taken to protect it from nonuniform stresses imposed by adjacent footings.
Hence
foundations under turbodynamos and electrical machines should be
tion

If

DYNAMICS OF BASES AND FOUNDATIONS

248

placed at such distances from adjacent foundations that the pearshaped

(i.e., the "pressure bulb") in the soil imposed by the

lines of stresses

do not distort significantly the symmetry of the lines of stresses

under the machine foundations in question. For the same reason, in
some cases it may be useful to increase somewhat the depth of machine
foundations with respect to the depth of adjacent footings under walls
or columns.
The lower foundation slab should be sufficiently rigid to secure proper
embedment of the foundation columns and prevent their nonuniform
In addition, the presence of a lower foundation slab having
settlement.
latter

common

considerable thickness decreases the height of the

gravity of the machine and foundation.

center of

Therefore the thickness of the

is usually taken larger than required by static

Tentative values of the height of the foundation slab,

lower foundation slab

computations.

depending on the power

For
For
For
For

machines
machines
machines
machines

with
with
with
with

of the

machine, are taken as follows:

power up to 6,000 kw:

power of 6 to 12,000 kw:
power of 12 to 25,000 kw:
greater power:

0.8 to
1
1

1.2m

to 1.6

m
4 m

6 to 2

2 to

Modern turbodynamos use steam of high temperature; consequently

proper thermic insulation of steam pipes and air lines conducting hot
air should be provided.
The pipes should be insulated at least until
they leave the foundation. The temperature at the outside surface of
the insulation should not exceed 40 to 50C otherwise considerable local
temperature stresses may develop in the foundation. Therefore the
installation of steam and air pipes directly inside the foundation is
;

objectionable.

Frame columns and beams are either reinforced according to design

computations or the reinforcement is fitted to field conditions.
In foundation slabs having a thickness of 1 .0 m, the vertical reinforcing
rods should reach the area in contact with soil. In higher slabs, it is
permissible to cut 50 per cent of the reinforcing rods at the half height
of the foundation slab.
Relevant chapters of the official Technical Rules
and Construction Code for Design of Reinforeed-Concrete Structures should
be used in the design of foundation units, and, in addition, the following
directions should be taken into account: All units of the foundation
should be provided with double reinforcement. A symmetric reinforcement should in all cases be employed in the columns. Reinforcing rods
should also be installed along the other two sides of cross sections of
beams and columns, even if they are not required by design computations.
The amount of reinforcement in separate foundation units should

FRAME FOUNDATIONS FOR MACHINERY

249

The distance between stirrups

less than 30 kg/m of concrete.
beams should not exceed 25 cm, and in columns 35 cm. To resist
stresses induced by settlement, reinforcing rods of 8 to 10 mm diameter
3

be no
in

are to be installed along three mutually perpendicular directions in

massive units of the foundation and are to be spaced 50 to 60 cm apart.
The upper and lower reinforcements of the lower foundation slab should
be tied together by stirrups (dowels) and spaced 50 to 50 cm apart in a
checkerboard pattern. Hooks are to be provided at the ends of steel
rods, subjected both to tensile and to compressive stresses.
When the
reinforcement for the columns is designed, it should be kept in mind that

the total steel area of vertical reinforcing rods in a column should be

smaller than the total cross-sectional area of the anchor foundation bolts.

Additional reinforcing rods should be placed in sections where the

foundation

is

weakened by openings, ducts, etc.

HOf is employed for the upper parts

of frame foundaused for the lower foundation slabs.

The construction of found. Instructions for Construction Operations.
dations under turbodynamos and electrical machines should proceed in
accordance with all requirements of the applicable official Technical
Rules and Construction Code.
It should be noted that large cracks observed in foundations under

Concrete type

tions,

and concrete type 90 f

is

operating turbodynamos and electrical machines are in most cases caused

by

careless construction work.

The construction

of foundations

under

these machines should be carried out with particular care, since the

performance of these machines affects the normal work of many plants.

Special care should be taken in meeting the following requirements in
regard to construction procedures:

Concrete employed for the erection of the foundation should be of

without excessive water; a slump test should show
that the cone slump is around 10 to 12 cm (4 to 5 in.). The same concrete mix should be used throughout the construction of the whole upper
part of the foundation. The forms for the upper part of the foundation
should be fitted with grooves and planed on their inner surface.
Concrete should be poured continuously in horizontal layers. In
an emergency, an interruption may be permitted at the level of the upper
edge of the lower slab, or at the level of one-third of the column height,
where the bending moment has a minimum value. If an interruption
in the work occurs, the following measures should be taken to secure the
monolithic character of the foundation:
1. Along the cross section of the foundation, where the pouring of
concrete was interrupted, 16-mm reinforcing rods should be added
to those installed according to the design.
Short dowels should be
plastic consistency,

See footnote in Art. IV-l-c, p. 132.

DYNAMICS OF BASES AND FOUNDATIONS

250

embedded
and

to a

depth

of

not

less

than 0.5
m.

on both

sides of the joint,

their spacing should not exceed 0.2

The

2.

surface of the joint should be rough.

Prior to placing a

new

layer of concrete, the previously laid surface should be thoroughly

cleaned,

washed by water, and covered with a

As
and a uniform

rich

cement mixture.

a rule, the placing of the foundation concrete should be

mechanized,

distribution of the concrete aggregates should be assured.

segregation of concrete aggregates into layers usually occurs in places

where

it

is

embedded

delivered from considerable height.

If

into the foundation to considerable depth,

anchor bolts are

it is

recommended

that pipes of corresponding cross sections be inserted for these bolts;

During the concreting

and of its aggregates
concrete are to be taken for investigation

these pipes remain in the concrete permanently.

of the foundation, the quality control of concrete

is

essential,

and sample cubes

of

of its strength properties in accordance with special instructions.

All

essential points in the process of foundation construction should be

recorded in special documents.

should be compiled:

made

In any case, the following documents

soil in the excavation

a record of the nature of the

(1)

for the foundation; (2) a record of changes in the type of concrete

used for the foundation it should be noted at what elevation such changes
took place (3) a record concerning the interruption in concreting, if such
an interruption occurred; the place where this interruption took place
should be noted with a description of measures taken to secure a proper
joint; (4) a record of the condition of the concrete after the forms were
;

removed; the length of time the concrete remained in the forms should
be noted.
In the process of machine assembly, prior to pouring cement under the
machine bedplate, the adjoining foundation surface should be cleaned
thoroughly.

This surface (erection platform) should be rough to secure

the best possible binding of the additionally poured cement to the

foundation.

The

location of

all

openings, recesses,

etc.,

should be carefully checked

against design drawings.

VI-2.
a.

The

Computations of Forced Vibrations of Frame Foundations

Exciting Loads Imposed by Turbodynamos and Electrical Machines.
exciting loads imposed

by turbodynamos and

electrical

machines,

unlike those of reciprocating engines and impact mechanisms, cannot

be established by computations.

The main moving

units of these machines are rotors which execute

simple rotating movements.

Theoretically the center of gravity of the

and consequently the theoretiunbalanced inertia forces equal zero.

rotor coincides with the axis of rotation,

cally established values of

FRAME FOUNDATIONS FOR MACHINERY

However, actual conditions are

rotating parts, even

if

different.

In any engine containing

this engine is well balanced,

certain unbalanced state caused

by the

251

there

remains a

fact that the center of gravity

of the rotating parts does not exactly coincide with the axis of rotation.

This residual unbalanced state cannot be completely eliminated, and in

the course of the machine operation there appear unbalanced inertial
forces

which induce foundation vibrations.

The magnitude

of these exciting loads

is

proportional to the eccen-

magnitudes of their masses, and the

square of the frequency of machine rotation. Rotors of high-power
turbodynamos and electrical machines weigh tens of tons, and their
speeds can be very large up to 10,000 rpm. Therefore even for minute
eccentricities of rotating masses the magnitudes of the exciting loads may
be very large. Consequently, their influence should be taken into
account in the design of foundations. For a long time the magnitudes
of exciting loads imposed by turbodynamos were unknown; therefore
in computations of foundations for turbodynamos some "temporary"
The static action of these loads was
loads were taken into account.
assumed to be equivalent to the dynamic action of actual exciting loads
caused by the unbalanced state of the engine.
Many suggestions were offered concerning the selection of these equivalent loads.
However, all these suggestions were equally ungrounded,
and design computations of foundations were reduced to static stress
tricity of the rotating parts, the

analyses of the action of arbitrarily selected loads.

However,

in the course of recent years,

voluminous material has been

turbodynamos and

collected in the U.S.S.R. concerning the balancing of

electromotors, as well as measurements of vibrations of these machines.

This material makes

it

possible to establish design values of exciting

by these machines with a degree of accuracy sufficient for

practical purposes.
Thus it is no longer necessary to introduce into
computations the previously mentioned static equivalents of loads. For
the same reason, the method of foundation design changes: instead of
forces caused

computations taking into account static-equivalent loads, computations

are performed of forced vibrations of foundations produced by exciting
forces and moments.
Consequently the foundation may be so designed
that forced vibration amplitudes do not exceed permissible values.
This method of design of foundations under turbodynamos and electrical machines does not differ in principle from methods of design
accepted, for instance, for foundations under reciprocating engines.
Let us assume that the exciting loads developed by the machine under
consideration can be reduced to one unbalanced centrifugal force F,
whose plane of action coincides with the plane of symmetry of the machine
rotor.
This unbalanced state is generally called the static unbalanced

DYNAMICS OF BASES AND FOUNDATIONS

252

The unbalanced

state.

state of the rotor

may

be caused by the fact that

an exciting moment (the

dynamic unbalanced state). Rotors of electromotor generators, as well
as rotors of steam turbines, are short if the machine under consideration
has one rotor whose plane of symmetry coincides with a vertical plane
in addition to the exciting force there exists also

passing through the center of gravity of the foundation, then the exciting

may be neglected.
unbalanced loads is much more difficult for
machines with several rotors. If a machine has two rotors (which occurs
in the majority of cases), then the force F acting in the vertical transverse
plane of the whole installation (the foundation and machine) can be conThen the exciting
sidered in computations as a design exciting load.
moment

will

be small and

The determination

moment

of

equals

M
where

ls

is

Fl s

main shaft between the

mass of the whole installation.

the distance along the axis of the

resultant of exciting forces and the center of

The

(VI-2-1)

exciting force of the rotor, being the unbalanced centrifugal

force, will rotate with the same frequency as the machine.

Therefore the vertical and horizontal components of the exciting force

inertial

will

equal

F =
Fx =
z

r
r

m
m

co
co

sin wt

cos

= eccentricity of machine rotor

= mass of rotor
oo = rotation frequency
F F x = vertical, horizontal components

where

(TI-2-2)

o)t

ra

z,

in a plane perpendicular to

of exciting force (which act

machine

shaft)

be resolved in the same manner into its

Under the action of the vertical
vertical and horizontal components.
component of this moment, the foundation will undergo forced vibrations

The

exciting

moment can

main shaft of the machine. Measurements

show that foundation vibrations often occur in this plane. However, the

in the plane parallel to the

amplitudes of these vibrations usually are small in comparison with the

amplitudes of vertical and horizontal vibrations in a direction perpendicuTherefore dynamic computation of the
lar to the shaft of the machine.
foundations under turbodynamos and electrical machines may be limited
to computation of the amplitudes of vibrations induced by the exciting
force and the horizontal component of the exciting moment, which equals

= Fz

l8

(VI-2-3)

FRAME FOUNDATIONS FOR MACHINERY

253

The frequency of rotation and the mass of rotating parts of a turbodynamo or an electrical machine are known; consequently, the eccentricity (unbalance) should be known in order to determine the exciting
The eccentricity can be tentatively
loads acting on the foundation.
determined only from the results of balancing of machines and from
measurements of vibrations before and after balancing.
Let us assume that the rotor of the machine under considerawhich
tion has a static unbalanced state, defined by the force F
Let us further
causes an amplitude of forced vibrations equal to An.
assume that in the process of balancing an additional mass was attached
This mass proto the rotor at some distance from the axis of rotation.
duced a centrifugal force F. Then we assume that as a result of the
balancing, the amplitude of forced vibrations decreased to the value A.
There exists a simple proportional relationship between the magnitude
of the exciting force and the amplitude of forced vibrations it produces;
,

therefore

Fq

F =
F =

Hence,

: Fo

An- A

With this relationship it is possible to determine the value of the initial

unbalanced exciting force from the results of balancing and measurements of vibrations before and after balancing. From the value of the
initial unbalanced exciting force, the mass of the rotor, and the frequency
of its rotation, it is easy to determine the eccentricity

Table VI-1 presents data on the balancing of several turbogenerators

and powers. From these data we computed values of

of various types

Fo and

All the data presented in Table VI-1 refer to balancing carried out

under operating conditions, at times when, in the opinion of workers,

machines vibrated with increased amplitudes. Therefore the computed
values of exciting forces and eccentricities lie within the range of maxi-

mum

permissible values.

The amplitudes

of vibrations were measured, not on the foundations,

but on the bearings, at the same place and in the same directions, both
before and after balancing.
Vibrations were measured by means of a
Geiger vibrograph, which in some cases yields considerably exaggerated
readings.
In spite of this, the presented values of F Q and fo permit some

DYNAMICS OF BASES AND FOUNDATIONS

254

conclusions which are of interest for dynamic computations of foundations under machines of the type being considered.

Table VI-1 shows that the eccentricity r depends on both the power
machine and its speed. For machines characterized by 1,500 rpm

of the

Tabile VI-1. Results of Balancing Machine Rotors

Amplitude

Weight
No.

vibrations,

of

mm

Power,

of

kw

rotor,

Prior to

After

tons

balancing

balancing

F,

Fo,

fo,

tons

tons

mm

Machines with 1,500 rpm

1

4
5

6
7

8
9

10
11

5,000
3,000
3,000
3,000
3,000
3,000
3,000
50,000
50,000
50,000
50,000

12.0
5.0
5.0
5.0
5.0
5.0
6.0
70.0
70.0
70.0
37.5

0.160
0.075
0.125
0.125
0.092
0.600f
0.062
0.260
0.350f
0.43f
0.150

0.035
0.042
0.023
0.025
0.027
058
0.026
0.017
0.030
0.035
0.058
.

Machines with
12
13

14
15

16
17

18
19

20
f

17,500
16,000
16,000
16,000
16,000
25,000
25,000
6,000
6,000

This figure

is

18.3
18.0

7.0
18.0
18.0
20.0
17.5
7.2
7.2

0.087
0.157
0.133
0.170
0.140
0.170
0.120
0.180
0.125

3,0C

0.066
0.060
0.056
0.042
0.025
0.058
0.040
0.030
0.021

1.86
0.43
0.92
0.76
0.30
1.23
0.52
11.41
23.5
25.4
3.97

2.38
0.97
1.12
0.95
0.43
1.36
0.80
12.2
25.8
27.3
6.5

0.086
0.085
0.097
0.082
0.038f
0.118
0.070
0.076
0.160
0.170
0.075

rpm
1.04
1.77
1.64

2.36
8.25
5.10
3.02
2.18
1.00

4.3
2.87
2.84
3.10
10.0
7.75
4.53
2.62
1.20

0.025
0.017
045
0.019
.

0.060f
0.042
0.029
0.040
0.018

not reliable.

and 3,000 kw power, the value of eccentricity lies in the relatively narrow
range from 0.070 to 0.118 mm. In only one case the eccentricity was
much smaller (0.038 mm). For machines with the same speed but
50,000 kw power, the value of eccentricity varies within a somewhat
wider range: from 0.075 to 0.170 mm. However, the order of eccenmachines remains similar to that for low-power
Consequently, the increase in the magnitude of the exciting

tricity values for these

engines.

FRAME FOUNDATIONS FOR MACHINERY

force observed with

an increase

in

power

is

255

mainly explained by the

increase in the mass of rotating parts.

The data of Table VI-1 which refer to machines characterized by

rpm show that the eccentricity for these machines lies in the range

3,000

0.017 to 0.045
1,500 rpm,

mm.

For these machines, as

the value of the exciting force

machines with speeds of

grows with an increase in

for

power, and consequently with the weight of the rotor.

Thus there is a difference between the orders of eccentricity for

machines with speeds of 1,500 rpm and 3,000 rpm. The maximum value
of eccentricity for the low-speed machines may be taken as 0.20 mm, but
for machines running at 3,000 rpm the maximum eccentricity does not
exceed 0.05 mm. Consequently, it can be held that the eccentricities
of the rotating masses of turbodynamos are approximately inversely
proportional to the squares of their speeds.

As the number

of revolu-

machine parts (provided the power

the same) decreases; therefore high-speed turbodynamos are better

tions increases, the weight of rotating

is

balanced.
Generalizing the above relationship between the eccentricity and the
of revolutions, and selecting 0.20 X 10~ 3
as the design value

number

machines having a speed of 1,500 rpm, we obtain the

following expression for a machine running at N rpm:
of eccentricity for

rN

This relationship

may

-T7y

meters

(VI-2-4)

serve as a basis for the selection of tentative

design values of eccentricity for rotating machine masses characterized

by

different speeds.

b.

Modulus

of Elasticity of Reinforced Concrete.

In the design of rein-

forced-concrete structures subjected only to the action of static loads, the

computations mainly determine maximum stresses and deformations

appearing under the action of primary loading. In this connection, it is
interesting to analyze the behavior of reinforced concrete subjected to

primary loading, or loading of the same sign.

The official Technical Rules and Construction Code gives values of the
modulus of elasticity of concrete established as a result of tests performed
on concrete samples under increasing loading. These tests established a
mean value of the modulus of elasticity of concrete Ec = 210,000 kg/cm 2
to be used in the design of reinforced-concrete structures.
When concrete is subjected to primary loading, the relationship
between load and deformation is nonlinear; therefore the modulus of
elasticity depends on the magnitude of the load and on its sign.
An experimental study of the behavior of reinforced concrete under
imposed loads shows that even small stresses result in a simultaneous

DYNAMICS OF BASES AND FOUNDATIONS

256

appearance of elastic and residual deformations. The relative value of

residual deformations grows with increase in load; it grows faster than
Therefore a "loading-strain" curve tends to turn in the directhe load.
tion of deformations,

and the

stress-strain relationship in concrete,

above

a certain value of stresses, has a nonlinear character; therefore the

modulus

of elasticity

depends on the magnitude of

stress.

Consequently, the modulus of elasticity established as a result of

investigations in which the irreversible part of deformation was not
separated from the total deformation is not the actual modulus of elasticity of the material, just as the coefficient of

subgrade reaction of

not the coefficient of elastic uniform compression of

soil.

soil is

The modulus

of

which is usually employed in design computations

for stresses smaller than the proportionality limit, represents its modulus
The modulus of elasticity may be established
of linear deformability.
after determining the relationship between stresses and the elastic part of
deformation. Corresponding static investigations should be carried out
by means of repeated loading and unloading of samples.
The amplitude of vibrations and natural frequencies of reinforcedconcrete structures depend on the elastic properties of the material, but
not on its characteristics corresponding to residual deformations; therefore the modulus of elasticity of concrete may be determined in the
simplest way from natural or forced vibrations.
As a result of measurements of vibrations of one frame foundation
under a pump and two frame foundations under turbogenerators, the
following values of the modulus of elasticity of reinforced concrete were
found
elasticity of concrete,

For the foundation under the

3

For the

first

pump

10 6 tons/m 2

foundation under the turbogenerator:

4.2

10 6 tons/m 2

For the second foundation under the turbogenerator:

5.78

These values of

are

much

10 6 tons/m 2

higher than those usually used for static

computations.

N. P. Pavliuk and O. A. Savinov investigated a two-column frame made

and found the modulus of elasticity to be some

of concrete type 160f

t

See footnote in Art. IV-l-c,

p. 132.

FRAME FOUNDATIONS FOR MACHINERY

10 tons/m
6

of concrete of

As a

2
.

of the

frame
was established that the value of
10 6 tons/m 2

result of the investigation of a four-column

about the same type,

lay within the range 4.62 to 3.5

From

257

it

beams made
was established to be

results of static investigations of reinforced-concrete

same

concrete, the

modulus

within the range 3.77 to 3.17

The modulus

of elasticity

10 6 tons/m 2

of elasticity of concrete

may

be established by the electro-

acoustic method, in which a sound generator excites in the sample

Table

Results of Acoustic Determination of

Young's Modulus of Concrete

VI-2.

Composition of concrete

Age

Young's modulus,
tons/m 2

of concrete,

days
7

1:2.55:2.55

28
7

1:3.0:3.0

28
7

1:1.93:3.23

28
7

1:2.6:4.05

28
7

1:3.76:3.0

28
7

1:4.65:6.18

28

3.6
3.81

X
X

10 6

3.02
3.81

X
X

10 6

3.53
4.11

X
X

10 6

4.32
3.96

X
X

10 6

3.10
3.67

X
X

10 6

2.95
3.31

X
X

10 6

10 6

10 6

10 6

10 6
10 6
10 6

under investigation longitudinal or transverse vibrations of varying

frequency.
In this way the frequency of natural vibrations of the sample
is determined, from which the modulus of elasticity can be easily established.
In the course of recent years, this method has been widely used
for the determination of elastic constants of very different materials.
Table VI-2 presents results of one such determination of the moduli of
elasticity of various types of concrete.
It is seen from the table that the
modulus of elasticity does not change much with changes in the composi-

show that the

an increase in the age of concrete.
Absolute values of the modulus of elasticity at an age of 28 days were in
no case smaller than 3.0 X 10 6 tons/m 2 the average value established
from six determinations was 3.78 X 10 6 tons/m 2
tion of concrete.

modulus

At the same

time, the test results

of elasticity increases with

DYNAMICS OF BASES AND FOUNDATIONS

258

Thus the experimental data show that the actual modulus of elasticity
2
6
i.e., it is larger than the
is much larger than 2 X 10 tons/m

of concrete

value established by the official Technical Rules and Construction Code.

The foregoing discussion makes it possible to consider that the actual
value of Young's modulus for concrete (at an age not less than 28 to 30
days) is not less than 3 X 10 6 tons/m 2
This
.

value of the modulus should be taken for design

computations.
lb

c.

-*-

Design Dimensions of

7777777?

Fig. VI-4. Design values

computations of
frame vibration frefor

quencies.

moments
(Fig. VI-4)

The

the

Upper Part

of the

dimensions of
foundation units are usually much larger than
the spans between them; hence the influence of
the rigidity of corner sections of the frame should
be taken into account. If one is to consider the
corner sections as being absolutely flexible, then
in the determination of deflections and bending
Foundation.

and the height

cross-sectional

of separate

frame elements, the span

ho should be introduced into the computations.

However, if one considers the frame corner sections as being absolutely

becomes necessary to use in computations the inside free span Zi
and the inside free height hi of the frame. For usual foundation sizes,
the value of U often exceeds the value of h by 25 per cent.
Formulas for
deflection computations contain the value of the span in the third or
fourth power. Therefore design values of the span and height considerrigid, it

ably affect the results of computations, in particular the value of the

natural vibrations of the frame. For example, if one is to calculate
frequencies of natural vibrations of a frame having U = 5.50
and
h = 4.00 m, then for different design values of the span (from I = U to
I = li),
will be
the frequency of vibrations computed for l = 5.50
(for the case in which the frame is loaded only by its own weight) approxi-

mately two times smaller than the frequency for the case where I = 4.00 m.
As a matter of fact, the frame corner sections are neither absolutely
flexible nor absolutely rigid; therefore design values of the span and
height of the frame should be smaller than l and ho, and larger than h
and hi. They should be determined from the following expressions:
I

= k - 2ab
= ho 2aa

(VI-2-4a)

where 2 a = height of frame beam (Fig. VI-4)

2 6 = width of column
The value of coefficient a is taken from the graph (Fig. VI-5). When
this graph is used, intermediate values of ho and l should be determined
by interpolation.

FRAME FOUNDATIONS FOR MACHINERY

259

haunches are provided in frame or beam corner sections, the values

b are taken as shown in Fig. VI-6.
The upper
d. Rigidity of the Upper Platform of the Foundation.
foundation platform on which the machine is placed is formed by longitudinal beams which tie the transverse frames together and a reinforcedIf

of a

and

0.40

0.30

ol

0.20

0.10

0.04

0.08

0.12

0.16

0.20

V/o
Fig. VI-5.

Graph

for

determination of coefficient

a.

in Eqs. (VI-2-4a).

concrete slab provided with openings required for pipes, machine parts,

The

depends essentially
bedframe of the machine.
The upper foundation platform together with the bedframe of the
machine represents a structure which is extremely rigid
in the horizontal direction; therefore, for an approximate
determination of the amplitude of horizontal vibrations
of the foundation, the upper platform may be considered
i
This assumption simplifies comto be absolutely rigid.
putations of horizontal vibrations of the frame foundaJ.
tion but does not involve large errors in the results of
condensers, and so on.

on the

rigidity

and

rigidity of this platform

relative position of the

computations.

Fig. VI-6. Val-

Computation of Forced Vertical Vibrations. The

foregoing assumptions reduce a dynamic computation
of the frame foundation to computations of the amplitudes of forced vibrations of a three-dimensional frame
system consisting of thin beams and columns embedded

ues of a and

e.

be used in
Eqs. (VI-2-4o)
if a frame has
haunches.

to

an absolutely rigid slab. The latter rests on an elastic base, i.e.,

on soil.
Although it is possible to obtain rigorous solutions to the problem of
forced vibrations of such a system, 30 the solutions obtained are so cumbersome and lead to such complicated calculations that they are of little

in

use for practical purposes.

Therefore several assumptions are necessary

DYNAMICS OF BASES AND FOUNDATIONS

260

which will simplify the solution of the problem without influencing

accuracy and will make the solution practical.

its

When vertical vibrations occur, the most intensive deformations are

observed in the beams of the transverse frames. The columns of the
frames are deformed less, and the corner haunch sections of the plane
transverse frames do not deform at all.
If a separate frame vibrates at a
frequency not too large in comparison with its smallest natural frequency,
then the form of vertical vibrations of this frame approximately corresponds to that shown in Fig. VI-7.
Let us assume that the transverse frames of the foundation are subjected to vertical vibrations in one phase.
Let us also assume that
differences in vertical deformations of the columns of separate frames are
Then

small.

sin

ut

/A,

T" J2

of their bending,

l*i

^C^j |L^c-^ivv
m

lr

the

elastic

resistance

of

longitudinal beams, developing as a result

T"

Ay

w ill be
T

small in compari-

W^n

^ ne elastic resistance against

longitudinal compression of the columns.
Longitudinal beams are usually fixed only

son

at the corner sections of the transverse

Therefore the vertical vibrations

frames.

r7^77777777777777777T/7T/

Fig. VI-7. Vibrations of a separate frame.

of a transverse

frame are also affected by

the

to

resistance

tudinal

beams

small, as

is

'

torsion

of

the longi-

This resistance

is

also

the resistance resulting from

the bending of the longitudinal beams.

Therefore the influence of longitudinal beams may be disregarded.

By

neglecting the influence of longitudinal

vibrations of the transverse frames,

independently of each other.

it is

beams on the

vertical

possible to consider the vibrations

The natural

frequencies of vertical vibra-

of such an assumption,
be somewhat smaller than the actual values. In order to compensate
for the influence of the longitudinal beams, we shall disregard the actions
of other factors which affect natural frequencies in an opposite manner.

tions of separate frames, calculated

on the basis

will

These factors include the shearing forces and the inertia

of rotation of

cross sections of the units of the frame.

Lur'ye 29 showed that the frequencies of natural vibrations of the

frames may be strongly influenced by the elasticity of the base under the
foundation.
Computations show that the frequencies of natural vertical
vibrations, computed with consideration of the elasticity of the base
A.

I.

under the foundation, may differ by 10 to 20 per cent from the values
computed without taking this factor into account.
The frequency of natural vertical vibrations fnz of the frame considered
to be a rigid solid resting on an elastic base is usually much smaller than

FRAME FOUNDATIONS FOR MACHINERY

261

We

the operational frequency of the turbodynamo.

denote by fn i the

smallest frequency of natural vertical vibrations of the frame; assuming

that the base

elastic

and

is

absolutely rigid, this frequency depends only on the

inertial properties of the frame.

For turbodynamos there

usually exists the following interrelationship

/,

<

< fm

(VI-2-5)

is the frequency of rotation of the machine.

where
For low-speed electrical machines (for example, motor generators) the
frequency of rotation may also be smaller than fnz therefore the following
<x>

interrelationship

may

exist

</< fm

(VI-2-6)

If one is to consider the elasticity of the base under the foundation,

then the two smallest frequencies of vertical vibrations of the system

"frame-rigid-slab-elastic-base" will have the following interrelationship

with the frequencies/^,

w,

ft

The

and/n i:

<fm<o>< fnl < f^

show that consideration of the elasticity of soil

an increase in the fundamental frequency of natural vibrations
based only on the elastic and inertial properties of the frame.
latter inequalities

leads to

If

requirement (VI-2-5) or (VI-2-6)

is

satisfied in the

foundation, then neglecting the elastic properties of

soil in

tion of vertical vibrations of the foundation contributes to

design of a

the computa-

an increase

in

the safety factor of the dynamic stability of the foundation.

Consequently, the computation of forced vibrations of a frame foundation may

be reduced to the computation of vibrations of plane frames resting on an
absolutely rigid base.
Then the columns of each frame may be considered
as being rigidly embedded in an immovable foundation slab.
Let us consider the loads acting on a transverse foundation frame.
Each frame usually supports one of the machine bearings. The width of
bearings supporting the rotating machine parts usually is small in comTherefore the load transmitted by
parison with the length of the beams.
the bearings may be considered concentrated and located in the middle
of the frame beam.
The static load is the part of the rotor weight resting
on this bearing. In addition, the same bearing transmits to the frame

an exciting

vertical force

sin ut.

The frame beam

the action of a uniform load imposed by

it

its

own

by an equivalent concentrated load located

beam

span.

is

also subjected to

weight.

Let us replace

at the center of the frame

In order that this change should not influence the results of

dynamic computations, the magnitude

of this equivalent load should

be

DYNAMICS OF BASES AND FOUNDATIONS

262

selected so that the kinetic energy of the system will not change.

out furnishing proof here,

let

With-

us note that the magnitude of an equivalent

mass, selected on the basis of the above condition, corresponds to 45 per

cent of the dead weight of the frame beam.

The frame columns are subjected to the following loads:

1. The loads imposed by adjacent longitudinal frame beams, including
their own weight.
These loads, falling on each frame, are computed
according to the laws of statics. Loads imposed by longitudinal beams

may

be considered concentrated at the tops of the columns, since the

beams are supported by the corner sections of the transverse

longitudinal

frames.
2. The loads imposed by the weight of the transverse beam, also
concentrated at the tops of the columns. On the basis of the same considerations which governed the selection of the equivalent concentrated

mass at the center

beam span, the dead weight loads imposed by

beam on each column are taken to equal 25.5 per

of the

the adjacent transverse

cent of the weight of this beam.

3.

The weight

by an equivalent weight load

of the column, replaced

concentrated at the top of the column. It follows from the theory of

longitudinal vibrations of prismatic bars that the value of this load
should be equal to 33 per cent of the column weight.

dead weight loads, we come to the considerawhose elements are weightless and
whose masses are concentrated in two places (Fig. VI-7) one mass m 2 at
the center of the frame beam span, and two other masses, each equaling
mi/2, at the tops of the columns. Vibrations of all frame units are
determined by the vertical displacements z\ and z 2 of these masses. In
this manner, the problem of vertical forced vibrations of transverse
frames of the foundation is reduced to the problem of vibrations of a
system with two degrees of freedom. It is assumed that the exciting
vertical force acts on mass m 2
Let us denote by c 2 the coefficient of rigidity of the frame beam; this
coefficient represents a vertical force which should be applied to the
center of the frame beam span in order to cause a deflection of unit length,

As a

result of reducing the

tion of vibrations of a plane frame

i.e.,

c2

The value

of

l2

is

= r

(VI-2-7)

determined by the formula

l

^ l+2k)

96/*(2

k)

Mi
II

+JL
SGA

(VI-2-8)

'

(VI-2-9)

FRAME FOUNDATIONS FOR MACHINERY

where E,

G =

Young's modulus,

modulus

beam
area, moment
area, moment

of

263

elasticity

in

shear

of

material of frame

A
Ah

frame beam
column
Let us denote by ci/2 the coefficient of rigidity of a column; this
coefficient represents the vertical force which must be applied to the
column in order to cause a unit change in its length it is evident that
i,

cross-sectional

cross-sectional

of inertia of
of inertia of

%=

(VI-2-10)

The

differential equations of forced vibrations of the system shown in

VI-7 will be exactly the same as Eqs. (IV-6-1), and the solutions for
the amplitudes are determined by Eqs. (IV-6-4) and (IV-6-5).
The differential equations (IV-6-1) do not take into account the influence of damping reactions; so the computations of amplitudes of forced
vibrations, presented in Chap. IV, will produce adequate results only in
cases in which the fundamental frequencies of the system (Fig. VI-7)
differ by at least 30 per cent from the frequency of machine rotation.
If this condition is not satisfied, and forced vibrations of the foundation
occur in the resonance zone, then the use of the foregoing equations leads
to large errors in the determination of the amplitudes of forced vibrations.
The natural frequencies of vertical vibrations of the foundation will be
determined as roots of Eq. (IV-6-8). The solution of this equation will
provide the two natural frequencies fn i and fn2 of the frame under consideration.
Let us assume that the frequency of excitation is close to
one of these frequencies.

Fig.

The

following relationship usually exists between the limiting fre-

quencies

fi

and/n2

fl

< fnz

Therefore the frequencies of natural vertical vibrations

< fl <

Jn2

fnz

<

lie

in the range

fnl

If the frequency of excitation lies close to the lower natural frequency,

then the form of the frame vibrations does not differ much from the form
of vibrations of frequency f, and it can be considered with sufficient
accuracy that the system under consideration has one degree of freedom
therefore the amplitude of forced vibrations of the system is determined

by the expression

V(/n2

(VI-2-11)
a?

-f 4c

DYNAMICS OF BASES AND FOUNDATIONS

264

where

c is

the

damping constant.

then, approximately,

A* =

If

the difference fn2 2

&

co

is

<

small,

VI - 2 - 12 >

An
of

/n2

approximate value of the damping constant

is

5 to 10 per cent

If the frequency of excitation lies close to the higher natural frequency

fn i, and the form of frame vibrations does not differ much from the form
of vibrations of frequency fnz then A 2 is determined from formula
(VI-2-11) or (VI-2-12), in which fn2 and m 2 are replaced by fn i and
,

(mi

+m

2 ).

Computation of Horizontal Transverse Vibrations. In the computation of forced horizontal vibrations of the frame foundation we neglect
the elasticity of the upper slab and the soil; i.e., we assume that the slab
is absolutely rigid and the frame columns are embedded in an unyielding
/.

foundation.

Let us consider, for example, a foundation having six columns and

We replace all vibrating masses of the foundation and machine by three equivalent masses mi, m 2 and m 3 each located
Each of these masses
at the center of one transverse frame beam span.
is computed by adding the following:
three transverse frames.

1.

The mass

of the concentrated

the frame beam, including

2.

its

and distributed deadweight load on

own weight

The mass formed by 30 per cent

of the

weight of the two columns of

the transverse frame

3. The mass
beams adjacent

of the

deadweight transferred by longitudinal frame

to the transverse frame under consideration, their

own

weight included

The amplitudes
these masses

of forced transverse vibrations

and on values characterizing the

depend on the

sizes of

rigidity of transverse

frames.

Let us replace each transverse frame by an equivalent spring (Fig.

VI-8), and the upper slab by a prismatic bar which is assumed to be
absolutely rigid.
The motion of the system (Fig. VI-8) is determined by
x, the lateral displacement of the center of mass of the prismatic bar,
and by <p, the angle of its rotation with respect to the center of mass.
Consequently the system has two degrees of freedom; i.e., it has two
natural frequencies of vibration.

The

differential equations describing the transverse horizontal vibra-

FRAME FOUNDATIONS FOR MACHINERY

265

by
<z 2 -*

cz

*-0
c

J.

Trn.
x

Fig. VI-8. Derivation of Eq. (VI-2-13).

tions of the foundation in the case

mx

under consideration read as follows:

sin ot

(VI-2-13)

Mm

lp

+ V Mi =
i

where

m = Y

A^ w

= T

= sum

m ai =
2

of all equivalent

moment

total

with respect to
di

M sin ut

=i

masses

of inertia of all equivalent

masses

common center of mass

common center of mass and mass m; we

distance between

consider these distances to be positive in one direction and

negative in opposite direction

Ri

elastic force acting

ment up

Mi = moment

on mass m during

its

forward displace-

to value xt
of inertia of force Ri with respect to

an

axis passing

through center of mass

P,

exciting force,

moment

The summation should be performed for all transverse frames, so that

n denotes the number of transverse foundation frames (usually n equals
3 or 4).

When

the mass nu

is

displaced by the value

the equivalent springs acting thereon equals

Ri

where

is

the rigidity of the ^th spring.

The moment

of the elastic force

Mi =
Since

Xi

is

CidiXi

Clip

Xi,

then the

elastic force of

DYNAMICS OF BASES AND FOUNDATIONS

266

we may

write

Ri

and

Ri

where

e =
C =

=
=

+ anp)
+ e<p)C

Ci(x

(z

distance between center of mass and center of rigidity

Sci

To determine

total rigidity of equivalent springs

the rigidity

of the ith transverse frame, let us

horizontal force to the center of the frame

unity and

(VI-2-14)

beam

span.

apply a
This force equals

directed along the axis of the frame beam.

is

that the lateral displacement

It is

known

caused by this force equals

+ )
+

._

12EI h (l

(VI . 2 . 15)

6fc)

Here, as before,

fc-Mf
k ~
ih

where h
I

= height of column
= length of frame beam
= moments of inertia of

h, Ii
Having determined the
the rigidity

beam
we find

cross sections of column, frame

deflection

for the ^th transverse frame,

characterizing the equivalent spring corresponding to this

frame

* =
In the same

way we

is

the

find

XMi = XaciiXi = C{x

y = 2cA

where

(VI-2-16)

&p)*

7<P

(VI-2-17)

moment

of a couple causing the rotation of the prismatic bar

through a unit angle.

Substituting the computed values of 2/2* and

we obtain two

t>

+ ceip
+ y)<p
x + fnx x + fnx
+ ~ fnJ + fn 9 \<f
mx

Mm

Cex

Cx

(Ce 2
2

or

e<P

The

XMi

into Eqs. (VI-2-13),

differential equations of forced vibration of the

pfnJz

= P sin cot
= M sin ut
= V sin oot

foundation

(VI-2-18)
(VI-2-19)

= RsmO0t

following designations were used in the foregoing formulas:

f*

(VI-2-20)

FRAME FOUNDATIONS FOR MACHINERY

267

where fnx is the limiting frequency of natural lateral vibrations of the

foundation when the center of rigidity of the foundation coincides with
the center of mass, i.e., when e = 0;

/--
where fn

<p

is

the limiting frequency of natural rocking vibrations of the

foundation when

0;

where

r is

M
m

the radius of gyration; finally

L
m

We

(VI-2-21)

*?
K

M,

Mm

seek solutions of the system of Eqs. (VI-2-19) corresponding only

form

to the forced vibrations of foundations in the

= Ax

sin

cot

<p

Inserting these values of x and

A9

sin ut

into Eq. (VI-2-19),

<p

we obtain

the

following two equations containing the amplitudes of forced vibrations

A x and A v

unknown

of the foundation as

(fnx *

/2

K/M

^fnx 2 A x

Solving these equations for

A
Kx
~

[(e

<**)A X

A(co 2 )

/r 2 )/n . 2 p

co

(afnx *

obtain
2

(VI-2-22)

(fn J

- ^)R

(VI-2-23)

+ /n

)co

+tV

(VI-2-24)

-!

~7i

A(co 2 )

above formulas

of foundation vibrations are deter-

co

differs

(VI-2-25)

for the determination of

of forced vibration of the foundation

frequency of excitement

The natural frequencies fni and /n2

mined as roots of the equation

It is clear that the

CO

-a>*]v-U R

+fn* 2

A(co

where

\
J

A x and A VJ we

/r*)fn X

= p
A = R

+ fnx hA

fn<p

A(co
(e

values:

may

amplitudes

be applied only

when

the

from the fundamental frequencies of

DYNAMICS OF BASES AND FOUNDATIONS

268

the

foundation.

Otherwise,

computed analogously

when one

vibrations

to the

the amplitudes of vibration should be

computation of the amplitudes of vertical

of the natural frequencies

is

close to the engine

frequency.

For example,

if co

to the lower frequency fn2 determined as the

lies close

fn2 does not differ much from fnx then the

foundation will be subjected predominantly to vibrations accompanied by
lateral displacements as a rigid body.
The amplitude of vibrations may
root of Eq. (VI-2-25),

and

if

be approximately established from Eq. (VI-2-11) where ra 2

equal m.

If

is

assumed

to

resonance occurs because the frequency of machine rotation

the second frequency fnX and the latter does not differ much
then
the foundation will undergo chiefly rocking vibrations.
f,p
In this case, an approximate value of the amplitude of vibrations may
be found from the same Eq. (VI-2-11), in which P, m 2 and/n2 should be

lies close to

from

replaced respectively by
g.

Design Values of

M,

Mm

and /nl

the Permissible

Amplitude of Vibrations.

Design

values of the permissible amplitude of vibrations of foundations for

turbodynamos and electrical machines should be established on the basis

from the study of operating machines. It is hardly
possible to establish these limits on the basis of any theoretical premises.
As a matter of fact, if a permissible amplitude of vibrations is established on the basis of permissible stresses for the foundation materials,
it is found that the computed amplitudes are tens of times larger than
those permissible for normal machine operation.
Therefore a selection
of the design value of vibration amplitude should be based on the amplitudes accepted by machine operators as permissible for a given machine
Table VI-3 presents data on the permissible values of amplitudes
type.
of data derived

of vibrations of turbogenerator bearings.

The absolute values

much

may be
3,000-rpm

of permissible amplitudes of vibrations

larger for machines running at

1,500

rpm than

for

one admits that the permissible amplitude of vibrations

be taken as the arithmetic mean of the values given in
Table VI-3 for machines with certain speeds, then the following values
may be accepted:

machines.

If

of bearings

may

For 3,000-rpm machines:

Vertical vibrations:

0.02 to 0.03

Horizontal vibrations:

0.04 to 0.05

mm
mm

For 1,500-rpm machines:

Vertical vibrations:

0.04 to 0.06

Horizontal vibrations:

0.07 to 0.09

mm
mm

Vibration investigations of 36 foundations under turbodynamos estabdo not exceed the above per-

lished that actual vibration amplitudes

FRAME FOUNDATIONS FOR MACHINERY

missible values.

Only

in

269

foundation was an amplitude found as high as

mm.

In 4 foundations, the amplitude of vibrations of the upper

In all
part of the foundation lay within the range 0.010 to 0.016 mm.
other foundations the amplitudes of vibrations were within the range
0.016

Table

Amplitude Values of
Turbogenerator Bearings

VI-3. Permissible Vibration

Amplitudes f of vibrations,

Type

of

vibrations

ments

Extreme

of:

Evaluation of
the engine
3,000

Vertical

mm,

corresponding to speeds, rpm,

Location of
measure-

Is

fit

for operation

2,500

2,000

1,500

1,000

and

less

0.02

0.03

0.04

0.06

0.08

0.03

0.05

0.06

0.09

0.11

0.04

0.08

0.09

0.13

0.15

0.01

0.02

0.03

0.04

0.05

0.02

0.03

0.05

0.06

0.08

0.03

0.04

0.08

0.09

0.13

0.05

0.07

0.08

0.09

0.12

0.08

0.10

0.11

0.12

0.15

0.13

0.14

0.15

0.17

0.20

0.03

0.04

0.05

0.07

0.09

0.05

0.06

0.08

0.10

0.12

0.08

0.09

0.13

0.14

0.17

bearings

No adjustment

is

needed
An adjustment

is

desirable

Central
bearings

Is

fit

for operation

No adjustment

is

needed
An adjustment

is

desirable

Horizontal

and

Extreme

Is

fit

for operation

bearings

trans-

No adjustment

is

verse

needed
An adjustment

is

desirable

Central
bearings

Is

fit

for operation

No adjustment

is

needed
An adjustment

is

desirable

The

largest permissible values of amplitudes are presented.

