LOAD-BEARING PROCESSES

IN
AGRICULTURAL WHEEL-SOIL SYSTEMS

OiYiVAKSEH

EüttJOTÏIMI^v
ï:ANDr-QTJ\Yu:sr>; Y'.STTEIT

0000 0248 9975

U0C(5\
Promotor: ir.H.Kuipers
hoogleraar indegrondbewerkingendegronddynamica
Co-promotor: Dr.ir.A.J.Kooien
universitair hoofddocent indegronddynamica
 ^O^IJC^W^Z,

F.G.J. Tijink

LOAD-BEARING PROCESSES
IN
AGRICULTURAL WHEEL-SOIL SYSTEMS

Proefschrift

ter verkrijgingvandegraadvan
doctor inde landbouwwetenschappen,
opgezagvande rector magnificus,
Dr.C.C.Oosterlee,
in hetopenbaar teverdedigen
opwoensdag 13 januari1988
des namiddagstevier uur indeaula
vande Landbouwuniversiteit teWageningen

k y X.
 ABSTRACT

Tljink, F.G.J. (1988). Load-Bearing Processes in Agricultural

Wheel-Soil Systems. Doctoral thesis, Agricultural University
Wageningen, The Netherlands, 173 p., 98 figs, 25 tables, 203
refs, English and Dutch summaries.
In soil dynamics we distinguish between loosening and load-
bearing processes. Load-bearing processes which can occur under
agricultural rollers, wheels, and tyres are dealt with in this
dIssertation.
We classify rollers, wheels, and tyres and treat some general as-
pects of these devices. Fundamentals of load-bearing processes,
I.e. kinematic, dynamic, and soil physical aspects are discussed
aIso.
Not only soil characteristics concerning load-bearing processes
are dealt with but also the suitability of these characteristics
for use In prediction methods. Special attention has been paid to
predicting some process aspects of a towed tyre under different
soil conditions and In different soil types. Under laboratory
conditions the use of characterizing processes (cone, vane, and
falling weight) and empirical prediction methods resulted in
accurate predictions of rolling resistance, rut depth, and com-
paction caused by a towed tyre.

Additional keywords: load-bearing processes, soil dynamics, tyre

performance, soil compaction, soil mechanical properties, soil
characteristics, prediction methods.

ISBN 90-9001968-5
Printed by: Krips Repro,Meppel

Copyright F.G.J. Tijink, 1988.

No part of this book (with the exception of the abstract on this

page)may be reproduced or published In any form and by any means
without permission from the author.
 JMÙ'?7.0\(\\^'

STELLINGEN

1
Het nauwkeurig voorspellen van roIweerstand, Insporingen ver-
dichting bij hetberijdenvan homogenegrond is mogelijk met
empirisehevoorspeIIIngsmethoden.
DitproefsehrIft

Insporing Isnleta Itijdeengeschiktemaat voor verdichting.

DltproefsehrIft

Voor het weergeven vandekarakteristieken van Iandbouwbanden

dient een "standaard" ontwikkeld teworden.
Dit proefschrift

De benuttingvan hetvoor ploegwerk beschikbare trekkervermogen

kan nogverbeterdworden.
DltproefsehrIft

Indien het gewenst isbovenover terijdenbij de hoofdgrond-

bewerkingdienen ergrondbewerkIngswerktuIgen ter beschikking te
komen die eenkerendewerking hebbenendoor deaftakas worden
aangedreven.

De specifieke ploegweerstand neemt af bijeen toename van de

werkbreedtevan eenploeg.

Voor ploegwerkzaamheden IseenslipregelingeengoedeaanvuIIInç

opdeweerstandsregeling.

Het gebruik van een schijfkouter bijhet voorste ploeglichaam

vereist een langer ploegframeeneen grotere hefkracht.

Eenwentelploeg,gebouwd volgenshetbouwdoossysteem, vereisteen

wentelsysteem dat hetwentelen over 180graden kanondersteunen.
 10
Bij onderzoek naar verdichting van bewerkte ondergrond behoort de
invloed van de massa van de bouwvoor niet verwaarloosd te worden.

11
Het gebruik van gepllleerd zaaIzaad komt de zaadverdeling ten
goede bij het zaaien van bleten.

12
Bij het ontwerpen van frames voor IandbouwwerktuIgen met een
lange gebruiksduur, dient meer aandacht geschonken te worden aan
het optreden van vermoeidheIdsverschiJnseIen.

13
Het verminderen van de hoeveelheid spuitvloeIstof in de ge-
wasbescherming dient samen te gaan met een beter gebruik van
persoonIiJke beschermIngsmIddeIen.

14
Conus, shear vane en valgewlcht zijn bruikbare hulpmiddelen bij
het karakteriseren van mechanische eigenschappen van grond.
Dit proefsehrift

15
Bij het ontwikkelen van een bewerkbaarheidstest verdient het
aanbeveling het onderzoek te concentreren op methoden, die geba-
seerd zijn op verkruImeIIng.
Dit proefsehrift

16
De invloed van de gelaagde opbouw van grond (toplaag, bouwvoor,
ploegzool, ondergrond) op afsteunende processen Is nog onvoldoen-
de onderzocht.
Dit proefschrift

F.G.J. TiJink
Load-bearing processes in agricultural wheel-soIIsystems
Wageningen, 13 januari 1988
voorAgnes
 VOORWOORD

Dit kleine stukje gebruik ik graag om iedereen te bedanken die

een bijdrage heeft geleverd aan dit proefschrift.

Hiervoor ben ik allereerst dank verschuldigd aan mijn promotor

prof. ir. H. Kuipers, die dit onderzoek mogelijk heeft gemaakt en
die mij enthousiast heeft gemaakt voor dit vakgebied. Zijn warme
belangstelling heeft mij erg geholpen bij het voltooien van dit
proefsehrIft.
Mijn co-promotor Dr. ir. A.J. Kooien wil ik vooral danken voor de
begeleiding van het onderzoek en de ondersteuning bij enkele met
deze dissertatie samenhangende publicaties.

De heren B.W. Peelen en A. Boers dank Ik voor hun hulp bij het
verrichten van metingen.

De volgende studenten ben ik erkentelijk voor hun medewerking

aan het promotie-onderzoek: Wim den Haan, Simon Hofstra, Harry
Swinkels, Gertjan van Dijk, Jan Broekhuizen en Pleter Vaandrager.

Voor het tot stand komen van het manuscript wil ik graag
danken: de heer B.W. Peelen voor het met zorg uitvoeren van al
het tekenwerk en Mevr. drs. A.S.R. Riepma voor het met grote
inzet corrigeren van de Engelse tekst.

De Stichting "Fonds Landbouw Export Bureau 1916/1918" en de

Landbouwuniversiteit wil ik dank zeggen voor het beschikbaar
stellen van subsidies ter bijdrage In de kosten verbonden aan dit
proefsehrIft.