0.004 to 0.010

mm;

i.e.,

they were considerably smaller than permissible

values.

Vibrations with an amplitude of 0.016

normal operation of the turbogenerator.

mm, had

no influence on the

Permissible values of foundation vibration amplitudes under electric

machines having speeds close to 1,500 or 3,000 rpm may be the same as

DYNAMICS OF BASES AND FOUNDATIONS

270

those for turbogenerators.

It is

very

difficult to select

even tentative

design values of permissible amplitudes of vibrations for electric machines

with speeds lower than 1,500 rpm, since scarcely any data are available
on the results of vibration investigations of these machines. For lowfrequency electrical machines (less than 500 rpm) such as motor genera-

and Leonard generators, a design value of permissible vibration

may be selected on the basis of values established for recip-

tors

amplitudes

rocating engines (around 0.20

mm).

Examples of Dynamic Computations of Foundations of the

Frame Type

VI-3.

Example

Illustrative design of a

1.

frame foundation under a 1,500-kw turbogenerator

= 3,000 rpm; i.e., the frequency of

is
9 X 10 4 sec" 2
a ~ 300 sec *; w 2
The foundation to be designed will have six columns and three transverse frames.
Table VI-4 gives the initial data required for the dynamic computations of the
foundation which are taken from the Design Assigment. Figure VI-9 shows the
geometry of the foundation with indication of the loads imposed by the stationary and
rotating parts of the machine.
The speed

Data.

1.

excitation

of the turbogenerator

is

Table

VI-4.

Design Data for Computations of Example VI-3-1

and Fig. VI-9

Frame

Dimensions and design parameters

Height of transverse frames ho,

Span of transverse frames l Qj
Height of cross section of transverse frame
Height of cross section of column 26, m
Area of cross section of column Ah, m 2

Moment

umn

Cross-sectional area of frame

h,

Weight
Weight

The

Frame

II

Frame

III

4.30
3.20
0.95
O.SO
0.76

4.30
3.20
1.00
0.83
0.78

4.30
3.20
1.00
0.80
0.76

0.039
0.58

0.041
0.83

0.039
0.64

0.0425
4.26
6.53

0.069
6.10
6.67

0.053
4.70
6.53

of inertia of cross-sectional area of col-

h,

Moment

2a,

beam

Ai,

of inertia of cross-sectional area of

beam

of

frame beam Wi, tons

of

column

Wh

loads imposed

tons

by the rotating parts of the turbogenerator act only on the beams

The loads imposed by the stationary parts (the stator of the

of the transverse frames.

generator and the cover of the turbine) are transmitted to the longitudinal beams.
The design value of the modulus of elasticity for the material of the upper part of
the foundation (concrete type 110f) we assume to equal = 3 X 10 6 tons/m 2
2. Computations.
The computation of forced vertical vibrations of the founda.

tion

is

begun by some preliminary computations.

Table VI-6 presents results

of

Table VI-5 gives their results.

computations of equivalent masses mi and m 2 for each

of the transverse frames.

t

See footnote, Art. IY-l-c,

p. 132.

FRAME FOUNDATIONS FOR MACHINERY

271

Results of computations of equivalent rigidities are given in Table VI-7, and those
of limiting frequencies of each

frame in Table VI-8.

0.60

To

&

<>y

0.64

0.83

h=

3.14

2.82

o
CO

ro
CD

CO

W/,

0.95

.0.95.

*-1. 57-^-1.57

<

1.41

O*

'///A

X-

41

0.95 *"

-*

4.5 tons

6.0 tons

7 tons

5 tons

3.5 tons

4.5 tons

6.0 tons

Fig. VI-9. Foundation of

Example VI-3-1.

Let us now compute the amplitudes of forced vertical vibrations of the foundation.
According to the foregoing data, the design value of the eccentricity of rotating masses
of the turbogenerator is taken to equal
r

The weights

0.05

of rotating parts falling

Wx =
The magnitude
tion

1.5 tons

10" 3

on each frame equal

2.0 tons

of exciting vertical load acting

1.0

ton

on each transverse frame of the founda-

is:

Frame
Frame
Frame

0.05

P =
P =
2

0.05

0.05

I:

Pi

II:

III:

X
X
X

10~ 3
10~ 3
10~ 3

X 1.5/9.81 X 9 X 10 = 0.69 ton

X 2.0/9.81 X 9 X 10 = 0.92 ton
X 1.0/9.81 X 9 X 10 = 0.46 ton
4

DYNAMICS OF BASES AND FOUNDATIONS

272

Table

Results of Preliminary Computations

for Example VI-3-1, Fig. VI-9

VI-5.

Frame

Design parameters
4.30

ho

Frame

4.30

0.40

h,

4.30

4.22

3.20

3.06

[fromEq.(VI-2-4a)]

<-?

0.17

0.48

0.80

4.30

4.22

0.17

3.20

3.06

4.22

3.06

3.06

0.17

II

0.0425

Ik

0.0390
hIt

1.09

0.125

X
X

0.50

0.83

4.30

4.22

0.17

3.20

3.06

0.17

1.38

0.069

1.51

1.62

0.50

0.80

1.39

1.94

=
0.053

1.62

1.36

0.039

0.041

3.06
1.91

1.09

1.39

1-35

0.17

4.22

1.94

0.40
=

0.130

0.17

4.22

0.17

III

3.20

0.17

[fromEq.(VI-2-4a)]

Reduced length

3.20

a (from Fig. VI-5)

Is

0.415
0.125

3.20

Reduced height

^^
3.20

1.35

3.20

4.30

1.35

3.20

Frame

II

1.38

2.23

1.36

1.39

1.89

Ilk

5.00

2.27

k*

We

3.57

then determine the amplitude of forced vibrations of each frame.

I.
We find the value of coefficient A(w 2 ) from Eq. (IV-6-8):

Frame
A(co 2 )

81

10 8

The amplitude

(1

0.55)(30.2

+ 75.0)9 X 10
+ (1 + 0.55)30.2 X
8

of longitudinal vibrations of the

30.2
1.28

X 10 X 0.
=
X 21.2 X 10 10

75.0

column we

10~ 6

The amplitude of vertical vibrations of the center of the frame

Eq. (IV-6-5):
_

(1

0.55)

75.0

4.7

10~

X 30.2 X
X21.2 X 10 10

10 4

0.71
6

find

0.71

0.55

10 4

10 8

21.2

10 10

from Eq. (IV-6-4):

m
beam span we

10 4

find

Q 69

from

FRAME FOUNDATIONS FOR MACHINERY

Table

273

Computations of Equivalent Masses

for Example VI-3-1, Fig. VI-9

VI-6.

Loads and

Frame

equivalent masses

Concentrated load
imposed on the
frame beam, tons.

Frame

Frame

II

7.0

5.0

III

3.5

Equivalent load from

frame beam weight,

0.45

Wi, tons.

0.45

Equivalent mass ra 2
reduced to center of
frame beam span,

5.0

4.26

1.91

1.91

0.45

7.0

0.71

9.81

6.10

2.74

2.74

3.5

0.45

2.12

1.00

9.81

4.70

2.12

0.58

9.81

sec 2 /m

tons

Equivalent load of
columns' weight,

0.33

h,

tons...

0.33

Load imposed by

6.0

longitudinal

6.53

2.4

3.14

2.20

0.42

0.33

6.0

9.16

beams, tons

6.67

2.25

+ 2.4
X (3.14

4.5

0.42

2.82)

0.33
4.5

6.53

2.4

2.82

2.20

0.42

7.36

16.50

Equivalent load

imposed on
columns by transverse

beams

of

frames, tons

Equivalent mass wi
reduced to top of
column, tons

1.55

1.09

2.20

9.16

1.09

2.25

1.55

2.20

1.28

7.36

1.20

9.81

9.81

9.81

16.5

1.20

2.08

1.11

sec 2 /m

Thus the amplitude

of total vertical vibrations of the

frame under consideration

equals

= Ai + A =

At

Frame
A(g) 2 )

(0.71

4.7)10- 6

0.005

mm

II

81

10 8

+ 49.8)9 X 10
+ (1 + 0.48)30.8 X 49.8 X 10 = 12.8 X
30.8 X 10 X 0.92
= 1.0 X 10- m
A
2.08 X 12.8 X 10
+ 0.48)49.8 X 10 + 0.48 X 30.8 X 10
X 10 0.92
(1

0.48)(30.8

A =

(1

1.0

5.7

At

10- 6

= A

m
+A =
2

(1.0

12.8

10 10

5.7)10- 6

0.007

mm

10 10

1I
11

to

ST

o
X
1

to

00

11

o
X
o
00
1

rH*

^ +
1
1-1

<3

OS
00
H

<M

CO
lo

X o
CM d
+ x

en
"HH

LO
Tt*

II

CO

II

1-.

d
X
CO
q
CO
X

II

O
CO
x
x lO
q
CO
to

CO CO

x^

o
o
00

CO

d
q
X
CO
X o
CO
X
CO

c5

to

oo

OS

>

LO
CO
<N

CO

c5
nH
o

X
Ci

to

CO

o
X
1

^
+
CO
Eh

1
h

8
3

g
c3

<N
c4

<m

iJ

CO 00

>

x
CO
OS

GT

II

00

II

II

X
X GO
q
CO

K
<)

11

CD*

+ x rO
O rH
to

E"

CO
LO

CO
CM
CO

00
CO
rH

Cft

CD
X o
(N

o
rH
X

to

cq
cn

**

II

CO
00

d
q
X
CO

X
CO
X

X
CO
q
X CO

00
CO

CO

>

00
00

co

a
c

,_5

''

00

>

CN

c-)

fa

o
m

Ph

o
X
00
q

LO
1

a;

LO
CM

X O
d
+ X

o
X

to

P
Eh

to

rl

Z
O

as
h

00
CO
1

CO

CM

fa

o
CO
X
q
XO
CO
CO
X
to

1>csi

CO
OS

II

00
LO

d
q
X
CO
X
CO
X

co

*h

d
o
QJ
+3

+3

d o
83
s
O U

S.6

CM

4-H

o^

+3

o
d * d sO
o
-5 d
d +3

'2
a
w
o3

CO

CO

6
00

Q
274

co

d
o

CD

-48

*
HT

CD

S-i

2^

CM

c"i

e+H

0)

rj-

oi

"-5

^3

>

o
t-.

00

3M

o CO
q
X CO

s o
d
S

X
to

00

CO

a;

T3

>>

c~^
Tt}

d
co

3bC ,0

3bC
s

+3

v..

11

1i

hH
OS

>

o3

00
tO

to

CO
00
00

||

II

b
o
X 00
d

b to
d
X +

TJH*

"tf

II

b
i
X

rfl

lO

'
X
r

CM

T-H
T-H
t-h'

00

ii

OS
i

CM

ll
i

CO
1

>
H
t-5

ft

00

b
pH
X

bH
X

oo

00

d
CO

II

II

'

CD

ft

II

O 00
p

03

t-h

CO

o
b q
X +
CO 00
iO o

o
1
X o
q
00
d
CO

cm

s
u
2

11

ii

iH

cm'

a
w

fe

g
H

iO
o

0)

ft

CM

d
CO

h3

03

l^

O
H
!
H
p
Oh

00

b
X

CM
1

i-h

b l>
d
X +
d
00
Th
*

,__,

i^.

t}h

CM

CM

co

o
CD

M
"^

-<

co

co

co

.2

d
O

+s

'+s

fa

CO

II

II

II

CO

*
O
X
o

b
X

+->

g a
Q ,5
'> o
-

T3

CD

-Q

CD

CO
ft

03

<D

ti

co

'a

c3

00

>

o3

<4H

o>
CO

^
T3

O d

"9
d ^r "*
03
a ^

n
!

81
CO

CT

rB

e+H

o
'+3

CD

--S

s ~>

-c

-s :

g d
|

>>

8^

CT

co

d
3
6

CD
tt

^o
^
co

o a 1

o3

P3

CD

""
03

mJ
d

5
03

ft

**

M
2h ~

-a

d ^
S
d
d >s
CO
4-3
fH

2 s,s

II g
53
II

ti_,

CD

a
CO

275

CO

DYNAMICS OF BASES AND FOUNDATIONS

276

Frame

A(co 2 )

III

81

10 8

(1

41.5

Al =
.

1.1

(1

0.48)88.3

1.6

It is clear

much

are

10

10- 6

X 10
X 37.8 X 10" ~
X 10 X 0.48 X 41.5 X
0.58 X 37.8 X 10 10
4

m
(0.46

1.6)10~

fi

10 8

37.8

10 10

= Ai + A 2 =

At

+ 88.3)9 X 10
+ (1 + 0.48)41.5 X 88.3 X
X0.46
_
n
Aa v
m
10
- 46
X in

0.48)(41.5

10 4

10 4

m S 0.002 mm

that the computed amplitudes of vertical vibrations of the foundation

smaller than the permissible values.

We

compute the forced horizontal (transverse) vibrations of the foundation.

First we determine the equivalent mass of each frame.
As has been indicated, the
magnitude of each of these masses is determined by the following
1.

Concentrated and distributed loading imposed on the frame beam, including

its

own weight
2.

30 per cent of the column weights

Loads imposed by longitudinal frame beams adjacent to the transverse frame

under consideration, the deadweight of longitudinal beams also included
3.

Thus we have,

frame

frame

for

m =

7.00

frame

III:

m3=
The

3.5

6.10

m+

+ 0.30j7jr=X

6.67

+ 4.70 + 0.30 X

^
X

1Q +tons
2.13

6.53

16.50

3.5

/m
sec 22/

mi

OK tons
+
3.25

.,

sec 2 /m

1.40 tons

sec 2 /m

6.78 tons

sec 2 /m

I,

along the foundation, to the

9*1

total equivalent

We

II:

and

5.0+4.26+0.30X6.53X2+9.16 =
p-^:

mi =
for

for

mass equals

+m +m =
2

2.13

3.25

1.40

determine the distance from the axis of frame

total equivalent mass:

ai

The

_
~

distances from the

3.14

common
a2

The moment

(3.14

+2.82)1.40

6778^

a,

of

3.25

=
=

_
"

Q
~- 95

center of mass to the axes of frames II and III are:

2.95

2.95

= -0.19
= -3.01

3.14
5.96

m
m

of inertia of all the equivalent masses with respect to the

mass equals

Mm

=
=
=

+ m a2 + m a
X 2.95 + 3.25 X
tons X m X sec

miai 2
2.13
31.2

0.19 2

1.40

3.01 2

common

center

FRAME FOUNDATIONS FOR MACHINERY

From Eqs.

the displacement

frame beam

5i

12

The equivalent

10 6

+3 X

"

1.51)

0.29

^XlP =
+3 X

4 22 3 (2

"

frame

12

10

3.12

3.44

10- 3

103

1U

tons/m

10

2 23)

0.041(1

0.32

" 6AZ X

2.23)

10 3 tons/m

III:

4.22 3 (2

53

12

c3

3 85

10

10 6
-3

1.89)

0.039(1

- 26

103

1.89)

tons/m

total rigidity of all three frames equals

C =

We

1.51)

0.039(1

II:
52

The

I,

rigidity equals

-"

for

frame

caused by a unit horizontal force directed along the axis of the

4.25 3 (2

and

277

find the equivalent rigidities: for

is

For frame

we

(VI-2-15) and (VI-2-16)

(0.29

0.32

0.26)10 3

0.87

10 3 tons/m

determine the distance to the center of rigidity, along the foundation, from the
frame I:

axis of

61

The

3.14

0.32

+ 2.82)0.26 X

10 3
(3.14
0.87 X 10 3

distance between the center of mass

e

2.95

and the center

2.95

10 3

2 nc
95
-

of rigidity of the

frame

is

Consequently, the center of inertia and the center of rigidity coincide. For this
and (VI-2-13) for determining the amplitudes of

particular case, Eqs. (VI-2-12)

vibration are simplified.

The amplitude

of lateral horizontal (transverse) vibrations of the foundation equals

The amplitude

of rocking vibrations

m(/ni 2 -

around the common center of gravity

A* "

M m (i

{fn<p

We

2
co )

5)
CO

determine the limiting natural frequencies of the foundation.

we have:

(VI-2-20)

^O^X.10^ 0.128X10* sec-

The

total exciting force equals

P =

0.69

0.92

0.46

2.07 tons

in the plane

(VI-2-27)

From Eq.

DYNAMICS OF BASES AND FOUNDATIONS

278

The amplitude

of transverse vibrations of the foundation

2.07

Ax =

To determine

6.78(0.128

10 3

is

we have

the rigidity of the foundation against torsion,

6i

2.95

62

a2

= -0.19

C363 2

b3

mm

- 034

10 3 )

= -3.01

a3

It follows that

=
=
=

Ci&l

(0.29

4.89

C 26 2 2

X 2.95 + 0.32 X
X 10 tons X m
2

19 2

0.26

3.01 2 )10 3

From Eq. (VI-2-21) we compute the frequency of natural rocking vibrations of the
foundation
4 89

w=

/v

'

103

- 157

3i 2

103 sec

"2

Let us assume that exciting forces in the generator and the turbine act in the same
Then the magnitude of the exciting moment will
equal
direction at each instant of time.

=
=
=

P101

P2CI2

+P

0.69

2.95

az

0.92

0.19

0.46

3.01

Xm

0.48 tons

The amplitude of rocking vibrations determined from Eq. (VI-2-27) is

A" =
The

31.2(0.157

Thus the

total

3.01

maximum

0.175

10~ 5

0.005

"5

radianS

ax

= Ax +

2.

Data.

a 3A v

latter value lies within the

Example

lO" 3

0.005

is

mm

amplitude of horizontal displacement of the foundation

in the direction perpendicular to the axis of the

1.

10

largest horizontal displacement as a result of rocking vibrations

a 3A v

The

' 175

9.0)10'

0.034

main
0.005

shaft of the turbine equals

0.039

mm

range of permissible design values.

Illustrative design of a

foundation for a 500-kw generator

The motor generator runs

at 750

rpm; consequently the frequency

of

forced vibrations will be

co

75 sec" 1

co

5.6

10 3 sec" 2

The foundation is designed to consist of six columns and three transverse frames.
Figure VI-10 gives a diagram of the foundation and the loads acting thereon. The
weight of the rotating part of the motor generator is 5.9 tons; its total weight is
13.2 tons.

"

C\J

0.75
-1.71

2.28-3.10

0.35_

Floor level of the

3.50-

^tiJd

machine room
Zj
L_JL
r/\'/\/
7^77777777
0-09
0.35

Ep.

0.73

0.73
-1.31-

<0.99^

0.40

0.40

h0.40

Level of the
basement floor

0.25 v

T777F7Z77777777777777

4.00
Cross section

I-I

Machine
center

Wi^

mtiMtJ

a,8

//
\y

0.18

-H05

-H'.IO-* 0.5

|--1.60--(

0.2^

777

777777)
0.2

2.6

Cross section H-II

Fig. VI-10. Foundation of

279

Example

VI-3-2.

line

DYNAMICS OF BASES AND FOUNDATIONS

280

Table VI-9 gives design dimensions of transverse frames of the foundation determined in accordance with Art. VI-2. The same symbols are used here as in the
preceding example

Table

VI-9.

Frame

Design Dimensions for Example VI-3-2 and Fig.VI-10

h,

I,

3.6
3.6
3.6

II

III

h,
4.2
4.2
4.2

1.5
1.5
1.5

X
X
X

h,

10-3

X
X
X

9.4
6.1
9.4

10-3
10-3

10-3

5.4
3.5
5.4

lO- 3
lO" 3

Computations.

Foundations under motor generators vibrate mostly in a

Therefore the dynamic computations of the foundation may be
limited to determining the amplitudes of forced horizontal vibrations.
2.

transverse direction.

Table

VI-10.

Main Loads Acting on Foundation of Example

and

VI-3-2

Fig. VI-10

Frame

Loads, tons

Concentrated load in center of frame beam span

Weight of frame beam
Loads imposed by longitudinal beams adjacent to
transverse frame
Weight of columns

Frame

II

Frame

1.2

2.3
0.7

1.7
1.2

2.6

4.5

3.6

1.7

1.7

1.7

1.9

Table VI-10 gives the magnitudes of the mass loads

(in tons) acting

III

on the

foundation.

The equivalent masses

mi =
for

frame

frame

are: for

I:

+ 1.2+0.30X1.7X2+2.6
^
=
1.9

n
aa \ton
0.66

9.81

n
00 +ton
0.88

for

frame

1.2

+ 0.30 X

1.7

+ 3.( =

0.76 ton

9.81
total equivalent

mass

m =
We now
from the

/m
see 22/

sec 2 /m

is

0.66

0.88

0.76

2.30 tons

sec 2 /m

determine the distance to the total equivalent mass, along the foundation,
frame I:

axis of

1.71

0.88

+3.10

0.76
1.67

2.30

The

v
X

III:
1.7

The

/m
sec 22/

II

L
=
m = 2.3+0.70+0.30X1.7X2+4.5
pr^j
9.ol
and

^
X

distances from the

common
a2
a3

=
=

center of mass to the axes of frames II and III are

1.67
1.67

1.71

3.50

= -0.04
= -1.83

m
m

FRAME FOUNDATIONS FOR MACHINERY

The moment
of

281

masses with respect to the

of inertia of all equivalent

common

center

mass equals

Mm
We

=
=
=

miai 2

0.66

10

IU ^

1Q

L68
for

frame

c2

for

4.2

1.83 2

each frame: for frame

10- 3 (1

0-59

0.76

5.4)

5.4)

3.5)

5.4)

1.68

I:

10~ 4

104 tons/m

cz

3.63(2+3X3.5)

12

frame
5l1

The

-4

0.04 2

II
5l1

and

10 6

sec 2

rigidities of

3.6 3 (2

Xm X

4.3 tons

compute the equivalent

12

+ma
+ 0.88 X

ra 2 a 2 2

1.67 2

10 6

_4

4.2

0.57

10- 3 (1

74

X 10-

10- 4 tons/m

III:

3.63(2+3X5.4)

=
12

L6g

1Q

10 6
-4

(0.59

10- 3 (1

10 4 tons/m

0.59)10 4

0.59

total rigidity of all frames

C =

4.2

is

0.57

1.75

10 4 tons/m

We determine the distance to the center of rigidity, along the foundation, from
the axis of frame I
.

hl

The

(1.71

0.57
1.75

To determine

1.67

3.10

10

distance between the centers of mass

e

=
=
=

6i

63

and

ei
= L61
i

1.61

a2
a3

=
=

(0.59

3.65

rigidity of the foundation

1.61

is

0.05

we have:

+ =
+ =
e

+ 0.06 =
+ 0.06 =

-0.04
-1.83

rigidity of the foundation against torsion

From Eqs.

0.59)10 4

the rigidity of the foundation against torsion,

62

The

X 1.61 + 0.57 X
X 10 tons X m
2

is

0.02

-1.77

then

0.02 2

0.59

1.77 2 )10 4

(VI-2-20) and (VI-2-21)

we compute the

limiting frequencies of natural

vibrations of the foundation:

/" =
fnv * =

75

104

2]
104

3 65
-

-2

7.6

103 sec

8.5

103 sec- 2

DYNAMICS OF BASES AND FOUNDATIONS

282

The square

of the radius of gyration equals

Mi =

We

4.30/2.30

1.86

find the value

1+4

r2

+0.06 2 /1.86 2

In the case under consideration, a does not differ much from unity; therefore the
and/n2 of the foundation, determined as roots of Eq. (VI-2-35),
differ very little from the limiting frequencies fnx and fn<p
According to Eq. (VI-2-4) we determine the design value of the radius of unbalance
(eccentricity) of the rotating masses of the motor generator:
natural frequencies fni

The

imposed by

exciting force

P =

0.9

500/750 2

10" 3

10~ 3

rotating masses of the motor generator equals

all

0.9

5.9/9.81

10 3

4.8 tons

Assuming, as in the preceding example, that the unbalanced state refers only to
the static loads,

I
The

we obtain

1.9

influence of this

may be

1.67

2.3

moment on

0.04

moment

1.7

1.83

0.06 ton

Xm

the amplitudes of forced vibrations of the foundation

neglected.

The amplitude

of horizontal displacement

A "
The

for the exciting

8
2.3(7.6

foregoing computations

frequencies

is

6.1)10

is

obtained from Eq. (VI-2-26):

= 1A X

10

show that the value

"3

of

= 1A

mm

one of the natural vibration

close to the operational frequency of the engine;

it

follows that the

amplitudes of vibrations considerably exceed permissible values. The foundation

should be brought out of the zone of resonance; this may be done by increasing the
cross sections of the columns to 0.6 by 0.7
instead of the dimensions 0.4 by 0.5
accepted in the design.

VII
MASSIVE FOUNDATIONS

Vll-l.

Massive Foundations

for

Motor Generators

Design and Design Computations. Motor

lower speeds than turbogenerators.
Therefore according to Eq. (VI-2-4) the eccentricity of rotating masses
in motor generators and analogous electric machines is considerably
larger than in turbogenerators.
This is indirectly confirmed by the
results of measurements of foundation vibrations.
If the eccentricities in motor generators were indeed of the same order
as in turbogenerators, the exciting forces imposed by motor generators
would be so small that they could not induce appreciable vibrations
even under conditions of resonance, when the frequency of vibrations
approaches one of the natural frequencies of the foundation. Then no
dynamic computations of foundation vibrations under motor generators
would be needed. Observations show, however, that the foundations
under low-speed motor generators (up to 300 to 400 rpm) often undergo
Foundations
vibrations with amplitudes of the order of 0.1 to 0.3 mm.
weighing several hundred tons may undergo forced vibrations with such
amplitudes only when exciting loads are large. If one is to take the
eccentricity of rotating masses in the motor generator as having the same
value as in turbogenerators (0.2 mm), then for a 75-ton weight of rotating
masses of a motor generator with a speed of 300 rpm, the value of exciting
forces generated will equal only 1.5 tons.
Such an exciting load, even
under conditions of resonance, cannot induce vibrations with an amplitude of the order of 0.1 to 0.3
in a foundation weighing several
General Remarks in Regard

to

generators usually operate at

much

mm

hundred

tons.

Thus the results of instrumental measurements of vibrations of lowfrequency motor generators provide a basis for the assumption that in
these machines the eccentricity of rotating masses is much larger than
in turbogenerators.
283

DYNAMICS OF BASES AND FOUNDATIONS

284

Stress analysis of a massive foundation

small magnitude of stresses imposed

by

is

not required because of the

static

and dynamic external

In addition to the computation of amplitudes of transverse

vibrations, it is necessary to center the foundation and check the magloads.

nitude of the pressure imposed on the

on the

soil

by

its

weight.

The

permissi-

may

be taken to equal the permissible pressure

under conditions of static loading only.
ble pressure

soil

The foundation is built of concrete; the upper part is made of concrete

type 100, f and the lower slab may be made of concrete type 60 f or
cyclopean concrete. In portions weakened by openings and grooves,
the foundation is reinforced to fit the field conditions; approximately
3
20 to 30 kg of steel are used for 1

Example.

Dynamic computation

1. Data.
A dynamic computation of the massive foundation under a 3,000-kw
motor generator running at 300 rpm is to be performed. Figure VII-1 shows the
main dimensions of the foundation selected on the basis of construction assignments
from the engine manufacturer and the client who ordered the design. The static
loads and points of application are also shown in the figure.
Geologic conditions are as follows: loessial clay with some sand extends to a depth
of 28 m; its moisture content is about 9 to 10 per cent; it is underlaid by dense brown
clays.
The ground-water level is at a depth of 14 m. The following coefficients of
elasticity of the soil have been established from investigations of its elastic properties:

Coefficient of elastic uniform compression:

c

Coefficient of elastic

5X

10 3 tons/m 3

nonuniform compression:
cv

10

10 3 tons/m 3

3.5

10 3 tons/m 3

Coefficient of elastic uniform shear:

cr

The foundation is to be erected in a machine room with several operating motor

The width of the building does not permit increasing the width of the
foundation beyond that shown in Fig. VII-1. The distance between footings under
columns and motor generators is 25 cm.
The following values necessary for dynamic computations were established from
generators.

calculations:

Weight

of the foundation (taking into account all cavities)

W
Mass

of the foundation

and engine:

m =
t

1,136 tons

See footnote, Art. IV-l-c,

115.7 tons

p. 132.

sec 2 /m

and engine:

MASSIVE FOUNDATIONS

285

Cross section I-I

^2 =
Wa =

Cross section

H-n

30 tons

8.1

tons

Wo,

=11.05 tons

(stotor of the synchronous motor)

^6 = 8.1
(frame of direct -current motor)

=10.75

(armature of electromagnet
of direct-current

motor

-^tons

weight of the bearing

1.5 tons)

W-,

.3

tons

(frame of direct-current motor)

Wn -

Wz = 30 tons

8.3 tons

=23.4

(foundation slab)

(rotor of the synchronous motor

=11.65 tons

5
(stator of the

tons

382
-

tons

armature of electromagnet of direct-current

motor + weight of the bearing)

synchronous

motor)

View from above

Fig. VII-1. Massive foundation for 3,000-kw 300-rpm motor generator of example of
Art. VII-1.

Foundation area

in contact with soil:

A =

96.0

Distance between the level of the foundation contact area with

center of mass:
h

Moment

4.6

and the common

of inertia of the foundation contact area

axis passing

soil

with respect
through the centroid of the contact area:
440 m'

to the longitudinal

DYNAMICS OF BASES AND FOUNDATIONS

286

Moment

of inertia of the

mass

and engine with respect

of the foundation

to the

same

axis:

Im

Xm X

3,974.6 tons

sec 2

Moment

of inertia of the mass of the foundation and engine with respect to the axis
which passes through the center of gravity and is perpendicular to the plane of

vibrations:

h =
Ratio between the

moments

sec 2

of inertia of masses:

7
2.

Xm X

1,510.6 tons

0.38

Using these data, we begin by establishing the frequency of

The frequency of natural vertical vibrations

Computations.

natural vibrations of the foundation.

(from Eq. Ill- 1-5)

is

5 ><

U? .
=

fnz

The number

4.4

sec-

10.

64.3 sec" 1

of natural vertical vibrations of the foundation is

N =
difference

9.55/

The

.0

ll*

9.55

between the numbers

64.3

614 min" 1

and forced vibrations

of natural vibrations

equals per cent

614

300

300

100

105 per cent

Hence, the design of the foundation is satisfactory in regard to vertical vibrations.

In order to determine the frequencies of natural vibrations of the foundation fni
and fn2 in a transverse plane, the limiting frequencies fnip and fnx of the foundation
should first be established. From Eqs. (III-2-6) and (III-3-2), we have

M
,

fnx

9
2

10

3.5

10 3

440

3974^

10 3

96

jjyj

_ v
LU
X
1

ins SGC
10

_2

im sec _ 2
= OA1
2.91 ^
X 10
2

Then we obtain

fnS+Jnx* =

1.11

10 3 +2.91
2 X 0.38

10 3
-

_
KQ v
X
" 5-3

jn^Ul =

1.11

10 3

10 3

2^

~-^r

2.91

R 5K
8
-

oT38

in3
10

Vin6
x 10

According to Eq. (III-4-8),

/nl>2 2

=
=

5.3
(5.3

Hence

10 3

/n2

10 6

8.5

10 6

/n2

Thus the minimum number

+ V28.0 X

4.43) 10 sec"
= 0.87
10 3 sec" 2

29.5 sec" 1

of natural vibrations of the foundation

N =
2

9.55

29.5

282 min" 1

is

MASSIVE FOUNDATIONS
This

differs

from

282
*72

300

300

= 3

100

287

by only

the operational speed of the engine

per cent

Therefore it can be assumed that if the rotating masses of the motor generator are
only slightly out of balance, considerable transverse vibrations of the foundation
may develop. It follows that the design of the foundation is not satisfactory in
regard to these vibrations.
The dimensions of the machine and the building do not permit
3. Modification.
any considerable changes in foundation
52 Strauss piles d = 0.3m, Z = 3.5m
Changes in the depth of the in
width.

foundation or an increase in the foundation length has very little influence on

the frequencies of natural vibrations of the
foundation; therefore the only way to
increase their value is to increase artificially the rigidity of the base under the
foundation. In the case under consideration, the best way to achieve an increase
in the rigidity of the base would be the
provision

of

short

+ + + + + + + + + + + + +

++

+ + + + +

I++

+ + + + + +

++++
+++++

+ + + +

in

+ + + + +

+ + +
+ + +

inf

-4, i.ok-

conical precast rein-

However, the use

would require a comparalong time for their casting and

forced-concrete piles.

Fig.

of these piles

concrete bore piles under the founda-

tively

tion of Fig. VII-1.

Distribution

VII-2.

of

situ-cast

In addition, the driving of piles

inside a building with machines in operation would cause considerable inconvenience.
Therefore the decision was taken to increase the rigidity of the base by the installation of 52 situ-cast bore piles system Strauss, each 3.5
Figure VII-2 shows
long.
the distribution of these piles in plan.
The coefficients of elasticity of such a pile base are about three times larger than
those of the natural base under the foundation.
One can take for the pile base:
curing.

<V
cT

=
=

30

10.5

10 3 tons/m 3

10 3 tons/m 3

Let us compute the forced vibrations of the foundation in a transverse plane when
used

piles are

V ==
/x

From Eq.

3
3

X
X

X
X

1.11

2.91

10 3
10 3

=
=

3.3

8.72

X 10 sec"
X 10 sec"
2

(III-4-8),

/ 1)2

/3.3

" \

(15.7

8.72

/3.3

- \

0.38

+
X

8.72 2

3.3

8.72
10

0.38

0.38

12.3)10 3

we have
/i

/2

2
2

=
=

X 10 sec"
x 10 sec"
3

29.0
3.4

58 sec"

We

find the multiplier,

A(o> 2 )

=
=

m/o(/i 2 - co 2 )(/2 2 - co 2 )
115.7 X 1510.6(29 - 0.9)(3.4

0.9)10 6

12.3

10 12

DYNAMICS OF BASES AND FOUNDATIONS

288

According to Eqs. (III-4-1) and (III-4-12), the amplitudes of forced vibrations of

the foundation will equal

Ax =
30 X

10 3

440

1136

4.6

12.3

+
~_ 10.5

10 3

12.3

(1.7LP

Thus the

X 96 X
X 10 12

4.6

10 5
-

rp

X 10 X
X 10
3

10.5

*, X

10.5
"+"

96

(4.6)

1510.6

X 0.9 X

10 3

12

10 3

4 6
'

*< =

(2-7P

- 115.7 X
12.3 X 10 12
96

0.9

+ 0.37^)10- m
X

10 3

MU

+0.00Mi)10- 6

total

amplitude of horizontal vibrations of the upper part of the foundation

equals

A = Ax +
where

hi

3.45

the distance from the

is

hiA<p

common

center of the engine and founda-

Inserting the numerical values of

tion to the top of the foundation.

A x and

A<p,

we

obtain

A =
The

exciting

moment

(2.7P

6.27M )10- 6
Z

equals

Mi =
where

5.2

is

PH

the distance from the axis of the machine shaft to the

common

center of gravity; thus

A =

27.7

10 6 P

According to the data from the plant, the design value of the rotor weight is
around 60 tons; an approximate value of the eccentricity r of rotating masses may be
taken as ten times that for turbogenerators with speeds of 1,500 rpm; i.e., r = 2 mm.

Then the design value

P =
Inserting this value of

A =

of the exciting force equals

10- 3

60/9.81

0.9

10 3

10.8 tons

into the expression for the amplitude,

27.7

10~ 6

10 8

0.3

10" 3

m =

we obtain

0.3

mm

For low-frequency motor generators, the permissible design value of amplitude of

may be taken to equal 0.30 mm. It follows that the foundation under
consideration satisfies the conditions of dynamic stability.

vibrations

VII-2.

Massive Foundations under Turbodynamos

turbodynamos are blocks with

cavities and grooves for individual parts of the machine or for mounting
auxiliary equipment.
Such a foundation consists of an upper part
designed as a very rigid box or as two walls with grooves and openings,
and a lower foundation slab transmitting the load to the soil. Special
design features of massive foundations for turbodynamos are seen in
Basically, massive foundations under

MASSIVE FOUNDATIONS

VII-3a and

289

which shows a general view and a longitudinal section

turbodynamo.
The upper part of this foundation consists of a complicated combination of individual structural units: girders, columns, walls, slabs, and
others.
Dynamic and static computations of this part of a foundation
Fig.

b,

of a foundation for a 1,200-kw

therefore involve a high degree of approximation.

the upper part of the foundation and

its

The dimensions

of

individual units are usually

determined by the construction assignment prepared by the machine

manufacturer. Thus a designer's task is limited to determining the
dimensions of the lower foundation slab and designing the reinforcement

Fig. VII-3. Foundation for a 1,200-kw

turbodynamo:

(a)

general view;

(6)

longitudinal

section.

for the foundation.

Essential points of instruction for the construction

frame foundations given in Art. VI- 1 should be followed in the design

of massive foundations for turbodynamos.
Concrete type 150f is used
for the upper part of a massive foundation, and concrete type 100 f for
the lower foundation slab.
All structural units of the upper part of the foundation are designed
so that their numbers of natural vibrations should not be smaller than
of

3,000 min-1.

Unbalanced

inertial forces of

turbodynamos may induce vibrations

of

the foundation as a rigid body on an elastic base, as well as vibrations

of the separate structural elements constituting the foundation.

Experience in operating high-frequency turbodynamos has not revealed

any cases
bodies on
t

of significant vibrations of massive foundations acting as rigid

elastic bases for the following reasons:

See footnote, Art. IV-l-c, p. 132.

DYNAMICS OF BASES AND FOUNDATIONS

290

The natural frequencies of foundation vibrations are usually smaller

than the operational frequencies of high-frequency turbodynamos, and
it is hardly possible that these two frequencies will coincide.
Highfrequency turbodynamos are well balanced in regard to both static and
dynamic loads. Their actual eccentricities (Art. VI-2) do not exceed
Therefore the exciting loads inducing foundation vibrations are
2 mm.
relatively small.

Even under the most unfavorable conditions, the exciting loads cannot induce vibrations with impermissible amplitudes, because the foundation mass is large in comparison with the mass of rotating machine parts.
In the case of high-frequency vibrations, there is considerable influence
In order to approximately evaluate this influence,
of damping forces.
us compute the amplitude of foundation vibrations under the most
unfavorable conditions at resonance (i.e., when co = fng ). The ampli-

let

tude of vertical vibrations of the foundation as a rigid body on an elastic

base can then be established from Eq. (III-1-21):

2mcc

m =

where

mass

and machine
constant whose value may be taken as proportional

of foundation

damping

to frequency of vibrations,

The maximum value

i.e.,

t)w

component

of the vertical

of the exciting force

equals

Pz

where

= eccentricity
mo = mass of rotating

foWoO) 2

Inserting expressions for

parts of machine

and

c in

the formulas for

z,

we obtain

the following expression for the amplitude of vertical vibrations of the

foundation at resonance:

where

/x

nto/m

is

the ratio between the rotating machine masses and

the total mass of the foundation and machine.

equals approximately 0.05; the value of the

For these values
coefficient of proportionality y\ may be taken as 0.5.

For turbodynamos

of

\x

and

77,

\x

we obtain

i = 5X
Even

for r

0.2

mm,

the

maximum

amplitude of vertical vibrations

under conditions of resonance will be on
Vibrations of such an amplitude are not

of the foundation to be expected

the order of only 0.010

mm.

10- 2 r

MASSIVE FOUNDATIONS

dangerous.

Actual

smaller, because

it is

amplitudes

of

291

foundation

vibrations

are

much

hardly likely that the frequencies of natural vertical

vibrations and the frequencies of the machine will coincide.

Therefore

computations of massive foundations under turbodynamos

with speeds greater than 1,000 rpm dynamic computation of the foundation as a rigid body on an elastic base is not required.
In order to prevent the vibration of individual units constituting the
foundation, it is advisable to check them as to danger of resonance and
then design them so that their natural frequencies will be larger than the
operational frequencies of the machine.
Investigations of resonance should be performed on transverse girders
of the foundation which support the machine bearings, because it is
these girders which carry the dynamic loads imposed by the machine.
In the computation of frequencies of natural vibrations of these elements,
formulas for single-degree-of-freedom systems may be used in most cases
with a sufficient degree of approximation. When computing deflections
one should consider only the dead loads carried directly by the element
in design

studied.

High frequencies are characterized by

large

as a result of vibrations of individual units.

damping

forces developing

Therefore, for foundations

under turbodynamos with speeds greater than 3,000 rpm, there

necessity to check individual units as to danger of resonance.

is

no

It suffices

computation of the foundation elements directly

The same computations should be made also for
machines having speeds below 3,000 rpm.
Dynamic stresses in bases under foundations are very small because
the amplitudes of vibrations of massive foundations under turbodynamos
are very small.
Therefore the permissible bearing value of soils under
foundations for high-frequency engines may be taken to equal about
to perform a static

supporting the loads.

80 to 100 per cent of the permissible bearing value for static load
only.
VII-3.

Foundations for Rolling Mills

In the process of hot-rolling operations, in addition to constant (with

respect to time) loads acting on the foundation, there appear also variable

These loads may induce foundation oscillations and dynamic

both the soil and the foundation.
The larger the rolling mill, the larger the alternate loads imposed on
the foundation and soil.
Of the heavy rolling mills, reversible double-

loads.

stresses in

most commonly used in engineering metallurgy.

Therefore the computations outlined below refer to this type of mill.

level mills are the ones

However, the data presented here

may

of rolling mills, such as three-level types

be easily applied to other types

and nonreversible

types.

DYNAMICS OF BASES AND FOUNDATIONS

292

Modern

rolling mills consist of the following

Driving

1.

main

units

motor, whose foundation in some cases

roll

to the foundation of the stands; occasionally

no

tie exists

rigidly tied

is

between these

foundations
Ilgner

2.

power system, which

always mounted on a separate

is

foundation

Operating

3.

and

drive-gear

usually

stands,

having

common

foundation

Dynamic Loads Imposed on

a.

the

Foundation by

the

Driving Roll Motor.

commonly used

Reversible direct-current motors are

for the operation

The operational speed of these motors is rather low

around 58 rpm. The maximum (switching-off) moment at the shaft of
the motor may reach several hundred tons X meters.
The power is
supplied from the Ilgner power system; the motor is mounted on a massive
of rolling mills.

foundation.

It will

be assumed in further discussions that the motor is

which is considered to be an absolutely

rigidly tied to the foundation,

body resting on a
known.
rigid

If

a torsional

moment

fully elastic soil

is

whose

essential constants are

applied to the rotor shaft, then the stator,

and consequently the foundation, will be under the action of a moment

whose magnitude equals \M\, but whose direction is opposite to that of M.
This moment is the only alternating load acting on the foundation.
Changes in the torsional moment
applied to the rotor shaft, and
consequently changes in the alternating moment acting on the foundation
and soil, are a complicated function of many independent variables whose

Therefore calculations are usually based

on several assumptions. First of all some assumptions should be made
concerning the distribution of the so-called reduced pressure of metal on
influence

the

is difficult

to evaluate.

The magnitude

rolls.

of this pressure essentially affects the

magni-

tude of
In computations of power consumed by the rolling

assumed
ingot and the

mill, it is

customarily

between the
assumed that the

that the reduced pressure along the arc of contact

rolls

remains constant.

angular speed of the rotor

magnitude

of the

moment

is

Usualfy

constant.

of rotation

it is

Under these

may

conditions, the

be expressed at any instant

as a linear function of time.

Figure VII-4 shows a graph of changes in

M for one of the

first

passes

plotted on the basis of the preceding

assumptions concerning the reduced pressure of metal on the rolls and the
angular speed of the mill. The horizontal axis of the graph shows periods

of the ingot through the

of time

rolls,

corresponding to successive stages in the passage of the ingot.