Dordrecht, 6 Juli 1987

 CONTENTS

CHAPTER 1 INTRODUCTION 13

CHAPTER 2 AGRICULTURAL ROLLERS, WHEELS, AND TYRES 15

2.1. ROLLERS 15
2.1, Roller-packers 15
2.1. 1. Single roller-packers 15
2.1. 2. Composed roller-packers 16
2.1. Roller-harrows 17
2.2, WHEELS 17
2.3, TYRES 18
2.3, 1 Hlghlights In tyre development 18
3.2. Tyre construction 18
3.3. Tyre s i z e s p e c i f i c a t i o n 20
3.4. Tyres used in a g r i c u l t u r e 22
1. Tyres for driven w h e e l s 23
2. Tyres for undrlven steered w h e e l s . . . . 24
3. Tyres for garden tractors 24
4. Implement tyres 24
5. Farm utIIIty tyres 25
6. SemI-tyres 25
7. O t h e r tyres used In a g r i c u l t u r e 25

CHAPTER 3 K I N E M A T I C A S P E C T S OF L O A D - B E A R I N G P R O C E S S E S 27

3 1. STATE OF M O V E M E N T O F A ROLLER, W H E E L O R TYRE 27

3 1.1 Slip 28
3 1.1.1. Zero-slip conditions 28
3 1.1.2. M e a s u r i n g m e t h o d s for slip 32
3 1.1.3. Wheel slip during p l o u g h i n g 36
3.1.2 M o v e m e n t s of a point at t h e rim of a
rol1er or wheel 38
3.1.2.1, T r a j e c t o r y of a point at the rimof a
rol1er or wheel 38
3.1.2.2. V e l o c i t y of a point at the rim of a roller
or wheel 39
3.2. M O V E M E N T S IN THE C O N T A C T AREA 39
3.2.1. Tyre d e f o r m a t i o n s 39
3.2.1.1. Radial tyre d e f o r m a t i o n and sidewaII
buIgIng 39
3.2.1.2. Tangential c a r c a s s and lug d e f o r m a t i o n s .44
3.2.1.3. Influence of tyre d e f o r m a t i o n s on slip .46
 3.2.2. Movements of a point at the circumference
of a roller, wheel, or tyre during motion
In the mutual contact area 46
3.2.3. Trajectories of soil particles In the con-
tact area between the soil and a roller,
a wheel,or a tyre 47
3.3. SUBSURFACE MOVEMENTS 51
3.3.1. Soil movements under rollers 51
3.3.1.1. Distribution of soil velocities relative
to the centre of a rol1er 51
3.3.1.2. Flow zones under rollers 52
3.3.2. SoIImovements under wheels 52
3.3.3. SoIImovements under a tyre 53

CHAPTER 4 DYNAMIC ASPECTS OF LOAD-BEARING PROCESSES 55

4.1. FORCES, MOMENTS, AND STRESSES ON ROLLERS,

WHEELS, AND TYRES ' 55
4.1.1. Mechanical equilibrium of rollers, wheels,
and tyres 55
4.1.2. Tractive performance 57
4.1.2.1. Analysis 57
4.1.2.2. Towed tyres 58
4.1.2.3. Driven tyres 60
4.1.2.4. Tractor tractive performance 64
4.1.2.5. Ploughing capacity 68
4.2. STRESSES IN THE CONTACT AREA 70
4.2.1. Stresses In the contact area between soil
and rigid wheel or rol1er 70
4.2.1.1. Radial stress distribution 71
4.2.1.2. Tangential stress distribution 71
4.2.2. Stresses In the tyre contact area . . . . 71
4.2.2.1. Average contact pressure 72
4.2.2.2. Radial stress distribution 73
4.2.2.3. Shear stress distribution 73
4.3. STRESSES IN THE SOIL 75
4.3.1. Stress distribution in the soil under
tyres, wheels, and rolIers 75
4.3.2. Parameters that affect the stress distri-
butions under agricultural driving equip-
ment 77
4.3.2.1. Soil parameters 77
4.3.2.2. Trailer tyre concept 78
4.3.2.3. Tractor type 79
CHAPTER 5 SOIL CHARACTERISTICS CONCERNING LOAD-BEARING
PROCESSES 81

5.1.RELATIONSCHIPS BETWEEN "TREATMENT" AND "BE-

HAVIOUR" OF SOIL 81
5.2. CHARACTERIZING PROPERTIES USED IN SOIL
SCIENCE 84
5 3 MECHANICAL PROPERTIES 87
5 3 1 CompactIbIIIty . 87
5 3 1 1 Measures for compaction 87
5 3 1 2 Measuring compactIbIIIty 88
5 3 1 3 The Influence of repeated loading, loading
speed, and vibrations on compactIbIIIty 92
5 3 2 DeformablIIty 94
5 3 3 Resistance against shear 95
5 4 CHARACTERIZATION PROCESSES 97
5 4 1 Comparison of different cones 99
5 4 2 Comparison of cone penetrometer, mlcro-
penetrometer, vane, and torvane 101
5.4.3. Relationships between cone-penetrometer,
shear vane, and falling weight 103

CHAPTER 6 RELATIONSHIPS BETWEEN SOIL CHARACTERISTICS

AND PROCESS ASPECTS AND THEIR SUITABILITY
TO PREDICT PROCESS ASPECTS 105

6.1. COMPARATIVE METHODS 105

6.2. EMPIRICAL METHODS 106
6.2.1. Empirical graphs 106
6.2.2. Empirical methods using relationships be-
tween soil characteristics and process as-
pects 107
6.2.2.1. Relationships between soil characteristics
and tyre rolling resistance 109
6.2.2.2. Relationships between soil characteristics
and soil compaction due to wheel action .113
6.2.2.3. Relationships between soil characteristics
and ruth depth 117
6.2.3. Empirical formulas based on dimensional
analysIs 117
6.2.3.1. Predicting off the road tyre rolling re-
sistance 117
6.2.3.2. Predicting tyre rolling resistance on a
hard surface 126
6.3. APPROXIMATE METHODS 128
6.4. EXACT METHODS 131
6.4.1. Slip linemethods 131
6.4.2. Finite element methods 131
6.5. CLOSING REMARKS ON PREDICTION METHODS . . .131
CHAPTER 7 SOIL PHYSICAL ASPECTS OF LOAD-BEARING PRO-
CESSES 133

7.1 THE INFLUENCE OF PHYSICAL PROPERTIES

ON MECHANICAL PROPERTIES 133
7.1.1 The Influence of aggregate diameter on
uniaxial compresslblIIty 134
The Influence of soil air on rebound . . .139
THE INFLUENCE OF MECHANICAL TREATMENT
ON PHYSICAL PROPERTIES 146
7.2.1 The Influence of mechanical treatment on
micro-factors and soil qualities 146
TlIlablIIty 147
TlIlablIIty test 147
7.2 TlIlablIIty tests based on processes other
than crumblIng 148
7.2.2.3. TillablIIty tests based on crumbling pro-
cesses 150

CHAPTER 8 SUMMARY 161

CHAPTER 9 SAMENVATTING 162

REFERENCES 163
 CHAPTER 1

INTRODUCTION

Research in agricultural soil mechanics focused first on soli

loosening processes. Scientific papers on load-bearing processes
have been published In fair amounts since the early 1950s. Gen-
erally these papers only deal with one single aspect of a par-
ticular load-bearing process.
Load-bearing processes can be Induced by rollers, wheels, tyres,
penetrating bodies, sliding and shearing bodies, or tracks. Agri-
cultural traction and transport devices generally have rolling
machine parts that touch the soil directly. Therefore, this
dissertation concentrates on systems with rollers, wheels, or
tyres.
The aim of this study Is to cover fundamental, characterizing,
and predicting aspects of load-bearing processes in agricultural
wheel-soIIsystems.
A classification of agricultural rollers, wheels, and tyres Is
presented In chapter 2.
Process fundamentals are kinematic, dynamic, and soil physical
aspects. The kinematic and dynamic aspects are discussed as
foIlows:
- the kinematics of the roller, wheel, or tyre (3.1.)
- the movements In the contact area (3.2.)
- the subsurface movements (3.3.)
- the dynamics of a roller, wheel, or tyre (4.1.)
- the stresses In the contact area (4.2.)
- the stresses In the soil (4.3.)
The kinematics of the "wheel" concentrates on wheel slip, meas-
uring methods for slip, and wheel slip during ploughing. Several
parameters that affect the kinematic and dynamic aspects are
discussed also.
Soli characteristics concerning load-bearing processes are
discussed In chapter 5. Special attention Is paid to character-
izing processes (5.4.). We have made comparisons of different
characterizing processes In order to choose those tests with
which maximum information about the soil can be obtained with a
minimum of tests.
Prediction methods have been divided Into:
- comparative methods (6.1.)
- empirical methods (6.2.)
- approximate methods (6.3.)
- exact methods (6.4.)

13
The relat ionshlps between soil characteristics and process as-
pects hav e been tested for their suitability to predict process
aspects. Predictions concentrate on predicting rolling resls-
tance, ru t depth, and soiIcompaction due to the passing of a
towed tyre
Chapter 7 deals with the soil physical aspects of load-bearing
processes. Both the Influence of soli physical properties on
mechanica1 properties and the Influence of soli mechanical prop-
ertles on physical properties are discussed. The former concen-
trates on tre Influence of soil aggregate diameter and soil air,
wh1Ie the latter pays special attention to tillability.

14
 CHAPTER 2

AGRICULTURAL ROLLERS, WHEELS, AND TYRES

A distinction needs to be made between rollers and wheels. There-

fore, it is necessary to give some definitions before a classifi-
cation can be made. Rollers have as most important functions:
pressing and firming of the surface soil. The most important
characteristic of wheels Is their transport function. So a roll-
ing device, with a narrow width relative to the diameter, Is
called roller when It Isused for firming a strip of soil. The
same device Iscalled wheel If it has a transport function.

2.1. ROLLERS

Rollers can be divided into two groups: roller-packers and

roller-harrows. Roller-packers have as most Important functions:
pressing and firming of soil. Roller-harrows are the Intermedi-
ates between roller-packers and harrows. Rollers are often compo-
nents of seed-bed preparing combinations.

2.1.1. ROLLER-PACKERS

Rollers can have one or more sections. A section consists of an

operating part that has been mounted on a rectangular frame. An
operating part can have one or more elements. Main parameters of
agricultural roller-packers have been given by Schilling (1962),
Bernackl et al. (1972), and Estler et al. (1984).

2.1.1.1. SINGLE ROLLER-PACKERS

These have operating parts with only one element. Plane rollers
have smooth cylindrical elements. The light plane rollers are
used for firming and smoothing of arable land. Clods are crushed
to some extent and pressed into the soil. Narrow plane rollers,
often fitted with semi-tyres, are used In drills to firm the
seed-bed. The heavy plane rollers are used on grassland for
recompactlon after frost damage. Driven plane rollers are used as

15
part of disinfection equipment for soil. Rollers fitted with
prickers can be used for maintenance of sports fields.
A sheep-foot roller can be used for the puddling of wet rice
fields. Sometimes amudroller Is used for this (Scheltema, 1974).

2.1.1.2. COMPOSED ROLLER-PACKERS

The sections of these rolIers are composed of a number of el-

ements which are either rings or discs. The elements can move
Independently of each other.

Smooth ring rollers. This type of roller leaves a rougher soil

surface than a plane roller. As a result there Is less danger of
erosIon.

Serrated rollers. Serrated rings have better ground contact than

smooth rings. These rollers are often used In grass seeders.

Toothed rollers have crust-breaking qualities.

CambrIdge rollers consist of smooth rings and flat-toothed discs

of greater diameter set alternately. The toothed disc can rotate
freely of the ring hub. Slight differences in number of revolu-
tions of rings and discs, resulting from different diameters,
cause the Cambridge rollers to be self-cleaning. Compared with
smooth ring rollers Cambridge rollers crush soil clods more
intensively and firm the soli more deeply, leaving shallow crev-
ices on the surface and a slightly pulverized soli.

Crosklll rollers consist of rings secured to the periphery with

several lateral lugs. Crosklll rollers provide aggressive brea-
king and crushing of clods and a somewhat shallower packing than
Cambridge rollers. Similarly to the Cambridge roller the Crosklll
rings can be separated by toothed discs.

Furrow packers. The above-mentioned roller-packers can not acce-

lerate the process of soil settlement. The use of very heavy
packers^ for this purpose would cause better packing of soil
sublayers, but the top soil would be packed too strongly.
Furrow packers are used when there is not enough time for natural
soil settlement between ploughing and seed-bed preparation. Work-
ing elements are narrow wedge rings, smooth or toothed, set on
spokes. The ring penetrates easily Into the soil to a depth of
ten or more centimeters and firms the lower soli layers. The
upper soil layer remains loose and can even be pulverized addi-
tionally by the ring spokes. Furrow packers are often used In
combination with ploughs.

16
2.1.2. ROLLER-HARROWS

These rollers have no pronounced packing task, but are used for
soIIcrushIng.

RoI ler-crumbIers or string-rollers form a thin weII-puIverIzed

surface on top of a slightly packed layer. Especially the roller-
crumbIerswith small diameters have good crushing qualities.
Roller-crumbIers generally are components of combined tillage
Implements and are available Inmany designs and constructions.

Rotary harrows with spiked teeth or knives move like rollers, but
according to their effects they should be classified as harrows.

2.2. WHEELS

The wheel, one of the greatest Inventions of mankind, meant a big

step forward In transportation. Archaeological evidence Indicates
that the essentially wheel-like form was Invented in Sumer in
ancient Mesopotamia around 3500 BC. The first wheels were solid-
disc wheels of wood with a 0.50 - 1.00 m diameter and a width of
0.10 - 0.15 m. They were towed by oxen.
Even the earliest wheels were fitted with a kind of wearing
surface (tyre). Copper was the first product used for this pur-
pose (Freitag, 1979). The spoked wheel towed by horses appeared
around 2000 BC.
The first-known concept of a powered vehicle Isa drawing of a
wind-powered vehicle, made by Valturo In 1472. In the 18th cen-
tury the first attempts were made to create steam-powered ve-
hicles.
The use of powered wheels In agriculture was much encouraged by
the Invention of the internal combustion engine. The first agri-
cultural tractors had smooth steel wheels and therefore, had
severe wheel-slip problems. The wheels were fitted with a few
lugs to lower wheel slip. In the early 1900s wheels were avail-
able with different lug patterns.
After the introduction of tractor and implement tyres In the
early 1930s, the rigid (steel)wheels almost completely disap-
peared from agricultural machines. Inmodern agriculture wheeled
agricultural tractors are fitted with pneumatic tyres. Sometimes
a remainder of the days before the agricultural pneumatic tyres
Is fitted to the tractor in addition to the driven wheel. One of
the remainders Is the strake :a device to improve traction under
poor surface conditions on cohesive soils. Cage wheels can be
used to reduce compaction of the seed-bed during seeding opera-
tions. A similar construction known as puddle wheels can be used
In paddy field preparation because of their puddling effect.
Sometimes steel wheels are used for depth control of tillage
tools, but In general the steel wheel has only a supplementary
task.
For transport purposes the rigid wheel has been almost completely
replaced by wheels with pneumatic tyres.

17
2.3. TYRES

2.3.1. HIGHLIGHTS IN TYRE DEVELOPMENT

The first patent describing a pneumatic tyre was granted to R.W.

Thompson In 1845 and the first practical tyred wheel was devel-
oped by J.B. Dunlop In 1888. His tyre was fitted on a bicycle
wheel. Further development resulted In passenger car tyres at the
turn of the century and availability of truck tyres In the early
1920s.
Because operating conditions In agriculture are considerably more
rigorous than on the road Itwas not until the early 1930s that
tyres with an adequate field performance were generally avail-
able. The first tractor rear tyres were Introduced by Fyrestone
and Continental, while Dunlop Introduced the first Implement
tyre.
A further highlight in history of agricultural tyres was the
introduction of the radial-ply tractor rear tyre by Pirelli about
1957.
Other Important factors in the development of modern low pressure
tyres are: - the use of nylon as carcass material
- wide base rIms
- tubeIess tyres.

2.3.2. TYRE CONSTRUCTION

Fig. 2.1 shows the most Important components of a pneumatic tyre.