MASSIVE FOUNDATIONS

293

These stages are as follows: (1) no-load speeding up of the rolls; (2) the
is gripped and forced through; (3) rolling with acceleration; (4)

ingot

rolling at constant speed; (5) slowing

down

of the rolls;

(6)

exit of the

ingot; (7) stoppage of the mill.

250

Fig-

ure VII-5 gives a graph of changes

in

200

during the whole process of

rolling of

150

an ingot.
from graphs VII-4 and

It is seen

VII-5

that

the

external

moment, and consequently the

moment acting on

ing

x100

torsional

* 50

excit-

the foundation
5| 6|

in the course of a pass of the ingot,

except for the periods of

and exit.
diagram

its

50

entry
100

Therefore, instead of the

of

changes in
shown
one can assume that
occur according to the

changes

will

diagram

in Fig. VII-6a.

When

the ingot emerges from the

Graph

Fig. VII-4.

in Fig. VII-4,

occurring in

*,sec

do not change much

of the motor,

torsional

moment

of changes in the

of the shaft during

one passage of an ingot on a rolling

rolls,

mill.

the absolute value of change

M in practice may be considered to equal the change occur-

240
ft

ft

160

Pi

E 80

x
M

20

qL

30JJL

40

50

70

180

fl00

240

\l

\l

Fig. VII-5.

Graph

rolling process of

when

^f 80
160

|l20M30

110

of changes
an ingot.

in the torsional

moment

of the shaft during entire

The exit of an ingot from

is gripped by the rolls.
accompanied by foundation vibration. Due to a decrease in
loading, the magnitudes of stresses induced by the vibration will not
ring

the rolls

the ingot

is

exceed those observed during the steady process of

it

rolling.

possible to base calculations not on Fig. VII-6a, but

In conformity with this diagram,

for the exciting
1.

When

moment
0,

M
0.

let

on

This makes
Fig. VII-66.

us set up the following conditions

DYNAMICS OF BASES AND FOUNDATIONS

294
2.

At the time the ingot

gripped by the

is

rolls (t

<

r),

>

constant

dt
3.

For the steady process of

M
Under the action

rolling
ilf m ax

of the torsional

(t

>

r),

constant

moment, the foundation

around an axis passing through the center

will rotate

of gravity of the foundation

Mr,

t,

Rolling

^k

sec

t,

rk

K Exit of

-Entry of
the ingot

se.c

the ingot
id)

(a)

Fig. VII-6. Simplified design diagram of changes in the torsional

moment

of a rolling

mill shaft.

area in contact with

soil,

perpendicular to the plane in which the

Therefore the stresses in the

acts.

and the maximum

stress

soil

p max at the foundation edge

_

Pmax

W
*

CtpCl(p ma -x

moment

along the contact area will vary,

will

equal
(VII-3-1)

=
A =
Cy =
2a =

weight of foundation and equipment thereon

foundation area in contact with soil
coefficient of elastic nonuniform compression of soil
foundation width in plane of action of moment
Let us compute ^ max for the interval of time corresponding to the
gripping of the ingot by the rolls.
The equations of motion of the
foundation are as follows:

where

X
ip

anX
a 2 iX

+
+

Mi =

where

and

+
+

is

the

foundation).

moment

Oi\2<f

a 2 2<P
L'l

=
= Mit

(VII-3-2)

max

M,

of inertia of the installation

mass (machine and

MASSIVE FOUNDATIONS

The

295

system of equations depend on the elastic

and the dimensions and mass of the foundation and

coefficients of this

properties of

soil

motor [Eqs. (III-4-5)].

Assuming that at time zero the displacement
tion

<p

general solution for

xo

and the angle

of the center of gravity of the foundation equal zero,

f 2ff 2
Jnl \Jn2

Jnl

we

of rota-

obtain the

<p:

2\

M Sm

fnlt

2( f

Jn2 \Jn2

M Sm

T^\
~ Jnl

fn*t

<P*

(VII-3-3)

where n and/n2 are the natural frequencies

from Eq. (III-4-8). The expression
i

au

Mil

fm%
gives the value of

a static

<p

(VII-3-4)

moment has only

Eq. (VII-3-3) evaluate the dynamic

for the condition that the torsional

The other terms

effect.

of the foundation established

in

moment

action of the external torsional

applied to the foundation.

Assuming

<p

we obtain

for the

dynamic

(an ~ jnv
fm )fn2
IV^-ii
ail(,/n2 Jn2 )fnlt

coefficient

Jjni.

V<Ps

>

(an
\"-ii

,,

ail(,/2

- jn\.
nl
/m )/,
/Jul
~ Jnl)Jn2t
2

(VII-3-5)
If

the gripping period

is

small in comparison with the periods Ti and

of natural vibrations of the foundation, then,

sin

fmt

sin

fn2 t

we obtain
In this case the

\x

=
=

assuming that

fmt

fn2 t
2

maximum rotation of the foundation

moment will not exceed the

under the action

twofold value of
displacement caused by the static action of the same moment.
As the ingot gripping time increases, the value \i 1 decreases,
approaching zero for high values of t. Consequently, if the gripping time
of the alternating torsional

is

and T 2 then the action

of the

the foundation does not differ

much

large in comparison with the periods Ti

alternating torsional

from the

moment upon

static pressures.

The value of ju may be computed with a comparatively high degree of

accuracy as soon as one knows the periods of natural vibrations of the
foundation in the plane of action of the alternating torsional moment;

DYNAMICS OF BASES AND FOUNDATIONS

296

know the time required to grip the ingots. In practice

time varies within a comparatively wide range. Only approximate
values of the periods T x and T 2 may be established by computations.

then one will also

this

The

calculation of 7\

and T 2 involves laborious arithmetic operations;

therefore practical design computations of the foundation under the

driving

roll

setting

fi

motor should be based on the most unfavorable conditions by

2.

t > t, i.e., for the steady process of rolling, <p will not
exceed the maximum value characterizing the times of gripping of the
ingot by the rolls and its emergence from the rolls.
Therefore the stresses

Evidently, for

along the foundation contact area during the process of rolling

not exceed the stresses developing at the time of entry of the ingot.
The values of these stresses should be used in calculations of the base
under the foundation for the rolling mill.
b. Dynamic Action on the Foundation by the Ilgner Power System.
The
purpose of the Ilgner power system is to feed power to the motor driving
the rolls.
The power system consists of one or several direct-current
generators and a flywheel, mounted on the same shaft.
The generators
are set in motion by an electric motor.
The power W\ of an asynchronous motor, taken from the line, remains
almost constant during power-system operation. It equals the average
quadratic power required for rolling during one cycle (15 to 19 passes).
The power supplied by the direct-current generators is almost constant.
The power
2 taken from these two generators by a motor driving the
rolls undergoes extremely sharp changes.
The range of these changes is
from zero, which corresponds to a pause in the rolling, to the maximum
power required by the motor. The maximum power may be considerably
larger than the power which at a given instant is supplied to the generators by the asynchronous motor.
The flywheel and other rotating masses
increase the amount of energy which is yielded by the generators, since
their kinetic energy changes as a result of changes in the speed of the
shaft.
Thus a deficiency in energy required for rolling is made up. In
addition, during the operation interval in which the generators do not
supply energy to the drive motor, the flywheel and other rotating masses
accumulate the energy which is taken by the asynchronous motor from

in the soil
will

the

line.

Let us investigate the dynamic loads acting on the foundation during

the operation of the power system.
If

= power taken by motor from line

= angular speed of aggregate shaft
M\ = torsional moment of motor shaft
i

then

(a

Wi = Mmi

varying value)

MASSIVE FOUNDATIONS

297

In electrical motors, the stators (and consequently the foundation) are

under the action of a reactive moment whose absolute value equals that
of the torsional moment Mi, but whose direction is opposite.
In addition to this moment Mi, a moment Mi is also acting on the
This moment has the same
foundation, induced by the generators.
If
sign as the moment of the generator shaft.
2 is the power yielded
by the generators, then

W
The
will

resulting external

moment Mi

2 co

acting on the foundation evidently

equal the difference between the moments;

Mi =

i.e.,

- Mi

Neglecting power losses in the engine, we obtain

= Wi -

^ Vh
co

where 27; = I is the sum of the moments of inertia of all the rotating
masses of the power system, i.e., of the flywheel, motor generators, and
armatures of electromagnets. Since

W
it

- Wi =

co(M 2

- Mi) =

ilfico

follows that

Mi =
when

^I

(VII-3-6)

the energy yielded by the power system equals the energy taken

from the

line, i.e.,

when
du)

ii =
the external

same

moment

At the
moments tending

acting on the foundation also equals zero.

time, the foundation will be subjected to internal

to produce torsion in

it.

Let us assume that the Ilgner power system consists of two alternatingcurrent generators, an asynchronous motor, and a flywheel.
The total
flywheel

870 tons

moment GD

Xm

1T

masses of the aggregate

of all the rotating

hence

GD

= ~4f =

870

4TX9M =

OOQ3 +t0nS
22
'

X m s/
X
x/

SeC 2

is

about

DYNAMICS OF BASES AND FOUNDATIONS

298
If

N is the number of rpm of the

W

= ?1n
60

M< =

Therefore,

power system, then

dt

I
%
60

60
dt

2.24

dt

^
dt

The rate of change of N, i.e., dN/dt, varies within the range 2.8 to 10.4.
The magnitude of the external moment acting on the power-system
foundation during the whole cycle of rolling of one ingot changes from
(periods of running idle) up to 24 tons X
2.6 tons X
(periods of

rolling).

The design value

of the exciting

moment

should be taken to equal

2Mi

most unfavorable case. The angle of foundation rotation,

induced by this moment, is determined from Eq. (VII-3-3).
In addition to the exciting moments caused by changes in the kinetic
energy of the power system, the foundation may be subjected to periodic
exciting loads caused by the unbalanced state of the engine with respect
The computation of forced
to magnetic forces and static equilibrium.
vibrations of the foundation caused by these loads is performed in the
same way as for foundations under motor generators.
c. Dynamic Loads on the Common Foundation of Working and Gear
In the process of the rolling-mill operation, the frame of the
Stands.
driving-gear stand, and consequently its foundation, are subjected to the
action of a varying exciting moment equal in magnitude and sign to
for

the

the

moment

The

may

of the shaft of the driving roll motor.

forces appearing as a result of the acceleration of the ingot

be neglected because of their minute magnitudes; hence

it

may

be considered that stresses occur only in the working stand during the
rolling operations.
These stresses have a tendency to rupture the stand.

The sum of all the external alternating loads equals zero.

The drive-gear and working stands may be mounted on a separate
foundation, not tied to that under the driving
the

dynamic influences

of external loads

roll

motor.

In this case,

on the foundations are evaluated

separately but similarly.

the drive-gear stand, working stand, and driving roll motor are
mounted on a common foundation, then the drive-gear stand is subjected
If

to the action of a torsional

moment whose

sign

is

opposite that of the

moment acting on the stator of the driving roll motor. Therefore the
sum of all the external dynamic loads transmitted to the foundation and
soil

equals zero.

torsional

moments

The foundation

will

be under the action of internal

whose magnitude equals the

moment

of the shaft of

the motor, as well as under the action of the equipment weight.

These

MASSIVE FOUNDATIONS

299

loads should be considered in the stress analysis of the foundation and

its

The dynamic nature of the internal moments is

account by introducing in the calculations the twofold magni-

separate elements.

taken into
tude of the maximum torsional moment of the shaft of the driving roll
motor.
In the case under consideration, stresses in the soil are determined for
a design load consisting of the combined weight of the foundation and the

equipment mounted thereon.

The foundations for the principal
d. Remarks concerning Design.
rolling equipment (stands, reducer, gear) are always built as massive

W/AWY//4

Fig. VII-7. Foundation for a stand of a sheet-rolling mill.

units which either are monolithic or are provided with deformation

joints.

As

illustration,

diagrams of massive foundations are shown as

follows: Fig. VII-7: a foundation for the stand of a sheet-rolling mill;

Fig. VII-8: a foundation for a light-section steel mill; Fig. VII-9: a
foundation for a drive-gear stand.

The main part

of the foundation under the drive-gear and working

always designed as one block. This part of the foundation
usually has two tunnels, located along the axis of the stand at different
heights.
The upper tunnel serves for the removal of mill scale and for
the runoff of cooling water under the working stand, as well as for the
inspection of equipment and the runoff of lubricant under the drive-gear
stand.
The lower tunnel serves for the inspection of anchor bolts; it is
provided with several recesses to facilitate access to anchor plates. In
the central part of the foundation are located spindle benches which are
provided with wells for counterweights and an appliance for changing the

stands

is

first roller

of the stand.

The foundation under the driving

roll

motor

(Fig. VII-10) is built as a

separate massive block or as a block forming one monolith with the

DYNAMICS OF BASES AND FOUNDATIONS

300

Fig. VII-8. Foundation for a light-section steel mill.

0.00

-h
g\W///.\\\V//A\\V//A\V

^Channel for
g.

cable

5000

Fig. VII-9. Foundation for a drive-gear stand.

foundation under the drive-gear stand. The foundation has a deep

groove for the inspection and mounting of the equipment and a channel
for the air-cooling of the motor.

On both

sides of the

working stand, along the

rolling-mill axis, are

located the foundations for manipulators and roller conveyors.

Usually

MASSIVE FOUNDATIONS
Axis of the

301

frame

of the rolling mill

driving

motor

Cement grout
during erection
745 250

250 745

Opening for
inserting bars

,-4500

Level of

basement floor

Fig. VII-10. Foundation for the motor driving the rolls

rB
,

1250

1250

1250

380

>.

1080

1620

1250

Cross section

B-B

Fig. VII-11. Block foundation for roller conveyors.

these foundations are also built as massive blocks with required channels

and grooves (Fig. VII-11).

foundations (Fig. VII-12).

Sometimes they are designed as frame

Foundations under rolling-mills equipment are made of concrete and

Concrete is employed for massive foundations which

reinforced concrete.

DYNAMICS OF BASES AND FOUNDATIONS

302

eJ

<*>250

.50

s^V ffi

340l^j^0
^1000-

I
250W250

jTl

m/j/wwv^w-i-

^W y//AW^\w^\\\w

-3000-1

>^AKW//,Wy//,WV///,

3000-

3000-

3000-

.w^^^s^

>7//A\\V//-A\^

Section

WTzmmmfc

A-A

^vzmmz?

Section

Fig. VII-12.

are not

on

weakened by

sufficiently

rigid

Frame foundation

for roller conveyors.

and channels and which are erected

and homogeneous bases. Otherwise, reinforced

large openings

is used.
As a rule, concrete type lOOf is employed.
The foundation area in contact with soil should be, as far as possible,
all on the same level.
Large differences in depths of separate sections of

concrete

See footnote, Art. IV-l-c, p. 132.

MASSIVE FOUNDATIONS

the foundation should not be permitted.

tact areas

on the same

If

303

locating

all

foundation con-

level leads to a considerable overexpenditure of

may

material, then deformation joints

be provided between sections

lying at different depths.

and settlement joints in foundaequipment is determined by the distribution of

the equipment, the depth of separate foundation sections, the soil bearing
Distances between
value, and the temperature regime of rolling.

The

location of expansion, shrinkage,

tions under rolling-mill

deformation joints are selected according to the official Technical Rules

and Construction Code. Joints should be located so that they divide the
foundation into separate sections which support units of equipment not
connected with each other. For example, in order to avoid uneven
settlement, the foundations under working and drive-gear stands should
not be separated.
and foundation sections
Continuous footings longer than 20 to 30
or
larger
than
15
by
15
20
by
20
should be provided
under stands
joints.
If
large
deformation
a
section
of
foundation
the
cannot be
with
in
deformation
joints,
then,
order
of
by
to
prevent
the
appearance
divided
shrinkage cracks, such a foundation may be divided by temporary joints
with reinforcement extending beyond the joints. Later these joints are
The projecting reinforcement is
filled with concrete of the same type.
overlapped and welded.
For the analysis of stresses within the foundation
e. Design Loads.
and for the determination of pressure on the base, the following loads
should be considered:

Weight
Weight

of the rolling-mill

of the driving roll

Maximum

disconnection

equipment
motor

moment

at the

motor shaft

Horizontal force transmitted to the footings under manipulators and

tilting devices

Erection loads

Foundation weight
Static computations of the foundation

1.

may

be limited to:

Stress analysis of separate units of the foundation, such as units

weakened by openings, cantilevers, and others

2. Computation of local stresses under supporting
3.

Analysis of stresses within the foundation

4.

Computation

slabs

of pressures transmitted to the soil

DYNAMICS OF BASES AND FOUNDATIONS

304

The foundation

is

on an elastic base.
For calculations

considered to be a girder of varying rigidity resting

listed in points

and

2,

a value of the dynamic

coefficient equaling 2 is introduced in the calculations of the

weight of
For calculations listed in
points 3 and 4 the actual weight of the same machines is taken, without

the rolling mill and of the driving

roll

motor.

introducing a dynamic coefficient.

If

a foundation

on an

is

treated in design computations as a

beam

resting

elastic base, then, in order to simplify operations, it is permissible

to consider separate comparatively rigid units of the foundation as being

An uneven settlement at the contact of the foundation

under the roller conveyers with the foundation under the rolling mill leads
To determine these
to the appearance of stresses along this contact.
stresses, it is permitted to consider the foundation under the rolling mill
to be an absolutely rigid unit.
The permissible pressure on the soil under the foundations of rolling
mills and driving roll motors for dynamic loads may be taken to equal the
absolutely rigid.

corresponding permissible pressure for static loads only.

In concrete or lightly reinforced foundations, the

by separate machinery units and established

soil

pressure imposed

for conditionally separated

foundation sections without considering the influence of other foundation

units should not exceed the permissible bearing value of soil.

Foundations subjected to horizontal impacts, such as those under

manipulators and tilting devices, should be designed for the double value
of the

maximum

horizontal force.

/. Data on Performance of Existing Foundations under Rolling Mills.

The author and B. M. Terenin investigated several foundations under
rolling mills at one of the Soviet plants.
These foundations were built
of concrete, and each consisted of a single massive block supporting the
driving roll motor as well as the drive-gear and working stands.
The foundations investigated were not reinforced at places weakened
by recesses, openings, and channels. Results of laboratory tests showed
that concrete had been used which, at the age of 28 days, had a temporary
compressive strength of 90 kg/cm 2 with slight deviations in some parts
of the foundation.
Concrete type 60 f was used for the foundations under
lifting platforms of rolling mill "750," and concrete type 130f for the
foundation under the first working stand of the same mill.
The foundations were placed on loessial clays with some sand. Owing
to the wetting of the soil, for different reasons the foundations underwent
uneven settlements resulting in the appearance of cracks. In the block
of the central part of the foundation under rolling mill "750" several
,

See footnote, Art. IV-l-c, p. 132.

MASSIVE FOUNDATIONS

305

cracks were observed in the tunnel under the driving gear and operation
stands, in the tunnels under the lifting platforms, in wells at the contact

between the foundations of roller conveyors and the foundations of

and in the foundation unit under the driving roll motor.
The appearance of these cracks was due to two causes:
1. A horizontal foliation of the foundation under the drive-gear and
The most
operating stands developed at the level of the anchor plates.
distinct crack was observed in the tunnel under the drive-gear stand.
Under the operating stands were found slightly developed small horizontal
These cracks indicate that a long
cracks coinciding with working joints.
interruption had occurred in the concreting of the foundation and that
no measures were taken to secure the monolithic character of the
rolling mills,

foundation.

There was a differential settlement of the foundation under the

and the foundation under adjacent auxiliary equipment.
This settlement was caused by the wetting of soil and resulted in cracks
in the tunnels of the rear and front lifting platforms, in the wall of the
middle platform of the staircase, in the arch near the lifting platform of
the second operating stand, and under the decelerator of the driving roll
2.

rolling mill

motor.

A vertical crack was observed approximately in the middle of the

tunnel of rolling mill "450." This crack ran along the walls in places
where they were weakened by niches, and along the arch.
A vertical crack was found in the tunnel of rolling mill "360" near the
inlet opening; two vertical cracks were found in niches, one of them running along the arch.
In the tunnel of rolling mill "280" a vertical crack was found under the
Channels
operating stand through which water was flowing abundantly.
of rolling mills "360" and "280," especially in their lower sections, were

with water.
instrumental investigation of vibrations of foundations under the
rolling mills was performed at several points along the foundation axis
and along its height: on the slabs of the operating and drive-gear stands,
at the level of niches where anchor slabs of the foundation were located,
and at points on the floor of the tunnels.
It is seen
Results of the measurements are shown in Table VII-1.
from this table that the largest amplitudes of vibrations were found
directly on the slab under the drive-gear stand of rolling mill "750."
filled

An

The measurements performed here showed that the foundation underwent extremely irregular high-frequency vibrations with amplitudes of
the order of 0.006 to 0.010 mm, caused by impacts of the gear.
These measured values of vibration amplitudes under rolling mills
show that the additional pressure on the soil and the stresses within the

DYNAMICS OF BASES AND FOUNDATIONS

306

foundation caused by dynamic loads are small in comparison with stresses

imposed by the weight of equipment and foundation. Therefore a value
of 3 for the dynamic coefficient, often taken in design computations of
such foundations, is exaggerated.

Table

Rolling

VII-1.

Results of Vibration Measurements on

Rolling-mill Foundations

Vibrations measured

at:

Nature of vibrations and amplitude

mill

"750"

Extremely irregular high-frequency vibrations

with amplitudes 0.005-0.010 mm. At time of
entry and exit of ingot, vibrometer records
impacts inducing vertical and horizontal vibrations with amplitude 0.030-0.050 mm.
Station 2 (at edge of High-frequency vibrations with amplitude less
foundation near slab of
than 0.003 mm. At time of entry and exit of
gear stand)
ingot, vertical impacts are recorded with ampliStation

(1

at slab of gear

stand)

"750"

"750"

Station 3 (housing under

"750"

Station 4

tudes of some 0.006-0.010

as for Station 2

mm.

The same

gear stand)
(on floor of

The same

as for Station 2

tunnel under gear

stand)

"750"

Station 5 (at surface of

foundation

near rolling-mill driving motor)

"450"

Station 6 (at surface of

foundation

near

roll-

ing-mill driving motor)

High-frequency vibrations with amplitude 0.003

mm. At time of entry and exit of ingot, impacts
are recorded inducing vibrations with amplitude
of 0.010 mm.
Quickly damped vibrations were recorded, with
amplitudes on the order of 0.0015 mm. Impacts
at time of entry and exit of ingot are only
slightly noticeable.

"360"

Station 7 (at surface of

foundation

near

Vibrations of same nature as those at Station 6

roll-

ing-mill driving motor)

"280"

Station 8 (at surface of

The same

as for Station 6

The same

as for Station 2

foundation

near rolling-mill driving motor)

Slab-

bing

Near working stand

Measurements of vibrations of the foundation under rolling mill "750,"

performed on the upper and lower parts of the foundation divided by a
horizontal crack, established that these two sections underwent vibraThis indicates
tions of the same character with the same amplitude.
It followed that
that the complete foundation vibrated as one block.
foliation of the foundation is not dangerous for rolling-mill operations.

MASSIVE FOUNDATIONS

VIM. Foundations

307

for Crushing Equipment

Design Computations of Foundations under Jaw Crushers. There

are many different arrangements of jaw-crusher operating mechanisms.
However, one common feature of these crushers is that, analogously to
reciprocating engines, they create unbalanced inertial forces varying with
These inertial forces form exciting loads which induce forced
time.
a.

vibrations of the foundation.

The most common arrangement

of the operating units of the jaw

one in which the motion of the mechanism is due to the action
Some typical arrangements of jaw
of so-called lower couples of rotation.
Approximate formulas
crushers of this group are shown in Fig. VII-13.
for the determination of unbalanced inertia forces are also given.
Accurate methods of computation of the exciting loads imposed by jaw
crushers may be found in specialized publications. 2
It follows from the equations in Fig. VII-13 that exciting loads imposed
by jaw crushers are of the same nature as exciting loads imposed by
Therefore all directives outlined in Chap. IV
reciprocating engines.
concerning the design of foundations for reciprocating engines may be
applied to the dynamic computation and design of foundations for jaw

crusher

is

crushers.

In
6. Computations
of a Foundation under a Gyratory Crusher.
gyratory crushers the ore is pulverized between the crushing head of the

main

shaft,

undergoing a rocking motion along a

circle,

and the armored

jacket of the upper stationary part.

Under the action

of frictional forces, the crushing cone

moves around

the axis of the crusher and develops an angular velocity whose value
close to that of the

movement but has

opposite sign.

As a

the frame of the machine, and consequently the foundation,

to the action of gyroscopic

and

inertial loads

which

may

is

result of this,
is

subjected

be approximately

expressed by one resultant exciting force:

R =
where

m =
m =
x

total

mass

mass

of

(wiri

- m

2r2

(VII-4-1)

of main shaft and crushing cone attached to it

camshaft and units rigidly connected with it (gears,

counterweights, and others)

ri

r2

distance between crusher axis and center of gravity of main

shaft

distance between another axis and center of gravity of

eccentric shaft

frequency of rotation of crusher

This force, rotating at a constant angular speed, acts
co

in a horizontal

plane

M
1

-tJ

03
->-5

d.
03

o3

.i-H

o3

O v

0.

3 ^O

>-c

Q>

OQ

3 ^

02

.0

(=1

02

t,_

03
(3

bfi

e+H

/-s
O & <
* g
a a

'dq

Q
02
02
03

02

a a

d
M

02
03
o3

11

11

-+^

II

cu

ha

d ^
3

*o

^o3

53

_g

-2

4S

qq

T3

fl
a>

fl
a>

a>

3M
o

o3

II
II

-5

**

T5

02

.2

+3

02

cq

0"S

II

I-H
'

"3
02

O
O
O

^
.-*

l__l

M
"S

"o3

>

2 m

+++i
^

++ + +

s***&
^^

.a

II

Dh

^^^

II

II

II

II

II

*
*
Ds Qh Oh

Oh Qs

(X,

.-

II

At
Ydn
Vfy

.a

s3

IsJ

*"

sJS^li

++

+ + + !

00

^^

'

II

..

'

'

II

II

^^N^j

>'

11

11

>

-fa4

(tf *

CO

<M

308

/
9

^0

(2/

VsS\V\\\U

11

oVkj

fcWsS^

5
6

Oh Qs h Oh

^V^c p

b
o3

3^

***ai

11

QsOh^Q,

(X,

R\\\^v\^

03

t^.

II

II

(VI

C4

02

+3

10

05

"=i

<5

'02
1
'

IS

*'*'>

*-

A ^ ^ ^^
^ J
J? lO
SS _:

o3

Oh

'to

"3

fi

<N

fe.

!a

.a ^

4^

8C

W %

JJ,

hH

1%

53

O
3

O
a

"3

1
.a

OQ

HH

09

4^

-*s

S~

o3

o3

3
Sh

+3

co
0)

co
QJ

*S

co

a,

03

fH

^H

03

03
co

d
C3

CO

03

fafi
#

'co

03

05

03
03

03

|
j d

cS

.5

III!
N

i)

10

ra

^^

l-H

g
K

03

.2

-2

c8

w
d

.3 fe

a,

.SP

'

d
*

43

ii

fts

03

CX,

CO

pd

h~

2 o

ii

03

3 s
cr "

HH
*

Hh-

"3

'

"0

*3

co
<u

o
o

co

3 O
d c

co

O
o

*-(

%
3

*&0

*3
0)

C-a

"S3

CO
0)

"3

"^

_d

'co

tj.

^
+

0j

^+

*S*S

++++

a
a
<

II

II

II

<

.-

^^

II

II

oj tf

&H

ftj

II

N
ft!

II

II

UN

<X,

IX,

IX,

u
03

d
03

,d

-m

4H

o
a
b

_/

p^r^

v^
V

ctfc

\\

^r

V\-*
H&
v
^y^p
v^^^r

o3

r\

^J

"3

\\\,

L^-^-

rn

10

CO

309

^3

i
3

'

^2*2
tf

^
N_^ ^-' m
O S
'

>
03

3 ^;

c3

i^

'

II

'

^ "3

'

II

'

II

II

Oh 0, Oh (X

^M
~^~

^^"^

^\

OjO
kj

CM

^h

!>

DYNAMICS OF BASES AND FOUNDATIONS

310

passing through the center of the main shaft

(in

crushers with a sharp

cone) or through the point of rest (in crushers with a flat cone).

Resolving

into

components along the horizontal axes x and

principal inertial axes of the installation,

Px R
Py = R

y,

the

we obtain

sin wt

cos ut

The dynamic computation of a foundation under a gyratory crusher

reduced to the determination of amplitudes of forced vibrations imposed
on principal vertical planes of the foundation by exciting forces P x and P y
is

and moments P x hi and P y h\.

Thus, dynamic computation

of a foundation under a gyratory crusher

not differ at all from the dynamic computation of a
foundation under a jaw crusher.
in principle does

VIII
PROPAGATION OF

ELASTIC

WAVES

IN SOIL

INTRODUCTION
As

stated in Chap.

I,

there are several reasons

why

the application of

For example, it has been indicated that

the elastic constants of soil depend on normal stresses and that elastic
deformations may affect the initial internal stresses which always exist in
It should also be noted that the solution of problems related to the
soil.
propagation of waves may be greatly influenced by dissipative properties
of soil which govern the absorption of wave energy.
When solving problems related to the propagation of waves in soils, one
has to start with models of the phenomenon, which are very far from
For example, the investigation of waves emanating from
reality.
machine foundations leads to a composite dynamic theory-of-elasticity
problem which starts with displacements in a certain section of the soil
Hooke's law to

surface

the

soils is limited.

while the

soil is

rest of the soil is free of stresses.

considered to be a semi-infinite elastic

In the simplest case

solid.

The

solution of

such a composite problem involves considerable mathematical difficulties.

is represented as an alternating force, either
concentrated or distributed over the given soil surface area. This model
Therefore, a source of waves

waves is far from reality, and the results of such a solution

(sometimes considerably) from the results of experimental
investigations of wave propagation from an actual source of waves such as
of the source of

may

differ

a vibrating foundation.

However, in spite of the indicated limitations, the development of the

theory of propagation of waves in soils on the basis of the theory of
elasticity, even for highly abstract conditions, gives us a chance to
investigate several very important specific features of
311

wave propagation

DYNAMICS OF BASES AND FOUNDATIONS

312

in

soils.

For example, the application of methods

of the theory of

made it possible to establish the influence of a free surface or a

layer on wave propagation in soil.
These investigations led to the
discovery of new waves, whose theoretically established properties were
elasticity

repeatedly confirmed by experimental data.

Experiments also

the principal conclusions concerning the dispersion of these waves,

relationship between the velocity of their propagation

As a

and
apply the phenomena

verified
i.e.,

the

their length.

even became possible to

of dispersion
to the study of the geology of construction sites.
Similarly, amplitudes of vertical vibrations of soil, obtained for vertical
sources of waves on the basis of theoretical considerations, are in good
agreement with the results of experiments.
By considering the influence of wave-energy absorption by soil, it
becomes possible to correlate experimental data with the theory of
amplitude changes in the vertical component of soil vibrations as a funcHowever, some theoretical conclusions concerning the
tion of distance.
propagation of elastic waves through soils do not agree with experimental
For example, according to the theory
data, and even contradict them.
of surface waves, the orbits of motion of particles on the soil surface have
Experithe form of ellipses with one axis perpendicular to the surface.
ments show that the orbits of motion of soil particles often deviate from
the elliptic shape and that the inclination of the axes of the orbit with
Accordrespect to the soil surface depends on distance from the source.
ing to theory, for a vertical source of waves the amplitudes of horizontal longitudinal components of vibrations should change monotonously
according to the same law which governs changes in the vertical components of vibrations. Actually this is not the case. Most of the experiments show that in cases of vertical sources of waves, the amplitudes of
result, it

of surface

waves

horizontal (longitudinal) components of vibrations of the

soil

surface

change with distance according to a much more complicated law. Similarly, theoretical data related to the distribution of amplitudes with
depth (for a surface source) do not agree with the results of some
experiments.

The above

discussion shows that the development of the theory of

wave

propagation, based on the theory of elasticity, should be carried out

simultaneously with numerous experiments which would provide data

for the

adjustment of theoretical conclusions.

Vlll-l.

Elastic

a.

Waves

in

an

Infinite Solid

Longitudinal and Transverse Waves.

the conditions of an absolutely isotropic

possible to use for the study of
differential equations of

Body
Assuming that soil satisfies all
homogeneous elastic body, it is

wave propagation

motion of an absolutely

in soil the general

elastic

body; these equa-

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

313

tions are as follows: 28

(X

where

w,

v,

+ mV

+ K=

w = components of elastic displacement

A = relative change in volume
dv
_ du
~ ~dx ^dy

+
,

(VIII-1-1)

along

x, y, z

axes

dw
dz

V 2 = Laplace operator

= iL

v*

dx 2

*L
d?/

11
dz 2

p density of soil
Z = components of body forces
X, n Lame coefficients, interrelated

X, Y,

E and

Poisson's ratio
x

=
(l

The

influence of the

fore

may

body

with Young's modulus

,)(l-2.)

E
(VIII-1-2)

forces, as in static problems, is small

and there-

be neglected.

one assumes that soil is a solid elastic medium extending to infinity

direction, then it follows that two types of waves independent of
each other are propagated from the source of vibrations; therefore the
displacement at any point in the soil is the sum of the displacements
caused by each wave; i.e.,
If

in

any

U =

The components
dWi
dy

which

will

dV!

Vi

V2

of displacements u,

v,

and w

U\

Ui

dUi

dz

dz

be valid

if

Ml

uh

vh

|?
dx

dWi
dx

and Wi have
1

= f?
dy

W =

W\

+w

also satisfy the conditions

<to\

dUi

dx

dy

/yjjT

o\

potentials; hence

Wl

=%?
dz

(VIII-l-4)

DYNAMICS OF BASES AND FOUNDATIONS

314

The components

u2

of displacements

A * = *ar+
da;

v2

and

p + f? =
dy

dz

satisfy the equation

(viii-i-5)

Thus, the components of the total displacement of

Substituting these values of u,

v,

and

soil

into Eqs. (VIII-1-1),

the following equations which should be satisfied by

a 2 VV

b 2V 2u 2

b 2 V 2v 2

b 2V2

dt 2

d*u 2
dt 2

d 2v2
dt 2

d 2w 2
dt 2

a2

equal

<p,

u2

v2

we obtain
and w 2

(VIII-1-6)

(VIII-1-7)

(VIII-1-8)
P

b2

(VIII-1-9)
P

Equations (VIII-1-6) and (VIII-1-7)

differ

only in the coefficients of V 2

Expressions (VIII-1-3) represent the components of the vector of

rotation; therefore

it

is

evident that a

wave propagating

in

soil,

described by Eq. (VIII-1-6), will not induce shear deformations.

mations corresponding to this wave are of such order that

only a relative change in volume which equals
Ai

soil

as

Defor-

undergoes

= VV

Hence, the wave corresponding to Eq. (VIII-1-6)

is

a wave of compression

and expansion. However, condition (VIII-1-5) shows that no changes in

volume occur for waves whose components satisfy Eqs. (VIII-1-7); when
these waves are propagated, the soil elements undergo only relative displacements whose components equal

2w x =

dw

dy

dv 2

dz

_
du 2
2w v =

dz

dw 2

dx

2w z =

dv 2

dx

du

dy

Hence, the waves satisfying differential Eqs. (VIII-1-7) are waves of

shear.

PROPAGATION OF

When waves

of compression

WAVES

ELASTIC

IN SOIL

315

and expansion are propagated,

soil dis-

placements are parallel to the direction of transmission; therefore waves

of compression and expansion are called longitudinal waves.
Shear waves may be called transverse waves, because their propagation
induces soil displacements only in a direction perpendicular to the direction of transmission.

Equations (VIII-1-6) and (VIII-1-7) are not interrelated. Hence, in

an infinite solid body longitudinal and transverse waves propagate independently of each other. If the conditions of appearance of the waves
are such that the components of soil displacement at the initial moment
correspond only to a change in volume, then transverse waves will not
appear, and vice versa.
The
b. Propagation Velocities of Longitudinal and Transverse Waves.
integral of a differential equation of the type of Eqs. (VIII-1-6) or
(VIII-1-7) describing waves may be taken in the form

H-iO
where

r =
F =
c =

radius vector of point under consideration

arbitrary function satisfying initial and boundary conditions

wave propagation;

velocity of

expansion

Using Eqs. (VIII-1-2),

a; for
it

waves

is

waves

of compression

waves

and

possible to bring Eqs.

(VIII-1-9), describing velocities of

to the

for

of shear c

of compression

(VIII-1-8) and

and expansion,

form

(1

)(1

\2(1-+
Hence

it

(VIII-1-10)

2v)p
(VIII-1-11)

v)p

follows that the velocities of propagation of longitudinal

and

transverse waves depend only on the elastic properties and density of the
soil.

The

ratio

between the velocities of the compression and shear waves

equals

a _

Ml

rViand always exceeds unity.

(VIII-l-12)

2v

Consequently, longitudinal waves have a

higher velocity of propagation than transverse waves.

between the two

Poisson ratio
is

v.

The

difference

velocities is directly proportional to the value of the

The

relative value of the velocity of compression

smaller in sandy soils than in clayey

soils,

waves

because the Poisson ratio

is

DYNAMICS OF BASES AND FOUNDATIONS

316

Table VIII-1 gives numerical values of velocities

smaller for the former.

of

waves

of compression

Table

VIII-1. Velocities of
Soil

and

p,

Moist clay
Loess at natural moisture
Dense sand and gravel
Fine-grained sand
Medium-grained sand

shear.

kg

Compression Waves a and Shear Waves

1.8
1.67

1.70
1.65
1 65
1.8
.

Medium-sized gravel

sec 2 /cm 4

10-6

X
X
X
X
X
X

m/sec

a,

b,

10-6

10-6
10-6
10-6

m/sec
150
260
250
110
160
180

1,500
800
480
300
550
750

10-6

Propagation of Waves Induced by Underground Explosion: Camouflet. |

an explosion occurs at a considerable depth (for example, in a mine),

c.

If

then to establish soil displacements at points located some distance away

from the surface the waves emanating from the explosion center may be
considered to be either longitudinal or transverse waves propagating in
an infinite solid body.

When a camouflet takes place, the initial zone of excitement of the

approaches the shape of a sphere. Then the radial components of
placements,

i.e.,

the components in the direction of

wave propagation,

are large in comparison with the tangential components.

for a camouflet explosion the

when

may

soil

dis-

Therefore,

of

wave propagation

f(t)

(VIII-l-13)

boundary conditions

be taken to be as follows:

=/;

w =

fit)

where

f(t) is the assigned function of time

Conditions (VIII-l-13) show that there occur only radial displacements
of the soil on the camouflet surface.
Therefore only longitudinal waves

be propagated from the explosion center.

will

may

The

following relation

be accepted for these waves:

If
r

H)

(VIII-1-14)

Since the radial component of displacements equals

u r = \/u 2
t

+w

Translation Editor's Note: "Camouflet" is a term of French and British

and refers to an underground blast, the effects of which do not produce

military origin

visible displacements of the soil surface.

PROPAGATION OF

we

WAVES

ELASTIC

IN SOIL

317

obtain, according to Eq. (VIII-1-14),

dF
1

Ur =

H) +

dt

In order to find the function F,

which

may

H)J

us use boundary conditions (VIII-1-13),

let

be rewritten as follows:
Ur

Setting r

in

f(t)

Eq. (VIII-1-15), we have

ndF(t-

roAv

~+

Fvnt
6~

gggi)

di

or

.,_=

~r

a)

(VIII-1-17)

e^o): d^l

(VIII-1-18)

-ro 2/6i

F(0

(VIII-1-16)

2/W

where

The

(VIII-1-15)

solution of Eq. (VIII-1-17) has the

c - na

-(a/r )<i

form

+ ^)

(h
f f

The displacement of the camouflet surface created by the

be expressed by the following function

explosion

may

f(t)

Values of a and

fi

Since at the initial

surface,

(VIII-1-19)

ate'*

depend on the properties of the explosive charge.

there are no displacements on the camouflet

moment

should be zero.

Substituting the expression for

integral of Eq. (VIII-1-18)

and

f(t)

from Eq. (VIII-1-19) under the

we

integrating,

= J<? aa
0r o - a

(t
\

obtain

_ \

(3r

e-pt

a/

(VIII-1-20)

According to Eq. (VIII-1-15), we find the following expression for

> at from the center of

the magnitude of displacement at a distance r

the explosion:

ur =
rG8r

-a) ja|/

+ t-

-a/ P J

5^H[.-K -i)]
f

(vm

- i -2i;

DYNAMICS OF BASES AND FOUNDATIONS

318

This expression makes

it

possible to evaluate the effect of an explosion

any distance from the charge. It is easy to see that at small distances
from the explosion center the displacements of the soil decrease rapidly;
at

the decrease

is

At

distance.

approximately inversely proportional to the square of the

large distances the decrease of the amplitudes

is

inversely

proportional to the distance.

The

computations depend on the assigned dimensions of

/? which characterize
changes with time of the soil displacement at the camouflet boundary.
These constants can be determined experimentally.
results of

the camouflet, as well as on the constants a and

VIII-2.

a.

Changes with Depth in Amplitudes of

Produced by Surface Waves

Soil Vibrations

General Expressions for Components of Soil Displacements Induced

by Waves Caused by a Vertical Exciting Force Acting along a Surface Line.

With the exception

industrial sources of

deep mines, practically

of sources located in

wave excitement

all

the

in soil lie close to the surface, as

receivers of waves, such as foundations under buildings.

Therezone close to the surface is of

practical interest in the study of the dynamics of bases and foundations.
In this connection, we must establish how the propagation of waves in
soil is influenced by the presence of the free surface of soil.

do

all

fore the study of

waves propagating

in a

Let us confine ourselves to the case in which

du

forces

and

dz

dz

Then if the action of body

of wave propagation are as

dv

w =

is

neglected, the differential equations

follows:

(VIII-2-1)

Let us take a general solution of these equations in the form

where

3>*

and

?*

dy

dx

>P* are analytic functions.

We direct the y axis perpendicular to the soil surface

in soil, y

>

0.

(VIII-2-2)

then for all points

Considering steady propagation of waves with frequency
;

PROPAGATION OF
o),

WAVES

ELASTIC

IN SOIL

319

we take
** =

e*"*(x,y)

(VIII-2-S)

Then the equations which should be

by the functions

satisfied

3>

and

are as follows:
(V2

(V 2

where

+
+

*">*
A;

)*

lVm

(VIII-2-4)

==

= %
b

(VIII-2-5)

2tt

Since

T is

where

the period of the propagated waves,

and

fc

it is

clear that

However, aT and bT represent the lengths of longitudinal and transverse

Hence, h and k are reciprocal values of the wavelengths, and k
is always larger than h.
waves.

Particular solutions of Eqs. (VIII-2-4) are taken as follows:

"

$ = Ae--<xyai&
^ = Be--PVgifr
~ h
a = e -/*
a^f
=

? - k
2

(VIII-2-7)

A, B, and

(VIII-2-6)

are arbitrary constants determined

by boundary conditions.

Let us assume that the border plane y =

is subjected to the action
of external normal forces distributed continuously over the entire plane;
these forces induce normal stresses equaling

ay

We
zero;

assume that the tangential

(VIII-2-8)

roe**
stresses

on the border plane equal

i.e.,

r yx

Stresses

<r y

and

(VIII-2-9)

ryx are expressed through functions

-22
jh

-^
dxdy

k 2V

dx 2

<f>

and

as follows:

dx 2

(VIII-2-10)

dxdy

Substituting, into the right-hand parts of these equations, expressions for

<

and

taken from Eqs. (VIII-2-6) when y

0,

we obtain

the

DYNAMICS OF BASES AND FOUNDATIONS

320

A and

equations for calculating the constants

B:

-2i&A + (2 - k )B =
(2 - k*)A + 2itfB
2

2? A =

Hence

F(Q

F()

Fit)
2ia (To

B =
where

/C

(2

(7q

(VIII-2-12)

- ky -

(VIII-2-11)

CTQ

4 2 a/

(VIII-2-13)

Using Eqs. (VIII-2-2), (VIII-2-6), and (VIII-2-12), we obtain

.