The carcass consists of cord layers (plies) laid In specific

directions. The flexible carcass carries the Inflation stresses
and gives strength to the tyre. Carcass strength depends on the
properties of the cords used, the angle between the cord layers,
and the number of pIIes In the carcass.
Originally the tensile strength of the carcass was specified by
mentioning the number of piles of cotton. Later on other ma-
terials with a higher tensile strength than cotton were used In
tyre construction, and therefore the strength is given in ply-
rating (PR). This isan index of tyre strength to cotton strength
and does not necessarily state the number of plies. Pfl-vaIues do
not give information about the allowed driving speed. In the near
future PR will be replaced by a Load Index {LI) and a Speed
Symbol. These new marks have been standardized in the E.E.C. for
agricultural tyres, tyres for passenger cars, and truck tyres.
The first tyres with these new marks are already on the market.
More details about LI and Speed Symbols are given In section
2.3.3.
Detailed Information about materials and their properties can be
found InClark (1971a).
Conventional tyres have a cross ply construction. Successive cord
layers of this tyre type cross each other at an angle of about 60
to 80 degrees (Fig. 2 . 2 ) . The cords of radial-ply tyres run

18
radially from bead to bead. A stiff belt for Improved stability
has been added to radial tyres. This construction makes for
relatively flexible sidewalls together with awelI-braced tread.

The bead must keep the Inflated tyre on the rim seat and press
the tyre against the rim flanges. Therefore, the bead usually has
a coIIof steel bead wires.

The sldewall protects the carcass. The rubber of the sIdewaII

must have excellent resistance against fatigue and aging.

The tread Is bonded to the carcass and the pattern Ismoulded In

the tread. Tyre operating performance and driving qualities also
depend on the type of tread pattern.

tread

carcass

sidewall

bead

beadwires
Fig. 2.1. Main components of a tyre.

radial ply cross ply

Fig. 2.2. Ply construct Ion of tyres.

19
2.3.3. TYRE SIZE SPECIFICATION

The size of the early agricultural tyres was specified by quoting

the sectIonwI dth b and the nominal rim diameter d, both spec-
IfIed In Inches (Flg. 2 . 3 ) . These tyres had an aspect ratio h/b
of about 1.0. An 11-36 tyre had a section width of 11 Inches and
a rim dIamete r of 36 Inches. This resulted In an approximate
overaII dIamete rD of d + 2h - 58 Inches.
Introduction o f the extra wide base rim In the middle of the
1950s resulted in awider cross section b for the same section
height h, and t herefore the aspect ratio h/b decreased from about
1.0 to about 0. 85. The new section width was added to the Indica-
tlon: 12.4/11-3 6.
Nowadays the old section width has disappeared from the size
designation so that the tyre size 12.4/11-36 Is now specified as
12.4-36.
The developmen t of tyres with even lower aspect ratios resulted
In the IncIus ion of the aspect ratio in the tyre designation:
9.0/75-18.
At present the re are many tyre size specifications In agrlcul-
ture. The rea son Is the continual Introduction of new types of
agricultural t yres and the use of tyres from other fields of
appIIcatIon In agrIculture.

Fig. 2.3. Tyre and rim descrIptIon.

b « tyre section width S» tyre deflectIon
D * overall tyre diameter h = tyre section height
d m rim diameter r'= static loaded radius

20
Recently standards for tyre size, Load Index (Lit, and Speed
Symbol have been accepted within the E.E.C.
Load Index Indicates the allowed tyre load W^In a number code.
Tyre catalogues generally contain tables with LI and Wf^.
LI can have values between 0 and 209; corresponding M^ vaIues
range from 450 N to 185,000 N. The relationship between W/^ and LI
can be expressed as:

U//80)
Wt 450.10

Speed symbols up until 40 km/h are Important f or agricultural

purposes. The speed symbol for this field of appI Icatlon is the
letter A and a number between 1and 8. This number mu111pied by 5
shows the allowed driving speed. A6, for example means an aIlow-
ed driving speed of 30 km/h.
Although there was no agreement about a symbol for Inflation
pressure, some manufacturers nowadays mark thel r tyres with a
symbol for this aspect as well. Inflation pressur e Ismarked by
one, two, or three stars. One star means that t he aIlowed tyre
load Is based on an Inflation pressure of 1.6 ba r. At two or
three stars allowed loads are based on Inflation p ressures of 2.4
bar and 3.2 bar respectively.
Fig. 2.4 shows an example of the marks that ca n be found on a
modern West European tractor rear tyre:

18.4 R 38 146 A8 *

Where,
18.4 tyre width of 18.4 Inches (=46.7 cm)
R Radial tyre; Cross ply tyres do not have letter R
38 rim diameter of 38 Inches (-96.5 cm)
146 Load Index: maximum allowed load of 30 kN
A8 maximum allowed driving speed of 40 km/h
* allowed load Isbased on an Inflation pressure of
1.6 bar.

Fig. 2.4. IndlcatIons on a modern West European tractor tyre.

21
2.3.4. TYRES USED IN AGRICULTURE

A wide variety of tyre types and sizes Is available. Each type

has Its own special fields of application. The tread pattern Is
based on the functional requirements of the tyre and Is therefore
a good visual guide to the expected use of the tyre.
Identification of the various types of agricultural tyres has
been facilitated by the adoption of a uniform code-marking system
(Table 2 . 1 ) . This code has been Initiated In the USA by the Tyre
and Rim Association and has been adopted by the American Society
of Agricultural Engineers (ASAE) and the European Tyre and Rim
Technical Organization (ETRTO). The Intention was to stamp the
code on each tyre, but In practice the manufacturers marked
their tyres with trade names. Sometimes the code marking Is given
Inmanufacturers' catalogues beneath the trade Identifications.
Amore simple identification code Isused inWest Germany. Trac-
tor rear tyres have the code AS (Ackerschlepper). The codes AW
(Ackerwagen) and AM (AckermachIne) are used for implement tyres.
Tractor front tyres have the code AS-Front.

Table 2.1. International tyre code.

tyre type code markIng

Rear Tractor: Regular Agr icu1ture

Rice and Cane
1ndustrla1 and Sand
1ndustr1a1- -Lug Type
Front Tractor: SIngIe Rib
Regular Agriculture
1ndustrla1 Rib
Garden Tractor: Regular Gar den
Intermediate Tread G2
Rib Tread G3
Implement: Rib Tread 1
Ut11Ity 2
Traction Implement 3
Plough 4
Plough 5
Smooth Tread 6

22
2.3.4.1. TYRES FOR DRIVEN WHEELS (Tractor Rear)

Tyres for driven wheels generally have a V-shaped tread pattern.

The normal direction of rotation of these tyres Is such that In
the "footprint" the top of the V points In opposite travelling
direction. Sometimes the rotation direction for driven front
tyres Is reversed to prevent excessive wear In road traffic.
Tread patterns can be classified Into groups with almost equal
characteristics. These tread pattern groups are discussed below.

Convent lonal field tread pattern (R-j)

This Is a general purpose open centre tread for maximum self-
cleaning. Tread density Isabout 30 %. One manufacturer supplies
a variant with a lug angle of 67 degrees Instead of the normally
used angle of 40 to 45 degrees.

Convent Ional f le Id-road tread pattern (R-|)

Tyres with an Increased tread density (of about 50 %) are useful
when the tractor Isoften driven on the road. The lugs are wider
In the middle than on the sides of the tread. These tyres are
Intended for use on grassland farms and on non-cohesive soils.

Half-high tread pattern (R^)

Tyres with lugswhich are about 30 % higher than those of tyres
with a conventional field tread have been designed for use on
cohesive soils. Modern tyres of this type often have a radlaI-ply
construction and are also useful when much time needs to be spent
on the road.

HIgh-speed road-f le Id tread pattern (R-|)

A tread with many low lugs characterizes tyres which can be used
when the tractor isdriven at a speed which Is higher than the
normal maximum of 40 km/h for agricultural tyres. Also, these
tyres are useful on non-cohesive soils when relatively low pulls
are needed.

High-lugged tread pattern (R2)

Tyres with extra high, often curved, lugs and a low tread density
are intended for use on very wet and cohesive solIs. On the road
they have uncomfortable driving qualities and excessive wear.
Harvesters are often fitted with these tyres.

Sports Fields tread pattern (R3)

Tyres with a continuous low tread pattern cause little turf
disturbance In recreational grassflelds. The self-cleaning abil-
ity and the power output of tyres with this tread are relatively
low.