--
v

(2

k 2 )e-y

-(2 2

2a&e~te

m+

k 2 )e- ay

^t

.,

2?erto

*t
.

t
%%x
p"<,^

Fit)

(VIII-2-14)

(TO

In order to transform the exciting force into one acting along the line

1/
Fig. VIII-1. Exciting force [Eqs. (VIII-2-15)].

0,

(Fig. VIII-1),

we assume
VO
2tt

Substituting this expression into the right-hand parts of Eqs. (VIII2-14)

and integrating with respect

to

from

oo

U=

iP

"2x M
2-K\i

[+

[(2g

k*)e- av

* -(2? -

k*)er<"

oo
f

we obtain

v:

2agg-^]e^

WW)

/_

/::

to

the following expressions for the displacements u and

2j 2 e-e]e i(

(VIII-2-15)
>

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

321

Equations (VIII-2-15) correspond to the forced waves induced by an

= 0, y = 0.
Free surface waves occur
b. Propagation Velocity of Surface Waves.
where the border surface is not subjected to the action of forces and the
waves are induced by some initial excitement. Assuming for this case
that <jq = 0, we obtain equations for determining the constants A and B:
exciting force acting along the line x

-2itaA
(2

A;

- k )B =
A + 2i^B =
2

(2?
2

will give solutions other

Equations (VIII-2-16)

(VIII-2-16)
1D;
lvmz

A and B
Hence we obtain

than zero for

only when the determinant of this system equals zero.

equations for determining :
F({)

(VIII-2-17a)

Instead of this equation, which contains irrational expressions, let us

consider the following equation, which does not contain radical signs:

*W(Q

(2

- wy - lea - w)(e 24 - 16
1 - 8
2

l+

/c

)^

p)g-

16

-S)l]

(VIII-2-17)

where

/()

(2^

/c

2
)

4| 2 a/3

one of the roots of Eq. (VIII-2-17) lies between 1 and + oo

It is easy to show that the other two roots, if they are real, lie between
and h 2 /k 2
The first root corresponds to positive values of a and (3; therefore this
The last two roots make
root does not satisfy the condition /() = 0.
a and (3 positive and imaginary; therefore they do not satisfy the equation
The latter equation has only one root ( 2 = x 2 ) which is
F() == 0.
For a Poisson ratio of 0.5, the real
Therefore x 2 > k 2
larger than 1.
root of Eq. (VIII-2-17) is
Since k

>

h,

For

value

1.04678

0.25, all roots of Eq. (VIII-2-17) are real; they are equal to

|
Of these

roots,

= i;i(3-V3);i(3

only the last

satisfies the

+ V3)

conditions of the problem;

is

V3 + V3 =

1.087664

its

DYNAMICS OF BASES AND FOUNDATIONS

322

Analogously to Eqs. (VIII-2-5),

us designate

let

CO

where

the velocity of propagation of the surface waves under con-

c is

sideration

it is

clear that

= -b

For v = 0.5,
for v = 0.25,

Thus

it is

=
=

0.95536
0.91946

seen that surface waves propagate with

somewhat smaller

velocity than transverse waves.

c. Approximate Expressions for Soil Surface Displacements Induced by
Waves Excited by a Concentrated Vertical Force. Let us return to the

consideration of the general Eqs. (VIII-2-15) for displacements induced

by waves excited by a

vertical concentrated force acting

on the surface.

Integrals on the right-hand sides of Eqs. (VIII-2-15), as they are written,

some values

are indeterminate, since the function F() for

of (e.g.,

In the general case, at y = 0, it is very difficult to

Therefore let us confine ourselves to the
disclose this indeterminateness.

= x)

equals zero.

investigation of soil displacements at points on the surface (y

we

0); then

obtain

uq==

iP

2^

[ + " (2 2

/c

2ap)ei * x Jy

~^

For

/_.

(VIII-2-18)

kW*

-- r

These integrals remain indeterminate; however, it is possible to single

out their principal value. Skipping details of this operation, which
includes integrating in the area of the complex variables along the
contour containing special points, we present here the following final
equations:

where

oV =

H
*

vo

= -

/i

uo

Hoe

(3S

'**

Su
(VIII-2-19)

Koe'-**

Sv

K ^= ~

^^
=

k*

F'(x)

W(x)

=*

(VIII-2-20)

PROPAGATION OF

For large kx and

mVttV

WAVES

asymptotic expansions of

hx,
k2

Su =

ELASTIC

IN SOIL

323

Su and Sv

are as follows:

f>i(a>tkxirl4)

(kx)*

W Vk- 2

J2

(k 2

*-?(-

te

h 2 lew-**-*'

2h 2 Y

(hx)*

'

(VIII-2-21)

{(utkxir/A)

(kx)*

h 2k 2

je

J2

(k 2

h 2) 2

i(o>t-hx-*i4)

(hx)'

Equations (VIII-2-21) show that not one wave, but a series of waves
be propagated from the site of the excitement of vibrations; the
velocity of propagation of these waves equals the velocity of propagation
of the longitudinal, transverse, and surface waves.
The amplitudes of
surface waves [described by the first terms in Eqs. (VIII-2-19)] do not
change with increase in distance from the source of waves. However, it
follows from Eqs. (VIII-2-21) that the amplitudes of waves, propagating
with the velocities a and b, decrease very rapidly (at least inversely
proportionally to x*) with an increase in the distance from the source.
Therefore at a sufficient distance from the source of the waves, the
amplitudes of the additional spectrum of waves are small in comparison
with the amplitudes of the surface waves.
will

Owing

to interference in the area close to the source of the separate

waves, the wave amplitudes do not change monotonously in proportion to

distance from the source, but, as this distance grows,

minimums of amplitudes are observed; the values of

and minimums become smaller as the distance grows.

maximums and
maximums

these

d. The
If one
Variation with Depth of Surface-wave Amplitudes.
assumes that x is not very small, then the additional spectrum of waves
may be neglected and it may be considered that only surface waves are
propagated which are described by the first terms in Eqs. (VIII-2-19).
Let us investigate the changes of amplitudes of these waves with depth

according to Eqs. (VIII-2-2) and (VIII-2-6)

we have,

substituting x for

x:
u

ttlv

where

Assuming y =

ai

0,

= ixAe~ aiy - PiBe-t iy )e iKX

= (-aiAe- + ixB-^y)e x
= K - h
0i = x - k

(VIII-2-22)

we obtain
uQ =
Vo

ixA

= ( aiA

^ 1 B)e
-f ixB)e

i>

(VIII-2-23)
l

DYNAMICS OF BASES AND FOUNDATIONS

324

Equating the right-hand parts

S u and S v we
,

and

ixA

/SijB

= - Ho

ixB

= iKq

A and

B,

we obtain

for determining constants

and (VIII-2-23), and

obtain two equations

of Eqs. (VIII-2-19)

disregarding a temporary multiple and

corresponding only to the surface

waves:

Solving these equations for

xH +

A =

cnHo .+

B =

ij8i

foKo

xK P
x2

ju

A and 5 into Eqs. (VIII-2-22), introducing a

and disregarding imaginary terms, we finally

Substituting these values of

temporary multiple

iut
,

obtain

=
\

a
aiPi

k 2~
1

~ &

a
arfj

x i~
2

cos
/

vff

4- A K-

aipi

otipi

(cot

,1

xx)

(VIII-2-24)

x"

sin

(cot

xx)

These expressions for components of soil displacement show that for

waves the orbit of motion of soil particles is an ellipse whose axes
coincide with the coordinate axes.
The interrelationship between the
amplitudes of the vertical and horizontal components depends on the
surface

depth.

Equations (VIII-2-24) make it possible to evaluate changes with depth

amplitudes of waves. We present below such an evaluation for
Poisson ratios of 0.5 and 0.25. Table VIII-2 presents values of several

in the

two

coefficients contained in Eqs. (VIII-2-24) for these

cases.

Substituting data of Table VIII-2 into Eqs. (VIII-2-24),

following amplitudes of waves: for Poisson's ratio

u
v

= (-0.1298e- ^ /L
2

<

= (0A298e- 2 * ylL <

0.0706<r (0

0.2387e- (0

2958)2

2958)2

the

0.5:

^ /L <)

M
-

we obtain

^ /L -

(VIII-2-25)

PROPAGATION OF

for Poisson's ratio v

ELASTIC

WAVES

IN SOIL

325

0.25:

u = (-0.2958e- (0

8474)2

^^ +

(0.3933)27T2//L

0.1707e"

M
v

where

(0

2507e

Poisson's

VIII-2.

8474)2,r2//Lc

0.4341e

-(0

Values of Coefficients

3933)2,ry/Lc
)

(VIII-2-26)

in Eqs. (VIII-2-24)

xHo

0.5

the wavelength.

is

Table

ratio

_(0

a\

/3i

ai/3i

0.05921

oo

00

0.10890

f5K-o

x2

aiHo+ xKo
aijSi

x2

-0.1298

-0.2387

-0.2958

-0.4341

0.2958x
X

0.25

V3

2.5

0.12500

0.18349 0.8474x 0.3933x

Figure VIII-2 gives graphs of u and

v for v

seen from these graphs that the amplitudes of

waves increase with an increase in the

depth to y = 0.2L C further on they
decrease, attaining, at a depth of the
;

order of

0.35L C

decrease.

in

depth,

Thus

and P \i = 1. It is
the vertical components of
0.5

Poisson
1.0

0.5

'/

'

ratio

= 0.5
-0.5

-1.0 jz_

values corresponding

With

to those at the surface.

increase

it

can

further

amplitudes

the

be

considered

an approximation that the amplitudes of vertical components of surface waves change relatively little to a
depth of the order of 0.4L C
Amplitudes of the horizontal longitudinal component of a wave change
with depth in an opposite manner: to a
depth of the order of 0.1 5L C the amplitudes rapidly decrease, approaching zero.
Then, to a depth of the order of 0.45L C
as

they increase; with further increase in

depth, the amplitudes gradually decrease.
Fig. VIII-2. Variations of wave
e. Experimental Investigations of Variamplitudes with depth.
ation with Depth of Surface-wave Amplitudes.
The foregoing theoretical data concerning the distribution
of amplitudes of soil vibration near the surface refer to surface waves

whose influence prevails at distances from the wave source which are

DYNAMICS OF BASES AND FOUNDATIONS

326

At

considerably larger than the wavelength.

from the

be influenced by

soil

vibrations produced

depth may
supplementary
vibrations with depth

by waves

spectrum; therefore the actual distribution of

may

relatively small distances

source, the distribution of vibration amplitudes with

of the

soil

greatly differ from that established

by Eqs. (VIII-2-24). The distribution may also be influenced by the fact that sources of waves actually
greatly differ from the concentrated vertical force assumed in the above
Thus, the assumption that waves produced by foundations
discussion.
undergoing vertical vibrations are excited by vertical exciting forces may
lead to considerable error in the determination of

soil

vibration ampli-

tudes at small distances from the source.

Great

difficulties are

involved in theoretical investigations of the influ-

ence of these factors on the distribution of soil vibration amplitudes

with depth. Therefore, experimental investigations are of special impor-

The author performed such

tance.

investigations under diverse soil

conditions in which foundations for machinery acted as sources of waves.

In one of the experiments, a foundation with a 1.0- by 1.0

-m

area

with soil acted as a source of waves. The foundation was

placed 2.5
below the soil surface, the usual depth of foundations under
machines.
It was not placed on the surface since, in an area close to
the source, the depth of the foundation may considerably influence the
distribution with depth of vibration amplitudes.
In order to measure the amplitudes, three test pits were dug, each
from the foundadeep.
Pit 1 was located at a distance of 3.0
2.5

in contact

tion; the other

15.0

m from

The

soil

two

pits (2

and

3)

were placed at distances of 7.30 and

the source.

at the site of investigation

was

fairly

homogeneous and con-

with some sand and silt and thin

layers of clayey sands; no significant difference was observed in the
mechanical properties of these varieties of soil. The ground-water level
sisted of alternating thin layers of clay

of some 5 to 6 m.
vibromachine was employed for the excitement of vibrations
of the test foundation.
Experiments were performed only for vertical
vibrations of the foundation at rates of 800, 1,000, and 1,200 oscillations
per minute.
The soil vibrations were recorded by an optical vibrometer
with a constant magnification of about 500 times. This double-component vibrometer made it possible not only to measure the amplitudes of
the horizontal and vertical components of waves, but also to fix the

was at a depth

field

Lissajous figures.

Measurements were performed only

points of different depths: the

the

soil

first

in a longitudinal

plane at eight

surface; the others were located each 0.30

ceding one; the last point was at a depth of 2.2

m lower

below
than the pre-

point was approximately 0.10

m from the surface.

The

PROPAGATION OF

N=

800 min"

WAVES

ELASTIC

/V

1000

min"

IN SOIL

/V

=1200

component
vibrations, mm x 200

Amplitude

of vertical

Test

12

12

pit

327

min"

of

12

12

0.1

0.4
0.7

^1.0
$1.3
1.6
1.9

2.2

Test

12

pit

2-345

0.1

0.4

0.7

E
-1.0
Q-1

1.6
1.9

2.2

Test pit 3
2 3 4

0.1

0.4

0.7

E
.1.0

f-1-3
1.6

1.9

2.2

Designations:

o Backfilled foundation
Exposed foundation

Fig. VIII-3. Experimentally determined variation with depth of the vertical component of oscillations induced by a vibratory machine.

measurements

in

each pit were performed for two cases:

the foundation were completely relieved from

soil

(1) side faces of

pressure;

(2)

the

foundation was backfilled. In the first case, only the foundation area
in contact with the soil served as a source of waves; in the second case,
both the contact area and the side faces acted as a source of waves.
Figures VIII-3 and VIII-4 give the results of the measurements as
graphs of changes in amplitudes with depth. It is seen from Fig. VIII-3

DYNAMICS OF BASES AND FOUNDATIONS

328

that within the pit depth changes in the amplitudes of the vertical com-

ponent are relatively small.

in all three pits,

and at a distance

The

This conclusion holds for measurements

immediate vicinity of the source of waves
from it.

in the

i.e.,

of 15

velocity of surface-wave propagation in the investigated soils

be considered to be of the order of 200 to 250 m/sec.

of the propagating waves will be:

For
For
For

N=
N=
N=

Thus

800 min"

1
:

min1,200 min" 1

1,000

L =
L =
L =

may

Then, the length

m
m
12.5 m

15.0 to 19.0

12.0 to 15.0

10.0 to

performed the lengths of the propagating

m.
It appears from the theory of surface waves that down to depths of
the order of one-half the wavelength, the amplitudes of the vertical component of surface waves change relatively little. Taking lengths of
propagating waves as being equal to 10 to 20 m, it is possible to assume
on the basis of the foregoing theory that down to a depth of 5 to 10
the amplitudes of the vertical component will not change much with
depth.
This conclusion approximately coincides with the results of
measurements of amplitude of the vertical component in pits with depths
to 2.5 m.
The graphs of Fig. VIII-4 show that in most cases the amplitudes of
the horizontal longitudinal component slowly decrease as depth increases.
The only exception was found in pit 1, where a relatively rapid damping
of amplitudes occurred; the source of waves there was a backfilled foundain the experiments

waves were

of the order of 10 to 20

much more
wave propagation near the free
surface of soil.
This discrepancy between experimental data and theory
is evidently explained by the fact that in reality the propagation of waves
is governed by a more complicated law than considered by the theory.
It should be noted that a similar discrepancy between theory and experimental data was disclosed by studies of changes with the distance from
the source of the same horizontal longitudinal component of a wave in
tion.

In

all

other cases the amplitudes decreased with depth

slowly than postulated by the theory of

the case of vertical vibrations of the foundation acting as a source of

waves

(see Art. VIII-3).

comparison of soil vibration amplitudes in the two types of foundations shows that the amplitudes are much larger in backfilled foundations
than in exposed foundations. This is explained as follows: in the first
case, the energy of the vibrating foundation is transferred to the soil
through the side surfaces and through the foundation area in contact
with soil; and in the second case, only through the base contact area.

PROPAGATION OF

800

min"

ELASTIC

/l/

WAVES

1000min

IN SOIL

/V

329
_1

1200min

Amplitude of horizontal component of

vibrations mm x 200
,

2345

Test pit

12345

12345

0.1

0.4
0.7

E
.1.0

1 1-3
1.6
1.9

2.2

Test pit 2

12

12

12

12

12

12

0.1

0.4
0.7

1.0

1 1.3
1.6
1.9

2.2
Test

0.1

0.4

pit

0.7

E
^-1.0
1-1.3
1.6

1.9
2.2

Designations:

o Backfilled foundation
Exposed foundation

Fig. VIII-4. Experimentally determined variation with depth of the horizontal component of oscillations induced by a vibratory machine.

Graphs obtained as a result of experiments carried out under other soil

show that the changes of amplitudes with depth differed
somewhat from the changes outlined in the preceding discussion. Figure
VIII-5 gives a graph showing the variation of vertical amplitude with
depth along the side face of a 4-m-deep ditch dug at a distance of 2.5 m
from a test foundation whose area in contact with soil equaled 0.8 m 2
This foundation was placed on the surface and was excited vertically at
conditions

DYNAMICS OF BASES AND FOUNDATIONS

330

On

the rate of 670 vibrations per minute.

the

soil

the site of the investigation

some sand. The graph shows that

the amplitudes were two times smaller than those on the

consisted of loessial clay with

at a depth of 4

Amplitude,
0.10

0.20

mm

Amplitude,

0.30

0.40

0.05

mm

0.10

0.15

0.20

2
<u

Fig. VIII-5. Variation with depth of the

Fig. VIII-6. Variation with depth of the

component of oscillations induced

by a vibratory machine operating at
670 rpm.

vertical

vertical

by a

component

of oscillations induced

pile driver.

and that the amplitudes of vibrations changed with depth

approximately exponentially. The relationship between amplitude and
depth, established from this experiAmplitude, mm
ment, agrees with the theory, provided
0.2 0.3 0.4 0.5 0.6
0.1
it is assumed that one of the terms in
soil surface,

the right-hand part of the expression

0.25

for v is large in

0.50

comparison with the

other term.

0.75

in

1.00
Q.

About the same character of changes

amplitude with depth was estab-

from experimental investigawaves propagated by a pile

A graph of the results
driver.j

a>

lished
1.25

tions of

1.50

given

is

1.75

soil

2.00

in

Fig.

VIII-6.

was found on the

site

Loessial
of

the

investigation.
Finally,

Fig. VIII-7. Variation with depth of

the vertical component of oscillations

induced by a forge hammer.

Fig.

VIII-7 gives graphs

the variation with depth of the

vertical component of soil vibration

of

amplitudes, plotted from the results

of the author's investigations of distribution of waves produced by the

foundation under a forge

f

The

hammer whose dropping

investigations were performed

parts weighed 7.5 tons.

by Ya. N. Smolikov.

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

331

Measurements were performed at a distance of 7 m from the source of

waves. At this site were found layers of clays with some sand and silt
alternating with clayey sands.

In spite of some contradictions between the experimental data discussed above and the theoretical data, the conclusion

is

possible that at

small depths, approximately in the range 0.2 to 0.5 wavelength, changes

in vibration amplitudes are relatively small.

This finding

practical importance, because an increase in the depth of

is

of a certain

wave

receivers,

such as the footings under columns of a building, may have significant

influence on the vibration amplitudes of the receivers only at considerable
depths.
A small deepening of the wave receiver in practice will not
influence the amplitudes of its vibrations.
Similarly, the vibration
amplitudes of a receiver will not be significantly affected by a small

depth in relation to the depth of the source of waves.

not necessary to place foundations under machines deeper
than foundations under adjacent structures (walls, columns, etc.).
The depth of the foundation under an engine may be selected without
consideration of the effect of waves produced by this foundation on
footings under walls and other structures.
This condition in many cases
(especially where there is a high level of ground water and in the presence
of previously constructed foundations) eliminates complications which
may develop in the process of construction of a machinery foundation
located near footings under walls or columns.
change

in its

Therefore

VIII-3.

it is

Dependence of

the Amplitude of Soil Vibrations on

Distance from the Source of

Waves

a. General Expressions for Soil Displacements Produced by Concentric

Waves. Of practical interest are waves emanating from an area of
excitement concentrated on a small section or distributed over a limited

soil surface.
Of special importance is the problem of wave
propagation with reference to displacements of a section of the soil
surface not subjected to external loads.
As indicated above, the solution
of a problem with such boundary conditions is very difficult and has not

area of the

yet been found.

ourselves to

Therefore in the following discussion

known

surface of the

soil,

we

will confine

exciting forces distributed over a section of the free

including the case in which the section

is

infinitely

small.

Assuming that the source

of

waves

shall seek solutions of Eqs. (VIII-1-6)

<p

U2

where

a>

is

i<tTj

a harmonic function of time, we

and (VIII-1-7) in the form

is

= e^Hx,y,z)
i*ty
V2 = e

the frequency of excitement.

(VIII-3-1)

W2

it

(VIII-3-2)

DYNAMICS OF BASES AND FOUNDATIONS

332

Substituting Eqs. (VIII-3-1) and (VIII-3-2) into Eqs. (VIII-1-6) and

(VIII-1-7),

we

find that

U, V, and

3>,

should satisfy the following

equations:

(V 2
(V

)U =

(v

h 2 )$

(VIII-3-3)

)V =

(V 2

k 2 )W

(VIII-3-4)

In addition, the following relationship should be satisfied:

+ JZ + =
da;

The

solution of Eqs. (VIII-3-4)

d 2 ^>

rr

U =

d:cd?/

(VIII-3-5)

dz

d?/

sought in the form

is

d ^
F = T^T
d?/dz
2

TF

d2

+ ^
^|
d2
/c

Substituting Eqs. (VIII-3-6) into Eqs. (VIII-3-4),

function

\I>

(VIII-3-6)

we

find that the

should satisfy the equation

(V 2

k 2 )*

(VIII-3-7)

Let us introduce a cylindrical system of coordinates, placing the origin

on the soil surface; then we have:

of coordinates

r cos 6

r sin

In this system of coordinates,

v , = il2
dr

Designating by

q a

+ il
+ ii
r dr
dz
2

displacement in the direction of the radius vector

r,

we

obtain

u -

y
a
r

According to Eqs. (VIII-l-5a) and (VIII-3-6), we have

+
dr

= |? + 5| +
dz
dz
2

dr dz

(VIII-3-8)

Let us assume that an exciting load of the type

P =
z

is
is

acting on the

soil surface, i.e.,

the Bessel function of the

stresses are

therefore,

(VIII-3-9)

-PJottr)

at z

first

0; in the

above expression J

type, of zero order.

No

(%r)

tangential

found on the surface;

rr

(VIII-3-10)

PROPAGATION OF

We

WAVES

ELASTIC

IN SOIL

333

take the solution of Eqs. (VIII-3-3) and (VIII-3-7) in the form

* =

$ = Ae-wJoi&r)

BfT+iJoUr)

(VIII-3-11)

where A, B, and are arbitrary constants.

Using the boundary conditions of Eqs. (VIII-3-9) and (VIII-3-10), we
obtain

A =

W^ ~

2J

B =

(VIH-3-12)

W*-

where the function F() is determined by Eq. (VIII-2-13).

The displacement components on the soil surface (when y
responding to Eqs. (VIII-3-11) and (VIII-3-12) are as follows:

_ -s(2p-fc.-2aff)
q

^o

where Ji

is

WTt\

J ni r)

0) cor-

(VIII-3-13)

k*a
P
= ^yJo(|r)-

the Bessel function of the

type and the

first

first

order.

In order to consider the excitation of waves by a vertical exciting force

ve

iu3t
,

acting at the origin of coordinates, let us assume that

Substituting this expression for

into the right-hand parts of Eqs.

(VIII-3-13) and integrating in the range from

displacements on the

<Zo

to

+ , we obtain for the

soil surface:

= p- [ "
T/i
27TM

m2 ~^~

2a/?)

Fft)

Jo

Jim d$

(VIII-3-14)

k 2 a
^K[k
b.

Displacement of Soil

For a

at

Small Distances from

vertical exciting source, the vertical

of practical interest.
tion of the value of

We

Jo

Therefore

let

the

component

Source of Waves.

of soil vibrations is

us confine ourselves to the investiga-

only, assuming that the value p

kr

assume that

fcWottr)

Let us introduce a

new

variable of integration, assuming that

where k

is

determined

in

kd

accordance with Eqs. (VIII-2-5).

is

small.

DYNAMICS OF BASES AND FOUNDATIONS

334

According to Eqs. (VIII-2-7) and (VIII-2-13), we have:

F(0 =

(2

J =

Hence

k 2) 2

Jo

4{*q0

l) 2

/&

40 2 \/d 2

40 2 V<? 2

#2

#2

Ve 2

(vm

Ve 2

1]

. 3 . 16)

# = r

where

and

l) 2

V^Wfl
-

(2d 2

k\(2d 2

is

The

determined by Eqs. (VIII-2-5).

wave propagation

velocity of longitudinal

wave propagation;

the velocity of transverse

is

always higher than

<

therefore h

<

1,

and

we have

J -

So

SI

- J<

/i"

Ji

Let us consider

Vo -

Jo 2(e

&2

e
2

i)

40

Jo(pd) de

Ve 2

$2

Ve 2

According to the meaning of the problem, the coefficients a and (3 should

have positive values for all values of 6. During changes of 6 within the
range to #, 6 < $; hence, in order to fulfill the condition a > 0, /3 >
in
the integral expression for J#,

-H \/# 2

J
For

and \/Q 2

-if
Jo

Ji, 1

>

-H \/l

>
J 1=

d2

46>

#2

for

Then we have

eV, -e J^e)de
- i) +
V& - e Vi 2

(2d

necessary to substitute y/o 2

it is

for

=
e2

#; hence

(20 2

J,,

l) 2

VE^mde
4i0 2

#2

(vm-3-18)

Vi -

Transforming the integral function, we obtain

/i

where

J,
Jl
Jl

/i (0)

+ t/i<

f (ggj^lilfgVgHZZoWjg
- 1) + 160
- ^ )(1 - 6

2
J (20

(2o 2

(<>

i) 4

iw(e*

*)(i

(VHM-19)

o2)

(vm

3 20)

As changes from 1 to co it takes the value xk, where x is the root of the
equation F(kO) = 0, corresponding to a surface wave. Therefore the
integral function J x for some values of 6, increases without limit, and the
f

integral

J^ becomes indeterminate.

As indicated

in Art.

VIII-2, in

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

335

order to disclose this indeterminateness, only the principal part of the

Cauchy integral should be considered. Let us denote this part of the
integral

by

CJoo] then

VO -

# 2 Jo(pd) dd

J M is
we have

Pk

w =
where

l) 2

Vo -

40 2

e*"(/i

(VIII-3-22)

^ (J/"' + CJ)

/,

Thus the determination

(VIII-3-21)
1

if 2 )

Using the results obtained with

/,

(J,

$ 2 \/e 2

a real value.

It is clear that

reference to J,

(2d 2

!,

J,)

+I

of displacements

distance from the source of the waves,

Jo

on the

e.g.,

soil

surface at a small

for a small value of

is

p,

reduced to the computation of the foregoing integrals.

The easiest way to compute these integrals is by expanding them into a
series.
Omitting the rather involved intermediate operations for their
computation, let us use the equation obtained by O. Ya. Shekhter 41 for/i
and /2
For the Poisson ratio v = 0.5, we have
.

/1

= -0.0796

0.0598 P

0.00607 P 3

0.000243 P 5

U =

0.0571/o(l. 047)

0.0474

0.00647p 2

= -0.119

0.25,

0.0895 P

0.000264p 4

For the Poisson ratio

/1

0.00000517 P 7

0.00000517p 6

we have

0.0104p 3

0.000466p 5

U =

0.0998

0.0484

= -0.159

ratio v

0.1392

0,

0.00595p 2

0.000240 P

0.163J (1.1441p)

0.00000484p 6

+ 0.000937p - 0.0000246p +
- 0.00585p + 0.000228p
- 0.00000445p +
7

0.0185p 3

0.0512

Returning to Eq. (VIII-3-22),

we

parts; neglecting the latter,

we have

U =

For the Poisson

/i

(1.08777p)

0.0000109p 7

us separate the real and imaginary

obtain the following simple equation for
let

DYNAMICS OF BASES AND FOUNDATIONS

336

the vertical component of

soil

displacement at small distances from the

source of waves:

w = A<&(p)

A,=

where

tan 7

(VIII-3-23)

7)

(VIII-3-24)

= V/i

*(p)

sin (ut

+/

(VIII-3-25)

(VIII-3-26)

j-

Figure VIII-8 gives graphs of the function

The use

values of the Poisson ratio.

"^(p)

plotted for different

of these graphs in design

computa-

0.50

0.40

\
0.30

-Poisson ratio v = ;/

(\

(/>)

V"

-Poisson ratio v - 1/4

\^

0.20

- Poisson ratio v =

0.10

0.5

2.5

2.0

1.5

.0

3.0

P
Fig. VIII-8.

tions

makes

it

Graph

facilitating the use of

Eq. (VIII-3-25).

possible to avoid calculations of /1

and/2 from

the foregoing

formulas.
c.

Displacement of Soil

at

Large Distances from

order to find approximate expressions for

soil

In
Source of Waves.
displacements at large

the

distances from the source of waves, the principal part according to

should be singled out of integrals (VIII-3-14) and (VIII-3-15).

Cauchy

Integrat-

ing along the contour of the area of the complex variable gives us the
following principal values of q and Wq\

Cq

-^Hy

Zfl

ixP
Cwo =
KoJo(p)
2/x

P)

(2? - k^apDid-r)
F(Qf(Q
1:

7TJU

ik*P
2ttm

(2

Di(xr)

(VIII-3-27)

Do(Hr) d

2tt M

JDo(xr)

F(Qf(Q
ik

where

/c

F(Q
- J (xr)

Do(i;r)

(VIII-3-28)

J
I.k

= - Y (xr)
= -7i(xr) -

iJi(xr)

(VIII-3-29)
(VIII-3-30)

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

337

Yq and Yi are Bessel functions of the second type, of the zero and first
are determined by Eqs. (VIII-2-20);
and
all other symbols have already been defined.
Equations (VIII-3-27) and (VIII-3-28) include standing waves. In
order to obtain a system of transient waves, it is necessary to impose on
Eqs. (VIII-3-27) and (VIII-3-28) the solutions with reference to free
surface waves; these may be taken as follows:

orders; the coefficients

<Zo*

wj =

#0/1 (*r)

Adding these solutions to the preceding

temporary factor e iwt we obtain

K<Jo(xr)

and taking

ones,

(VIII-3-31)

into account a

w = '

Pik 2

fJm D

;2

Jh

7TM

W-^f^

Ic^apD-ftiAofat

dj

(VIII-3-32)

(VIII-3-33)

ik 2

f (2P

2^ Jh-

W)
mm
fe)

(fr)e*"

Since the above expressions consist only of transient waves, they give a
solution of the problem for the case of the application of a normal periodic
force
If

Pe

io3t

at the origin of coordinates.

the distance from the source of waves

large, expressions (VIII-3-32)

is

such that the value xr

and (VIII-3-33) may be reduced

is

to the

form
go

= -

w =

HoDi(xr)^*

^KD
Q

(xr)e^

Sq

Sw

(VIII-3-34)
(VIII-3-35)

Here S q and Sw contain terms which are inversely proportional at least

and Qir) 2
Since Di(xr) and Do(xr) are inversely proportional
to xr
it is clear that, at sufficiently large distances from the source of
waves, the values of S q and Sw will be small and may be neglected.
Replacing Di(xr) and D (xr) by their asymptotic values, we obtain the
to (kr) 2

1/2

following final equations:

q

= -

Ho

J^ ^<---W4)

w = ^-Ko x f^- c<*-/

A\i
\irxr

(VIII-3-36)

(VIII-3-37)

DYNAMICS OF BASES AND FOUNDATIONS

338

It is seen

from Eqs. (VIII-3-36) and (VIII-3-37)

that, at sufficiently

large distances, the vibration amplitudes decrease at a rate inversely

proportional to the square root of distance from the source.

Neglecting the imaginary parts of Eqs. (VIII-3-36) and (VIII-3-37),

we have
go

Wo

=
= kP Ko
2m

J^

Lit

kt

J^-r cos (cot -

xr

sin

|J

|J

(VIII-3-38)

(VIII-3-39)

These expressions show that the orbit of motion of soil particles is an

which the ratio between the axes, as in the case of plane waves,

ellipse in

equals

Ko/H

By

equating the vibration amplitudes computed by Eqs. (VIII-3-23)

and (VIII-3-39), we obtain the distance to points at which the amplitudes
of vertical displacements of soil particles as computed by each of these
equations have the same value; as a result we obtain an equation for
k

k:

= Ko^F

(p) \/p

For the Poisson

ratio v

0.5,

*(p)

(VIII-3-40)

0.109; hence

Vp

8.7

10- 2

Using the graph of the function ^(p) for

0.5 (Fig. VIII-8),

we

obtain
P

2tt

therefore follows that for r

It
v

0.5),

= L

the amplitude of the vertical component of

>

/2ir

(for the

Poisson ratio

Eqs. (VIII-3-23) and (VIII-3-39) will give the same value for
soil

vibrations.

For distances

amplitudes of vibrations should be calculated from Eq. (VIII-3-39)

for smaller distances, from Eq. (VIII-3-23).
By the use of Eq. (VIII-3-40) it is also easy to determine limiting
distances for other values of the Poisson ratio.
The author performed experimental
d. Experimental Investigations
investigations of propagation of waves under various soil conditions.
Water-saturated sands were a subject of the most detailed investigations.
The results of these investigations permit some conclusions concerning the
validity limits of theoretically established relationships between soil

displacements and distances from the source of waves.

A test foundation was used as a source. The area of this foundation in

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

339

Vibrations were excited by a special vibrator

contact with soil was 1
with a controllable speed. Investigations were mostly performed in
cases in which the foundation (the source of waves) underwent vertical
Since the area of the foundation in contact with soil is small,
vibrations.
it may be considered as representing a concentrated source.
The foregoing theoretical investigations took as the source of waves a
vertical exciting force whose magnitude was determined by a certain
If the foundation area in contact with soil serves as a
assigned value P.
source of waves, then the value of P may be taken as follows:
2

P =
where

cu

A =
A =

coefficient

of

elastic

cu

AA

(VIII-3-41)

uniform compression of base under

foundation
foundation area in contact with

soil

amplitude of vertical vibrations of foundation

During the experiments under discussion, the value of c u was established
from the resonance curve of the vertical vibrations of the foundation;
3
Soil vibrations were measured by means of an
it equaled 5 kg/cm
oscillograph which made it possible to record the vertical and horizontal
components of vibrations and the orbits of motion of soil particles.
According to the theory of surface waves, soil particles located on the
soil surface move along vertical ellipses in which the ratio between the
For soils
vertical and horizontal axes is constant and equals Ko/Ho.
satisfying the Poisson hypothesis (v = 0.25), this ratio equals 1.47; for
a

incompressible

soils (v

0.5),

K /H =

1.85.

Figure VIII-9 shows the orbits of motion of

soil particles

recorded at

from the wave source and at vibromachine speeds

The vibration amplitude of the
equaling 800 (a) and 1,200 rpm (b)
foundation (source of waves) remained constant and equaled 0.30 to
Experimental orbits of motion of the soil particles differ con0.32 mm.
siderably from theoretical orbits, especially for relatively small distances
from the source. At small distances the orbits of motion of soil particles
may have a very complicated shape (a figure eight or a deformed ellipse).
With an increase in distance, the orbit approaches an ellipse; however,
different distances

the angle of

its

inclination with respect to the vertical axis does not equal

t/2, as should be the case according to the theory of surface waves.

It

appears that the considerable distortions of the orbit observed at

small distances from the source of waves are explained

by the influence

supplementary spectrum of waves propagated through the soil

simultaneously with the surface waves. It will be shown later that, at
large distances from the source of waves, the displacement of soil particles
is affected by the absorption of the wave energy by soil.
Consequently,
the phase angle between the horizontal and vertical components does not
of the

DYNAMICS OF BASES AND FOUNDATIONS

340

remain constant and equal to t/2, but changes from point to point; theredepends on the distance of
the point under consideration from the source.
In order to verify Eq. (VIII-3-23), amplitudes of vertical vibrations of
the soil surface were measured at small distances from the source of

fore the inclination of the axes of the ellipse

Distance,

m
T 2.55 t3.60t4.15

1.65

Distance,

t 4.80 t 5.75 t6.60 t7.50 t7.90

t8.70t9.70 tIUO t13.10

t 14.90 t 16.(

Fig. VIII-9.

Measured

orbits of

vibro machines operating at:

(a)

motion of surface particles at varying distances from

800 rpm; (b) 1,200 rpm.

Table VIII-3 gives the results of these measurements performed

The amplitude of vibrations of the
foundation (source) was equal to 0.30 mm.
waves.

at different vibromachine speeds.

Table VIII-3. Comparison of Theoretical and Experimental

Vibration-amplitude Values of Soil Surface
Amplitudes,

N,

min -1

sec

-1

r,

Ao,

cm

62.5
84.0
104.0
156.0
187.0
218.0

1.6
1.6
2.4
1.6
1.6
1.6

2.24
4.3
5.3
6.25
6.70
8.80

X
X
X
X
X
X

10-2
10-2
10-2
10-2
10-2
10-2

0.90
1.12
2.10
1.92
2.20
2.70

Design values

^(p)

Computed

600
810
1,000
1,500
1,800
2,100

mm

0.115
0.090
0.057
0.060
0.053
0.050

2.5
3.9
3.10
3.75
3.50
4.4

X
X
X
X
X
X

10-2
10-2

lO- 2
10-2

lO" 2
lO" 2

Measured

2.5
3.9
3.25
3.7
3.6
4.2

X
X
X
X
X
X

lO" 2
lO" 2
lO" 2
lO" 2
lO" 2
lO" 2

b,

P,

m/sec

kg/cm 2

150
120
120
130
135
130

374
245
245
287
312
287

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

341

When amplitudes of soil vibrations were computed from Eq. (VIII-3-23),

the velocity of transverse waves was taken to equal 120 to 150 m/sec 2
This value is fairly accurate, since it was confirmed by other experiments
.

site of the investigation.

The values of the modulus in shear
corresponding to these velocities of transverse waves were found to

on the

kg/cm 2

equal 245 to 374

The value of ty(p) was found from the graphs

mean between
0.5 and v = 0.25.

of Fig. VIII-8 for the arithmetical

8.0

7.0

E
E

6.0

5.0

?4.0

*3.0

^H N.
"^

(fl)

Styoot

coqs/cfe,"t'ng

nhc orbtion-

>,^s
Co
s6

^2.0

^rir^

1.0

uo^ Vrf,

Von

E
E

-^-~^
3

(&)

"

V-

7
8
9 10 11 12
Distance from the source,

1
13

14

15

-o
16

17

18

19

Experimental check of Eqs. (VIII-3-42) and (VIII-3-43) with

component A v of vibration amplitude on soil surface; (6)
measured variation of horizontal component A h of vibration amplitude. In both
Fig. VIII-10.

(a)

reference to the vertical

cases vibrations at the source are vertical.

Amplitudes of vertical soil vibrations computed from Eq. (VIII-3-23)

agreed with measured amplitudes. It follows that Eq. (VIII-3-23) may
be used with confidence for small distances from the source.
In accordance with Eqs. (VIII-3-38) and (VIII-3-39), the amplitudes

waves at relatively large distances from the source of waves

decrease at a rate which is inversely proportional to the square root of the
of surface

distance.

and

is

Hence,

if

is

the amplitude at a distance r from the source,

the amplitude at a distance r

then

(VIII-3-42)

Figure VIII-10 presents a graph of experimentally established values

of vertical amplitudes of vibrations at different distances from a foundation subjected to 810 vibrations per minute.
The experimental data are

DYNAMICS OF BASES AND FOUNDATIONS

342

little circles.
The same figure gives a graph (dashed line) of
changes in amplitude in accordance with Eq. (VIII-3-42); the initial
values of r and A r are taken from Table VIII-2.
Comparison of the
theoretical curve with the experimental data shows that only within the

shown by

range of small values of r (as compared with r ) do the amplitudes computed from Eq. (VIII-3-42) coincide with the experimental data. With

an increase in the ratio r/r the discrepancy between the two increases.
The measured amplitudes are always smaller than those computed from
Eq. (VIII-3-42).
Experiments conducted on different soils revealed a similar discrepancy
between computed and measured amplitudes of surface waves. Thus
the conclusion is possible that actual changes with distance of the vertical
component of soil vibrations (when the source of waves is also vertical)
are more intensive than might be concluded from the theory of surface
,

waves.

The discrepancy between theory and experimental data is explained

by the presence of factors which are not taken into account by the theory.
Therefore

may

it

only so far as

soil

be considered that the theoretical findings are correct

In reality, even

represents an absolutely elastic body.

such small deformations as occur in soils when elastic waves are propagated therein induce nonconservative processes accompanied by partial
absorption of the energy of elastic waves.
Therefore with an increase in
distance from the source of waves, the amplitudes drop off somewhat more
rapidly than in the inverse proportion to distance stated by the theory of
wave propagation in an absolutely elastic body.
Some assumptions should be made in order to take into account the
absorption of wave energy by soil and to introduce a correction into the
relationship between amplitude and distance.
These assumptions
concern the dependence of wave absorption on the properties of the
propagating waves and the distance from the source of waves. Hereafter

we

shall

assume

that,

when

a spherical

wave

is

propagated over the

soil

surface from a concentrated source, the absorption of the energy of this

wave

is

proportional to the

as well as to

its

amount

of energy entering a given soil layer

thickness; however, the absorption will not depend on the

radius of curvature of the wave.

Then an approximate formula for
amplitudes of spherical waves propagating over the soil surface may be

written as follows:

A = A
r

where

Ar A =
,

amplitudes of
coefficient of
-1

meters

soil

Jy e-o-^
vibrations at distances

wave energy

or centimeters

-1

(VIII-3-43)

r,

ro

absorption, having dimensions

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

343

If the soil were an absolutely elastic medium, no energy would be

absorbed; in this case a would equal zero. Hence the coefficient a is a soil
constant determining the deviation of its physical properties from the

properties of an absolutely elastic body.

The continuous line in Fig. VIII-10 shows a relationship between the

component of soil vibrations and distance as computed from

vertical

It is seen from this graph that

Eq. (VIII-3-43) for a = 0.100 m" 1
calculated and experimental amplitude values fully agree for all distances
from the source of waves. Several analogous graphs obtained as results
of experiments with other frequencies and other sources of waves are
.

given in Art. VIII-8.

These graphs

also

show good agreement between

the results of computations from Eq. (VIII-3-43) and the experimental

data, as do investigations carried out under different soil conditions.

Hence the conclusion is possible that the equations for the computation
vertical component of vibration amplitudes as a function of

of the

distance (in the case of a vertical source) are in good agreement with

experiments

if

the absorption of

wave energy by the

soil is

taken into

account.

Energy absorption by the

soil is

an irreversible process, similar to the

formation of loops during elastic hysteresis. Therefore it may be

assumed that these two processes are caused by similar factors inducing
deviations in the behavior of

soil (in

cases of small deformations) from the

behavior of an absolutely elastic body.

Very

little is

known about

nature of these factors, not only for such complex material as

such materials as concrete, wood, and steel.

soil,

the

but for

There are two main points of view in regard to these phenomena.

According to the first, observed deviations in the behavior of materials are
explained by the existence of forces of internal friction or resistance,

which are dissipative forces; i.e., they depend not on the magnitude of
deformation, but on the rate thereof.
The influence of these forces on
free and forced vibrations is fairly well established in systems with one
degree of freedom, such as the vertical vibrations of massive foundations

under machines.