IndustrIal-lugged tread pattern (R4)

Tyres with a high tread density (of about 70 %) are mainly in-
tended for road work, but perform welIoff the road when high
pulls are not required. Besides, these tyres can be used to
advantage for earth-moving and Inmining operations.

23
2.3.4.2. TYRES FORUNDRIVEN STEERED WHEELS (Tractor Front)

Tractor front tyres generally have high circumferential ribs to

provide easy steering andresistance toside slipping. Sometimes
tractor front wheels arefitted with tyres having asports field
tread pattern.

Single Rib (F^)

Marked byasmooth tread andhaving onecircumferential rib, this
tyre hasbeen especially designed foruseInextrememud condi-
tions found Inwetriceandcane fields.
In the Netherlands some potato planters have been fitted with
these tyres.

Regular Rib (F 2 )
These tyres haveonetothree longitudinal ribs. Sometimes there
are "climbing lugs"ontheshoulders tomake iteasier togetout
of a furrow.

Industrial Rib (F 3 )
Front tyres with five longitudinal ribsandawider tread than
the previously mentioned twotypes ofpatterns, characterize this
tyre.

2.3.4.3. TYRES FORGARDEN TRACTORS

Regular Garden (G-|)

This isgenerallyasmaII-sIzed tyre with conventional field or
field-road tread. There aresometimes narrow connections between
the lugstoprevent grass from winding around thetyres.

Intermedlate Garden (G2)

This Isatyre with atread pattern similar tothesports field
tread ( R 3 ) .

Garden Rib (G3)

This tyre hasaribbed tread pattern foruseon front wheels.
Compared with a normal front tyre (F1.F2.F3) this tyre has a
wider tread forbetter flotation onsoft wet lawnsor on loose
soIIs.

2.3.4.4. IMPLEMENT TYRES

Usually thetread ofan Implement tyre Israther wide becauseof

the following requirements:
- lowInflation pressure forgood field performance
- small diameter forlowtrailer platform height
- high load-carrying capacity.

Implement Rib (I1)

A general purpose Implement tread hasseveral longitudinal ribs

24
for good direction stability and limited side slip. A variant Is
the tyre with serrated grooves for useon n o n - c o h e s l v e soils and
when much road traffic Is necessary.

Implement UtI 11ty (l 2 )

This tread is similar to the sports field tread( R 3 ) .

Traction Implement (I3)

Tyres of ground-driven Implements usually have a conventional
field or a conventional field-road tread p a t t e r n . The direction
of rotation Iso p p o s i t e to theo n eof driven tractor rear tyres.

Plough(l4>I5)
Here w e deal with a tyre for the rear or tall wheel of ploughs.
The tread hastwo ( l 4 ) or three ( l 5 ) heavy ribs. This tyre runs
In the plough furrow at an angle of about 15 to 3 0 d e g r e e s . O n e
shoulder takes the vertical stress due to load and the other one
takes the side stress from ploughs on the vertical side of the
furrow.

2 . 3 . 4 . 5 . FARM UTILITY TYRES

This group c o n t a i n s tyres for small agricultural equipment such

as hay tedders, e l e v a t o r s , and w h e e I - b a r r o w s . These tyres are
a v a i l a b l e In many different tread p a t t e r n s .

2.3.4.6. SEMI-TYRES

These special tyres are used without e f f e c t i v e p r e s s u r e and,asa

c o n s e q u e n c e , are highly flexible and s e l f - c l e a n i n g . The tread has
ribs or very small lugs. Semi-tyres a r e used on narrow rollers of
seed d r i l l s and on w h e e l s of Inter-row c u l t i v a t i o n Implements.

2 . 3 . 4 . 7 . OTHER TYRES USED IN A G R I C U L T U R E

Tyres, o r i g i n a l l y designed for other fields of a p p l i c a t i o n , are

also used ln a g r i c u l t u r e . Farm w a g g o n s a r e s o m e t i m e s fitted with
second-hand tyres from passenger cars and t r u c k s . S o m e t i m e s , when
high load-carrying c a p a c i t i e s are required earth-mover tyres,
aircraft tyres or terra tyres are used .

Earth-mover tyres are flexible and have a high load-capacity ata

relatively low Inflation pressure. Earth-mover tyres used In
a g r i c u l t u r e are generally of the f l o t a t i o n - t y p e ( E 7 ) : moulded
with longitudinal ribs or with a sports field pattern.

Aircraft tyres usually have a w i d e tread, a high load-carrying

c a p a c i t y and are relatively c h e a p . They have a stiff c a r c a s s and
a rather round c r o s s section. Therefore, aircraft tyres generally

25
show lower field performance than tyres designed for agricultural
purposes. Size specification deviates from normal agricultural
specification. When aircraft tyres are remoulded with an agricul-
tural tread pattern, they normally get an agricultural tyre size
speelfIcatIon.

Terra tyres have been developed especially for applications where

very high flotation properties are required. In comparison with
conventional tyres they have awider cross section, a larger air
volume, amore flexible carcass, and they operate at lower infla-
tion pressures. Because of the high flotation effect they have a
rather good go-anywhere performance. The tread pattern of terra
tyres is available In different designs: smooth, ribbed and
lugged. They can be used, depending on tyre size and loading, at
Inflation pressures of 0.35 bar and higher. Tyre size specifica-
tion Isdifferent from normal agricultural Indications. In the
Netherlands some heavy self-propelled slurry tanks, combine har-
vesters, and sugar beet harvesters are fitted with terra tyres.

26
 CHAPTER 3

KINEMATIC ASPECTS OF LOAD-BEARING PROCESSES

3.1. STATE OF MOVEMENT OF A ROLLER, WHEEL, OR TYRE

A roller, w h e e l , or tyre can have two basic velocities: angular

v e l o c i t y ^ and forward velocity v (Fig. 3 . 1 ) . Moving traction and
transport devices can be In five different situations:

- free rolling: the rolling circumference Is equal to 2Ttr0

slip: the rolling circumference < 2TCr0
skid (negative s l i p ) : the rolling circumference 2Ttr,
100 % slip: when cj é 0 and v = 0
100 % skid: when u = 0 and v * 0.

In normal road and field traffic a traction device always has

travel reduction and a transport device skids. Free rolling Is
only possible when Internal resistance and motion resistance are
negligible. At 100 % skid the behaviour of a transport device
will have properties of a bulldozing blade. A traction device at
100 %slip In the field digs Itself m o r e deeply into the ground.

Roller/Wheel

^ ^
^W^%
Fig. 3.1. PrIncI pal velocIt les of rollers, wheels, and tyres.

27
3.1.1. SLIP

Slip Is the relative movement In the direction of travel at the

mutual contact surface of the traction or transport device and
the surface which supports it (ASAE Standard S296.2; Hahn et al.,
1984). Travel reduction Isdefined as one minus the ratio of
distance travelled per revolution of the traction device to the
rolling circumference under the specified zero conditions. Slip
and travel reduction are often used synonymously, and are fre-
quently expressed In percentages.
Slip S (Bock, 1952; Söhne, 1952) Isdefined as:

S= 'o - °a 1 [3.1]
S,
5
o

where, s a = real travelled distance

s 0 = travelled distance at zero slip.
Slip can also be defined In velocities (Bailey et al., 1974):

co - v/rQ
S m [3.2]
co

where, co = angular velocity of the traction device

v = linear velocity of the traction device
r 0 = rolling radius under specified zero condition.

Analyses of Söhne (1969) and Steiner (1979) show that wheel slip
of a tyre on deformable soil Iscomposed of three components:
- tangential carcass and lug deformation
- tangential soil deformation
- slipping In the contact area.

3.1.1.1. ZERO-SLIP CONDITIONS

For a flexible device such as a pneumatic tyre It isdifficult to

define zero slip and no one has been able to determine the exact
position of the zero-slip point. The problem is that there is
always relative movement in the mutual contact area when a ve-
hicle system Is inmotion. Using the slip formulas has the fol-
lowing problems: rQ can not be measured directly and s 0 can not
be measured at zero slip. According to Gill and Van den Berg
(1967) the problem of measuring wheel slip Isa problem of de-
fining and measuring zero slip.
Fig. 3.2 shows the principle path of gross traction T/rQ ,
with T = torque, and pull P to slip after Schuring (1968). Slip
Isequal to zero somewhere between T/r0 = 0 and P = 0. The exact
place of the zero-slip point Is unknown, but according to Söhne
it is somewhat closer to the towed position (T/rQ = 0) than to
the self-propelled position (P = 0) because of the flow of soil
particles In the run-in zone of the tyre.

28
 P> 1
£c , r/f 0
o
o
o
•**

Q. in
-p
_ w
•=: o
D t.
Q. o

/ 0 / SI p^

Flg. 3.2. PrInclpal path of gross traction and pull to slip

(Schuring, 1968).

The zero condition generally Isa performance condition or Is

calculated from performance conditions. Conditions used to define
zero slip are: the self-propelled position, the towed position,
and the point halfway between the self-propelled position and the
towed position. Other zero conditions may be used, but no other
practical solutions have been found In literature. The specific
zero conditions chosen should always be mentioned.

Self-propel led position (SPP)

The SPP on a hard surface was used by Wlsmer and Luth (1973),
Melzer (1976), and others. Terpstra and van Maanen (1972) used
the SPP on a grass-field as zero condition. Bailey et al. (1974),
Dwyer et al. (1974), Gee-Clough et al. (1977), and many others
used the SPP on the test surface.
The distance between SPP and towed point (TP) on the slip axis
will be longer when the field conditions get worse. The defini-
tion slip Is zero In the SPP will then be undesirable. A single
wheel operating In the SPP on a field, where the performance
conditions change gradually from good to bad, only has to over-
come an increas Ing rolling resistance. If even In the worst
conditIons slip Is supposed to be zero as long as P - 0 the
wheel can not get stuck. However, a spinning wheel has 100 % slip
and stIII P - 0.
On sandy loam with amoisture content of 23.5 % Holm (1969)
measured with a 12 4/11-28 buffed tyre that pull was zero at 15 %
slip (r. calculated from tyre deflection measurements) At a
moisture content of 16.5 % and a somewhat lower wheel oad he
found zero pull at zero slip.
Steinkampf (1971) shows the Influence of the way of deter-
mining rolIing r a d i u s rQ on the pulI and rolIIng resistance
curves. In Fig. 3.3 curve P? ,which can be compared with Holm's
curve (1969), has a value o f P - 0 at S - 15% (rolling radius

29
r
o2 = °-78 calculated from tyre deflection measurements). Calcu-
lating the rolling radius from the rolling circumference In the
SPP, gives a value of r 0 i - 0.67 m and the path of pulI and
rolling resistance versus slip would be like P-\and R^ .

Fig. 3.3. Influence of rolling radius on pull and rolling resis-

tance curves (Stelnkampf, 1971).

Gee-Clough et al (1978) found nearly 10 percent higher no-

slip speeds on ther oad than In the field. These differences must
be due mainIy to si Ipwhen the tractor Is in the self-propelling
pos11Ion In the fiel d.
We drove an unba Ilasted tractor under different good driving
conditions on grass land (Wageningen sand), on a maize stubble
field (WagenIngen sa nd), and on concrete. The distance of 20 tyre
revolutions was mea sured whIIewe were driving at 2000 rpm In6
different gears.
At an Inflation pre ssure P| = 0.8 bar the ratio travelled dIs-
tance grass/concret e had values ranging from 0.981 to 0.998. On
the stubble field t he ratio travelled distance stubble/concrete
varied from 0.963 to 0.999. At p\ = 1.5 bar this ratio had values
ranging from 0.975 to 0.990. On concrete the ratio distance
traveIIed of p\ » 0.8 bar to p\ » 1.5 bar varied from 0.978 to
0.981. The average travelled distances in the field were shorter
than on concrete: 1 .2 %and 1.5 % on grassland and stubbIeflelds
respectIvely.
The preceding res ults show clearly that the SPP as zero condI-
tlon isonly useful under good traction conditions. The use of
the SPP as zero con ditlon under less Ideal traction conditions
may give mlsleadIn g and even useless results. The best SPP as
zero condition Is th e one measured on a hard surface.

30
Towed position (TP)
Analogous to the self-propelled position the use of the towed
point as zero slip condition Is realistic under good traction
conditions. The use of the TP as zero condition under less ideal
traction conditions Isundesirable. The most useful TP as zero
condition Is the one measured on a hard surface.

The point halfway between SPP and TP

Bock (1952) measured pull versus slip of a tractor in the field.
As travelled distance he used s 0 In n revolutions at zero slip,
the mean of the self-propelled dis tance s^ and the distance
travelled S2 when towed by another tr actor.
Thus s 0 = (s-|+ S2)/2.
At the Technical University of Mun Ich (Stelner, 1978 and 1979)
and the FAL of BraunschweIg-VoIkenrod e (Steinkampf, 1971), the
rolIIng radius r 0 of a tyre was foun d by Iteration to that value
for which the linear regression lines through measuring points of
gross traction and pull Intersect t he slip-axis at equal dis-
tances from zero (Fig.3.4).
The advantage of choosing the zero -sIIpcondition at the point
halfway between the SPP and the TP, above the zero conditions
measured at SPP and TP, Is that th s point is nearest to the
theoretical zero condition.

U=T/r0 P

Fig. 3.4. Position of the zeros IIp point (Stelner, 1978).

The way of determining the zero-slip condition used InMunich and

Braunschweig-VoIkenrode Is, in theory, better than Bock's method
because of the absence of front tyres using a single-wheel
tester, but has as a disadvantage the need of very advanced test
equipment. The presence of front tyres In the method of Bock can
be the reason that there Is still a force In the horizontal

31
direction at s 0 . Under good traction conditions this force can be
neglected. Under worse conditions this force can cause small
deviations from the theoretical zero-slip point. The advantage of
the method of Bock is Its simplicity of making measurements and
caleulatIons.

3.1.1.2. MEASURING METHODS FOR SLIP

There are many known slip-measuring methods. Two groups can be

distinguished: methods measuring Instantaneous slip and methods
measuring mean slip over a travelled distance.

MeasurIng Instantaneous slip

To measure Instantaneous slip advanced measuring equipment is
needed. In wheel testers as used by IMAG In Wageningen (Werk-
hoven, 1975) and NTML in Auburn (Bailey et al., 1974) electronic
pulses for forward velocity and angular velocity, processed by a
computer, can give instantaneous values of slip. Thansandote et
al.(1977) used a microwave Doppler radar to measure the true
ground velocity of the tractor and the circumferential velocity
of the driven wheel. They concluded that the Doppler radar slip
monitor seems to be feasible as a practical device for use on
agricultural tractors. Since the late 1970s Doppler radar sensors
are being installed In automatic sprayer control systems. In such
a system the radar sensor, mounted on the tractor, is used for
measuring the true driving speed. If the forward speed changes,
the control system automatically alters the sprayer settings to
maintain the target application rate. Since the early 1980s many
makes of tractors can be equipped, standard or as an optInaI,
with a tractor monitor. Such an instrument can display the true
driving speed or the slip percentage of the driven wheels.
A new generation Doppler radar sensors was Introduced In 1985.
According to Kellermann (1985) these sensors can achieve an
accuracy of + 1 %. BolIand Isensee (1987) found a much lower
accuracy In farm fields.
A radar sensor must be mounted on the tractor so that It Is
pointing to the ground at a correct angle. This angle is crucial
concerning the accuracy of the instrument: a change In this angle
results in a deviation of about 1 %per degree (Schmitt, 1986).
A tractor linkage-control system with slip control was Intro-
duced at the 1985 agricultural machinery show "Agritechnica" In
Frankfurt. In this system (Fig. 3.5),the electronics calculate
slip by measuring the true speed of the tractor with a Doppler
sensor and the theoretical (no slip) speed by measuring a shaft
rotation in the final drive. During ploughing operations the
electronic draft control system operates up until the pre-set
slip ceiling (of 15% slip). If slip exceeds this limit ploughing
depth Iscontrolled by the slip control which gives a signal to
raise the plough Inorder to reduce wheel slip. The number of
centimeters the plough Is raised Is less than 10 percent of
ploughing depth. According to Hesse (1986) this system saves fuel
and time at ploughing.