It

is

usually assumed that dissipative forces

directly proportional to the velocities of deformation.

are

For continuous

systems it may also be considered that in addition to stresses

depending on deformation, there appear stresses depending on the rate of
deformation at a given moment. In the simplest case (for low velocities)
it may be considered that there is a linear relationship between dissipative
stresses and the rates of deformation.
Elastic systems which permit the
existence of stresses depending on the magnitude and rate of deformation
elastic

are called viscoelastic systems.

We

shall

not deal here in detail with the theory of wave propagation

body, but shall only indicate the principal conclusions of

in a viscoelastic

DYNAMICS OF BASES AND FOUNDATIONS

344

this theory concerning the influence of the viscosity of the

and the damping

velocity of propagation

medium on

of three-dimensional

the

and surface

waves.

Let us use the following symbols:

a
v
a, b

= damping constant of wave which depends on distance

= velocity of propagation of wave in viscous medium
= velocities of propagation of longitudinal, transverse waves
medium
medium

in perfectly elastic
7]'

rj/p
t]

p
co

=
=
=
=

kinematic viscosity of

coefficient of viscosity

density of

medium

frequency of wave

Assuming that the medium

is

incompressible and

is

characterized

by a

assuming that r]/a 2 T and T]/b 2 T are small, we obtain the

following expressions which establish the dependence of v and a on the

low viscosity,

i.e.,

viscosity of the

medium:

For longitudinal waves:

4t?V
3 3
For transverse waves:

ab

-b>

For surface waves:

OLb

low viscosity, surface waves are damped in the same

way as transverse three-dimensional waves; the ratio between the
constant a a of a longitudinal wave and the constant a& of a transverse
Hence, in

soil of

wave equals
6

In
is

soils,

usually 3a 3 /4

smaller than the

>

b3

damping

3a 3

therefore the

damping

of longitudinal

waves

of transverse or surface waves.

It follows from the expressions for damping constants that in all three
types of waves they increase proportionally to the square of the frequency
Hence, according to the theory of wave propagaof propagating waves.
tion in soils characterized by viscosity, waves produced by high-frequency

machines are damped in the soil much more rapidly than waves created
by low-frequency machines.
According to the second point of view, the damping of waves in soil is
due to the influence of elastic recovery. The application of the theory

PROPAGATION OF

of elastic recovery to the question of

by B. Deryagin, 12

considered

'

13

principal conclusions arrived at

of seismic

waves

WAVES

ELASTIC

IN SOIL

damping

of

345

waves

in soil

was

V. G. Gogoladze, 19 and others.

by B. Deryagin

in regard to

The
the damping

in soils follow.

For three-dimensional waves of not too small periods,

where

Hence

Ci a

Ub

62

r- a*TG
B
2bTG
7T

==

constant of elastic recovery

==

modulus

it

follows that

of rigidity

OLb

If

1.86,

then a a

-t

.23o*.

from the theory of elastic recovery that the damping

Thus it
of longitudinal three-dimensional waves occurs much more slowly than the
follows also

damping of transverse waves.

The theory of wave propagation
conclusion that damping constants

body leads to the

waves are proportional to the

in a viscoelastic

of

However, the theory of wave propagation

medium characterized by elastic recovery leads to the conclusion
that the damping of waves is proportional to the first power of the frequency. This means that the rate of damping for one wavelength is

squares of their frequencies.

in a

constant.

For surface waves, the damping constant determined by B. Deryagin

on the basis of the theory of elastic recovery does not differ much from
In this regard, conclusions of the elastic-recovery and viscoelastic
ah.
theories agree.
It is seen that in the question of damping of waves in soil
the theories differ only in regard to the relationship between the damping
constant and the period of propagating waves.
How far either theory

may be shown only by experimental investigations

propagation in soil of waves with different periods.
The author experimentally investigated under winter conditions
the propagation of waves with periods of vibrations of 0.111 sec
corresponds to reality
of the

(N = 540 min- ), 0.083 sec (N = 720 min- ), and 0.067 sec (N = 900
min -1 ) in water-saturated brown clays with some silt and sand. He
1

found 4 that a decreases at a rate which is inversely proportional to the

square of the period this rinding agrees with the conclusions of the theory
for a viscoelastic medium.
However, it should be noted that theoretical
curves computed from Eq. (VIII-3-43) differ considerably from experi;

DYNAMICS OF BASES AND FOUNDATIONS

346

mental values of soil vibration amplitudes. Therefore the values of a

obtained for waves with the periods indicated above are approximate,
and the dependence of a on frequency is confirmed only with a high degree
of approximation, especially as the investigations were conducted for three

wave periods

only.

For the determination of the relationship between damping constant

and frequency, one can use data obtained from investigations of vibrationamplitude changes on the soil surface as a function of distance. These
data were obtained from experimental studies of screening of waves by
line and closed sheet-pile rows (see Art. VIII-8).
Data on amplitudes of
the vertical component of soil vibrations were obtained during vertical
vibrations of the foundation (source of waves) before the sheetpiling was
installed.

From measured amplitudes of soil vibrations, values of a were computed for different numbers of oscillations per minute, and then curves of
vertical soil vibration amplitudes were plotted.
Comparison of the
theoretical and experimental curves of soil vibration amplitudes shows
satisfactory agreement.
Table

VIII-4. Soil

Damping Constants Computed from Two Tests

Foundation

Oscillations,

min

Damping
l

constant,

m _1

600
800
1,200
1,500
1,800

0.100
0.100
0.100
0.100
0.090

600
800
1,000
1,200
1,620

0.040
0.050
0.050
0.040
0.040

Table VIII-4 gives approximate computed values of a damping conwave propagation from two test foundations. This table
shows that in both cases, i.e., for waves emanating from foundation 7
(with the foundation area in contact with soil equaling 1 m 2 ) and for
2
waves propagating from foundation 8 (contact area equaling 4
), the
waves.
propagated
of
period
the
constant
not
on
the
damping
does
depend
Hence, at least for the case of wave propagation in water-saturated finegrained sands, the conclusions of the theory of wave propagation in a
medium with elastic recovery agree more closely with the experimental
stant for

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

347

data than the conclusions of the theory of wave propagation in a viscoelastic

medium.

The hypothesis that

on the
seem to be supported by experimental
It is possible that the assumption concerning the dependence of
data.
the forces of internal resistance on deformation, f when extended to
continuous media (which include soil), gives more satisfactory results in
regard to damping of vibrations than the theory of a viscoelastic body or a
body characterized by so-called recovery.
Values of the constant a in waves propagated from foundation 7 were
about two times larger than in waves propagated from foundation 8.
Apparently this can be explained by the fact that for foundation 7 the
measurements of wave propagation were made along a ridge of yellow
sand, and for foundation 8 along gray medium sands containing organic
dissipative forces are linearly dependent

velocity of deformation does not

silt.

These experimental investigations of wave propagation from foundamade partly after the bulk of the soil had already thawed;
From these
others were made at a time when the soil was still frozen.
_1
experiments an average value of a equaling 0.058
was obtained for
frozen soils; this is about one-half the value for a thawed soil.
Hence,
other conditions being equal, waves from machine foundations propagate
to larger distances in winter than in summer.
Table VIII-5 gives approximate values of the coefficient of absorption
of wave energy (damping constant) obtained by the author as a result of
investigations of wave propagation in different soils.

tion 7 were

Table VIII-5
Coefficient of

No.

2
3

Soil

absorption,

Yellow water-saturated fine-grained sand

Yellow water-saturated fine-grained sand in a frozen state
Gray water-saturated sand with laminae of peat and organic

5
6
7

Clayey sands with laminae of more clayey sands and of clays

with some sand and silt, above ground-water level
Heavy water-saturated brown clays with some sand and silt

Marly chalk
Loess and loessial

0.100
0.060
0.040

silt

m -1

soil

0.040
0.040-0.120
0.100
0.100

Figure VIII-106 shows a graph experimentally obtained of amplitude changes in the horizontal longitudinal components of vibrations.
It is seen from this graph that, while the amplitude of the vertical comf

This assumption was advanced by

I.

L. Korchinskiy. 25

DYNAMICS OF BASES AND FOUNDATIONS

348

ponent (Fig. VHI-lOa) changes with distance more or less monotonously,

no regularity is observed in changes of the amplitude of the horizontal
longitudinal component.
Graphs of relationship between the amplitude
of the horizontal longitudinal

VIII-8,

show

this

same lack

component and

distance, given in Art.

of regularity.

Thus experiments demonstrate that the application

of the theory of

surface waves to the computation of amplitudes of the horizontal longi-

tudinal component of vibrations

The

is

not possible in the case of a vertical

above (i.e., the supplementary

spectrum and the absorption of wave energy by soil) greatly affect the
distribution of amplitudes of the horizontal component of vibrations;

source of waves.

factors indicated

4.0
3.0

o\

2.0

(a)

CI

2.0

(&)

r,
gl.U
a

9
7
8
6
5
4
3
Distance from the foundation, m

10

11

12

Fig. VIII-11. Vibrations at the source are induced by a rocking motion:

with distance of horizontal component Ah of vibrations at soil surface;
for the vertical

component

(a)

variation

(6)

the same

v.

therefore the results of computations conducted without taking into

account these factors cannot be applied to the investigation of wave
propagation in soil.
From the principle of reciprocity, the assumption is possible that an
analogous conclusion will be justified for the vertical component of soil
vibrations if the source of waves consists of a foundation undergoing
horizontal or close to horizontal vibrations.
This is confirmed by
experiments.
Figure VHI-lla shows an experimental graph of changes

component of vibrations
shows an experimental graph of changes of amplitudes
the vertical vibration component when the source of waves was the

in the amplitudes of the horizontal longitudinal

Ah
of

Fig. VIII-116

same foundation undergoing rocking vibrations with a frequency of 30

-1
sec
These graphs show that changes in the horizontal longitudinal
component follow Eq. (VIII-3-43); the amplitudes of the vertical component change with distance in the same complicated way as the horizontal longitudinal component when the foundation (source of waves)
.

undergoes vertical vibrations.

PROPAGATION OF
VIII-4.

Influence of the

WAVES

ELASTIC

IN SOIL

349

Area of the Source of Waves on the Amplitude

of Vibrations of the Soil Surface

Let us assume that the distance r of the point under consideration from
is large by comparison with the lengths of propagating waves.

the source

In this case the influence of the supplementary spectrum may be negand it may be considered that displacements on the soil surface
Let us also assume that the exciting
are created by surface waves only.
force producing waves in soil is uniformly distributed along a limited
lected,

Assuming that the

surface having the shape of a circle with radius R.

p on the

exciting force imposes pressure

and (VIII-3-35), replace the value

we may,

soil,

in Eqs. (VIII-3-34)

of the concentrated force

by a

force

acting on an element of the area of the circle and equaling

ps ds

dip

According to Eqs. (VIII-3-34) and (VIII-3-35), displacements on the

surface at a distance r from the point of application of the con-

soil

centrated force

equal

= AD (xr)ps ds d<p
= BDi(xr)ps ds d(p

wo
qo

A =

where

-=r-

B = All other

Ke

iut

~ Foe-'

symbols have been previously defined.

dA=sdsd(p

Fig. VIII-12.

Diagram

It follows directly

from

illustrating Eqs. (VIII-4-1)

Fig. VIII-12 that

r

The

horizontal

cos

component

and (VIII-4-2).

<p

=
=

Vx +
2

r cos

s2

2xs cos

<p

(VIII-4-2)

\p

of soil displacement along the x axis equals

Uo

qo cos

\p

Displacements produced at the point

located at a distance x from the
center of a circle whose whole area is subjected to the action of an evenly
distributed exciting force of intensity p are as follows:

w = L
o

Uo

L ApDo(xr)s

(J( R
=
]

ds d<p

(VIII-4-3)

BpDi(xr) cos

\f/s

ds

d<p

DYNAMICS OF BASES AND FOUNDATIONS

350

From

the theory of Bessel's functions,

>

0,

>

under the condition that

we have

Jo(xr)

Jq(xs)Jq(xx)

Jo(xs)Yo(xx)

+2 V

J n (xs)J n (xx)

COS

Tlif

(VIII-4-4)
Q (xr)

J n (xs)Yn (xx)

cos n<p

=l

In addition,

-n(o =

Ji(--Ji(a

-Fi()
and taking into account

Differentiating Eqs. (VIII-4-4) with respect to x

the fact that

dr

cos

dx

<p

we obtain
00

Ji(xr)x
,

cos

= xJ o{xs)J i{xs)

<p

r /

COS

^M
J
=2

Or, using Eqs. (VIII-4-2),

\p

(xs)cosn(p

(xs) cos n<p

=l

we have

= J

(xs)Ji(xx)

+ V
n

Yi(xr) cos

=l

= xJ o{xs) v
Y i{xx)\,oV
+2 >

<p

Ji(xr) cos

v
Yi(xr)x

dJJxx)

+2o 'V>

^ = Jo(^s)Fi(%x)

C cos up

=l

+ V
n

Cn

(1)

cos

n^

=l

where C n and C n (1) are the expressions under the summation signs in the
two penultimate equations.
Substituting these expressions into Eqs. (VIII-3-29) and (VIII-3-30),
which give expressions for D (xr) and Di(xr), and integrating with respect
to <p, we have
Wo
uq

= 2irAp
= 2-wBp

f R
/

J o(xs)[Y o(xx)

iJo(xx)]s ds

iJ\(xx)]s ds

After integrating with respect to

(xs)[Yi(xx)

s,

we obtain
xRJi(xR)

Wo

2>irApDo(xx)

Uo

2irBpDi(xx)

xz

xRJi(xR)

(VIII-4-5)

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

351

It follows from a comparison of Eq. (VIII-4-5) with Eqs. (VIII-3-38)

and (VIII-3-39) that, if a source of waves is distributed over a circular
area with radius R, changes in amplitude of surface waves are related to
changes in distance from the source with the same regularity as was
observed in the case of a concentrated exciting force. In addition, however, for the same value of the intensity p of the exciting force, the
amplitude will be greatly influenced by the dimensions of the area of the
The size of this area influences the amplitudes of both vertical
source.
vibration components.
horizontal
and
consider
two cases:
Let us
1. The source of waves is the foundation under a low-frequency engine;
consequently X is small, and the wavelength L c is large. If one assumes

that the dimensions of the foundation area in contact with

soil

are small

comparison with the wavelength, then xR will also be small. By

expanding the function Ji into a series with respect to the argument for
this boundary condition, we obtain approximately

in

Ji(*m =
Then

where A'

Thus

is

= JPD

( K X)A'

the area of the source of waves in contact with

soil.

under low-frequency engines (where the value of

xR is small) the amplitudes of soil vibrations grow in proportion to the
area of the source of waves in contact with soil.
2. The source of waves is a foundation under a high-frequency engine
(turbogenerator, forge hammer, and others) for which the product xR is
relatively large.
For these values of the independent variable, the function J i may be replaced by its asymptotic value; i.e.,
for foundations

Ji(xR)

V2R

sin

xR

Therefore, for this case,

Wo

^^ VW

sin

Lr

|)
(VIII-4-7)

Uq

Hence the amplitudes

of surface

waves radiated from high-frequency

sources will undergo periodic changes

when the

radius

continuously

DYNAMICS OF BASES AND FOUNDATIONS

352

varies.

Maximum

magnitudes
in

i.e.,

or

xR ~ =
4

when

when

L ~

Amplitudes at

all

of amplitudes will be attained

fxR

~i)=

n =

~~2~ T
2w + 3

when

1, 2, 3,

points on the soil surface equal zero

if

sin

kR j J =

0,

i.e., if

R =
L ~

27i

If

a source of waves, for example, a vibrator placed on the

as a solid

body with one degree

soil,

vibrates

then theoretically the resonance curve of forced vertical vibrations has one maximum. However, if

10

15

of freedom,

20
Number of

30

25

35

40

vibrations /sec

Fig. VIII-13. Resonance peaks established

by Koehler. 24

one simultaneously measures the resonance curves of the source of waves

and the soil surface, the curve obtained will have several maximums in
addition to the principal maximum corresponding to the maximum on the
resonance curve of forced vibrations of the source of waves. Some
investigators explain these additional

maximums by

coincidences of the

frequency of natural vibrations of the soil layer and the frequency of

propagating waves.
Figure VIII-13 shows a resonance curve plotted by R. Koehler, 24 who
suggests that the presence of several maximums and minimums can be
explained by the phenomena of interference and resonance (with the
exception of a large maximum lying between 21 and 31 hertz and produced

by natural vibrations

of the engine).

It is

seen from curve

that the

PROPAGATION OF

maximum

ELASTIC

WAVES

IN SOIL

353

component of vibrations corresponds to

Resonance curves recorded at distances of 60 to 90 m
from the source also demonstrated the presence of this maximum. On
the basis of this observation Koehler came to the conclusion that the
first

for the vertical

11.2 to 11.5 hertz.

recorded frequency of vibrations corresponding to the first maximum of

the resonance curve represents the frequency of natural vibrations of a

layer of carbonaceous gray clay having a thickness of 5.8

and located

Koehler does not explain the natures of the

other maximums on the resonance curve.
Let us assume that a resonance curve is recorded in the soil by a device
installed at a constant distance r from the source of waves, and that the
When the frequency co of the
radius of the source remains constant.
at a depth of 4.8 to 10.6 m.

source changes, x will also change, since the latter

is

proportional to

co:

CO

where

c is

the velocity of the propagating waves (independent of the

frequency).

We assume that the amplitude of vibrations of the source is constant;

consequently p is constant. Thus the only variable will be co or x.
Let us consider the influence of changes in the frequency of the source
on the amplitudes of vibrations (for example, on the amplitude of the
vertical

component

w = If

Using Eq. (VIII-4-5), we have

of the soil vibrations).

-K pRJi(xR) VJo

the distance from the source of waves

YJ(xx)

(*x)

is

+ Y

(xx)

large, then,

approximately,

2
J<?{xx)
TTXX

and consequently,

where

Wo

J
a

-^B-

R
ix

It follows

(VIII-4-8)

\ x

from Eq. (VIII-4-8) that when

soil

vibrations are excited

by

a source with a constant intensity p and varying frequency, the amplitudes of soil vibrations change with the frequency of vibrations.
It is

noteworthy that the amplitude of soil vibrations does not change in a

monotonous way, but has several maximums and minimums analogous
to those occurring in the experimental curves presented in Fig. VIII-13.

Of the greatest practical importance

is

the computation of the vertical

DYNAMICS OF BASES AND FOUNDATIONS

354

component

produced by a vertical source. Assuming

is large enough so that we may
asymptotic value, we obtain the following expres-

of soil vibrations

that in Eq. (VIII-4-5) the value of xx

replace

by

(xx)

its

sion for the amplitude of vertical soil vibrations:

l^^J^R)

Aw =

(VIII-4-9)

\TXX
\7

\X

Values of the function Ji(xR) are taken from the tables.

Computation of amplitudes of the vertical component of
produced by shocks to a foundation under a forge hammer

Example.

1.

Data.

The foundation under the hammer

tions apply to

it

and the

is

soil

oscillations

a block; the following specifica-

soil:

Foundation contact area:

A =

65.6

Coefficient of elastic uniform soil compression:

cu

Amplitude

10 3 tons/m 3

of vertical vibrations of the foundation:

A =
z

Frequency

6.0

0.85

10~ 3

of natural vertical vibrations of the foundation:

co

70 sec -1

Velocity of propagation of transverse waves:

b

Modulus

G =
2.

Computations.

V!

10 3 tons/m 2
circle:

Vt!=-

find coefficient x:
<

xR =

Hence,

We

17

Let us determine the equivalent radius of the

* =

We

100 m/sec

of rigidity of soil:

find

from the

b
X

0.7

I_
100

4.46

o 7

3.13

tables,

JiixR)

The dynamic pressure on the

p

Cu A =
t

soil will

10

0.30

be

We take the value of the coefficient

Poisson ratios of 0.5 and 0.25:
Ao =

0.85

5.1

tons/m 2

as the arithmetic

0.108+0.183
s

10~ 3

,_
= ni
0.145

mean

of its values for

PROPAGATION OF
Substituting

all

3.14

5.10
1.7

355

0.145
0.30

y3.14 X 0.7x

<\

by soil is not great at

Therefore for small values of
R, the influence of energy absorption may be neglected. For example, assuming
6 m, we obtain an amplitude of the vertical component of soil vibrations of

As indicated

in Art. VIII-3, the influence of energy absorption

relatively small distances

>

10- 3

1.79

X 4.46 X
X 10

IN SOIL

we obtain

the above values into Eq. (VIII-4-9),

x
x

WAVES

ELASTIC

from the source

of waves.

0.9

0.7

0.6
1

0.5
o

0.4

o
fl

0.3

oOV
<JO

0.2

*"

SoC

oo8

SN ^
o
o

10

15

"^***

0.1

20

o ffBf^ 8^~~TJ
30

25

35

Distance from the axis of foundation,

Fig. VIII-14. Comparison of measured

and computed

40

o
45

'j

8
8
' vj

50

55

vertical

components of vibra-

tion amplitudes.

mm.

A we can determine from Eq.

For the given soil conditions, the
value of the coefficient of energy absorption by soil may be taken as 0.04 m -1
Then
for the amplitudes at distances r > 6 m from the foundation axis, we have
0.73

Taking

amplitude

this as the initial

(VIII-3-42) the amplitudes at distances r

>

Ar =

7Q 6-0.040-6)
^
1

Figure VIII-14 presents a graph of amplitude changes with distance, computed from
the above equation.
The same figure shows measured amplitudes at different distances from the hammer foundation.
If one does not consider the interference maxi-

mums and minimums

in the experimentally obtained distribution of amplitudes, then

a comparison of the graph of computed values with the measured values of amplitudes

shows good agreement.

DYNAMICS OF BASES AND FOUNDATIONS

356

VIII-5.

Dependence of
Source of

Soil Vibration

Amplitudes on the Depth of the

Waves

Foundations under engines, as well as other industrial sources of

seismic waves, are usually located below the soil surface.

Therefore

interesting to investigate the influence of the depth of the source

amplitudes of

it is

on the

soil vibrations.

Let us consider as a source of waves a vertical exciting force acting at a

depth / below the soil surface. We shall place the origin of coordinates

downward

direction.

Assuming

that vibrations proceed steadily at a frequency w,

we seek

solutions of

on the

surface, directing the z axis in the

Eqs. (VIII-3-3) and (VIII-3-7) for the functions

form:

P f"
$ = - ?r-n / <r* cosh as/
XM

<E>

and

^ in the

ftr)fd{

Jo

2ttm/c

rvin-5-n

P
2

following

e-M sinh
sii
PzJ*(&r)Z d
f* e-v
P
Jo

proved by means of direct substitution that these solutions

< z < /.
and (VIII-3-7), but only f or
According to Eq. (VIII-3-8), the components of soil surface displacements (the radial q Q and the vertical wo) corresponding to Eqs. (VIII-5-1)
It is easily

satisfy Eqs. (VIII-3-3)

are as follows:

* = 2S? /."

{e

~ a!

~ **>**<

(VIII _ 5 . 2)

W =
It is

assumed that the

soil

surface

is

free of stresses; therefore the

bound-

ary conditions are:

z

r r the shear stress,

,

and

^ by

rr

a,

and a z the normal

,

stress, are

(VIII-5-2a)

expressed through

3>

the following equations:

a2

\6

2M

^=
dz

/d 2V
\dr*

dr

d 2 $\
dz 2

+-(S+S+*S)

into the
which correspond to z =
when
that,
find
we
(VIII-5-4),
Eqs. (VIII-5-3) and

Substituting the values of

right-hand parts of

<V,,I 5 '4,

<f>

and

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

357

Eq. (VIII-5-3) satisfies the conditions of Eqs. (VIII-5-2a), the normal

do not equal zero, but are equal to

stresses

[(/c

2rf /;

2 2 )e-'

(VIII-5-5)

2per]Jo(Zr)idl

In order to obtain a solution for a surface free of normal

stresses, it is

necessary to superimpose on solutions (VIII-5-2) a solution obtained for

a surface under the action of stresses of magnitudes equal to those of

Eq. (VIII-5-5), but of opposite sign. For this purpose, we take, in

Eq. (VIII-3-13),

p = pL [(fcj _
Integrating from

components

to

2?)e- af

we obtain the

<,

2?erV]

(VIII-5-6)

dt;

following expressions for the

of displacements of the soil surface

produced by

this force:

p
f- e(2e

Wq

Pv

-k*~

2ff)[(/c

"-a[(fc
2ttm

90

2?e-f]J 1 (l;r)

2ee-e f ]J

(Hr)

rfj

(vni . 5 . 7)

total displacements of soil at the soil surface equal

*>

Wo

2?)e-'

F(0

q*

Jo

Wo

2?)ert

Thus the

-f

w% =

(VIII ~ 5 ~ 8)

^)

(VIII-5-9)

Pv^-

7r
Ztt/jl

/;

For/ = 0, Eqs. (VIII-5-8) and (VIII-5-9) coincide with Eqs. (VIII-3-14)

and (VIII-3-15).
The integrals in the right-hand parts of Eqs. (VIII-5-8) and (VIII-5-9)
and the integrals of Eqs. (VIII-3-14) and (VIII-3-15) are to a certain
degree indeterminate; in order to single out their principal values,

it is

necessary to integrate in the area of the complex variable along the

determinate contour.

If,

free surface waves,

after determining the principal values of the

and (VIII-5-9), we superimpose on this result

then taking into account a temporary factor, we will

integrals in Eqs. (VIII-5-8)

DYNAMICS OF BASES AND FOUNDATIONS

358

obtain, for sufficiently large values of

placement components

Qo

Wo

where the

coefficients

K, =

dis-

soil surface:

H Di{xr)e wt
f

(VIII-5-10)

K f Do(xr)e

iat

i/ are as follows:

--r

the following formulas for the

r,

the

= - 2~

#/ and

Hf = -

w on

and

(2* 2

(2* 2

/c

/c

-^ -

)^' -

2a 1 /3 1 e-^ /

(VIII-5-11)

2* 2 e"^

(VIII-5-12)

If / = 0, then
[see Eq. (VIII-2-20)], and Eqs.
f = #o,
f =
(VIII-5-10) identically coincide with Eqs. (VIII-3-34) and (VIII-3-35) for

the displacement components

when

the exciting force

is

located on the

soil surface.

The author conducted special investigations to verify Eqs. (VIII-5-10).

The soils on the site of the investigation consisted of water-saturated
gray sands. Three foundations were built for these experiments; all
had the same weight (W = 6.8 tons) and the same area in contact with

The base area of

), but they were placed at different depths.
foundation 5 was placed 2.0
below the soil surface, that of foundation
below the soil surface, and that of foundation 7 was on the
6 was 1.0
The soil vibrations were measured on the surface of the sand
soil surface.
soil (1.0

layer; the overlying peat layer

During

all

these

was removed.

investigations,

the foundations which served as

sources of waves were subjected to vertical vibrations only, with the

vibrator running at about 950 rpm.
vertical vibrations of the foundations

The average amplitude

was about 0.42

mm;

soil

of forced

vibration

amplitudes recorded for other amplitudes of vibrations were reduced to

this magnitude.
Soil vibrations were measured by a vibrometer in a
vertical longitudinal plane in almost all cases only the vertical component
of vibrations was recorded.
The investigations were conducted on
;

frozen

soil.

Small circles plotted in Fig. VIII-15 represent experimental values of

reduced double amplitudes of soil vibrations at varying distances from
foundation 7; the dashed line is a computed curve plotted on the basis of
the assumption that the wave is cylindrical and is characterized by
_1
damping. For a damping constant of 0.045
the computed curve is
Measin fairly good agreement with experimental values of amplitudes.
urements were taken twice at each point: while moving the device away
from and toward the foundation.

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

359

some 7.5
length was cleared of the peat
In the remainder the peat was removed immediately
before measurement; therefore the sand was much less frozen here.
The
section of the profile of

layer in autumn.

boundary between the two sections of the profile lies between points 11
and 12. In the first section (points 4 to 11) there is some regularity in
the distribution of amplitudes; this is disturbed at point 12.
There is a
certain irregularity in the distribution of amplitudes of the second section.
11.0
o
10.0

\
o

9.0

8.0

HL

17.0

\
N3

6.0
5.0

^ S?

,-4.0

^&>

3.0

^^
-

2.0

o
o

1.0

i *
!

-7 8 9 10 11 12 .13
Distance from foundation 7,
10

13

12

11

15

14

14

15

16

17

18

19

20

m
17

16

18

19

Nos. of points
Longitudinal

as

profile

^^^^^^^^m^^^^mmmmmmm?.

Fig. VIII-15. Comparison of measured and

tion amplitudes for foundation 7 (1.0

computed

area on

vertical

components

of vibra-

soil surface).

Values of the horizontal component of soil vibrations are shown in

by black circles. It is noteworthy that between points 4

Fig. VIII-15

and

11,

where the

horizontal

soil

component

was uncovered

in the

autumn, the amplitudes

of the

of vibrations decrease fairly smoothly, but in the

unfrozen section the changes in amplitude have a different character:

with a decrease in distance from the source of waves, the amplitudes
increase

up

to point 17 (at a distance of 14.9

then, at points 18

and

19,

the amplitudes decrease.

vertical vibrations of the foundation

on a

soil

from the foundation);

It follows that

during

which was frozen to a

certain depth, one wave is propagated through the soil.

Its vertical
component is much larger than its horizontal component. When this
wave reaches the section of unfrozen soil, there appear supplementary

DYNAMICS OF BASES AND FOUNDATIONS

360

waves which

interfere with the principal

tions of amplitudes of both the vertical

components

wave and
and

distort the distribu-

(especially) the horizontal

of soil vibrations.

Figure VI 1 1- 16 presents a distribution curve of vertical vibration

7.0
1

o \

6.0

o
5.0

K<

* 3.0

O '-<x.
2.0

-au

"**9r

1.0

7
10
11
6
8
9
Distance from foundation 6,
i

12

15

14

17

16

13

12

10
11
12
Nos. of points

Fig. VIII-16. Comparison of measured and

tion amplitudes for foundation 6 (1.0

computed
area 1.0

components of vibrabelow the soil surface, sides

vertical

backfilled).
9.0

8.0

I
7.0

E6.0

in 4.0

^3.0

rfs.

J'Nq

^o^
^fc--

2.0

--o~. *-fi. ^o

1.0

"o"
5
6
7
8
9
10
11
Distance from foundation 6, m

12

10

11

12

13

12

14

15

16

13

Nos. of points

Fig. VIII-17. Comparison of measured and computed vertical components of vibration

amplitudes for foundation 6 (1.0

area 1.0

m below the soil surface,

not backfilled).

amplitudes of soil during the vertical vibration of foundation 6, whose

base area in contact with soil lies at a depth of 1.0
and whose sides are
backfilled with soil.
Figure VIII-17 presents an analogous curve for the
same foundation, but for the case in which its sides were exposed along

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

361

the ground water was not pumped out of the excavaThe dashed curves indicate amplitudes obtained by computations
taking into account a damping constant of 0.050 to 0.060 m -1
These
their whole length

tion.

curves are in good agreement with the experimental data; in both cases
the interference of waves is slightly noticeable.
The peat was removed

from the soil surface

about 1 m.

in the fall; therefore the soil froze to a

depth of

Figure VIII-18 presents analogous experimental data on the distribuwaves from foundation 5, placed at a depth of 2
and backfilled.

tion of

was found to equal 0.064 m-1

Figure
VIII-19 gives experimental data on the distribution of vertical vibration

The damping constant

of soil

8.0
\

7.0

E 6.0
E

\
fi

5.0

^4.0
X

<3.0

V
y*

"-8
"-n ~y

2.0

^-s

O-

1.0

o
C)

12

>

7
5
6
8
9
10
Distance from foundation 5,
i

Fig. VIII-18. Comparison of measured and

tion amplitudes for foundation 5 (1.0

10
11
.12
Nos. of points

12

11

13

computed

14

13

16

15

m
i

15

vertical

area at a depth of 2.0

components of vibram and backfilled).

amplitudes of soil for the same foundation, but not backfilled and with
water filling the excavation; in this case, the damping constant of soil was
_1
found to equal 0.060
Figure VII-20 gives the results of measurements when the water was pumped out of the excavation and its level
was kept constantly some 20 cm higher than the foundation area in contact with soil.
In this case the damping constant was found to equal

0.121

m_1

i.e.,

it

was approximately two times

larger than in all the

foregoing cases.

The

profile

was measured

along which the propagation of waves from foundation 5

is identical to that of foundation 6.
It was cleared of the

peat layer early in the

when measurements were

was frozen to a depth of some 1 m.
exclude the influence of damping of vibrations on the propafall

therefore at the time

started along this profile the soil

If

one

is

to

DYNAMICS OF BASES AND FOUNDATIONS

362

gation of waves from nonbackfilled foundation 5 when the water was

pumped out of the excavation, then the average value of the damping

-1
constant for the other five measurements equals 0.058
In order to find out how the amplitudes of soil vibrations are affected
by the depth of the foundation area in contact with soil, a comparison
should be made of amplitude values obtained at points located at the
.

E4
E
<*>

1 *^ -a..

o_

C-,

<1

-0--

2.

CL__

5
6
7
8
9
10
Distance from foundation 5,
9

10
11
12
Nos. of points

-o-

12

11

14

13

m
14

13

16

15

Fig. VIII-19. Comparison of measured and

tion amplitudes for foundation 5 (1.0

with water

filling

computed vertical components of vibraarea at a depth of 2.0 m, not backfilled and

excavation).

\
\ a
St

X^

^T

-8^

=tfi_

"Ty- -8.

~^8- .

5
8
9
6
7
Distance from foundation
i

11
10
12
Nos. of points

10
5,

& 814

13

12

11

15

15

13

Fig. VIII-20. Comparison of measured and

tion amplitudes for foundation 5 (1.0

with water

pumped out

computed vertical components of vibraarea at a depth of 2.0 m, not backfilled and

of excavation).

same distances from the foundations but at

different depths.

For this

purpose, the graphs of the distribution of amplitudes at varying distances

from the source of waves may be used. On the basis of these graphs, one
can plot curves of the relationship between vibration amplitudes and
depths of the foundation base contact area. These curves are given in
Fig. VIII-21a for backfilled foundations and in Fig. VIII-216 for nonbackfilled foundations.

Let us denote by
the source

is

A and A f
Q

placed on the

soil

the amplitudes of

surface

soil

vibrations

and the foundation

is

when

placed at a

PROPAGATION OF

IN SOIL

363

According to Eqs. (VIII-2-20) and (VIII-5-12), we have

depth/.

(2^
k

I =
At-0

We

WAVES

ELASTIC

shall

compute the

ratio

l) er"t
)

9
2

A f/Ao

~ e-M

k2

(VIII-5-13)

for the conditions of the experiment.

According to the graphs of measured changes in amplitudes with distance, the velocity of transverse wave propagation for the given soil
Taking for sand the Poisson ratio
conditions is about b = 100 m/sec.
v = 0.35, we obtain from Eq. (VIII-1-12) the velocity of longitudinal
12

0.5
.1.0
1.5
Depth of foundation h,

2.0

0.5
1.5
1.0
Depth of foundation /?,

(a)

2.0

id)

Fig. VIII-21. Effect of depth of foundation on the vibration amplitudes at varying dis-

tances

from the foundation:

wave propagation a = 210 m/sec.

gation
of

may

be taken as

foundation;

(a) backfilled

0.936

The

(6)

nonbacknlled foundation.

wave propaThe foundation (source

velocity of surface

93 m/sec.

waves) vibrates at 950 oscillations per minute therefore the frequency

Thus we have
waves co = 100 sec -1
;

of propagating

2 2
(0.93) &

1.15

m-

,.,2

h*

fc

a=i

0i

=
=

p
a

1.00

m-

0.25

m"

V* -

V*

h2

fc

=
=

0.95

0.39

m"
m-

DYNAMICS OF BASES AND FOUNDATIONS

364

Consequently for the

conditions under consideration,

soil

we

obtain,

according to Eq. (VIII-5-13),

4^ =

1.33e-- 95 '

Calculations from this formula show that lowering the foundation 2

change compared to the wavelength) will not essentially influence

the amplitudes of vibrations of the soil surface.
This conclusion has
much in common with theoretical data concerning the distribution of
amplitudes with depth when the source lies on the surface. As indicated
in Art. VIII-2, under these conditions of wave excitement, amplitudes
computed for small depths (as compared with wavelengths) do not differ
much from amplitudes on the soil surface. This is contradicted by the
results of some experiments; investigations show that amplitudes decrease
with depth more intensively than is postulated by the theory of surface
wave distribution. An analogous contradiction is found in the question
under consideration. Contrary to theoretical conclusions, graphs plotted
on the basis of experimental data (Fig. VIII-2 la and b) show a considerable influence of the depth of the source of waves; this influence also
manifests itself at large distances from the source.
An empirical relationship between the amplitude of vibrations and
depth may be written as follows:
(a small

Af = A
For nonbackfilled foundations the
filled

Thus the

foundations, 0.654.

tion (source of waves)

lessens

when

its sides

e~ mf

coefficient

soil.

equals 1.00; for back-

influence of the depth of the founda-

on the decrease

touch

(VIII-5-14)

This

in amplitude of soil vibrations

is

equivalent to a decrease in the

depth of the foundation.

If one assumes that m ~ k ~ k and sets m = 1 in accordance with
experimental data for nonbackfilled foundations, then the velocity of
transverse waves will equal 100 m/sec; i.e., it will be close to the magnitude obtained from the analysis of the graphs of relationship between
Then Eq. (VIII-5-14) may be
vibration amplitudes and distances.
effective

written as follows

A f = Aoe-^' L^

(VIII-5-15)

This expression coincides with the theoretical formula if the vibrating

is considered not as an exciting force acting at a depth / below

foundation

the surface, but as a local exciter of

soil vibrations.

determined by the following values of

f x cosh
Jo

* =

<f>

and ^: 27

These vibrations are

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

365

machine foundations usually are such that the ratio f/L b is

For example, for low-frequency
smaller than unity.
machines (reciprocating compressors, diesels, etc.) the depth of the
foundation usually does not exceed 4.0 to 5.0 m, while in soft soils
the lengths of waves propagated from these foundations are usually of the

Depths

of

considerably

m; consequently,

order of 40 to 100

for these engines

f/L b

will

be of the

order of 0.10 to 0.05, and

e -2Tfiu

For high-frequency sources

(for

75

t0

hammer

example,

foundations) the

depths of foundations are approximately of the same order, but the waves
propagated are considerably shorter. For soft soils, the lengths of waves

propagated by hammers are 6 to 12 m; for these values of

will be of the order 0.3 to 0.8, and

b,

the ratio

f/L b

Thus

it is

-2^//L6

o.l5 to 0.007

even a small increase in the depth of the source has

clear that

considerable effect on the decrease of amplitudes of

soil

surface vibrations,

especially for high-frequency sources.

The Spreading of Surface

VIII-6.
a.

The Influence

Waves.

Many

of a

soils

Waves

in

Layered

Layered Soil Structure on

are formed of layers

Soils

the Distribution of

characterized

by

Surface

different

The layered structure of soil is responsible for

wave propagation which cannot be explained if the
be a homogeneous body.

mechanical properties.

many

peculiarities in

soil is

considered to

For example, the layered structure of

propagation of surface waves.

soil

may

essentially affect the

Theoretical analyses can handle mainly

mass of homogeneous soil. The

on the propagation of surface waves has
been studied by many authors (see, for example, Ref. 15). As a result of
these investigations it was established that in the process of surface-wave
propagation in a layered medium, there is a dispersion of waves; thus the
velocity of wave propagation is not constant and is determined not only
by the elastic and inertial properties of the medium, but also by the
lengths of propagating waves and their frequencies.
In addition, it was established that the presence of a layer overlying a
homogeneous soil mass and having properties different from this layer
does not produce the monotonous change of amplitude with depth which
takes place when surface waves are propagated in a homogeneous semicases in

which a

soil

layer overlies a

influence of a layered structure

mass.
If such a layer is present, graphs of the distribution of
amplitudes with depth may have several maximums and minimums
whose values depend on the ratio between the wavelength and the layer

infinite

DYNAMICS OF BASES AND FOUNDATIONS

366

thickness.

For a certain

ratio

between these values, the

possibility

is

not

may increase in the layer with an increase in

above may partly explain the discrepancies between

excluded that the amplitude

All of the

depth.

experimental and theoretical graphs of the distribution of amplitudes, as

indicated in Art. VIII-2.

To

simplify

all

calculations, let us confine ourselves to establishing the

influence of the layer

on the distribution

of free plane surface waves.

We

assume that the elastic properties and densities of the layer and the
underlying soil mass are determined
respectively by the constants Xi, mi,
Let us place the origin
pi and X, /x, Pof coordinates on the bottom surface
of the layer (Fig. VIII-22) and direct
the y axis vertically upward.
We take

Puf>\
I'l

iolS

the solutions of Eqs. (VIII-2-4) for the

underlying

soil

Fig. VIII-22. Coordinate system for

mass as

follows:

$ = Ae ave^ x

Eqs. (VIII-6-1) and Fig. VIII-23.

SI>

(VIII-6-1)

= BePve^x

According to Eq. (VIII-2-2) the components of displacements corresponding to these values of $ and ^ are as follows:

u
v

where a and

/5

=
=

(iAe ay

(aAe ay

are determined

^BePy )e^x

(VIII-6-2)

i^Be ?y )e^ x

by Eqs. (VIII-2-7)

and

are arbitrary

constants.

For waves propagating in the

will

layer, the solutions of Eqs. (VIII-2-4)

be as follows:

= (C cosh aiy
=
(E cosh fay
^i
ai = ^2 _ hl
3>i

where

+ D sinh aiy)e^
+ F sinh /3iy)e^
ft

h!

^2

kl

(VIII-6-3)

(VIII-6-4)

CLi

and cii and 6i are propagation velocities of longitudinal and transverse

waves in the layer; it is clear that hi and ki are the reciprocals of the
wavelengths.

For the displacement components in the layer we obtain

Ui

ii-(C

cosh aiy

+D

sinh aiy)e^ x

fa(E sinh fay

+ F cosh fay)e^

vi

ai(C sinh aiy

+D

cosh aiy)e^ x
~ i(E cosh fay

+ F sinh fay)?**

VIII . 6 _ 5)

PROPAGATION OF

The

arbitrary constants A, B,

WAVES

ELASTIC

IN SOIL

367

F should be selected in such a way as

boundary conditions of the problem.

It follows from the conditions of continuity of stress and deformation
that deformations and stresses in the layer equal deformations and
to satisfy the

When

boundary.
be written as follows:

stresses in the underlying soil at their

contact conditions

may

u
V

It is

*y

V\

Ty x

a yi

these

0,

(VIII-6-6)

Tyxl

assumed that the upper surface of the layer

boundary conditions for y H will be

is

free of stresses;

therefore

<ry

The components

rvx

(VIII-6-7)

through

of stresses are expressed

<

and

by Eqs.

(VIII-2-10).

Using the solutions of Eqs. (VIII-6-2) and (VIII-6-5) and the boundary
conditions of Eqs. (VIII-6-6) and (VIII-6-7),
relating the constants

we obtain

equations

six

A, B, C, D, E, and F:

iA + (3B = iC + faF
aA - i^B = a D - i%E
x

(2

fc

)A

2itfB

- [(2 2

kS)C

2i(fi t F]

fc^E

2^iD]

ii

2iaA
(2

fci

(2?