32
 electronic control unit control panel

position sensor

lifting cylinder

draft sensor
hydraulic pump control valve radar sensor

Fig. 3.5. Electronic 11nkage-controI system with add 11lonal slip

control (Bosch).

Mean slip over a travel led distance

The most popular method of measur ng slip Is the one that
measures thetravelled distance s a In n revolutions and compares
this with the travelled distance s 0 underthe zero condItIon. A
variation „.. on this Is „ the revolutions
.„counting
„„_ , v^.w..-... -
within a certaIn
distance. Sometimes electronic equipment Is used for counting the
number of tyre revolutions.
For our Investigation of slip during ploughing on farms, we
needed a slip measuring method with thefollowing qualities:
- easy touseand transport
- easy tofitup
- non-tIme-consumIng forthefarmer.
Except providing theself-propelled position, none ofthemethods
from literature meet thefirst tworequirements. Because measure-
ments under bad field conditions were also envisaged, theSPPas
zero condition was notuseful. It would a so have cost the
farmer too much time. Sowehadto find new method for our
measurements.
Tyre manufacturers provide tables with tyre dimensions, From
tables given byContinental, Dunlop, Fulda, B.F.Goodrich, Good
Year, Kléber, Michelin, Trelleborg, Veith, and Vredeste n, tyre
diameter D and therolling circumference RC0 under the zero
condition ona hard testsurface were used. For298 tractor rear

33
tyres the ratio RC0/D was calculated. This resulted In a mean
value of 3.000with a standard deviation of 0.032. So slip can
also be defined as:
s s
a a
S - 1- — . 1- — [3.3]
3D Br

where, s a » the actual travelled distance per revolution

D » the tyre diameter
r - the tyre radius.

The calculated rolling circumference values (using RC0=3D) fitted

the manufacturers values (r-0.997) very well. The largest devi-
ations were ± 2.5 %. These maximum deviations decIIned IInearIy
with increasing slip from + 2.5 % a t S = 0 % t o 0 % a t S = 100 %.
To calculate slip with formula 3.3 the tyre diameter Dor radius
r and the travelled distance per revolution have to be known. In
practice this means measuring the radius of the tyre and the
travelled distance in n revolutions.
The rolling circumferences RC0, given In manufacturers'
tables, are determined when the wheel Isself-propelling on a
hard surface with a load equal to half of the maximum load and
an appropriate Inflation pressure. At a maximum load the rolling
circumference can deviate 1to 2 %, depending on tyre construc-
tion. Our measurements with a 13.6-38 cross ply tyre at different
Inflation pressures on concrete and sand show little difference
between calculated slip with formula 3.3 (S = 1 - s a /6r) and
calculated slip with S - 1- S^/SQ (see Table 3.1).For practical
purposes there Is no need to correct formula 3.3 because of the
use of a lower or higher Inflation pressure than the inflation
pressure used In the manufacturers'flC0-determInations. In gen-
eral it Isclear that measuring slip while using formula 3.3 Isa
simple and rather accurate method.

Table 3.1. Slip values calculated under different condltIons with

two different slip calcuI at Ing formulas .

Slip A In» Slip B InX

concrete concrete sand sand

Pl=0.8 bar D|=1.5 bar P|=0.8 bar p ( =1.5 bar

10.00 9.18 11.05 8.68 10.15

30.00 29.36 30.82 28.98 30.12

A: calculated with s - 1- s a /6r

B: calculated with S = 1- sa/s0

It Is also possible to measure slip with a stationary cine-

camera. The cine-camera Is Installed In the field as follows: the
camera-1ens must be at axle-height of the wheel and the viewing
direction perpendicular to the travelling direction.
Because only relative values of sizes and distances are needed an
appropriate size of projection can be chosen for Illustration of
the results. On the film screen tyre radius r, travelled distance

34
of theaxle, andtherotated angle ofthewheel canbemeasured.
From the travelled distance andtheangle of rotation the
travelled distance s a per revolution iscalculated. Substitution
of s a and r In3.3 gives theslip percentage.
We drove a tractor at different slip-levels onWagenIngen sand.
Slip wasmeasured bya 16mm cine-camera; thetravelled distance
In tenwheel revolutions wasmeasured simultaneously. Table3.2.
shows theresults ofthetwomethods used. Theconclusion Isthat
slip canbemeasured with acine-camera, provided thecamerahas
been Installed correctly.

Table 3.2. Comparison of three s11p-measurIng methods used on

WagenIngen sand.

Track Slip A SlipB SlipC

1 8.4 % 7.6% 8.1%
2 15.7 % 15.0% 15.3 %
3 24.8 % 24.2% 23.8 %
A: s a and r measured Inthefield; Scalculated with
S - (1 - s a /6r).100%
B: s a ands 0 measured inthefield; Scalculated with
S = (1 - sa/s0).100 %
C: s a and r measured on film-screen; Scalculated with
S = (1 - s a /6r).100%.
In principle twoframes suffice when thecine-camera method is
used. Sowe tried slip measuring with 35mm color-slides. At a
concrete test surface thecolor-slide method was tested for
different camera-Ienses (50mm, 85mm, 105mmand 135mm focal
distance) andviewing positions. Thebest measuring procedurewas
the foIlowIng:
- a camera with a 50mm lenswasmounted at axle height (ona
trIpod)
- a distance of8meters waskept between thecamera andthe
track
- the camera viewing direction was perpendicular to the
direction of travelling
- the first slidewastaken atthemoment that the wheel
came completely Inthepicture
- the second slidewastaken Just before the wheel disap-
peared from thepicture.
We measured rands a atthescreen (color-slide method) and at
the test surface andcompared them afterwards. The color-slide
method showed maximum absolute deviations Inslip percentagesof
i 2.5% slip. Themean values forslip measured atthescreenand
at thetest surface were equal. Soforpractical purposes the
color-slide method Isaccurate enough.

35
3.1.1.3. WHEEL SLIP DURING PLOUGHING

There Ismuch discussion about wheel slip but information on slip

In normal field work seems to be lacking completely. There Is
some literature about slip measurements taken during ploughing
demonstrations inWest Germany (Traulsen and Splngies, 1978) and
Great Britain (N.N., 1980 and 1981a). Some results of these
demonstrations can be found In Table 3.3 and 3.4.