)(C cosh ai

H+

(2

k*)B

E [(2 2

VIII-6 - 8)
(

sinh a x H)

&# +

2^/5i(' sinh

k^){E cosh faH

Z)

cosh frtf)

+D

cosh ai #)

+ F sinh ftff)

2iai(C sinh

!#

solution of the system of Eqs. (VIII-6-8) for the constants A,

F
be other than zero only if the determinant equals zero. Hence we
obtain an equation for the determination of the velocity of propagation

The

will

of surface

waves

in a layer overlying a soil mass:

12(f)

In the general case this equation

is

=
very complicated;

(VIII-6-9)
it

contains irra-

and is very difficult to solve. It is, however, solved 40

for the case in which the densities of the soil mass and the overlying layer
do not differ much (p = pi) in addition, both the layer and the underlying
soil mass must satisfy the Poisson hypothesis
Figure
(y = 0.25).
VIII-23 presents graphs of the relationship between c a /c and the ratio
between layer thickness and wavelength for several values of n/pi (c a is
the velocity of surface waves with overlying layer, and c is the velocity
tional expressions

DYNAMICS OF BASES AND FOUNDATIONS

368

same waves without overlying layer). The wavelength depends

on the frequency of propagating waves; therefore the graphs show the
dependence of the propagation velocity on the frequency in a case in
which waves propagate in a layer of constant thickness.
It is seen from Fig. VIII-23 that with a decrease in the wave frequency
(i.e., with a decrease in H/L) the velocity of surface waves increases.
A
particularly intensive change in velocity depending on changes in H/L
takes place when H/L < 0.5.
For these values of H/L, the velocity of
waves increases approximately in proportion to the wavelength, i.e.,
of the

2.30

2.20 \

\
\

2.10
2.00
1.90
1.80

\ vV

fr
V \-

1.70
1.60

1.50

1.20

(\ c
\ ^c --
NN \

1.10

\,

1.40

-W

1.30

1.00
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

H/L
Fig. VIII-23. Effect of the ratio of the surface layer thickness

on the velocity
lying

soil

ratio c a /c for different ratios of the

mass) to

(of

Lame

H to

the wavelength

coefficients

/*

(of

the under-

the upper layer).

inversely proportionally to the frequency.

The

coefficient of propor-

depends on the ratio m/mi- The larger the difference between

the mechanical properties of the overlying layer and the underlying soil

tionality

mass, the larger the value of the coefficient of proportionality and, conThe waves are
sequently, the larger the velocity of wave propagation.

propagating mostly in the overlying layer, and the influence of the undersoil on the velocity of wave propagation is insignificant when the

lying

frequency of vibrations

is

so high that the lengths of propagating

waves

are equal to or smaller than the layer thickness.

As indicated

in Art. VIII-2,

when waves

are propagated in a

homo-


'

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

369

geneous, elastic, semi-infinite mass, the ratio between the amplitudes of

the horizontal and vertical components of displacement of particles on

and equals 0.6811. If one sets y = H

and determines the ratio between the horizontal and
vertical components of vibrations, an expression is obtained which shows
that the ratio depends on H/L.
Figure VIII-24 presents graphs of the relationship between u a /v a on
the soil surface and L/H; it is seen from these graphs that for short waves
(L/H > 1) the ratio between the horizontal and vertical components of
vibration does not differ much from a value typical for waves propagating
in a homogeneous semi-infinite mass.
Within a range of frequencies
the

soil

surface remains constant

in Eqs. (VIII-6-5)

1.4
i

1.2

5^H

^.
=$"
4

/ S'

0.8

"a
-

0.6

T^T

1.0

\t
^ Lj4
V/

'

"TT~

fi=2fi,

0.4

0.2

10

12

14

18

20

22

/H

L to the upper layer thickness

displacements on the surface of the upper

Fig. VIII-24. Effect of the ratio of the wavelength

on the

ratio of horizontal

u to vertical

layer.

corresponding to wavelengths approximately two to four times larger

than the layer thickness, the ratio u a /v a becomes smaller than that for
waves propagating in a soil that is not layered; i.e., the ellipsis of vibrations becomes elongated in the vertical direction.
When the frequency
decreases so that the wavelength is four to five times larger than H, the
relative value of the horizontal

the

ellipsis of

component

of vibrations increases

vibrations expands in the horizontal direction; for

at a certain value of the frequency,

The foregoing data concern the

mass with properties

it

becomes a

/xi

>

and
3m,

circle.

influence of one layer, overlying a soil

from

own, on surface waves of the

in a many-layered
soil is much more difficult, and not much has been done in this area.
However, there is no doubt that the presence of several soil layers showing
Raleigh type.

An

different

investigation of

its

wave propagation

DYNAMICS OF BASES AND FOUNDATIONS

370

a diversity in mechanical properties may influence considerably the

properties of propagating waves.
A theoretical treatment of problems

wave propagation in many-layered soils is too difficult; therefore

experimental investigations should be conducted for the solution of
problems of this type.

of

Transverse Surface Waves in Two-layered Soil Systems.

b.

structure

soil

layered

may

infinite or finite

change the laws governing wave propagation in an

solid body; in addition, the layered system may cause the

formation of new waves characterized by specific properties. It was

found, for example, that where a soil layer is underlaid by a more rigid
base, specific transverse waves may occur in the soil, with the same
velocity in both layers.

With

all

symbols

and

for elastic constants

soil densities

as defined in

by t =
waves are steady with respect to
Eqs. (VIII-1-1) as follows: for the upper layer:

Art. VIII-6-a, let us assume, as before, that the plane described

is

Assuming

a contact plane.

time,

we

seek solutions of
t*i

and

*;i

also that

wi

w =

F(y)e^ x

-Ut)

(VIII-6-10)

for the underlying soil:

u = o
Substituting Eqs.

(VIII-6-10)

Eqs.

into

(VIII-6-11)

Ce/fo-KC**-**)

and taking

(VIII-1-1)

into

account that
dz

we obtain an equation

for the determination of F(y)

^=

(fcx

- ?)F =

(VIII-6-12)

As

before,

ki

= 1
t

Ox

(bi is

the velocity of transverse waves in the upper layer).

Eq. (VIII-6-12)

is

F(y)

where

The

solution of

as follows:

= A

r =

hi

sin aty
2

+B

cos cay

(VIII-6-13)

Substituting Eqs. (VIII-6-11) into (VIII-1-1) for the underlying

we

find that this solution will satisfy the equations of

2
/3

where

k2

= e = r-r
o2

k2

motion

soil,

if

(VIII-6-14)

PROPAGATION OF

and

b is

ELASTIC

WAVES

IN SOIL

371

the velocity of transverse waves in the underlying

The

soil.

requirements of continuity of deformations and stresses provide us with

the boundary and contact conditions; therefore for y = 0, it is necessary
that

Tyz

Wi

(VIII-6-15)

Tyzl

In addition, since the upper surface of the overlying

layer

soil

is

not

subjected to the action of stresses, the following condition should be also

observed:!

If v

= H,
ryzl

(VIII-6-16)

Since
Ty Z i

we

dwi
m

-7T

dy

ni<xi(A cos aiy

sin

2/)e

**~ a ' )

obtain, according to Eq. (VIII-6-16),

cos

aH - B

sin

aH =

(VIII-6-17)

According to Eqs. (VIII-6-15) we have

B = C

fnaA =

(VIII-6-18)

n(3C

Excluding from Eqs. (VIII-6-17) and (VIII-6-18) the constants A, B,

C, we obtain an equation for determining the constant :

and

tan

aH = ^-

(VIII-6-19)

Via

In order that the wave amplitudes in the underlying soil may decrease
To satisfy
fi should be positive.
this condition, the following interrelationship should exist:
with depth, the real part of coefficient

But

>

= -

(VIII-6-20)

(VIII-6-21)

ch

where

Cb is

wave propagation.

the velocity of transverse

this expression for into the left-hand part of

cb

<

(VIII-6-22)

the velocity of propagation of transverse surface waves

i.e.,

Substituting

Eq. (VIII-6-20), we obtain

is

lower than

the velocity of propagation of transverse waves in the underlying

It is
t

easy to prove that

The

c&

>

&i.

As a matter

in

if

we assume

soil.

that

component magnitudes, i.e., of r yx and a y is satisfied

and the underlying soil, the displacethe direction of the x and y axes equal zero.

equality of other stress

automatically, because for both the upper layer

ment components

of fact,

DYNAMICS OF BASES AND FOUNDATIONS

372
cb

<

then according to (VIII-6-21) we would obtain

61,

on

j <
or

<

hi

Then a 2 <
example,

(VIII-6-23)

(VIII-6-24)

which would be possible

0,

let

61

us assume that a

a were an imaginary value.

if

For

Substituting this expression into

iy.

Eq. (VIII-6-19), we obtain:

ttmh(yH) =

- ^- <0
Mi<*

which

is

impossible; hence c b

>

bi.

Thus it is clear that the transverse waves under discussion may appear
and propagate only when b > b%. Since the densities p and p x usually do
not differ much from one another, the condition governing the propagation of surface transverse waves will be as follows:
M
i.e.,

the underlying

soil

the overlying layer.

The

limiting values of propagation velocity of the

be

will

bi

We now
mi

(VIII-6-25)

mi

should have a higher elastic rigidity in shear than

waves under consideration

and

>

<

<

Cb

transform Eq. (VIII-6-19); taking into account that

bfp,

\x

b 2p

we have
tan

(VIII-6-26)

Substituting into Eq. (VIII-6-26) the expressions for

and assuming that

pi,

after

and

a,

various transformations

we obtain

Eq. (VIII-6-26) in the following form:

This equation establishes the dependence of the velocity c& of the surface
61, but also on the ratio between the
layer thickness and the length of propagating waves.
It follows that the
roots of Eq. (VIII-6-27) depend on the frequency of the waves.
The expression governed by tan in Eq. (VIII-6-27) is always positive,
transverse waves not only on b and

but,

depending on the magnitude of H,

it

may

attain

any

value.

For

PROPAGATION OF

example,

it

ELASTIC

WAVES

IN SOIL

may equal ir/2 or may even be larger,

may acquire negative values; but

in consequence of

the tangent

Eq. (VIII-6-27) will always remain positive.

aH lying within the range

373

which

the right-hand part of

Therefore

the values

all

of

2n

should be eliminated.

Eq. (VIII-6-27)

H
2tt

where k

Hence

may

< aH <

1, 2, 3,

As a consequence

of the periodicity of the tangent,

be written as follows:

&

0, 1, 2, 3,

n-K

...

kir

arctan

\c
bi2

'

cb

yv

/V -

(VIII-6-28)

waves under consideration is

determined by a multiterm expression; consequently, theoretically there
are several curves of dispersion which correspond to each set of values of b, b h and H.
Figure VIII-25 presents a graph of wave
dispersion.
This graph gives the relationship
between c 6 /&i and H/L for
<
< tt/2. 'Va, 2 -
It is seen that surface transverse waves, as in
the case of waves of the Raleigh type,f are
characterized by a normal dispersion; i.e.,
their velocity increases with an increase in
wavelength.
Fig. VIII-25. Effect of the
Let us consider the nature of changes in
wave amplitudes with depth, confining our- ratio of the upper layer thickness H to the wavelength L on
selves to cases in which y < H, i.e., conthe ratio of transverse wave
sidering points located only in the overlying
velocities in the upper layer (c&
it

follows that the dispersion of the

layer.

The

at its surface

distribution of amplitudes along the

and

6i

within the

layer).

depth of the overlying layer depends essentially on the ratio between layer thickness and wavelength.
In the general case the amplitudes are not distributed in a monotonous manner,
but form nodes and protuberances. In the nodes the amplitudes
of vibrations equal zero; therefore by equating the right-hand part of
Eq. (VIII-6-13) to zero, we obtain an equation for the determination of
the location of the node line along the depth of the overlying layer:

A
or

sin

ay

+B

tan ayo
t

In the presence of a

properties.

soil

cos ay

= B
-j

layer overlying a soil

mass characterized by

different

DYNAMICS OF BASES AND FOUNDATIONS

374

According to Eq. (VIII-6-17) we have

A
-5

COt

aH

>

hence

From

tan

aH =

the above expression

cot ay

it is

tan

2/o

(VIII-6-29)

seen that nodes in the distribution of amplitudes

computed from the contact

yQ

Specifically,

-\

= aH

along the depth of the overlying layer only

of the node,

we obtain
ay

Thus

when aH =

x/2, y

H-

when aH >

may

w/2.

appear

The depth

surface, equals

la

and the surface transverse waves

and do not penetrate into the

are propagated only in the overlying layer

underlying
VIII-7.

soil

mass.

Screening of Elastic

Waves Propagating

through Soil

Elementary Concepts of Physics concerning the Diffraction of Elastic

In order to decrease vibrations of structures it is sometimes
recommended to protect the structure from the influence of the energy of
a.

Waves.

elastic

waves propagating

in soils.

The

following methods are used for

open trenches, employment of

and provision of compacted zones.
There is a widely shared opinion that the best effect is obtained by
placing a barrier around the source of vibrations.
For example, if it is
necessary to protect the walls of a forge shop from the harmful influence of
waves propagating from a hammer foundation, it is recommended to place
barriers around the foundation.
If a vibrating diesel foundation induces
this purpose: provision of backfilled or

sheetpiling or barriers of special design,

considerable vibrations of a structure located at a distance of even several

tens of meters from

it, then it
around the source of the waves.

is

also recommended to place a barrier

Only when the source of vibrations is

moving

(transport) is it recommended to provide barriers near the receiver.

Experience with various types of barriers has shown that they are often
of no use at all, or their effect is very small.
However, cases are on

record in which barriers were very effective.

sheetpiling, trenches, etc.,

is

It appears that the use of

not always effective; these installations are of

use only under certain conditions rarely encountered in practice.

This is explained by the nature of the propagation of elastic waves

encountering an obstacle in the form of a screen. In such cases the
concept of "wave beams" cannot be used, since the propagation is no

PROPAGATION OF
longer linear.

ELASTIC

WAVES

IN SOIL

In the presence of an obstacle such as a screen (or a

in a screen) the propagation of elastic

waves

375
slit

accompanied by a phenomappears that in cases in which

is

enon known as the diffraction of waves. It

barriers were of no use. designers did not take into account elementary
features of the theory of
If

wave propagation

in the presence of a screen.

plane steady waves propagating over the earth's surface meet a

then (in the case of linear wave

is formed behind the screen, in
which no vibrations are observed. There is a sharp boundary between
this screened zone and the area of wave propagation behind the screen.
At the same
It looks as if vibrations are interrupted at this boundary.
time, the character of wave propagation behind the screen remains the
same and does not depend on the wavelength or the screen width.
As a matter of fact, the character of propagation changes after the
waves have passed the screen. For example, no sharp decrease is
observed in the amplitude of vibrations at the boundaries of the screened
The amplitudes decrease as the waves approach the screened
zone.
zone; inside the zone they decrease in such a way that the sharp contour
The depth of the screened zone is limited; as the
of the zone vanishes.
distance from the screen grows, the waves penetrate more and more into it.
At some distance from the screen, the waves which were propagated
After that the
separately in areas to the right and left of the zone merge.
process of wave propagation basically does not differ from propagation
where no screen is present. The depth of the screened zone is finite; it
increases with an increase in the ratio of screen length to wavelength.
If screen length is large in comparison with wavelength, then the nature
of the propagation approaches that of linear propagation.
As the screen dimensions decrease in relation to the wavelength, the
depth of the screened zone decreases. If the screen dimensions are small
in comparison with the length of the propagating waves, no screened zone
at all is formed, or else this zone is very small.
Then the screen just
dissipates the waves falling upon it and partly reflects them.
The phenomenon of wave propagation in the screened zone represents
one of the special features of wave diffraction. However, the phenomenon of diffraction is not limited to the distortion of the boundaries of the
screened zone; it also changes the nature of the distribution of amplitudes.
At the zone boundary the amplitudes do not change in jumps as they
should in the case of linear propagation of waves, but, as indicated above,
the amplitudes change continuously, and the character of the changes in
amplitudes behind the screen may be very complicated. As the distance
from the screen grows, the average amplitude of the wave within the
screened zone increases and, at some distance from the screen, attains
the amplitude of a wave that has not been diffracted.
screen placed parallel to the

wave

front,

propagation) a screened zone ("shadow")

DYNAMICS OF BASES AND FOUNDATIONS

376

The screen

creates a screened zone of a certain finite depth; therefore

the screen affects the distribution of amplitudes over the surface only
The largest effect of the screen is observed
within the screened zone.
directly behind

tion of

waves

it.

As the distance from the screen grows, the propagait increases, and the effect of the screen

in the area behind

on the amplitudes

of vibrations gradually decreases.

Thus, if one evaluates the effect of the screen on the basis of relative
changes in amplitudes, f then the following relationship may be established
for the propagation of a plane undamped wave: as the distance from the
screen grows, the relative change in amplitude continuously increases

from a minimum

in the section directly behind the screen to unity at

points located at distances larger than the depth of the screened zone.
It

appears that the amplitude of

soil

vibrations directly behind the

screen depends mainly on the relationship between the screen dimensions

and the length

of the propagating waves.

As the screen dimensions

increase in comparison with the wavelength, the initial value of the ratio

between the amplitudes decreases; if the screen dimensions are large in

comparison with the wavelength, then the waves directly behind the
screen will not get into the central portion of the screened zone, and consequently the amplitudes of

soil

vibrations there will equal zero.

the waves were propagating only along the surface and not penetrating into the soil, then, under conditions of wave diffraction, it would be
If

For this purpose, it would suffice

an enclosing screen around the source. However, even surface
waves penetrate to depths on the order of one wavelength; therefore the
diffraction phenomenon is observed not only on the surface, but in depth
The placing of an enclosing barrier around a source would be
as well.
accompanied by the creation of a continuous screened zone behind it only
on the condition that the barrier cut through the entire soil zone in which
waves are propagating.
possible to isolate the source of waves.
to place

It is practically

impossible to satisfy this condition.

Therefore in the

should be remembered that the dimensions

of the screened zone formed by the barrier depend not only on the dimensions of the barrier in plan, but also on its depth.
Thus, if one provides a
design of a screen (barrier),

it

barrier with a large length along the surface but with a small depth in

comparison with the wavelength,

its effect will

be negligible.

The foregoing considerations concerning the diffraction of elastic waves

by a screen are based exclusively on some elementary concepts borrowed
from experimental investigations
of waves propagating in water.
t

of diffraction of

The relative change in amplitudes is defined as

when the screen is present to the amplitudes

point

present.

sound waves in

air

and

the ratio of the amplitudes at a

at the point when no screen is

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

377

There is no doubt that the diffraction of elastic waves in soil is somewhat different from the diffraction of sound waves or waves traveling
through water.
tatively,

if

The

differences could be established, especially quanti-

the theory of diffraction of elastic waves by means of a screen

were worked out.

However, no theory

a basis for quantitative computations.

tions

have appeared in

this field,

is

available

now which would

give

Several mathematical investiga-

but solutions obtained on the basis of

these investigations are of a form which does not permit the practical

evaluation of the influence of the screen on the

decrease in

4243
soil vibrations. 26
-

Obstacles in the direction of

-Vibrating

foundation

wave propaga-

tion produce not only the diffraction of waves,

but also their dissipation and reflection. The

dissipation may occur even if the dimensions of
the obstacle are very small in comparison with
the wavelength. This phenomenon is used in
special branches of physics, particularly in the
study of minute particles within the field of
light or sound waves.
However, in comparison with other processes
taking place in the soil when elastic waves are
propagating (for example, by comparison with
energy absorption by the soil), it appears that
dissipation is of secondary importance for the
protection of structures from waves in

No.O

h*-8 and

11

Trench
No. 2-'

No.3-<
No.4-<

No.

5-6

No.6-

soil.

Nevertheless, this phenomenon, combined with

the reflection of waves,

may

contribute some-

what to the efficiency of a barrier, although the

amount of energy dissipated and reflected by
the obstacle apparently makes up only a small
part of the total energy which produces vibra-

No.

,1.

Fig. VIII-26. Plan of experiment on wave-screening action of trench.

tions of structures.
b.

The author,

Experimental Screening of Elastic Waves in Loess.

in

cooperation with P. A. Saichev and Ya. N. Smolikov, investigated the

screening of waves

by means

of a trench.

reinforced-concrete foundation with

the

soil.

an area

The source
of 0.81

of

waves was a

in contact with

A vibrator installed on this foundation produced forced

vibrations at the desired frequency

and amplitude.

In

all

vertical

the experi-

ments, although the frequencies changed, the amplitude of vibrations

remained more or less constant, equaling some 0.19 mm.
Figure VIII-26 shows the location of the source and the points at which
amplitudes were measured. Two profiles were selected for the investigation of the distribution of amplitudes: (1) a longitudinal profile, perpen-

DYNAMICS OF BASES AND FOUNDATIONS

378

and

dicular to the trench;

The

(2)

a transverse

profile, parallel to

the trench.

distance from the trench to the source of vibrations was 2 m.

experiments the length of the trench was 8 m. At

1
wide; the width somewhat decreased
with depth and at a depth of 4.0
equaled 0.7 m; the walls of the trench
were not braced.
The distribution of amplitudes over the surface was studied in the

In the

first series of

the surface the trench was some

m
m

following order:

Before the trench was excavated, the distribution of amplitudes was

investigated along the longitudinal
50

^_^__^_^_^__^_^__^_^
X

when h= 0.0 m
o when h = 0.8 m

40

when h = 1.8
x when h- 2.5

m
m

o
o

and transverse profiles for frequencies

of 730 and 930 oscillations per minute.
Then the trench was excavated to
a depth of 0.8 m, and for the same
frequencies the vibration amplitudes
were measured along the longitudinal
profile.
These measurements were
successively repeated when the trench
reached depths of 1.8, 2.5, and 4 m.
Further excavation without bracing

x 30

was dangerous, but the use

1
1

could introduce
20

10

it

Distance,

12

3 4

VIII-27.

account.

take these
After

the

studied not only for the above-men-

10

tioned frequencies, but also for 810

12

14

16

m
6

Vibration

and 1,030

Then

L_

oscillations per minute.

the length of the trench was

increased by 3 m, and in addition to

Nos. of points

Fig.

difficult to

the distribution of amplitudes was

rr
1

into

trench reached a depth of 4.0 m,

s ^m

would be

factors

>4- C

c
in

of bracing

factors into the

phenomenon under investigation, and

1%
INk
<1

new

amplitudes

measured during experiment illustrated

by Fig. VIII-26, for different depths h
of the 8.0-m-long trench and a 730-rpm
vibrator speed.

the investigation of amplitude distri-

bution along the longitudinal

profile,

measurements along the transverse

profile were conducted for 670, 730,
930, and 1,030 oscillations per minute.

Figure VIII-27 presents graphs of amplitude distribution along the

rpm at different depths of the 8.0-m-long

longitudinal profile for 730

The

amplitude of vibrations was one four-hundredth of the

from Fig. VIII-27 that there are no sharp
changes in the amplitude distribution for 730 oscillations per minute.
The amplitude distribution graphs are smooth curves, generally typical
for the amplitude distribution along the soil surface when no screening
trench.

values given.

real

It is seen

PROPAGATION OF
is

present.

ELASTIC

Therefore the conclusion

is

WAVES

IN SOIL

379

possible that for this frequency,

a trench of any of the depths investigated has no significant influence on

Analogous results
the distribution of amplitudes over the soil surface.

were obtained for other frequencies and for trenches shallower than 4 m.
Figure VIII-28 shows graphs of the amplitude distribution when the
deep and 11.0
long.
It is seen from these graphs
trench was 4.0
that at the trench the amplitudes underwent a sharp change at 930
Changes in amplitudes in the area beyond the
oscillations per minute.
trench are so insignificant that in the first approximation the amplitudes
may be considered as being invariable over the entire area investigated.
It is noteworthy that even at a fairly large distance from the trench
the soil vibration amplitudes are considerably smaller than the vibration
deep and the
amplitudes at the same points when the trench was 4
frequency was less than 930 oscillations per minute. This leads to the

assumption that in this case the influence of the trench on the decrease of
the amplitudes of soil vibrations is observed at distances much larger
than the depth of the trench.
depth
An investigation of the screening properties of a trench of 4
length was also carried out for 810 and 1,030 vibrations per
and 8
minute. Figure VIII-29 shows graphs of amplitude distribution for
these frequencies and, for comparison, the graphs for 730 and 930 oscilla-

tions per minute.

A comparison of the amplitude distributions beyond the trench during

930 and 1,030 oscillations per minute leads to the conclusion that an
increase in the frequency by 100 oscillations per minute has almost no
However, an increase from 730 to
effect on the amplitude distribution.
810 oscillations per minute resulted in considerable decrease in ampliIt appears that when the trench had a depth
tudes beyond the trench.
of 4

and the waves propagated at 810

oscillations per minute, the

trench screened the vibrations, but not to such a degree as to cause a

rupture in the amplitude distribution as in the case of 930 and more

oscillations per minute.

To

between vibration amplitudes at the soil

shows the seismograms recorded at points 1 and 2. Let us note that while vibrations were
propagated at a frequency lower than 930 vibrations per minute, they
were felt without any instrument, by "sensory perception," on both
sides of the trench; when the frequency exceeded 930, the vibrations were
felt only on the side of the trench nearest the source.
An increase in the trench length by 3 m did not have a significant effect
on the amplitude distribution beyond the trench for frequencies of 930
and 1,030 oscillations per minute. However, an increase in the trench
length had considerable influence on the decrease of amplitudes of vibraillustrate the relationship

surface

on opposite

sides of the trench, Fig. VIII-30

DYNAMICS OF BASES AND FOUNDATIONS

380

50

40
40

at

670 rpm
730 rpm
930 rpm

at

1030 rpm

at
at

30

at

730 rpm
810 rpm
930 rpm

at

1030 rpm

at

at

V
-20

ii
\i

10

s
"*-

<
1
(

c/>

Distance,

2 3 4

10

12

16

Distance,

5
6
Nos. of points

3 4

Nos. of points

Fig. VIII-28. Vibration amplitudes measured during experiment illustrated by Fig.

VIII-26 for a 4.0-m depth h of the 11.0-mlong trench and varying vibrator speeds.

Fig.

VIII-29.

Vibration

amplitudes

measured during experiment illustrated

by Fig. VIII-26 for a 4.0-m depth h of
the 8.0-m-long trench and varying vibrator speeds.

On the

730 rpm

wall

face adjacent to the source

On the opposite

730 rpm

wall face

930 rpm

Fig. VIII-30. Seismograms recorded at the

Fig. VIII-26.

(point 1)

1030 rpm

930 rpm

soil

(point 2

1030 rpm

surface on both sides of the trench of

PROPAGATION OF
tions

beyond the trench

for

ELASTIC

waves

WAVES

IN SOIL

381

of 730 oscillations per

minute

(see

Figs. VIII-28 and VIII-29).

In addition to the investigation of the distribution of amplitudes over
the soil surface, an investigation of the distribution of amplitudes along
For this purpose, cavities were
the depth of the trench was conducted.
of 1
from each another.
distances
trench
of
the
at
walls
both
in
dug
Measurements of the vertical component of vibrations were conducted
by a seismograph and recorder. These measurements were performed
A distinct screening effect of the
at 1,050 and 670 vibrations per minute.
trench was observed at 1,050 vibrations per minute, but not at 670

Amplitude,

10

20

mm

40

30

o at

(a)

at

400
20

10

670 rpm
1050 rpm

(b)

Fig. VIII-31. Variation of vibration amplitudes along the depth of the trench of
Fig. VIII-26: (a) trench wall adjoining vibrating foundation; (6) opposite trench wall.

vibrations per minute.

tribution of amplitudes.

Figure VIII-31 shows the graphs of the dis-

Figure VII 1-3 la refers to the wall of the

trench nearest the source of vibrations; Fig. VIII-316 refers to the

opposite wall.

At the trench wall adjacent to the source of waves, vibrations at the

two frequencies do not differ much one from the other; as was to be
expected, the vibration amplitudes at this wall decrease with an increase
in depth in approximately the same manner as vibrations on the soil
surface.

On

the wall opposite the source, the distribution of amplitudes

at 670 vibrations per minute basically does not differ

from the distribution

amplitudes on the wall adjacent to the source. On the other hand,

the distribution of amplitudes for the screened frequency (1,050 vibrations per minute) is distorted by diffraction.
This is explained by the
of

fact that

on the

side of the trench opposite the source,

two reciprocally

DYNAMICS OF BASES AND FOUNDATIONS

382

opposed processes influence the amplitude distribution:

(1) the screening

amplitudes; this effect increases
a decrease in amplitudes due to the increase in depth of

effect of the trench, causing a decrease in

with depth; (2)

the point under consideration.
Depending on the relative significance of
the screening effect in comparison with the depth effect, the distribution
of amplitudes in depth may be of a different nature.
For example, it
may happen that the amplitudes will not decrease, but will even somewhat
increase with depth.
If no screening effect is observed, then the amplitudes should decrease
with increasing depth; this was so when waves were propagated at 670

oscillations per minute.

After

all

backfilled

the experiments described had been finished, the trench was

and the amplitudes were measured at points
to 4 of the

longitudinal profile.

It

turned out that the backfilled trench

lost all its

screening properties.
Investigations of a linear sheetpiling screen were conducted on the

but on a different section. The decision was made first to dig

deep and then to drive sheetpiling along the trench axis to a
Then the
depth of 2.8 m; 25-mm planks of 3.0
length were used.
and the sheetpiling was driven
depth of the trench was increased by 1.0
Thus the maximum depth of the screening installation, consisting
1 m.
in the 1.8-m-deep trench and the 2.8-m-deep sheetpiling, was 4.6 m.
The length of the sheetpiling was 6.0 m.
The same foundation which was used for the investigation of the trench
was taken as the source of vibrations; the amplitude of excitement of
vibrations was the same.
Measurements of amplitude distribution along the longitudinal profile
were made prior to driving the sheet piles (when the trench was 0.8
deep), at the time the sheetpiling was driven to a depth of 3.6 m, and

same

site,

a trench 0.8

m
m

finally when the trench was 1.8

deep and the sheetpiling was at a
from the soil surface. Measurements were made at
depth of 4.6
730 and 1,030 oscillations per minute.
Figures VIII-32 and VIII-33 show graphs of the amplitude distribution

along the longitudinal profile for the experimental frequencies. In spite

of some scattering of points, it may be seen from Fig. VIII-32 that prior
to sheet-pile driving and when the sheet piles were at a depth of 3.6 m, no
rupture occurred in the distribution of amplitudes at the barrier (this
refers to the waves characterized by 730 and 1,030 vibrations per minute).

An

analogous conclusion

amplitudes

when

characterized

may

be drawn in regard to the distribution of

deep and the waves were

the sheet pile was 4.6

by 730 vibrations per minute

(Fig. VIII-33).

However, at

1,030 vibrations per minute, at a sheet-pile depth of 4.6 m, a rupture in

the distribution of amplitudes occurred near the barrier (points

and

2).

PROPAGATION OF

The amplitude

ELASTIC

WAVES

IN SOIL

383

distribution at this depth of the barrier

and at 1,030

vibrations per minute clearly shows the presence of the screening effect

which the wave frequency was the

deep (without sheetpiling).

of the barrier, analogous to the case in

same and the trench was

4.0

Experimental Screening of Waves in Water- saturated Soils. It is

employment of a screen in the form of a trench is not
Therefore
possible in some soils, for example, in water-saturated sands.
c.

clear that the

sheetpiling should be used in such

The author experimentally

soils.

40

ol

30

1
20

1
o /V=
/V =

\\

730 rpm
1030rpm

\a

of

\|

piling

10-

O \

o\

line

\(
c
Center

o
the

i>

Si
c

i>

1
l

a>

>

sheet

H ^
6

10

12

Distance,
1

10

12

14

16

Distance,

Fig. VIII-32. Vibration amplitudes measured at the soil surface on both sides of a
trench and 3.6-m-deep sheet-pile wall.

investigated such screens 10 in the

4 5
6
Nos. of points

2 3

Fig. VIII-33. Vibration amplitudes measured at the soil surface on both sides of a
trench and 4.6-m-deep sheet-pile wall.

field, where a water-saturated dense

sand was found down to a depth of some 8 m. The sand was underlaid
by a thick layer of organic silt, below which sands were again found.
Foundation 7 was used as a source of waves in the investigations of the
screening effect of linear sheetpiling.
The foundation area in contact
2
with soil was 1.0
Prior to the sheet-pile driving, amplitudes of
vertical soil vibrations were measured at several different frequencies of
forced vertical vibrations of the foundation.
The vibrometer recorded
soil vibrations in a vertical plane which passed through the foundation
and the point where vibrations were measured. Lissajous curves

'

8.0

7.0

-1

N-- -

\
\\
\h

6.0
5.0

4.0

600 mi'n

W^
^*^
Y-

3.0

2.0

^^
i

^C;

1.0

jr-- "^.r*J

8.0

-*>
*-*

-o

7.0

6.0

5.0

= 810 min

v
\3

4.0
3.0

2.0

>>

;-

^<
> *"v>

<*>

th
nter

1.0

i V of
SI

*-o

>.^^

p->_

~r

_a_

i
i

~o

10.0
<

>

9.0

N-- 10C)0rr

8.0

6.0
5.0

\
\
\

7.0

in

q
i

> S>

4.0

- Jjo

3.0

-ex " -4

)^_

2.0

T~

-o
.

1.0

-0-

9.0

8.0

t_

7.0

N = 135 0m

'

\\

6.0

'

5.0

\ t\

4.0

S
t.

3.0

^U
">o

2.0

1.0

-\

K/

^v s

**

-o .5?

-^

3": .r-

9 10 11 12 13
7
8
Distance from foundation 7,

14

15

16

17

18

19

20

Fig. VIII-34a. Variation along a longitudinal profile of amplitudes of the vertical

component

of vertical surface soil vibrations before (open circles)

circles) sheet-pile driving (see Fig.

VIII-37, foundation
384

7).

and

after (black

'

PROPAGATION OF

ELASTIC

'

WAVES

IN SOIL

385

8.0
1

7.0

N = 1620 min"

\\

6.0

,1

|\

5.0

4.0

c >t

E2.0

<

1.0

,-,N

>

i\V

+.

i_

a>

^.

--?^

10

11

12

Distance from foundation

13

7,

^. p

14

15

16

17

18

19

8.0

7.0
1

/l/

189 Dmi i-'

6.0

5.0

_4.0

V.

a>
1

3.0

2.0
I
<

^r>

1.0

> f'
4

"J>

"\^_ -j_.
5
6
7
8
9
10
11
12
1 3
14
Distance from the axis of foundation 7. m

15

16

17

-o-

19

18

9.0
I

8.0

A/ =

6.0
5.0

/
'\

!\

'

o
>

3.0

^ <<o

i
i

<

i"

4.0

215 Omi

7.0

2.0
1.0

v
!\

f
/

Fig. VIII-346.

-^

<r

Same

*^'-~

o
b"""*'

8
9
10 11
12
Distance from foundation 7,

13

14

15

16

17

18

19

as Fig. VIII-34a, but for other vibrator speeds.

obtained by means of the vibrometer were recorded at each point. Thus

data were collected not only on the values of the vibration amplitudes,
but also on the shape of the orbit of particle motion at a given point on
the soil surface.
The results of the measurements are shown in Figs.

VIII-34 and VIII-35, where small open

circles

show the amplitudes

of the

-,

DYNAMICS OF BASES AND FOUNDATIONS

386
5.0

600 min"

4.0
1

3.0
1

2.0
1.0

<s

J_

-9*-""

-c

-o

jUr**

6.0

N
5.0

'

810mm"

1
1

4.0
1

3.0

<
1

\\

O2.0

o
x

1.0

lL

V.

C\J

yj

y
'1

<l

NT*

~^

s^

o- -cf

^-^>-

-o

~^~(S
HI

-o

oT6.0

5.0

<

4.0

/V = =

100 Omi

.si

i_'o-|

CO

*>

It

3.0
J

2.0

'\

; N^

w_y

^> y~-0<

1.0

^.>o

'"o --CX.

-O

-o

"

-o-o

6.0

# = 135 D

^
5.0

mir

1
'

4.0

3.0

1\
1

2.0

1^

*\ <x
> krw-^f

o-

- ^=-

7
8
9
10
11
Distance from foundation

12
7,

13

14

15

16

Fig. VIII-35a. Variation along a longitudinal profile of the horizontal

vertical surface soil vibrations before (open circles)
pile driving (see Fig.

vertical

VIII-37, foundation

17

18

19

and

component

of

after (black circles) sheet-

7).

and horizontal longitudinal vibration components obtained

prior

to sheet-pile driving.

The same figures indicate by dashed lines the theoretical curves computed from Eq. (VIII-3-41) for a given value of the damping constant a
of the waves.
It is seen from these graphs that the theoretical curves coincide fairly

PROPAGATION OF
6.0

WAVES

ELASTIC

IN SOIL

387

.1

Si

5.0

4.0

i\

= 1620

mm"

i\

3.0

i\

2.0

i
i

a.

^jO d^T^r-""

>-o

7.0
,<->-

6.0

'

5.0

= 18$ lOm n

'

4.0

s
i.o

2.0

IV

--A

I*

4_

<L>

3.0

-o-- --CX

- "9

>-_
D-"~

'

o-^"

>o

^10.0

<

9.0

<

/!/

215 Omi n"

8.0
7.0

I
I

6.0

\l

5.0

4.0

l\

II

\
I

I
*

3.0

b~w <*

2.0

\-

s,

V*

'
I

-r^

1.0
I

CX

__

^^

Fig. VIII-356.

well

7
8
9
11
10
Distance from foundation

Same

12
7,

13

14

15

16

17

18

19

as Fig. VIII-35a, but for other vibrator speeds.

with experimentally established points,

with the exception of

measurements at a frequency of 2,150 oscillations per minute, when the

computed amplitudes for average distances from the source of waves
differ considerably from the experimentally established values.
The
decrease in amplitudes of the vertical component of vibrations with an
increase in the distance from the source of waves has a more or less
monotonous character which in some places is distorted, apparently by
the interference of waves.

DYNAMICS OF BASES AND FOUNDATIONS

388

The

component changes with an increase

horizontal longitudinal

in

distance from the source of waves according to a law which has not been
established sufficiently clearly as yet.
oscillations

1,000

change
10 to 15

much

At frequencies

of 600, 800,

and

the longitudinal component does not

per minute,

along the profile; therefore the amplitude at a distance of

m from the foundation

close to the foundation.

is

approximately of the same order as that

When

the frequencies are higher (1,350 oscillations per minute and more) there may be observed a more or less regular
5.0

o c

4.0

!1

<d"q.I

3.0
(a)

2.0
1

\f\

1.0

o^>

r
E 6.0

o.E|

5.0

4.0
r

"l

3.0
I

<

>

2.0

1loV

(b)

1.0

"S
1

""*

r ^-L
*-*

B"

~1
>

5
6
7
8
9
Distance from foundation

10
7,

12

11

13

14

15

Fig. VIII-36. Variation along a longitudinal profile of (a) the vertical, and (6) the
horizontal components of surface soil vibrations for induced horizontal transverse
oscillations of foundation 7 (Fig. VIII-37) before (open circles)

and

after (full circles)

sheet-pile driving.

decrease in the horizontal vibration amplitudes, especially at relatively

In some cases the amplitudes

do not decrease with an increase in distance, but on the contrary they
= 1,890 min-1 ).
= 1,000 min-1 and
grow (for example, see data for
In addition, an investigation was made of the distribution of amplitudes
along the same profile when the foundation underwent transverse forced
The frequency of these
vibrations in a plane parallel to the profile.
oscillations was taken to equal the frequency of natural transverse
-1
oscillations of the foundation (some 300 min )Under these conditions
small distances from the source of waves.

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

the form of the foundation vibrations did not differ

389

much from

rocking

vibrations.

Graphs of Fig. VIII-36 present results of these vibration-amplitude

measurements over the soil surface. Small open circles indicate the
experimentally obtained amplitudes of the vertical (Fig. VIII-36a) and
Unlike the
horizontal (Fig. VIII-366) components of soil vibrations.
of
the
foundation
(source
of
waves),
vibrations
a monotocase of vertical
of
horizontal
component
of
soil
change
with
distance
the
vibrations
nous

Linear sheet piling

4.10 -5.20m deep

10.0
I

II II II

II

3.4-

TTTTTT

Closed
sheet

Foundation 7

piling-

Foundation 8

X-

Fig. VIII-37. Plans of foundation 7 (used in the experiments described

by

Figs,

VIII-34 to VIII-36) and foundation 8 (used in the experiments described by Figs

VIII-39and VIII-40).

was observed here; the

vertical

component remained constant at almost

all points.

After the distribution of amplitudes had been measured over the

soil

and transverse vibrations of the foundation, the sheetpiling was driven, for which 6- to 7-m-long timber beams having a cross
The sheetpiling was driven
section of 20 by 20 cm 2 and more were used.
In plan it
at a distance of 1.7 m from the foundation (source of waves).
was located perpendicular to the profile along which the amplitude distribution was measured.
Figure VIII-37 shows the arrangement of the
sheetpiling; the total length of the linear sheetpiling in plan was 10 m
The depth to which
(5 m in each direction from the source of waves).
individual piles were driven varied within the range of 4.10 to 5.2 m.

surface for vertical

DYNAMICS OF BASES AND FOUNDATIONS

390

Jetting

was used during the

The space between

sheet-pile driving.

individual piles at the soil surface reached 5.0 cm.

Of 44 piles, 1 was
driven to a depth of 6.10 m, and 2 to a depth of 2.7 m.
Measurements of the distribution of amplitudes for vertical vibrations
were performed along the same profile before and after
The frequencies in both cases were approximately
The amplitudes of soil vibrations were also measured when
the same.
the foundation underwent transverse vibrations under conditions of the
-1
close to 300 min
first resonance
The results of these measurements are given in Figs. VIII-34 to VIII-36.
Small full circles designate amplitudes of soil vibrations after sheet-pile
driving; their curve is a continuous line.
It is seen from these graphs
that the curves describing changes in vibration amplitudes with distance
from the foundation undergo some distortions. While prior to sheet-pile
of the foundation

sheet-pile driving.

driving these curves had a

more or

driving they are not monotonous

monotonous character, after the

anymore but are characterized by

maximums and minimums.

This distortion in the distribution

several

of amplitudes

is

found at

all

less

frequencies.

At points located in front of the sheetpiling, the amplitudes change

very intensively. This is illustrated by Table VIII-6, which gives amplifrom the
tudes of vertical soil vibrations at distances of 1.0 and 2.0
foundation center line; both points are located in front of the sheetpiling.
Table VIII-6 shows that over a distance of 1 m, the amplitudes on the

average decrease to one-half their value. At frequencies of 1,620, 1,890,

and 2,150 oscillations per minute, maximums are observed at a distance
of 1.5

from the foundation; thus the vibration amplitudes at these

1 m from the foundation.

points are larger than at a distance of

Table

VIII-6.

Decrease of Vibration Amplitudes over a Distance of

Amplitudes,

mm X

200,

600
810
1,000
1,350
1,620
1,890
2,150

1.0

7.4
7.6
8.6
8.5
7.4
3.1
5.0

when
Ratio between

N, min -1
r

2.0

amplitudes

2.1

3.6
4.0

1.9
2.1

4.1
2.8
4.1
3.1
3.9

3.0
1.8
1.0
1.5

The smallest vibration amplitudes beyond the sheetpiling were

from the source, i.e., at a
recorded at a point located at a distance of 2.5
from the
distance of 0.5
from the sheetpiling, and a distance of 3

PROPAGATION OF

WAVES

ELASTIC

IN SOIL

391

foundation center line. This indicates that during wave propagation

sheetpiling undergoes vertical vibrations at a smaller amplitude than the
surrounding soil. To show how the vibration amplitudes decrease
immediately beyond the sheetpiling, Table VIII-7 gives amplitude values
before

and

after sheet-pile driving.

It is seen

from Table VIII-7 that with an increase in the number

of

with a decrease in the wavelength, the sheetpiling

for points located directly behind it.
effective
A decrease
became more
vertical
vibrations,
caused
by
the
installation
of the
in the amplitude of
sheetpiling, was observed for all frequencies investigated in the course of
Therefore it may be stated that linear sheetpiling of
the experiments.
decreases the amplitudes of vertical soil vibrations
studied
dimensions
the
oscillations,

i.e.,

during vertical vibrations of the foundation, at least for

and

600 min -1

greater.

Table

VIII-7. Effectiveness of Sheet-pile Screen (Fig. VIII-37)

at Higher Frequencies of Vibration (Foundation 7)

Amplitudes

N, min

mm X

20
Ratio between

Prior to sheet-

After sheet-

pile driving

pile driving

4.5
6.5
6.5
4.6
4.6
6.3
7.0

2.0
2.9
2.9
1.5
1.6

600
810
1,000
1,350
1,620
1,890
2,150

After the amplitudes of vertical

amplitudes

2.2
2.2
2.2
3.1
2.9
5.7
6.4

1.1
1.1

soil

vibrations attain a

minimum, they

begin to increase with an increase in the distance from the sheetpiling,

and hence from the source

attain a

and

maximum

For

at a distance of 1.0 to 1.2

600 to 1,620 min -1 they

m;

for

1,890

min-1

-1

min
at 4 m.
from the graphs of amplitude distribution that at a distance

2,150

It is seen

of waves.

behind the sheetpiling (corresponding to the maximum),

between the vibration amplitudes measured
before and after sheet-pile driving was observed at the smallest number
= 600 min-1 At wavelengths corresponding
of oscillations, i.e., when
of 1.0 to 1.2

the

smallest

difference

to this
soil

number

of oscillations, a decrease in the amplitudes of vertical

vibrations after

(as

compared

to

observed only immediately behind the

before)
piling.

sheet-pile

driving was

At a distance

of 1.0

DYNAMICS OF BASES AND FOUNDATIONS

392

from the sheetpiling and

further, the difference in amplitudes before

and

may

be considered that at these distances

the sheetpiling does not have a noticeable effect on the decrease in
amplitudes. Thus the zone of decreased amplitudes, or "screened zone,"
= 600 min-1 extends behind the
formed by the sheet piling when
after driving

so small that

is

it

screen only for some 0.2 m.

When
much

higher frequencies are studied, the depth of the screened zone

is

Table VIII-8 gives values of the depth of the screened

zone for different values of N. The depth of the screened zone is taken
as the distance from the sheetpiling at which the amplitudes after
sheet-pile driving are no larger than one-half the amplitudes at the same
points measured prior to sheet-pile driving.
larger.

Increased Depth of Screened Zone behind Sheetpiling

at Higher Frequencies for Foundation 7 (Fig. VIII-37)

Table

VIII-8.

N, min"

Depth of screened

When

2,150

zone,

0.2-0.3
0.5-0.7
0.5-0.7
2.5-2.7
2.7-3.0
10.0-12.0

600
810
1,000
1,350
1,620
1,890

min-1 the amplitudes

,

of soil vibrations decrease in

the area directly behind the sheetpiling; for example, at a distance of

0.7

m, the amplitude

was oneAs the distance from the sheet-

of soil vibrations after sheet-pile driving

fourteenth the amplitude before driving.

piling grows, amplitudes increase,

and within the range

of distances of

m, the ratio between the amplitudes before and after sheet-pile

driving changes within the range of 1.0 to 1.75.
Thus, at these distances,
4.2 to 9.5

when the number of oscillations equals 2,150 min-1 the sheetpiling does
Hownot much affect the decrease in the amplitudes of soil vibrations.
ever, as the distance from the sheetpiling increases further, a new sharp
,

decrease in amplitudes occurs after the driving, so that at a distance of

10 to 13

m from the piling the ratio of amplitudes before and after sheet-

pile driving is of the

order of 10 to 12.

Thus, due to a varying distribution of

soil

vibration amplitudes after

sheet-pile driving, the following zones are observed in the area adjacent
first, directly behind the sheetpiling, a zone is found
where a noticeable decrease in amplitudes is observed; then follows a zone
of relatively small influence, and further again a zone of considerable
influence.
The presence of similar zones of decreased and increased
= 2,150 min-1 but also
screening effect is noticeable not only when
at lower frequencies.
However, the differences in amplitudes then are

to the sheetpiling:

smaller.

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

393

Table VIII-8 shows that with an increase in the number of oscillations,

with a decrease in the length of propagating waves, the depth of the
screened zone increases, and consequently the sheetpiling is effective at

i.e.,

larger distances.

Summarizing the foregoing discussion with regard

to the influence of

the dimensions studied for vertical vibrations of the

source of waves, the following may be stated concerning the distribution
of amplitudes of the vertical component of soil vibrations at different
line sheetpiling of

numbers

of oscillations,

at different wavelengths: the sheetpiling

i.e.,

causes a decrease in amplitudes at points located directly behind

it

for

600 min -1 and greater. However,

-1
= 600 min
the depth of the screened zone is very small,
when
As the number of oscillations
being of the order of 20 to 30 cm.
increases, the depth of the screened zone also increases (at the same
time, the amplitude directly behind the sheetpiling decreases), and when
= 1,000 to 1,620 -1 the depth of the screened zone is some 2.5 to
= 810 to 1,000 min-1 the depth is some 0.5 to 0.7 m;
3.0 m; when
= 1,890 min-1 the depth of the screened zone extends
and when
behind the sheetpiling. Hence, if a foundation which
some 10 to 12
should be protected from propagating waves has a width exceeding
0.5 to 0.7 m, its total area in contact with soil will lie within the screened
equals or exceeds 800 to 1,000 min -1
zone if
If the oscillation numbers of the propagating waves are smaller than
those indicated, then the sheetpiling will screen only part of the foundaIf oscillation base contact area which is to be protected from waves.
-1
then the contact dimensions of
tion numbers are less than 1,620 min
the protected foundation should not be larger than 2.5 to 3.0 m.
All that has been said above concerning the influence of linear sheetpiling on the distribution of amplitudes over the soil surface refers to the
vertical component of soil vibrations under vertical vibrations of the
all

frequencies studied,

i.e.,

foundation; the influence of sheetpiling on the horizontal longitudinal

component is much more complicated. This is why no regularity has

been established in the distribution of amplitudes of the horizontal
component after sheet-pile driving. When
was equal to 600 min -1 the
amplitudes directly behind the sheetpiling were reduced to approximately
one-half the previously observed values; then, to a distance of some 8 m
from the sheetpiling, vibration amplitudes were about the same before
and after sheet-pile driving. At distances larger than 8 m, there followed
again a zone of relative decrease, where amplitudes of vibrations were
1.5 to 2 times larger before than after sheet-pile driving.
When was 810 min-1 in the 8-m zone, the sheetpiling induced, not
a decrease in amplitudes, but even an increase in the amplitudes of
horizontal soil vibrations; at distances larger than 8
from the sheet-

DYNAMICS OF BASES AND FOUNDATIONS

394

piling, there followed

a relative decrease in the vibration amplitudes:

prior to driving the amplitudes were approximately 1.5 to 2 times larger

than the amplitudes at the same points after driving.

was 1,000 min -1 the sheetpiling influence on horizontal

vibrations was about the same as when N was 810 min -1
When oscil-1
lation numbers were larger than 1,000 min
in all cases, the sheetpiling
caused a decrease in the horizontal component in a zone extending directly
behind it for 1.5 to 2.5 m; then followed a zone of increase in amplitudes,
where absolute values exceeded the values of amplitudes prior to sheetpile driving; finally came a zone where the sheetpiling had only an
insignificant influence on the amplitudes of horizontal longitudinal soil

When

vibrations.

In comparison with the influence of sheetpiling on the amplitude of

the vertical component,

component
tion

is

its

influence

on the horizontal longitudinal

small therefore the use of sheetpiling as a screening installa;

may

vertical

be recommended mainly for the purpose of decreasing the

component of soil vibrations when the exciting source of waves

also undergoes vertical vibrations.

Such a decrease

of vertical vibration amplitudes

cases in which the source of

waves causes

is

necessary in some

vertical vibrations of founda-

tions of adjacent structures f at impermissible amplitudes.

In addition to the investigation of the distribution of amplitudes of the

horizontal and vertical vibration components along a profile perpen-

an investigation was also performed of

the distribution of amplitudes induced by vertical vibrations of the source
Measurements were
of waves along profiles parallel to the sheetpiling.
made for oscillation numbers equaling 1,620, 1,890, and 2,150 min-1 at
points directly adjacent to the sheetpiling, both in front of and behind it.
The results of these measurements are presented in the form of graphs
(Fig. VIII-38) which characterize changes in soil vibration amplitudes in
front of and behind the sheetpiling.
These graphs reveal the complexity
It can be
of the law governing amplitude changes around the piling.
dicular to the line of sheetpiling,

seen that the screening effect of the sheetpiling changes along

In

all cases,

the ratio between the

and behind the sheetpiling has the

quarters of

its

length.

decreases rapidly and

its

Toward

soil

largest value along the

the

its

two central

edges of the piling, this ratio

value differs relatively

little

from unity, which

indicates a decrease in the screening effect close to the edges.

basis of these data, the conclusion
piling to protect a structure

is

See Art. IX-3.

On

the

possible that the design of sheet-

from waves propagating through the soil

is greater than the length of

should provide that the length of the piling

the protected structure.
f

length.

vibration amplitudes in front of

PROPAGATION OF

The

ELASTIC

WAVES

IN SOIL

395

on the distribution of soil amplitudes

under conditions in which the foundation underwent
transverse vibrations at the same frequency before and after sheet-pile
driving.
Amplitudes of soil vibrations, obtained as a result of these

was

influence of line sheetpiling

also studied

Horizontal component

/V

Vertical

=162.0 min

component

20
Behind sheetpiling .

10
O

'

"

kr*2\

O^o5

5-O-C

r
5

7jy<>O 3
(

10

ront of
she et piling -

20

o
o
c\J

N = 1890 min"

X
E 20
E
10
co

Behind sheet^ ipi/ing-y

o
'a

I>
-

>
;

10

20

XL7 ^s

s^_
_^j

<>

V,

30

V,
\J

"40

iI

_J

<*-

<

'

-In front of

50

Behind sheet piling

J
V

k,

Y
1/

Dm

^- -^

->

<\

!5

t-l

'
!

10

20

In fron tofs heet piling -J

V V/

60

"o.

<

2150 min"

/V

40
Behind sheet piling

30

^y \

f 7 \\
^
-\ /

20

10

t
'

10

Behind sheetpiling-

Dm

Tr s>4^n
1

2/

Vj

K<>^>^

I>

kK

/s

8/

ii

>

-//? front of

20

r'

30

40

{^

'A

In front o
sheet pili ng
I

50

Fig. VIII-38. Variation of vibration amplitudes in front of and behind the sheet-pile
line (Fig.

VIII-37) at different oscillation frequencies of foundation

measurements, are plotted as small

7.

These
VIII-36.
have any influence on the

full circles in Fig.

results indicate that the sheetpiling did not

component of vibrations over the soil surface.

The horizontal component of soil vibrations was found at all points to be
somewhat smaller after than before sheet-pile driving.
Some influence on the decrease of horizontal soil vibration amplitudes
distribution of the vertical

DYNAMICS OF BASES AND FOUNDATIONS

396

was shown by the

fact that the amplitudes of the horizontal

component

of transverse foundation vibrations after sheet-pile driving

were somewhat
smaller than those prior to driving; an average value of the amplitude
prior to driving was 1.6 mm; after driving it was 1.57 mm; i.e., the former
was 1.04 times larger. However, the amplitudes of soil vibrations after
sheet pile driving decreased much more intensively (approximately 1.3 to
1.5 times), and this decrease was noticeable at all distances from the
foundation, both in front of and behind the sheetpiling.
This leads to
the assumption that the decrease was not caused by a screening effect,
but by some other reasons not taken into account during the experiment.
If this is so, then the conclusion is possible that the presence of the sheetpiling did not have in this case any significant effect on the distribution of
the amplitudes of horizontal longitudinal soil vibrations.
A closed sheet-pile wall (Fig. VIII-37) was installed around foundation
The sheetpiling had in plan
8 with a 2- by 2-m area in contact with soil.
The depth of the sheet piles,
the shape of a square with 3.4-m-long sides.
as in the previous case, was in the range of 4 to 5 m.
Prior to sheet-pile driving, an investigation was conducted of the distribution of amplitudes of the vertical and horizontal components of soil
vibrations along a selected profile.
Soil conditions were different here.
While the profile of the line sheetpiling ran across yellow medium-grained
sands of a ridge, the profile on the site of the closed sheetpiling crossed
gray water-saturated sands with some organic silt.
Measurements were performed onty for vertical vibrations of the
foundation when the oscillation numbers successively equaled 600, 800,
1,220, 1,620, 2,000,

and 2,280 min-

1
.

Figure VIII-39 presents the results of these measurements in the form

of graphs

showing the changes of

soil

vibration amplitudes as a function of

distance from the foundation (source of waves).

Figure VIII-39a and b

presents curves describing the vertical vibrations; Fig. VIII-40a and b

Reduced values of the vibration amplitudes are shown by small circles. As before, the value of the reduced
amplitude is taken to equal 0.30 mm. The amplitudes of soil vibrations,
recorded by the vibrometer, were reduced to this amplitude.
The following may be stated in regard to the graphs showing the relaAs
tionship between the amplitudes and distances from the foundation.
in the case of wave propagation from foundation 7 (when line sheetpiling
was investigated), the amplitude of the vertical component of soil vibrations changes with distance from the foundation fairly monotonously, espeWhen N = 2,000 and 2,280 min-1
cially at low and medium frequencies.
the changes in vibration amplitudes are somewhat more complicated,
probably due to the influence of wave interference.
Theoretical curves of changes in the amplitudes of the vertical comrefers to horizontal vibrations.

PROPAGATION OF

ponent of

soil vibrations,

WAVES

ELASTIC

600, 800, 1,220,

that theoretically obtained values of

fairly well

397

computed from Eq. (VIII-3-41) which takes

wave energy are shown by broken lines
and 1,620 min" 1
It is seen from these graphs

into account the absorption of

for

IN SOIL

with experimental values.

soil

vibration amplitudes coincide

As

in the investigation discussed

previously, the relationship between the horizontal

vibrations
plicated.

component of soil
and the distance from the foundation is much more comThe measurements performed did not provide sufficient data

for generalizations concerning the nature of the decrease in

of the horizontal

component with changes

in distance

amplitudes
from the source of

waves.
After the measurements of the distribution of vibration amplitudes
along a profile running from the foundation (source of waves) had been
completed, this foundation was surrounded by timber sheetpiling of the

type described above.

Then measurements of

the distribution of amplitudes of the vertical and

made along the

Figure VIII-39 gives reduced

horizontal longitudinal components of soil vibrations were

same

profile

and

for the

same

values of the amplitudes of

circles joined

The curves

frequencies.

soil

by a continuous

vibrations

shown by means

of small

open

line.

component
from the source of waves after sheet-pile
driving, for all oscillation numbers, are not monotonous, with the exception of the smallest value investigated.
These curves have a complicated
shape with maximums and minimums located at different distances from
of the distribution of amplitudes of the vertical

of vibrations versus distance

The

the foundation.

first

minimum

lies,

as a rule, directly behind the

sheetpiling; then, at a distance of 1.5 to 2.5

m from the foundation,

there

comes a maximum.

As the distance from the foundation grows, the

amplitudes reach their smallest value as compared with all amplitude
values along the investigated profile section.
The location of this
"principal minimum" depends on the frequency.
The following relationship was established between the number of oscillations and the distance
from the foundation of the "principal minimum."
T
,

min

1,220
1,620
2,000
2,280

Distance of "principal minimum'

from foundation, m

6.4
4.4
3.0
3.7

Hence, with an increase in the number of oscillations, there is a decrease

from the foundation to the principal minimum.
At all frequencies, the amplitudes of the vertical component of soil
vibrations decreased after the enclosing sheetpiling was installed.
Table

in the distance

12.0
11.0

10.0

600 mi f

9.0
11

8.0
7.0

\
1

6.0

5.0

\
1

\n

4.0
3.0

^,

*.. ~.

2.0

1.0

*-0

3
1

13.0

12.0

11.0

,10.0

Af = i300 min

9.0
8.0

\
1

7.0

"

6.0

\\

5.0

4.0

i
i

u
*

3.0

2.0
1.0

^*J

\P-VK,

""""

J \T

,_.

r-P

'

<;

14.0
13.0

12.0

11.0

N = 15>_20

mirr

10.0

(01

9.0

>

8.0
7.0

SI

6.0 *!

s
\

5.0 si

4.0

Vn

91

3.0

2.0

1r^

1.0

[""

L\

>__.

BT

(J

10

11

12

13

14

15

Distance from foundation 8,

16

17

18

19

20

21

22

Fig. VIII-39a. Variation of vertical vibration amplitudes as a function of distance

from vertically

oscillating foundation 8 (see Fig. VIII-37) before (broken line)

after (full line) driving the sheet-pile enclosure.

398

and

14.0

<

-r

13.0

l
1

12.0

i
i

A/

11.0

1620

min"

\
1

10.0

\
1

8.0

\
1

7.0

6.0
1

5.0

4.0

v
'

-^

3.0

r-..

c)

2.0

o-

1.0

16.0

15.0
14.0

//

= IOC Orr

in

'

13.0
12.0

3
gn.ob
i

x 10.0 Jr
_
E 9.0k
.
Q)

8.0

7.0
l

=t 6.0 \

5.0

4.0

*.

/& 1/

t
\/
r

"

ii

3.0

\
i

2.0
I

1.0

*"

>

14.0
"?

13.0
i

12.0
1/

11.0

= 226JO

nin"

/
\

10.0

9.0

8.0

7.0

6.0

A
'

\
\

ij

5.0

4.0

3.0
2.0

A\
'x>-

1.0
I

>

VV
23456789

>-

10

12

13

14

Distance from foundation

Fig. VIII-396.

Same

15

8,

<4 >

16

17

18

19

20

21

as Fig. VIII-39a, but for other vibrator speeds.

399

22

<

DYNAMICS OF BASES AND FOUNDATIONS

400

VIII-9 gives the values of the depth of the screened zone. As in the case
by "depth" is meant the length of the space behind the
sheetpiling in which the amplitudes of the vertical component after

of line sheetpiling,

do not exceed one-half the amplitudes observed prior to

sheet-pile driving

driving.

Table

VIII-9.

Effect of Oscillation Frequency of Foundation 8

on Depth of Screened Zone

(Fig. VIII-37)

Depth of screened zone formed by

N.
600
800
1,200

closed sheetpiling,

1.3
10

Screened zone was found along entire

from foundation.
distances of 10-12.5
Screened zone was found along entire
from foundation.
distances of 9-10.5

profile,

except for a section at

profile,

except for a section at

1,620

Along section at distances of 1.5-2.5

from foundation, the ratio of
amplitudes before and after sheet-pile driving was 0.5; screened zone is
distributed over rest of profile.
Screened zone is distributed over entire profile, except for a section at

2,000

2,280

m from foundation.

distances of 0.5-7.0

comparison of the depth of the screened zone, presented in Table

VIII-9, with the depth of the screened zone formed

by the

line sheetpiling

leads to the conclusion that the screened zone extends

behind closed than behind

much

further

line sheetpiling.

4.0

A/

600

min

3.0

2.0

-*

\\

-o

1.0

v>
I

4.0

/V= 800 min"

3.0

-2.0
1.0

1 Av

^k

K\

/>-

>f

<

"-l

Vr

91
ft

3.0

1.0

<^

2.0

>--
--<

5.0
4.0

ol

rf

LA

s> vw
<r
y^
-Ce nter

o- -s

/V

>

>--

( >

""""

line of

^v

th< J she et p ling

= 220'min-

7
8
9 10 11 12 13 14
Distance from foundation 8, m

15

16

17

18

19

20 21

Fig. VIII-40a. Variation of horizontal vibration amplitudes as a function of distance

from vertically

oscillating foundation 8 (see Fig. VIII-37) before (broken line)

after (full line) driving the sheetpiling.

and

PROPAGATION OF

c<

ELASTIC

WAVES

IN SOIL

401

13.0

12.0

11.0
/

10.0

A /=1

620

mirr

9.0

\
i

8.0

7.0
\
1

6.0

\
1

5.0

9i
/

4.0

3.0

2.0
1.0

\K
\

4 AS^kV A. K
Y

-o

>o-

>

-or

11.0

\
I

10.0

9.0

/ 1=1.000 mir

8.0
7.0

V
rl

1
1

-6.0

1!

a>.

5.0

4.0

si

3.0 -1
2.0

/v

uV

i\
#r

N^

/
1

-1

N.

^ >- ^

<ul

~~~

- -

1.0
1

9.0
i

8.0

6.0

5.0
4.0
3.0

7.0

rV

V\
V

1/

,F

1'

= 2

28C min

\
2.0
1.0

Fig. VIII-406.

Same

9 10 11 12 13 14
8
Distance from foundation 8,

15

16

17

18

19

20

as Fig. VIII-40a, but for other vibrator speeds.

21

22

402

DYNAMICS OF BASES AND FOUNDATIONS

can be seen from the graphs of Fig. VIII-39 that, as in the case of line
between the amplitudes of the vertical component
after and before sheet-pile driving increases with an increase in distance.
This means that the screening effect of the closed sheetpiling decreases
with an increase in distance from this piling.
On the strength of the results of the investigations, one may consider
that the protection of structures from vertical vibrations by the use of
closed sheetpiling will be effective only when the protected structure is
from the source of waves.
located at a distance not greater than 10 to 15
It was found that the influence of the closed sheetpiling on the horizontal longitudinal component of soil vibrations was much smaller than
As is the case before sheet-pile
its influence on the vertical component.
driving, changes in the amplitudes of the horizontal component of
soil vibrations are very complicated and do not lend themselves to
It

sheetpiling, the ratio

generalization.

For all the oscillation numbers investigated, at almost all distances

from the sheetpiling, the amplitudes of the horizontal longitudinal component were found to be smaller than the corresponding amplitudes
observed prior to sheet-pile driving. This shows that the use of the
closed piling is effective with respect to the reduction of horizontal
vibration amplitudes.
However, closed piling (as is line piling) is less
effective against horizontal vibrations than against vertical soil vibrations
induced by a source undergoing vertical vibrations.
d. Some Examples from the Practice of Screening Waves; General
Conclusions derived from experimental investigations of
Conclusions.
the possibility of screening waves propagating in soil contradict many
cases encountered in practice where screening installations did not fulfill
their purpose.
Therefore some authors are of the opinion that screens
are of no use.
A case was recorded where a group of buildings was insulated from
the street by a detached retaining wall having a length of some 400 m with
short breaks at the entrances.
Figure VIII-41 gives a diagram illustrating the general design of the retaining wall.
Its height was some 4.3 m;
the width at the base was 1.7 m.
The retaining wall and the buildings
were supported by wooden piles. The retaining wall was insulated from
the building by sheetpiling.
It was found, however, that in spite of the
presence of the retaining screening wall the vibrations produced by street
traffic were transferred to the buildings.
Several cases were recorded in practice indicating that the use of
trenches for the protection of buildings from shocks produced by machinis less expedient than the employment of other screening
For example, there was a case in which residential buildings
were separated from a plant by a river; in spite of this, it became necessary

ery foundations
devices.

PROPAGATION OF

ELASTIC

WAVES

IN SOIL

403

to insulate the machine shop of this plant because vibrations transferred

from the shop to the residential buildings were too great and had to be
In another case, a trench 7 m deep and 8.5 m wide was
decreased.
provided at a distance of 400 m from a plant in order to decrease shocks
created by the operation of a 5-ton crushing hammer; however, no positive results were obtained.
Following are four cases in which the use of trenches did not have a
noticeable effect on the decrease of vibrations beyond the trenches.
The First Case. The necessity arose to decrease vibrations produced
by horizontal reciprocating engines and transferred to adjacent buildings
For this purpose, a
located at rather large distances from the source.

mzwmwzm

W^ffizmmfm.

was not successful

an adjoining building.

Fig. VIII-41. This retaining wall

of traffic vibrations to

in preventing the transmission

m depth was provided between the source of vibrations

and the buildings; this, however, did not have any effect.
The Second Case. Here also no effect was obtained by providing a
trench of about the same dimensions as above to protect buildings from
vibrations produced by a horizontal reciprocating engine.
The soil
between the engine and buildings consisted of a mixture (fill) of sand and
trench of some 6

organic

silt.

A deep and wide river was located between the

foundation of a horizontal reciprocating engine and some buildings. In
The Third Case.

were subjected to considerable vibrations.

Trenches were installed, but were found to be of no

spite of this, these buildings

The

soil

was sandy.

use.

The Fourth Case. A deep trench was located between a foundation

under a vertical reciprocating engine and buildings subjected to vibrations.
This trench, however, did not prevent the transmission of vibrations

DYNAMICS OF BASES AND FOUNDATIONS

404

through soil which consisted mostly of organic silt. Thus the trench
was of no use in this case either.
Let us discuss one more case of the use of screening installations at one
of the plants in the U.S.S.R. in which again no positive results were
obtained.

-Sheet piling

Fig. VIII-42. General plan of attempted screening of a compressor plant.

compressor plant located among residential houses and auxiliary

buildings produced vibrations of several buildings at different distances

from

Of greatest nuisance were vibrations in 2 five-story apartment

intensive vibrations were observed in a building
from the compressor plant. In this buildlocated at a distance of 45
ing, vibrations were distinctly perceptible "to the touch." On the upper
floors, the vibration amplitudes naturally were larger than on the lower
it.

houses.

The most

raw
WMT^^y^W^
:0.64-H

Cross section

Plan

Fig. VIII-43. Details of attempted screening

floors;

Slag

shown

in Fig. VIII-42.

mm

they reached 0.3

on the fifth floor. An inspection of the outboth buildings did not reveal any cracks or deformations.

side surfaces of

In order to decrease the vibrations of the buildings, the decision was

taken to provide a screen formed by a trench filled with cinders and by
sheetpiling.
A diagram of the screen as designed is shown in Fig.
VIII-42; details of the design are given by Fig. VIII-43. The width of
The sheetpiling
the trench filled with uncompacted cinders was 0.64 m.
was to consist of 7-m-long wooden piles of 16 by 16 cm cross section. The
design provided 1 1 short arches supported by master pile clusters. Eight
arches were to be located parallel to the longitudinal wall of the compressor plant nearest the vibrated buildings; one arch was to be located

PROPAGATION OF
at the left-hand end,

ELASTIC

WAVES

IN SOIL

and two at the right-hand end

405

of the wall.

The

pile

wide section it had four piles; a

In the opinion of the authors of
long was provided behind.
trench 1
separate
provision
of
arches
with master pile clusters
the
design,
this
would contribute to an increase in the general rigidity of the screen; this
would increase the screening effect of the installation.
This project was not fully completed, since after eight arches, designated in Fig. VIII-42 by the numbers 1 to 8, had been constructed, it was
found that the finished section of the screen had no noticeable effect on
Since there was not much hope
the decrease of vibrations of the building.
that the installation of the remaining three arches would change the
situation, the decision was taken to stop construction of the screen.
Thus again in this case, the sheetpiling did not prove its value.
In order to understand why in all the foregoing cases the use of screens
cluster consisted of ten piles.

In

its

was ineffective, it is necessary to remember that the damping of vibrations

by a screen depends on the relationship between the smallest dimensions
of the screen and the length of the propagating waves.
At the same
time, the wavelength depends on the velocity of wave propagation and the
period of excitation.
Wavelengths for different periods of the source
and velocities of propagation are presented in Table VIII-10.
Table VIII-10. Wavelengths for Different Wave Velocities,
Periods of Waves, and Numbers of Oscillations of Source of Waves

Number

of

oscillations

of source,

Period
of

wave,

length of wave j, m, for

velocity, m/sec, of:

Notes

-1

sec

50
100
200
300
500

0.075
0.060
0.050
0.040

3.75
3.00
2.50
2.0

7.5
6.0
5.0
4.0

min

800
1,000
1,200
1,500

300

50.0

100

200

1.2

60

120

0.6
0.3
0.2
0.12

30
10

60
30
20
12

360
180
90
60
36

600
300
150
100

6.00

240
120
60
40
24
15

22.5
18.0
15.0
12.0

37.5
30.0
25.0
20.0

15

13

10

8.0

500

60

Low-frequency
machines (diesels,
compressors, gas
motors, etc.)

High-frequency
machines (mainly

turbodynamos and
hammers)

forge

Knowing the wavelengths

taking the

minimum

ratio

for different sources in different soils and

between screen dimensions and wavelength in

the presence of which the effect of screening of vibrations

able,
its

one

may compute

is

already notice-

the smallest dimensions of the screen, as well as

smallest depth, at which the effect of screening will be felt.

Experience with the installation of trenches and sheetpiling in

loessial

DYNAMICS OF BASES AND FOUNDATIONS

406

showed that these devices had a screening effect when the ratio
between the depth and the length of the propagating waves was about 0.3.
This lower limit is approximately the same in other soils as well. It

soils

follows that in order to secure the screening of vibrations,

to

make

the depth of the screen not less than 0.3L, where

it is

is

necessary
the length

of the propagating waves.

The depth

of a screening device should be

the source of waves.

The maximum

counted from the depth of

practical depth of the screening

is about 5 m, since machinery foundations are usually placed at a

and more.
depth of the order of 3
It appears that in most soils the velocity of propagation of surface
waves is about 200 m/sec.
Taking into consideration both these factors, it may be held that in
practice it is possible to damp vibrations with a number of oscillations
For example, in order to screen vibranot less than 600 to 700 min -1
tions propagating from a diesel running at 200 rpm when the velocity of
wave propagation in the soil is 200 m/sec, it would be necessary to install
a screen to a depth of at least 17 to 25 m; when the velocity of wave
propagation is some 300 m/sec, the depth of the screen should be increased
It is hard in practice to construct screening devices at such
to 27 to 30 m.

device

depths.

Thus

it is

impossible in practice to screen waves propagating from low-

frequency engines.

This

is

the reason for the failures recorded in the

damping by means of screens.

more than 600 to 700 rpm, it

practice of vibration

is possible to screen
For engines of
if the depth of the screening device equals 5 to 7 m.
Thus in most cases it is impossible to screen vibrations from such
Vibrations
engines as diesels, piston compressors, and gas motors.
propagating from high-frequency engines, as well as from hammer
foundations with fairly high natural frequencies, might be screened
without too much difficulty. When the propagating vibrations have a
whole spectrum of frequencies (for example, if vibrations are excited by
traffic), not all frequencies will be screened, but mainly the higher

vibrations

frequencies.

Trenches may be employed as screening devices only if there is no

ground water in the trench. If the ground-water level is close to the
surface, then there is no use employing trenches as screening devices.
In such cases, wooden or metallic sheetpiling might be used; if the piling
then it is recommended to provide rigid junctions such
as clusters of master piles.
Finally, if a sand of sufficient porosity is found on the site, then a screen
may be formed by chemical or cement stabilization of a narrow strip of
is

sufficiently long,

soil.

IX
EFFECT

ON

STRUCTURES OF

WAVES FROM

INDUSTRIAL

SOURCES

IX-1.

Classification of Industrial Sources of Elastic

Some 30

Waves

in Soil

was attached to
by seismic waves from
industrial sources.
At present these problems are very important, and it
often happens that the normal work of a whole plant depends on their
solution.
For example, cases are on record in which shocks, produced by
hammer foundations or by traffic and transferred to soil, led to failures
in the work of molding shops or precision machines.
Mining procedures
employing explosives often had to be changed because of the harmful
effects on structures of the seismic explosive waves.
Several examples
are cited in Art. IX-2 which show that waves propagating from lowto 40 years ago, no great practical importance

the effects on structures of vibrations produced

frequency engines

may

lead to considerable vibrations of high structures.

Studies of vibrations of residential houses, caused by waves from the

above sources, show that they have a very harmful physiological and
psychological effect on the residents of such houses.
As indicated in
Art. IX-3, seismic waves from hammer foundations in some cases produce damage in forge shops.
Thus seismic waves from various industrial sources may have a harmful effect on structures, technological processes, and people.
Therefore
the study of the action of seismic waves from industrial sources on structures is of great practical importance.
Equally important is the development of methods to decrease or eliminate the harmful influence of seismic
waves.
407

DYNAMICS OF BASES AND FOUNDATIONS

408

With

respect to their actions on buildings, industrial sources

may

be

divided into the following main groups:

Machinery located within the buildings on floors of one or several

supported by special constructions such as cantilevers tied to
Machinery of this type includes
the main elements of the building.
machines in spinning and weaving mills, transmissions, printing presses,
and others. The absence of strict synchronization and periodicity in
work is a characteristic feature of these engines. They produce local
vibrations of structural elements, e.g., of separate floor spans, of some
columns, and of wall sections. These vibrations are not periodic and
have a complicated nature due to the superposition of vibrations and
shocks which are produced by many sources and which are often unstable
Vibrations produced by the type of engines under consideration
in time.
1.

stories, or

are often transmitted through the soil to adjacent structures.

Powerful high-frequency engines, such as turbogenerators, turboSuch machines usually have only rotating
parts of high static and dynamic stability running at more than 500 to
These machines may cause only local vibrations of some
1,000 rpm.
structural elements; the amplitudes of these vibrations are very small,
No cases of impermissible vibrations
not exceeding 0.05 to 0.10 mm.
caused by such machines have been recorded in practice. Also no cases
are known of vibrations of entire structures having been caused by the
action of high-frequency engines, even where these engines were located
Thus it can be held that high-frequency engines
close to the structures.
This is partly
are safe from the point of view of vibrations of structures.
explained by the fact that vibration amplitudes of high-frequency engine
foundations are very small therefore the amplitudes of soil vibrations are
also small and even in areas adjacent to the foundation usually do not
exceed 0.1 micron (0.0001 mm).
From the point of view of the safety of structures with respect to their
reception of high-frequency waves, it is important that the natural vibration periods of structures be much lower than the working periods of highfrequency engines so that the phenomenon of resonance is prevented.
3. Automotive and railroad traffic exciting high-frequency vibrations
with unstable regime and complicated spectrum. A relatively large
vibration amplitude at the source is a specific feature of these vibrations.
Therefore in spite of the high frequency of such vibrations, they may
produce highly perceptible vibrations of some structural elements located
close to roads with heavy traffic.
Measures against such vibrations in
structures comprise an important contemporary task of public interest
2.

compressors, and turbofans.

in cities.
4.

Low-frequency engines with periodic stable regime of work, including

horizontal piston compressors, diesels, and reciprocating steam engines.

EFFECT

ON

STRUCTURES OF WAVES FROM INDUSTRIAL SOURCES

409

The speeds of these engines rarely exceed 500 rpm, and in the most
Engines of this type are more
powerful engines drop to 75 to 100 rpm.
dangerous for structures than any other machines with a steady work
regime.

Cases are

known

in

which foundations under such engines

produced vibrations of structures located at distances of several hundred

meters from the source.

This

is

explained by the fact that the periods of

natural vibrations of structures in most cases do not exceed the

periods of low-frequency engines; therefore resonance
result of

which vibration amplitudes

may become

is

minimum

possible, as a

very large.

vibration work, cases of vibrations of structures caused

In anti-

by low-frequency

engines are most frequent.

Forge hammers and other installations with impact actions on their

The amplitudes of foundation vibrations under such units
and more; the vibration amplitudes of soil close to a
may reach 1.5
hammer foundation also may reach large values. Hammers do not exert
a periodic action on foundations; therefore the vibrations of hammer
foundations have an unstable character with respect to time. Waves
propagating through soil from a hammer foundation in practice do not
cause resonance with the natural frequencies of a building, especially
since these foundations induce waves always higher in frequency than
500 periods per minute. Waves induced by hammer foundations may
lead to dangerous shocks only in structures located in the immediate
5.

foundations.

mm

vicinity of these foundations.

Vibrations of hammers are dangerous because they lead to additional

nonuniform settlements caused by waves propagating from their foundaVibration amplitudes of footings under walls or columns of
tions.

same order
which in the vicinity of hammer
foundations may reach fairly high values. These vibrations of wall or
column footings are characterized by fairly large amplitudes and result
in dynamic pressure on the soil which may attain considerable magnitude
and may act in addition to static pressure. Under certain conditions this
dynamic pressure may cause considerable settlement of walls or columns,
structures located close to the source of vibrations are of the
as the amplitudes of soil vibrations,

damage to a building (see Art. IX-3).

Explosions inducing waves, the nature of whose action on structures

often leading to serious

6.

has much in common with waves propagating from foundations under

impact-producing machines.
IX-2.

Action on Structures of Seismic

Waves Propagating from

Foundations under Low-frequency Machinery

Resonance as the Main Cause of Vibrations of Structures. In most
which foundations under low-frequency machines cause vibrations of structures, the foundations themselves undergo vibrations at
a.

cases in

DYNAMICS OF BASES AND FOUNDATIONS

410

Seismograph measurements in areas

columns of a vibrating structure show that
the amplitudes of soil vibrations are ten or a hundred times smaller than
Thus, the energy of waves
the amplitudes of vibration of the structure.
which sometimes cause dangerous vibrations of structures is very small.
In spite of this, amplitudes of structures receiving the waves may reach
considerable values, sometimes several millimeters.
If the magnitude of excitement is small, the vibration amplitudes of a
structure, and consequently the stresses therein, may reach large values
only when the structure is in resonance with the excitation; then the
damping forces acting in the structure are small in comparison with the
entirely permissible amplitudes.
close to the footings of walls or

elastic forces.
0.40
p

0.30

0.20

:i

0.10

100

120

140

160

180

Rpm
Fig. IX-1. Resonance curve of a 45-m-high structure.

Let us discuss two cases from practice which illustrate how structures
by a coincidence of one of the natural frequencies of the
structure with the frequency of the propagating waves.
The First Case. Vibrations of the foundation under a steam engine

are influenced

The vibration
of 650 hp caused vibrations of a 45-m-high structure.
amplitude in the upper part of the structure was 0.37 mm; it decreased
downward, and on the second floor the vibrations were so small that they
could not be perceived.
The vibrations of the structure were measured within the range of 55
For this engine, the normal speed is 150 rpm. Figure IX-1
to 155 rpm.
gives a curve of the relationship between the vibration amplitude of the
structure and the speed of the engine, plotted on the basis of results of
measurements. It is seen from this graph that the resonance curve of the

EFFECT

structure has

ON

STRUCTURES OF

WAVES FROM

INDUSTRIAL SOURCES

411

an extremely sharp peak with a maximum corresponding

to

If the speed increases to

the operating speed of the engine (150 rpm).
155 rpm, i.e., becomes larger by 3 per cent than the speed at resonance,

then the vibration amplitude is only 30 per cent of the amplitude corresponding to 150 rpm; if the speed decreases to 140 rpm (by 6.5 per cent),
then the vibration amplitude is only 7 per cent of the vibration amplitude
at resonance.

from Fig. IX-1 that in the case under consideration the

structure was at resonance with waves propagating through soil from
In addition, it is seen from the graph that the
the engine foundation.
vibrations
zone
of
is located within a very narrow range of
resonance
values
do
not
differ by more than 3 to 5 per cent from
whose
frequencies
It is seen

the resonance frequency of the structure.

Such a sharp resonance curve shows that the damping

forces of the

structure corresponding to the low frequencies of vibration are very small

>3=2.23mm

T = 0.623 sec
/V=96.5 oscillations/min.

w = 10.10

sec

-1

r = 0.628 sec
/V=96 oscillations/min.
a;

=10.05 sec"

Fig. IX-2. Vibrograms of a 20-m-high structure during the operation of an adjoining

compressor.

in

comparison with the

elastic forces.

This

is

the reason

why

the struc-

ture reacts only to excitement having a period which does not differ

from

its

much

own.

The Second

Case.

The

vibrations of a frame structure of 20

height were caused by vibrating foundations under compressors operating

Figure IX-2 presents a diagram of vibrations
at a maximum of 100 rpm.

when one compressor was operating at a speed of 97 rpm

measured by a tachometer. Actually, the speed of the compressor
was not constant, but, due to some irregularity in speed typical of
The
reciprocating engines, changed within a range of 3 to 4 per cent.
of the structure

as

changes in vibration amplitudes of the structure, as recorded by the

vibrograph, are explained by the fact that the indicated speed of the

engine (97 rpm) was close to the lower natural frequency of the structure.
Due to the irregularity in the speed of the engine, the period of propagating waves at times approached the natural period of the structure and

from it. Thus the structure at times was in resonance

with waves propagating through the soil, and at times was not. The
vibration amplitudes recorded by the vibrograph at times decreased to
zero, then grew to values dangerous to the safety of the structure; this is
explained by the fact that even 3 to 5 per cent changes in the period of
at times departed

DYNAMICS OF BASES AND FOUNDATIONS

412

excitement,

by comparison with the period

of natural vibrations of the

an extremely sharp decrease in its vibration amplitudes.

The time intervals during which vibrations were not observed at all

structure, lead to

alternated with intervals of intensive vibrations of the structure; these

changes were irregular and the length of the intervals reached several
When the speed of the compressor decreased to 90 rpm, the
vibrations of the structure ceased completely.
Other similar cases of vibrations confirm that the resonance of structures with waves propagating from foundations under low-frequency
engines is the main cause of their considerable vibrations.
This is why
vibrating foundations under low-frequency engines often do not affect
minutes.

structures located in the immediate vicinity, but induce considerable

vibrations of structures located at a large distance.

The

latter structures

waves propagating through soil from foundations

under machines, while neighboring structures do not only because the
periods of natural vibrations of these structures differ from the period of

get into resonance with

the waves.

In order to decrease vibrations of a structure caused by low-frequency

In
it is necessary to take the structure out of resonance.
practice this is facilitated by the fact that the resonance zone of the
structure, corresponding to the low frequencies, lies within an extremely
narrow range of frequencies located close to its frequency of natural
vibrations.
Therefore it suffices to change the ratio between the exciting
engines,

by 3 to
and the vibrations of the structure will completely die away.
Changing the operating speed of the engine is the easiest way to change
the ratio between the frequencies of an engine and the natural frequencies
frequencies and the frequencies of natural vibrations of structures
5 per cent

The maximum amplitude of vibrations of a structure

under the most unfavorable conditions, and the decrease in the amplitude
of its vibrations with changes in engine speed, may be established from
the resonance curve of the structure obtained by means of a vibrograph
of a structure.

It is simpler to record the resonance

curve at the time of starting or stopping the engine. It is desirable to
record diagrams of vibrations when the speed of the engine is somewhat

for different speeds of the engine.

greater than the operational speed.

The speed

cannot always be changed, because this causes

When
its work.
becomes necessary to change the natural frequency

of the engine

a change in the power of the engine and in the regime of

this is inadmissible, it

This is generally done by increasing its rigidity. Here

the condition should be satisfied that the natural frequency of the vibrat-

of the structure.

This may
is larger than the operating speed of the engine.
be checked by measuring the vibrations of the structure at the time of
starting or stopping the engine.
If possible, this check may also be

ing structure

EFFECT

ON

STRUCTURES OF WAVES FROM INDUSTRIAL SOURCES

413

performed by calculation. The increase in the rigidity of the structure

must depend on the ratio between the frequency of natural vibrations of
the structure and the frequency of the engine, i.e., on whether the first
frequency

is

larger or smaller

than the engine speed.

For example,

established that the natural frequency of the vibrating structure

is

if it is

3 per

cent smaller than the operating frequency of the engine causing vibrations,
then, in order to decrease the amplitude of vibrations of the structure,
is

necessary to increase the rigidity of the structure so that

its

it

natural

frequency becomes 7 per cent larger than the initial frequency. If, however, the natural frequency of the vibrating structure is 3 per cent larger
than the operating frequency of the engine, then, in order to obtain the

same

effect, it suffices to increase

the rigidity of the structure so that

its

natural frequency will be larger by 2 to 3 per cent.

In many cases calculations of natural frequencies of structures may be

performed only with a high degree of approximation; therefore, in such
cases the measures to change the natural frequency of the structure should
provide for an increase in the frequency of its natural vibrations by some
5 to 10 per cent.

The

selection of measures for changing the natural frequency of the

vibrating structure depends on

in

it,

its

design and on the equipment installed

as well as on the design of the foundation under the low-frequency

engine.

We will now discuss several examples to illustrate the structural

methods employed for decreasing vibrations.
b. Measures to Decrease Vibrations of a Power-station Wall Induced by
Operating Saw Frames.
Structures adjacent to a sawmill underwent
noticeable vibrations created by waves propagating through soil from the
foundations under saw frames. Three frames ran at 320 rpm and one at
300 rpm. Vibrations were noticeable even on the other side of a river of
The foundations underwent mainly vertical vibrations with
8 m depth.
amplitudes of some 0.25 mm. The horizontal component of rocking
vibrations equaled only some 0.1 mm.
Due to an inexact coincidence of
the frame rotations, vibrations over the entire plant territory were of a
pulsating type.

The foundations under the saw frames were located in a layer of

The soil conditions were such that they
soil.
contributed intensively to the development of foundation vibrations and
water-saturated peaty

the propagation of vibrations.

In spite of this, although the vibrations

were noticeable on the territory of the plant, they did not
cause serious apprehensions.
However, during the construction of a
power station which had been designed as a wooden frame structure with
brick filling, considerable transverse horizontal vibrations of a 20-m-long
fireproof wall developed after it was erected to a height of about 4 m.

of structures

DYNAMICS OF BASES AND FOUNDATIONS

414

These vibrations reached such dimensions that it became dangerous to

continue the work. A brick placed on its narrow side would fall off the
wall, the set of the brick mortar was unreliable, and there was no assurance that the vibrations would not increase as the construction of the wall
proceeded further, and that it would not lead to the collapse of the wall.
The newly erected wall was joined to the existing structure of the
machine shop by interlocking masonary courses and was located at a
distance of approximately 17 to 20 m from the nearest foundation under a
saw frame. Measurements of vibrations showed that the wall vibrated
almost exclusively in a horizontal (transverse) direction. The curve of
the distribution of measured amplitudes along the center line of the wall
was bent in such a way that the vibration amplitude was smallest near
the wall edges, and the largest amplitude was found at a distance of
0.55

0.59
0.51

0.065

0.045

Fig. IX-3. Amplitudes (in millimeters) of transverse horizontal vibrations of a newly

it reached 4 m
frame engine operating 17 m away.

built brick wall, after

in height, as a result of resonance with a saw-

approximately 40 per cent of the total length from the end adjoining the
existing

The

machine shop.
vibrations on the
and the amplitude

soil

ceptible,

hundredths

of a millimeter.

site of

of

construction had been hardly per-

vibrations had not exceeded several

Therefore

it

may

be assumed that con-

siderable wall vibrations, as indicated above, appeared only because the

lowest natural frequency of the wall (corresponding to the shape of the

wall vibrations as

shown

in Fig. IX-3)

was

close to the operating fre-

quencies of the saw frames.

resonance,

it

Hence, in order to take the wall out of

was necessary to change its rigidity, which was done by

providing a reinforced-concrete belt and counterforts (Fig. IX-4).

At first two abutments were installed which divided the wall into three
uneven sections. However, these abutments did not exert any noticeable
influence on the vibrations.
Then a reinforced-concrete belt with double
The
reinforcement was installed along the entire length of the wall.
construction of this belt proceeded at a time
after

when

the sawmill did not

and consequently the wall did not vibrate. Three to five days
the installation of the belt was completed, the sawmill resumed

operate,

EFFECT

ON

STRUCTURES OF

WAVES FROM

INDUSTRIAL SOURCES

operations; the wall vibrations did not start again,

415

and the construction

was successfully completed.

Vibrations of a Boiler-shop Wall Induced by Operating Compressors.
almost analogous case of wall vibrations occurred due to the influence
3
/hr running at
of a horizontal piston compressor with a capacity of 60
were
unfavorable;
conditions
an
upper
layer of alluvial
167 rpm. Soil
and was underlaid by clays with
soil extended to a depth of 1.5 to 2
c.

An

some

silt

and sand and by loose sands.

The ground-water

level

was at a

Reinforced concrete
belt with double
reinforcement

Channel iron

20

Counterfort

^-Foundation beams

Fig. IX-4. Cross section of counterforts

the excessive wall vibrations

depth of some 3 m.

shown

and reinforced-concrete

in Fig.

belt used to eliminate

IX-3.

Although the foundation vibrations were noticeable

(especially horizontal vibrations in the direction of the piston motion),

they were within the range of permissible values, since their amplitude
did not exceed 0.2 mm.
The observed vibrations did not exert any
noticeable influence on machine operation or on the state of the foundation.
In spite of this, considerable vibrations of several structures were
observed in a zone with a radius up to 150 m. Extremely intensive
vibrations of one of the walls of the boiler shop nearest the compressors
were observed. This boiler shop was located at a distance of some 80
from the compressor plant. In addition to this structure, several other

structures located at various distances from the compressor plant also

DYNAMICS OF BASES AND FOUNDATIONS

416

vibrated, but with smaller amplitudes.

wall vibrations reached 1.0

As

The maximum amplitude

of

mm.

in the foregoing case, a reinforced-concrete belt

was

cast along the

entire wall in order to decrease the vibrations.

Curve of variation of
amplitudes with height

475

Amplitude, /z(microns)

Cross section

Fig. IX-5. Cross section of structure described in Art. IX-2-d.

d.

Measures

Foundations.

to

Decrease Vibrations Induced by Waves from Compressor

had plan dimensions of 33.0 by 22

A vibrating structure

and an average height of some 5.5 m; it was divided by a brick wall into
two sections. In the basement of one of these sections, five synchronously operating horizontal piston compressors were installed, each on a

Fig. IX-6. Plan of structure described in Art. IX-2-d.

Figures IX-5 and IX-6 show a cross section of the

schematic plan.
The vibrations of the building depended to a considerable degree on

separate footing.
building and

its

the season.

In most cases of vibrations of buildings, and of sources of

ON

EFFECT

vibrations as well,

STRUCTURES OF

an

WAVES FROM

INDUSTRIAL SOURCES

instability of vibration amplitudes

417

is

observed.

This can be explained by minute variations in the elastic properties of

materials which may cause minute changes in the natural frequencies
and thus (especially under conditions of resonance) may cause changes in
vibration amplitudes.

If this is

the case, then amplitudes which change

with atmospheric conditions are an indirect indication of the fact that

Changes in
the structure vibrates under conditions of resonance.

amplitudes of a structure's vibrations may depend on changes in soil

moisture and on fluctuations in the ground-water level, especially in areas
An increase in soil
close to the foundation and the source of vibrations.
moisture contributes to a decrease in the absorption by soil of the vibration energy; other conditions remaining equal, this leads to an increase in
the transmission of wave energy to the structure.
An investigation of the structure's vibrations by a vibrograph provided
the following data:

The

structure underwent vibrations mostly in the direction of motion of

In Fig. IX-6, the arrows show the main direction

to the longitudinal walls of the
structure; the walls are characterized by low rigidity in the transverse
Therefore, from the point of view of vibrations of the strucdirection.
ture under consideration, the direction of motion of the compressor
pistons (Fig. IX-6) is the least favorable, because the waves propagating

the compressor pistons.

wave propagation, perpendicular

of

from the compressor foundations exert maximum influence on the walls.

This fact is not often taken into consideration when designing plants with
engines inducing horizontal foundation vibrations (which
the

main cause

of structural vibrations).

tion the direction of piston motion

was

If in

may become

the case under considera-

parallel to the longitudinal axis

then the wall vibrations would undoubtedly have had

considerably smaller amplitude.
In general, it is necessary to avoid as

of the structure,

far as possible the installation of reciprocating engines so that the direcis perpendicular to high walls elongated in plan and
having no rigid sections.
Measurements showed that almost no vibrations were observed in the
transverse walls of the structures when the external and internal longitudinal walls underwent vibrations with an amplitude reaching 0.475 mm.

tion of piston motion

Points of measurements are indicated by numbers in Figs. IX-5 and IX-6.

Vibrations were of a sinusoidal nature, and a superposition of components
was observed at a frequency equaling the double frequency of machine
rotation.

The curve
Fig. IX-5,

shown

of variation of vibration

and the curve

in Fig. IX-6.

of vibration the

amplitudes with height

is

shown

of variation of amplitudes along the roof ridge

It is

in
is

seen from these curves that during the process

frames of the structure bend somewhat along their height.

DYNAMICS OF BASES AND FOUNDATIONS

418

Therefore the variation of amplitude with height has a curvilinear

If the amplitudes had grown proportionally to height, the
conclusion would have been possible that the frames do not bend during
character.

vibrations, but oscillate as a solid

body rocking around a horizontal

axis

located in the plane of the foundation surface in contact with soil or

somewhat below it.

The distribution

of amplitudes along the length of the central wall of

the building reveals that the presence of the transverse wall (Fig. IX-6)

There is almost
distorts the regularity in the distribution of amplitudes.
no difference between the amplitudes at points 1 and 0. At other points,
the distribution of amplitudes has a character which is usual for the
vibrations of walls of buildings.
The amplitudes increase toward the
center of the wall and attain a maximum there; then they decrease
toward the transverse wall; there they equal some 40 microns.
Thus it was established that the transverse wall decreases the amplitude
of vibrations of the ridge at the place where the wall and the ridge are
joined.
Hence, the installation of a similar rigid brick or reinforcedconcrete wall in the planes of other frames, for example, frame 4 or 3,
would considerably decrease the amplitude of vibrations. There is no
necessity to erect the wall along the whole height of the frame; it would
suffice to install the wall in the upper part of the frame and support it
with two or three columns.
In addition, vibration amplitudes of longitudinal walls of buildings
could be decreased by the installation of two or three heavy counterforts
outside the external walls.
Finally, vibration amplitudes could be
decreased considerably by the construction of extensions adjoining the
longitudinal external walls of the vibrating building and properly tied to

them.

shown

The

method was employed to stop vibrations of the building

IX-5 and IX-6. After the construction of these extensions

latter

in Figs.

adjoining the building, the vibrations ceased.

e. Measures to Decrease Vibrations
Waves from a Compressor Foundation.

of a Refrigerator Plant

The construction

Induced by

of the refrigera-

tor plant proceeded as follows: first the building was erected, then the
frame structure was placed within this building and separated from it by
thermoinsulation.
After the building was erected, it was found that one
of its walls underwent considerable vibrations which caused cracks at
several places.
This wall was standing perpendicular to the direction of
piston motion of the compressor located near it.
Attempts were made to decrease the wall vibrations by strengthening
the foundation under the compressor and thus decreasing the amplitudes

These measures did not help at all, and

During this
interruption, work on the construction of the frame structure continued.

of foundation vibrations.

operation of the compressor was temporarily interrupted.

EFFECT

ON

STRUCTURES OF

As soon as each story

of the

WAVES FROM

frame was

INDUSTRIAL SOURCES

built, it

was connected by

419
ties

with

reinforced-concrete antiseismic belts installed in the wall at the level of

each story of the frame (see Fig. IX-7). After three stories of the frame
were constructed, the compressor was set into motion again. No
vibrations of the wall were then observed.

External
wall

Projecting steel

reinforcement
tying the wall to

the frame structure

Mfmmmmmmm
53
Direction of wave propagation
from the compressor

Fig. IX-7. Cross section of refrigerator-plant wall described in Art. IX-2-e.

In this

case, the elimination of the vibrations is explained

by the

fact

that the wall went out of resonance with waves propagating through the
soil

due to an increase in

its rigidity after it

was joined

to the frame

structure.
IX-3.

The Action of Shocks Caused by

Hammer Foundations

Waves from

Unlike foundations under low-frequency machines with steady speeds,

hammers and other machines with impact action

foundations under forge

are characterized

by a considerable amplitude

tively high frequency (70 to 120 sec -1 ).

of vibrations

and a

rela-

induced by
shocks on hammer foundations differ greatly from vibrations caused by
waves from foundations under machines with a settled regime of work.
Figure IX-8 shows a typical seismogram of vertical soil vibrations
induced by a hammer operation; it indicates that, as in the case of
hammer foundation vibrations, the soil vibrations are of an unsettled
nature.
Because of this, even if the frequency of the building vibrations
Soil vibrations

420

DYNAMICS OF BASES AND FOUNDATIONS

coincides with the frequency of waves induced

by the hammer foundation,

the amplitudes of the building vibrations cannot reach considerable values

if the amplitudes of soil vibrations on the site of the building are small.

Therefore the amplitudes of forge-shop vibrations are usually of the same

Fig. IX-8. Seismogram of vertical soil vibrations induced

order as the amplitudes of

soil

by hammer

action.

vibrations at the location of the footings

under the walls or columns of the forge shop.

Figure IX-9 shows

several experimental curves of the distribution of amplitudes along the

height of the columns and along the girder.

These curves illustrate the

nature of vibrations of a forge shop induced by waves from the foundation
under a heavy hammer.

Amplitudes are shown

in

It is

seen that the columns undergo transverse

mm
)

00-4 3000 -4<-3400-H

9800

Fig. IX-9. Cross section of

first

forge-hammer shop described

in Art. IX-3.

EFFECT

ON

STRUCTURES OF

WAVES FROM

INDUSTRIAL SOURCES

421

vibrations in bending, while the girders undergo mainly vertical vibra-

The vibrogram of Fig. IX-10 shows that the vibrations caused

by waves from the hammer foundation are superimposed on the natural

tions.

damped to a
comparatively slight degree.
Amplitudes of forge-shop vibrations are not large; therefore they are
not dangerous with respect to the strength and stability of the structures.
In addition to the horizontal transverse vibrations, the shop columns,
and consequently their footings, undergo vertical vibrations with an
vibrations of the structure under consideration; the latter are

amplitude which often does not

vertical vibrations of the

but dynamic loads as

the acceleration of

hammer

differ

much from

foundation.

well, are transferred to the

soil

the amplitude of
Therefore not only static,

foundation bases.

vibrations at small distances from the

Since

hammer

Fig. IX-10. Vibrogram of the frame structure in Fig. IX-9.

foundation

may

real pressure

the

static

reach very high values

on the base under columns

pressure.

(of

the order of 0.7 to l.Og), the

be 1.7 to 2 times higher than

Therefore the footings under columns or wall
will

sections subjected to impacts will undergo larger settlements than foundations subjected to the action of static loads only.

Figure IX-11 shows

graphs of the amplitudes of vertical vibrations and the residual settlements of the columns. It is seen from these graphs that the settlements
are proportional to the amplitudes (or to the acceleration) of the vertical
vibrations.

To

illustrate the

harmful

effect of

action of waves propagating from a

foundation settlements caused by the

foundation, several cases are

hammer

discussed below, derived from experience in the operation of forge shops.

Impacts transferred from a foundation under a 4.5-ton hammer were

the reason for the complete destruction of a three-story building which

was attached to a forge shop, but

which no manufacturing was done.

foundation was about 1.0 mm;
the period of the waves propagating from the foundation was 0.075 to
0.070 sec. The brick auxiliary building was located at a distance of 6 m
from the hammer foundation and was erected much later than the forge
shop.
It was supported by a continuous foundation built of cyclopean
concrete; the pressure on soil was 1.75 to 2.0 kg/cm 2
The foundation rested on fine-grained sands characterized by a higher

The

vibration amplitude of the

in

hammer

DYNAMICS OF BASES AND FOUNDATIONS

422

than medium density; within this soil there occurred lenses of clay with
The ground-water level was at a depth of 4.0 m, i.e.,
silt and sand.
approximately at the elevation of the hammer foundation surface in
contact with soil.
The vibration amplitude of the foundation under the wall nearest the
hammer foundation of the auxiliary building was 0.65 mm; hence, the

some

acceleration of vertical vibrations of this wall

The

was

0.54gr.

vibration amplitudes of the building foundations at different

points varied from 0.05 to 0.65

mm;

Columns

therefore the

of the forge

-I

shop
i

J-

Cv90

280
Fig. IX-11. Vibration amplitudes
of the

the

forge-hammer shop shown

soil

dynamic pressure on

and settlement
in Fig.

of footings

under different columns

IX-9.

was nonuniform and the foundations underwent considerable

and

differential settlements resulting in the failure of the foundations

later of the walls.

There was another case in which shocks to a foundation under a 2.5-ton

hammer were the cause of the destruction of the forge shop building.
The foundation under the hammer underwent vertical vibrations with an
amplitude of 0.55 to 0.60 mm; the period of these vibrations was 0.085 to
0.075 sec.
The one-story brick forge shop rested on a continuous foundation built of cyclopean concrete; the static pressure on soil was about
forge

2.0 to 2.5

kg/cm 2

The foundations under the walls and hammer rested on gray and
yellow fine-grained sands of medium density. The ground-water level
was

at a depth of 8.5 m.
Operation of the hammer for 200 hr caused differential settlements of
the foundations under the walls of the forge shop and the formation of

numerous cracks

in the

masonry

of the walls.

EFFECT

ON

STRUCTURES OF

WAVES FROM

INDUSTRIAL SOURCES

423

In a third case, the source of vibrations was the foundation of a 3-ton

hammer supported by 6-m-long timber piles. Fine-grained uniform
sands of higher than medium density were found beneath the hammer

forge

foundations and the column foundations of the forge shop. Two groundwater horizons were found on the site of the shop: one was at a depth of

m, the other at a depth of 6 m. The pressure on the base of the

columns of the forge shop was taken to equal 1.50 to 1.75 kg/cm 2
The forge shop was a reinforced-concrete frame structure which consisted of columns tied together by continuous edge beams and cranesupporting beams. Soon after the 3-ton forge hammer was set in operation, a differential settlement of the shop columns nearest the foundation
was noticed. This settlement led to the appearance of cracks in the
reinforced-concrete structures of the shop, in the edge beams, in the
crane-supporting beams, and in the masonry of the brick filling of the
frame structure and the fireproof walls.
In all three cases, the development of a considerable differential
settlement was caused by the combined action of static pressure and
The foundations rested on noncohesive soils sands with
vibrations.
somewhat higher than medium density.
Static investigations of such soils by means of load tests or in consolidometers show that they are characterized by a resistance much
higher than that in cohesive soils (moist clays). Therefore sands of
medium and higher than medium density present good bases for foundations transferring static loads only.
However, if static loads and shocks
(or vibrations) act simultaneously, sands and similar noncohesive soils
may offer much less resistance to external loads. This was indicated in
Chap. II and is illustrated by the above examples.
It was established in Chap. II that settlements caused by a simultaneous action of static loads and shocks (or vibrations) are proportional
to the static pressure and inversely proportional to the coefficient of
2.5

vibro viscosity.

The value

of this coefficient

sideration.

is

inversely proportional to

at the point under conHence, the settlements of foundations subjected to vibra-

the acceleration of vibrations of

soil particles

and static loads are proportional to the accelerations.

Forge-hammer shops, similarly to other structures, are usually designed

tions

on all columns or shop walls remains constant.

However, the acceleration of vibrations depends essentially on the distance from the source of waves. Therefore columns or wall sections
so that the static pressure

located near

hammer

foundations are subjected to higher accelerations of

vibrations than building elements lying at larger distances from

hammer

This is the main reason for the differential settlements of

foundations under columns, with all the consequences which follow.
In order to decrease the harmful influence of vibrations on the settle-

foundations.

DYNAMICS OF BASES AND FOUNDATIONS

424

ments

of foundations

under buildings,

it is

necessary not to assign uniform

values of the static pressure on the foundation base.

depend on the distance

of a

The

values should

column or a wall section from the source

of

waves. The smaller that distance, the lower should be the permissible
bearing value. Thus permissible bearing values of soils beneath the
foundations of a building undergoing vibrations induced by hammers
should represent a certain function of the soil vibrations at the site of the
foundation.
sions of the

The amplitude

hammer

of soil vibrations, depending on the dimenfoundation area in contact with soil the amplitude
;

and frequency of foundation vibrations; and the soil properties and

distances from the foundation may be established similarly to the
computation presented in Art. VIII-5.
Hence, in order to avoid differential settlements of foundations, it is
necessary to assign permissible static pressures on the soil (depending on
the amplitude of soil vibrations at the location of foundations under
columns or walls of the shop) according to the directions presented in
Chap. II.

illustrative

REFERENCES

Notes:

1.

The

reference

number used

in the original

Russian text

parentheses at the end of each item of the present

is

given in

list; e.g., for

item

1,

(32).
2.

The

transliteration of Russian

words follows the system adopted by

the United States Board on Geographic Names.

1.

Anikin, S. G.: Opredeleniye raschetnoy velichiny dinamicheskoy nagruzki,

deystvuyushchey na fundamenty turbogeneratorov (Determination of

Design Values of Dynamic Loads Acting on Foundations under Turbogenerators), Sbornik informatsionnykh materialov Mosenergo, part

1,

1946.

(32)
I. I., S. I. Artbolevskiy, and B. V. Edelshteyn: "Teoriya i
metody uravnoveshivaniya shchekovykh drobilok" ("The Theory and
Methods of Balancing Jaw Crushers"), ONTI, 1937. (35)

2.

Artbolevskiy,

3.

Barkan, D. D.: Eksperimentalnoye issledovaniye vibratsiy sploshnykh

fundamentov, ustanovlennykh na svyaznykh gruntakh (An Experimental
Investigation of Vibrations of Continuous Foundations Resting on Watersaturated Cohesive Soils), Symposium 4, VIOS, 1935.
(25)
rasprostranenii ustanovivshikhsya kolebaniy v grunte (Distribution of Steady Vibrations in Soil), Symposium 4, VIOS, 1935.
(40)
"Raschet i proektirovaniye fundamentov pod mashiny s dinamicheskimi nagruzkami" ("Computations and Design of Foundations under
Machinery with Dynamic Loads"), Moscow, 1938. (15)
voprosu o raschete i proektirovanii fundamentov pod turboagregaty (On the Question of Computations and Design of Foundations
under Turbopower Systems), StroiteVnayaPromyshlennost\ no. 3, 1941. (31)
vybore glubiny zalozheniya istochnika i priyemnika voln, rasprostranyayushchykhsya v grunte (On the Selection of the Depth of Placement of the Source and of the Receiver of Waves Propagated in Soil), Zhurnal
Tekhnicheskoy Fiziki, vol. XI, no. 11, 1941.
(37)
Proektirovanie i raschet fundamentov pod kuznechnyye moloty i
kopry v usloviyakh voyennogo vremeni (Design and Computations of
Foundations for Forge Hammers and Large Drop Hammers under Wartime

4.

5.

6.

7.

8.

Conditions), Stroitel'naya Promyshlennost' no.

,

9.

6,

1942.

(30)

"Primeneniye vibrirovaniya pri ustroystve osnovaniy sooruzheniy"

425

DYNAMICS OF BASES AND FOUNDATIONS

426

("The Use of Vibrations for Treatment

Moscow, 1943. (16)

of

Bases under Buildings"), Gosstroy-

izdat,
10.

11.

Eksperimentalnyye issledovaniya ekranirovaniya voln, rasprostranyayushchikhsya v grunte (Experimental Investigations of Screening

of Waves Propagated in Soil), Inzhenernyy Sbornik Akademii Nauk SSSR,
Moscow-Leningrad, vol. II, no. 2, 1946. (51)
Eksperimentalnyye issledovaniya kolebaniy fundamentov kuznechnykh molotov (Experimental Investigations of Vibrations of Foundations
under Hammers), Nil, Sbornik trudov no. 12, Vibratsii sooruzheniy i funda:

mentov, 1948.
12.

Deryagin, B.:

(29)

zatukhanii

dispersii

seysmicheskikh voln (Damping and

Dispersion of Seismic Waves), Zhurnal Geofiziki, no. 1-2, 1931.

(38)
Zatukhaniye seysmicheskikh i akusticheskikh voln i yego zavisimost
13.
:

(Damping of Seismic and Acoustic Waves and Its Dependence on

Frequency), Zhurnal Geofiziki, no. 3-4, 1932.
(38)
Die Anwendung dynamischer Baugrunduntersuchungen, Veroffentlichungen
ot chastoty

14.

des Institute der Deutschen Forschungsgeselschaft fur Bodenmechanik, no.

1936.

4,

(5)

A. P.: Vynuzhdennyye kolebaniya neogranichennoy plity lezhaschchey na uprugom poluprostranstve (Forced Vibrations of an Infinite Slab
Resting on an Elastic Semi-infinite Space), Prikladnaya Matematika i Mek-

15. Fillipov,

hanika, vol. IV, no.

2,

1940.

(45)

16.

Gerner: "Geologie und Bauwesen," H.

17.

Gersevanov, N.

Moscow, 1937.
18.

19.

1932.

(10)

of Soil

Mass"),

(4)

Giproavtoprom (Technical Division): "Rukovodstvo po proektirovaniyu i

vozvedeniyu fundamentov pod kuznechnyye moloty i pressy" ("A Manual for

Design and Erection of Foundations under Forge Hammers and Presses"),

compiled by engineer A. N. Cheshkov; D. D. Barkan consulted; mimeographed edition, 1945. (28)
Gogoladze, V. G.: Uprugi3*e kolebaniya v srede s uprugim posledeystviyem
(s nasledstvennost'yu) (Elastic Vibrations in a Medium Characterized by
Elastic Recovery),

1941.
20.

3,

M. "Dinamika gruntovoy massy" ("Dynamics

Trudy Seysmologicheskogo

Instituta

AN

SSSR,

no. 109,

(39)

Gorbunov-Posadov, M. I.: Osadki i davleniye pod zhestkimi pryamougol

'nymi fundamentnymi plitami (Settlements and Pressure under Rigid Rectangular Footings), StroiteVnaya Promyshlennosf no. 8, 1940.
(9)
Ivanov, N. N., and T. P. Ponomarev: "StroitePnyye svoystva gruntov"
,

21.

("Engineering Properties of Soils"), Lengostransizdat, 1932.

(1)
N. N., and M. D. Telegin: Uplotneniye dorozhnykh nasypey (The
Compaction of Highway Embankments), Novosti Dorozhnoy Tekhniki, Sym-

22. Ivanov,

Moscow, 1939. (17)

and S. V. Puchkov: Rasprostraneniye kolebaniya v grunte ot
istochnika tipa impuPsa (The Propagation of Vibrations in Soil from a Source
of the Impulse Type), Trudy Seysmologicheskogo Instituta AN SSSR, no. 59,
posium

8,

23. Kirillov, F. A.,

1935.

(14)

24. Kohler, R.: Z. Geophysik, vol. VIII, no. 46, 1932.

(43)

REFERENCES
25. Korchinskiy,

I.

L.:

Dinamicheskiye

427

kharakteristiki

drevesiny,

betona,

zhelezobetona (Dynamic Characteristics of Wood, Concrete, and Reinforced

Concrete), Symposium Dinamicheskiye Svoystva StroiteVnykh Materialov, Yu.
A. Nilender,
26.

27.

Stroyizdat, 1940.

(ed.),

(41)

Kupradze, V. D.: "Osnovnye zadachi matematicheskoy teorii diffraktsii

(ustanovivshiyesya protsessy)" ("Basic Problems of the Mathematical
Theory of Diffraction"), Moscow-Leningrad, 1935. (50)
Lamb, H.: On the Propagation of Tremors over the Surface of an Elastic
Solid, Phil. Trans. Roy. Soc. London, series A, vol. 203, September, 1904.
(44)

28.

"Kratkiy kurs teorii uprugosti" ("A Brief Outline of the

Chap. XIII, Moscow-Leningrad, 1942. (36)
Lur'ye, A. I. Vliyaniye uprugosti grunta na chastoty sobstvennykh kolebaniy
prostoy ramy (Effect of the Soil Elasticity on Frequencies of Natural Vibrations of a Simple Frame), Symposium Vibratsii Fundamentov Ramnogo Tipa,
Leybenson, L.

Theory

29.

S.:

of Elasticity"),
:

ONTI,
30.

Gosstroyizdat, 1933.

(34)

Vliyaniye uprugosti osnovaniya na chastoty svobodnykh kolebaniy

turbofundamentov (Effect of Elasticity of the Base on Frequencies of Vibrations of Foundations under Turbodynamos), Symposium Vibratsii Funda31.

mentov Ramnogo Tipa, ONTI, Gosstroyizdat, 1933.

(33)
Lyav, A.: "Matematicheskaya teoriya uprugosti" ("Mathematical Theory
of Elasticity"),

32. Pavliuk,

1931.
33.

N.

Moscow, 1935.

(13)

Byulleten' Leningradskogo Instituta Sooruzheniy, no.

P.:

15,

(24)

and A. D. Kondin: O pogashenii vibratsiy fundamentov pod mashiny

(Damping of Vibrations of Foundations under Machinery), Proekt i Standart,
no. 11, 1936.

(26)

I., and others: Novyye metody issledovaniya szhimaemosti

vnutrennego treniya v gruntakh (New Methods of Investigation of the
Compressibility and Internal Friction in Soils), Vestnik Voyenno-Inzhenernoy
Akademii RKKA, no. 6, 1934. (3)
35. Pol'shin, D. E.: "Ugol povorota fundamenta pri ekstsentrichnoy nagruzke
yego" ("The Angle of Rotation of a Foundation when Loaded Eccentrically").

34. Pokrovskiy, G.

(12)
36.

Ramspeck,

A.,

and G. Schulze: Veroffentlichungen des

Instituts der Deutschen

Forschungsgesellschaft fur Bodenmechanik (Degebo) an der Technische Hochschule, no. 6, 1936.

(47)

37. Reissner, E.: Stationare axialsymmetrische,

durch eine schutternde Masse,

Schwingungen eines homogenen elastischen Halbraumes, Ingenieur

Archiv, December, 1936.
(22)
Sadowsky, M.: Zweidimensionale Probleme der Elasticitatstheorie, Z.
Angew. Math, und Mech., Bd. 8, 1928. (7)
Schleicher: Zur Theorie der Baugrundes, Der Bauingenieur, Heft 48/49,

erregte

38.

39.

1926.

(8)

Sezawa, K.: Dispersion of Elastic Waves, Bull. Earthquake Research Inst.,

Tokyo Imperial University, vol. 3, 1927. (46)
41. Shekhter, O. Ya.
Ob uchete inertsionnykh svoystv grunta pri raschete
40.

DYNAMICS OF BASES AND FOUNDATIONS

428

vertikal'nykh vynuzhdennykh kolebaniy massivnykh fundamentov (Consideration of Inertial Properties of Soil in the Computations of Vertical

Forced Vibrations of Massive Foundations), Nil, Symposium 12, Vibratsii

Osnovaniy i Fundamentov, Moscow, 1948. (23)
diffraktsii sfericheskikh uprugikh voln vblizi poverkhnosti
42. Sobolev, S. L.:
sfery (Diffraction of Spherical Elastic Waves in the Vicinity of a Sphere
Surface),
43.

44.

Trudy Seysmologicheskogo

Instituta

AN

SSSR,

no.

7,

1930.

(49)

Teoriya diffraktsii ploskikh voln (Theory of Diffraction of Plane

Waves), Trudy Seysmologicheskogo Instituta AN SSSR, no. 41, 1934. (48)

"Tekhnicheskiye usloviya na raschet i proektirovaniye fundamentov pod
mashiny s dinamicheskimi nagruzkami (proekt)" ("Technical Rules for
Computations and Design of Foundations under Machinery with Dynamic
Loads"), Gosstroyizdat, 1939.

(27)

mekhanika grunta na osnove yego fizicheskikh

Mechanics Based on Physical Properties of Soil"), translated

45. Terzaghi, K.: "Stroitel'naya

svoystv" ("Soil
from German, Gostekhizdat, Moscow-Leningrad, 1933. (2)
46. Tsytovich, N. A.: "Mekhanika Gruntov" ("Soil Mechanics"), Gosstroyizdat,
47.

Moscow, 1937. (6)

"Ukazaniya po proektirovaniyu i vozvedeniyu v usloviyakh voyennogo
vremeni fundamentov pod mashiny s dinamicheskimi nagruzkami," U-56-42
("Instructions for the Design and Erection of Foundations for Machinery
with Dynamic Loads under Wartime Conditions," U-56-42), Gosstroyizdat,
Moscow, 1942. (11)

NAME INDEX

Kondin, A. D., 165, 167, 427

Korchinskiy, I. L., 347, 427
Kupradze, V. D., 427

Anikin, S. G., 425

Arnold, R. N., 95
Artbolevskiy,

I. I.,

Artbolevskiy, S.

I.,

425
425

Lamb,
Barkan, D. D.,

vii, viii, xi, xii, xvii, 13, 16,

H., 427

Lessels, J. M., 239, 427

27-29, 31, 32, 39, 49, 62, 123, 125, 202,

Leybenson, L.

220, 304, 345, 347, 377, 425

Lur'ye, A.

Berdichevsky, G. I.,
By croft, G. N., 95

ix

Lyav,

A.,

I.,

S.,

427

260, 427

427

Maslov, N. N., 74
Mikhalchuk, A. I.,

Deryagin, B., 345, 426

Edelshteyn, B. V., 425

27, 32, 122

Florin, V. A., 74

Pavliuk, N. P., 110, 165, 167, 256, 427

Pokrovsky, G. I., 12, 55, 427
Polshin, D. E., 19, 25, 30, 427
Ponomarev, T. P., 12, 426
Puchkov, S. V., 45, 426

Gerner, 26, 426

Ramspeck,

Gersevanov, N. M., vii, xii, 426

Gogoladze, V. G., 345, 426
Goldshteyn, M. N., 74
Gorbunov-Posadov, M. I., 25, 426

Reissner, E., 95, 427

Fillipov, A. P.,

Ivanov, N. N.,

426

12,

426

Katzenelenbogen, R. Z.,
Kirillov, F. A., 45, 426
Koehler, R., 352, 426

13, 16

13,

427

427
Sadovsky)
Saichev, P. A., 27, 125, 377
Savchenko, I. A., 56, 67
Savinov, O. A., 256
Schleicher, 24, 427
Schulze, G., 427
Sezawa, K., 427
Shekhter, O. Ya., xvii, 95, 427

Sadovsky, M.,
Sadowsky, M.

429

22,

(see

DYNAMICS OF BASES AND FOUNDATIONS

430

Smolikov, Ya. N., 27, 39, 45, 49, 123, 125,

330, 377
Sobolev, S. L., 428
Suppiger, E. W., viii
Telegin,

M.

Timoshenko,

S. P.,

239

Tschebotarioff, Florence B.,

viii

Tschebotarioff, Gregory P.,

viii,

Tsytovich, N. A.,

13,

Tumarkin, D. M.,

ix

D., 426

Terenin, B. M., 31, 304

Terzaghi, K., 12, 428

Worburton, G.

B., 95

428

xvii

SUBJECT INDEX

Absorbers of vibrations, 172-184, 220231

Amplitudes of vibration (see Vibrations)

Dampers, dynamic, 167-172

Damping
Dynamic
Dynamic

properties of

soils,

65-68

pressures, 76-81, 339, 354

stresses (see Stresses)

Balancing machine rotors, 254

Base, coefficient of rigidity c r

of, 86,

87

modulus (see Young's

modulus)
Explosion, underground, 316-318
Elasticity

definition of, vii

Camouflet (underground explosion), 316318

Chemical stabilization of soils, 161
Clays, saturated, coefficients
30, 33, 49, 124,

Coefficient, of elastic

pression
of elastic
of elastic

c<p

Forge hammers (see Foundations)

Foundations, design of, crushing equipment for, 307-310
frame type, data required for, 243
examples of, 270-282
field of application, 242
instructions for, 243-250
natural frequency of, 260-263,

of, 27, 29,

126

nonuniform com-

(see Soils)

nonuniform shear c^,

uniform compression

40, 41
c u (see

Soils)

of elastic uniform shear c T (see Soils)

278, 281

66-68
43, 46, 48

of energy absorption ^,

of proportionality Cs,
of restitution

e,

vibrations
for

subgrade reaction

cp ,

250-255, 259-270

classification of, 185

construction procedures, 189, 190

195, 196

data for design, 185-187

directives, general, 187-189
efficiency of hammer, 191-195
examples of, 216-225

of rigidity c r , 86, 87
of

of,

hammers,

17-19

Coefficients, of elastic resistance of piles

(see Piles)

of loess, 28-30, 33, 49, 124

materials of foundation and pad,

of sands, 26, 29, 30, 33, 49, 125, 126

187
natural frequency

of saturated clays, 27, 29, 30, 33, 49,

of,

198-200,

207, 219, 229

124, 126

Compaction, vibratory, 69-74

Compressive strength of concrete, 132,

selection of weight

and base

area,

211-216
vibrations of, 196-211
massive type, example, 284-288
experimental studies of vibrations
of, 122-130

155, 187, 190, 219, 228, 234, 249,

256, 270, 284, 289, 302, 304

Counterbalancing exciting loads, 160

Crushing equipment, 307-310
431

DYNAMICS OF BASES AND FOUNDATIONS

432

Foundations, design

of,

massive type,

general instructions, 283

natural frequency
for, rolling mills

of, 286,

Load

tests, piles, 42, 46, 47, 51,

soils, 17, 18, 20, 21, 25, 26,

287

Loess, coefficients

of,

52

37-40

28-30, 33, 49, 124

design loads on,

303

dynamic loads imposed by,

Massive foundations

(see

Foundations)

292-299
general instructions, 299-303
for reciprocating engines, ampli-

tudes, permissible, 131

examples of dynamic analyses,

149-160
forces of belt pull, 147-149
general comments, 133, 134
machine data required, 132
material of foundation, 132

unbalanced forces of, 135-146

with vibration absorbers, examples
of, 181-184, 227-231
theory of, 172-180, 220-227

by

existing, decrease of vibrations,

chemical stabilization of

soils,

Natural frequency, experimental data

on, 123-130
frame foundations, 260-263, 278, 281

hammer

foundations, 198-200, 207,

219, 229

massive foundations, 286, 287

pile foundations, 51, 52, 128
simultaneous rocking, sliding and vertical motion of foundation, computations of, 153, 158
theory of, two principal values, 112115, 118-122

vibrations, pure shear, theory of, 106

rocking, theory

of,

105

computations
theory of, 87, 92-94

vertical,

161

by counterbalancing

of,

152

exciting

loads, 160

by dynamic dampers, 167-172

structural measures for, 161-167,

287
performance data on, 162, 203-205,
245, 246, 254, 304-306
for scrap-crushing, examples, 240-241
theory of, 233-239

Frame foundations
Frequency

(see

(see Foundations)
Natural frequency)

nonuniform

Piles, coefficient, of elastic

resistance

K v of pile group,

50

of elastic resistance C& of one pile in

a group, definition

values

of,

of,

48

49

of elastic resistance

of pile group,

definition of, 46

values

of, 47,

128

of proportionality Cd for single piles,

definition of, 43

Gyratory crushers, 307-310

values

of, 46,

improvement

Hammer

foundations (see Foundations)

Hooke's law, 1, 6-8

48

elastic lateral displacement, 51, 52

of base properties, 47,

287
load tests on, 42, 46, 47, 51, 52
of, 51, 52, 128

natural frequencies

vibrations during driving, record

Instructions for the Design
tion of

and Construc-

Machinery Foundations

(official

U.S.S.R.), 29

Poisson's ratio, definition

of soils, effect

on forced

of, 5,

of,

vertical vibra-

tions, 96-99, 102

values

of,

11-13

Pressures, dynamic, 76-81, 339, 354

Jaw

crushers, 307-310

Lame's constants,
(See also Shear)

4,

313, 368

44

10

(See also Stresses)

Reciprocating engines, foundations for

(see Foundations)

SUBJECT INDEX

Resonance, experimental data on, 117,

123, 127, 129
observed cases of, 409-424
theory of, 93, 94
Restitution, coefficient e of, 195, 196
Rheological models, 75
Rolling mills, 292-303
Rules

(official

U.S.S.R.), technical,

and

construction code, 245, 248, 255, 303

Sands, coefficients of, 26, 29, 30, 33, 49,

125, 126
Scrap-crushing foundations, 233-241
Settlements, residual, dynamic, 75-81

methods

to decrease,

82-84

433

load tests, results

21, 25, 26,

of, 17, 18, 20,

37-40

permissible load on, values

shear tests

of, 29,

30

56-61

of, 8,

stabilization of (see Foundations, existing, decrease of vibrations)

vibro viscosity

of,

61-64

dynamic, horizontal impact on

foundation, 232
vertical, in pad under anvil, 200
in soil under hammer foundation,
211-216

Stresses,

(See also Soils,

dynamic pressures on)

Turbogenerator foundations (see Foundations, design of, frame type, massive

permissible, 76

Shear,

Soils,

modulus in, 5, 313, 341, 345, 354

Lame's constants)

type)

(See also

Shear tests of soils, 8, 56-61

Shear waves, velocity of, 321, 328
absorption (see

Soils, coefficient, of

damp-

of elastic

nonuniform compression
30-32

values

of, 32, 33, 149, 284,

nonuniform shear

nition

values
of elastic

of,

of,

cv,

287

c$, defi-

40

uniform compression

c,

definition of, 19-21

of,

28-30, 124-125,

149, 181, 207, 209, 218, 228, 240,

amplitudes, frame foundations, 263,

264, 267, 277, 278, 282

hammer and

foundation, 198, 205,

massive foundations, 288, 290

permissible values of, frame foundations, 268-270
hammer foundations, 212
reciprocating-engine foundations,
131

284, 287, 354

effect of size

with vibration absorbers)

209, 210, 219, 229

41

design values

of,

acceleration ratio of, 60, 64, 69, 74

definition of,

of elastic

Vibrations, absorbers (see Foundations,

design

ing constant, below)

and shape

of founda-

tion base on value of, 21-25

experimental investigations of,
25-28, 47, 124, 125, 208, 339
of elastic uniform shear c T definition

pure shear, 107

rocking, 105
simultaneous, sliding and vertical
motion, 116, 117, 154, 159, 160
soil surface,

340-348

of, 34,

values

vertical, 77, 89, 90, 93, 100-102, 152,

35

of,

damping constant, a
c,

66-68

247
dampers, dynamic, 167-172

definition

damping

of,

effect of,

on consolidation

36-40, 130, 149, 284

energy absorption
of subgrade reaction
of, 17-19
of

\p,

cp

of waves,

344-347

(See also Stresses)

layered, in waves, 365-374

of soils, 69-

74
threshold of vibratory compaction, 72

92

reduced, A, 93, 94
, 100-102, 126-128

damping properties of, 65-68

dynamic pressures on, 76-81,

65-68, 77

339, 354

on permeability of soils, 74, 75

on pore pressures in soil, 74
on shearing strength of soils, 54-61
experimental studies on massive foundation, 122-130

DYNAMICS OF BASES AND FOUNDATIONS

434

Vibrations, frame foundations, computations, examples

282
theory

of,

Waves, compression

270-

(see longitudinal,

below)
diffraction, general concepts,

259-270
exciting loads, 250-255

hammer and

374-

377
general theory of, 318-322
layered soils in, 365-374

of,

foundation, experimental

studies of, 200-211

longitudinal, velocity of, 315, 316

theory of, 196-200

impact effect on, 77
natural frequency of, 105, 106
during pile driving, 44
rocking, 103-106

screening

of,

examples from practice,

402-405
377-382
383-402
general conclusions, 405, 406
experiments, in

loess,

in water-saturated soils,

simultaneous, sliding and vertical,

107-122

shear, velocity of, 96, 315, 316, 371

with damping, 91-94

without damping, 85-91

theory of, general, 311-315

transverse (see shear, above)

forced, effect of soil inertia on, 94-

vibrations caused by, amplitude of

surface, velocity of, 321, 328

vertical,

surface, source of

103

waves produced by

(see

Waves)

soil

influ-

ence on, area of, 349-354

depth of, 356-365
amplitude variation, with depth,
experimental data on, 325331
theory of, 323-325
with distance from source, experimental data on, 338-348
theory of, 331-338

Vibratory compaction, 69-74

Vibroviscosity of soils, 61-64

Wavelengths, table, 405

Waves, action on structures, from lowfrequency machinery, 409-419
from shocks from hammers, 419-424
caused by explosion underground, 316318

Young's modulus, definition of, 5, 10

values of, for concrete, 255-258
for soils, 13-16

classification of industrial sources of,

407-409

438F11

waves

BB\31\99 347B5

SOB

MARSTON SCIENCE LIBRARY

Date Due

Due

Returned

Due

Returned

UNIVERSITY OF FLORIDA

1262

07017

0427