Table 3.3. Slip during plough!ng demonstratI on with 11 four-

wheel drive tractors In West Germany ( Traulsen and
Splngies, 1978).

tractor lowest slip % highest slip % mean slip %

front wheel 14 39 24
rear wheel 13 31 22

Table 3.4. Some results from Tractor at Work Events (N.N. 1980
and 1981a).

tractor year lowest hIghest mean number of

siip % siIp % siip % entrants

two-wheel drive 1980 13.4 28.0 17.1 9

two-wheel drive 1981 11.8 19.4 16.3 3
four-wheel drive 1980 8.0 18.4 12.2 25
four-wheel drive 1981 6.9 16.9 13.2 24

These ploughing demonstrations giveI Ittle information about slip

during regular farm work, because all tools are working under the
same field conditions and because the participants are weII-
prepared manufacturers or Importers o f tractors.
To collect more Information about slip on the farm, we meas-
ured wheel slip during ploughing o peratlons In 189 different
places all over the Netherlands. For this slip measuring it was
necessary to develop a new measuring method. The first Idea was
to measure slip "at a distance", b ecause we did not want the
ploughing behaviour of the farmer to be Influenced by our pres-
ence. Thus the cine-camera method w as born. Because the cine-
camera method is rather expensive and its use time-consuming, we
decided to use the color-slide method
During spring 1980we drove aroun d in the Dutch regions with
light soils until we saw a ploughing farmer. With his permission,
slip was measured with the color-sli de method. Driving speed,
ploughing depth, ploughing width, drI vlng conditions, and type of
tractor, plough and tyres were alsor ecorded.
It soon became clear that the plough Ing behaviour of the farmer
was not notably Influenced by the pre sence of the measuring team,
Therefore, and also because the co lor-slide method was tlme-
consuming, the measurements of tyre r adlus and distance travelled

36
on the heavy soIIswere carried out on the spot. Slip measure-
ments on heavy solIswere taken In autumn 1980. At that time the
measuring programme was extended with tyre Inflation pressures.
Detailed Information about ploughing depth, ploughing speed, and
tyre Inflation pressures Isgiven In TIJInk and Den Haan (1981).
On light soils, ploughed In spring, 100 slip measurements were
taken. In autumn slip was measured on heavy soils In 89 places.
The observed values for wheel slip have been plotted and are
shown as a cumulative curve In Fig. 3.6. The median slip value Is
14.6 % on light soils and 17.0 %on heavy soils. The highest
measured slip on light and heavy soils Is 54% and 37 % respect-
Ively. In spring as weIIas In autumn more than 60 % of the
participating farmers plough with slip values between 10 % and 20
%. On the light soils 23 % of the farmers plough with more than
20 %slip and even 33 % do this on heavy soiIs.
The wheel-slip measurements have also been analysed with re-
spect to tyre construction. Some results of this analysis are
shown in Table 3.5.
Tractors equipped with radial tyres had lower slip values on both
types of soil. These differences between slip values of tractors
with radial and those with cross ply tyres are not only due to
tyre type. Among the group of tractors with radial tyres were
many new and heavy tractors with awell-adjusted tractor-plough
combination. So It Isa combination of factors that leads to the
lower slip values of tractors fitted with radial tyres.
The Investigation of TIJink and den Haan (1981) concluded that
a lot of farmers can plough with lower slip values provided they
pay attention to the causes of wheel slip.
100

80

60

=Light soils,ploughed in spring

=Heavy soils,ploughed in autumn
40

20

10 20 30 40 50 60
Slip
Fig. 3.6. Wheel slip during pi oughtng on farms.

37
Table 3.5. Wheel slip during plough!ng In relation to tyre con-
struction .

soIItype percentage mean slip mean slip

radial tyres radial tyres % cross ply tyres %
IIght 26 13.7 22.8
heavy 37 16.1 20.1

3.1.2. MOVEMENTS OF A POINT AT THE RIM OF A ROLLER OR WHEEL

3.1.2.1. TRAJECTORY OF A POINT AT THE RIM OF A ROLLER OR WHEEL

Söhne (1952) has given a description of the trajectory of a point

at the rim of a rigid wheel. Such a point of a free-rolling wheel
on a rigid surface has the following motion parameters (Fig.
3.7):

x - r Q («t - si nut)
y - r 0 (1 - cos<ut) [3.4]

where, rQ = radius of the wheel CJ= angular velocity

rQ«*t = forward travelled distance t = time

The rigid wheel on the rigid surface can also have a slip S. The
parameters for a slipping wheel are:

x-r 0 [ ( l - S ) « t - sln<ut]

y = rQ (1 - coswt) [3.5]

The forward-travelled distance of the axle is now:

r Q (1 - S)6Jt = (1 - S)vt.
Fig. 3.7 shows remarkable changes In the trajectory because of
slip and skId.

(1+0.25)2Ttr0 2TtrQ (1-0.25)2rtr0

Fig. 3.7. Trajectory of a point at the rim of a rigid wheel on a

hard surface. A=free rolling, B=sIIpplng wheel (S=25 % ) , C=skld-
dlng wheel (S=-25 % ) .

38
3.1.2.2. VELOCITY OFA POINT ATTHERIMOFA ROLLER OR WHEEL

A moving wheel with a radius r 0 and n revolutions per second,has

a theoretical forward velocity v t n =2TTr Q n andan actual forward
velocity v a = (1- S)2nrQn. Theeffective radius r e is the
actual displaced distance s a percircumference divided by2TT(ASAE
Standard 296.2;Hahn et al., 1984). Ina formula:

sa (1-S)2TTr0
r e -— = (1- S)rQ [3.6]
2it 2Tt

Thismeans that re<rQ for a slipping wheel and re>r0 fora skidd-
ing wheel. Awheel with a radius re moves without slip, with the
same n, onan imaginary plane. Thepoint where this plane touches
the equivalent wheel Isthe instantaneous centre of rotation I.
All the points of this wheel rotate around this instantaneous
centre I.
It ispossible bymeans of this instantaneous centre of rotation
Iandby using instantaneous kinematics togive thedirectionand
the length of theabsolute velocity vectors of points at therim
of awheel. This Isshown InFig.3.8.Note the influence of slip
and skid on thevelocity vectors.
All the points at therimofa skidding wheel have a forward
velocity above zero. A slipping wheel hastwozones. The upper
one with forward velocities above zero, while inthe lower zone
it istheopposite.

3.2. MOVEMENTS INTHECONTACT AREA

3.2.1. TYRE DEFORMATIONS

A tyre under a static load shows radial deformation and sidewall

bulging, whIleamoving tyre canshow tangential carcass andlug
déformâtions.

3.2.1.1. RADIAL TYRE DEFORMATION ANDSIDEWALL BULGING

In general radial tyre deformation isaccompanied by sidewall

bulging. Measurements of Knight andGreen (1962) show deflection
and sidewall bulging ofamoving tyre on surfaces ranging from
fIrm to soft (Fig. 3 . 9 ) .
In addition to inward carcass deformation (deflection) there
can beoutward deformation. Fig.3.10shows radial carcass defor-
mation versus rotation angle ofa radial andcross plytyre. The
radial tyre hasanoutward deformation ina rotation angle range
from about 220to 140degrees. This outward deformation can be
explained as follows. Thecross ply tyre shows only Inward defor-

39
Flg. 3.8. Length and dl reet Ion of velocity vectors of points at
the rim of a wheel.

40
 (a] (b) (c)
Pj=1.03bar

'i
- unloaded
Fig. 3 . 9 . Tyre deflect ton and sIdewal I bulging at different In-
f lat ion pressures on (a) asphalt-paved concrete, (b) sand, and
Cc) si It (Knight and Green, 1962).

601
mm
0°/360°
^ 50-
c
o
o ^0-
E
o
"ai 3 0 -
•o
l/>

8 20-

- 10
o
cc

-10
0 60 120 180 240 300 360
Tyre rotation angle 9
Fig. 3.10. Radial carcass deformations of a cross ply and
radial tyre (SteIner, 1979).

matIon . T h e r e f o r e , Its c a r c a s s will be s h o r t e n e d . T h e radial tyre

has a stiff belt In the tangential d i r e c t i o n . T h i s belt d l s t r i -
butes the d e f o r m a t i o n s over thew h o l e c i r c u m f e r e n c e and Is b e -
cause of Its s t i f f n e s s the c a u s e of the outward c a r c a s s d e f o r m a -
tIon.
MostIy radlal d e f o r m a t i o n is m e a s u r e d In a s t a t i c s i t u a t i o n and
is ca led d e f l e c t I o n S . Tyre d e f l e c t i o n has been m e a s u r e d by
H o f f e r b e r t h and Relnhold (1969) Sitkel (1969) Sonnen (1970)

41
and others. Static tyre deflection depends on Inflation pressure,
load, tyre construction, and on the character of the supporting
surface. Sonnen measured remarkable differences In the deflection
curves for loading and unloading. Sltkel (1969) measured tyre
deflection and sinkage. Tyre deflection decreases with Increasing
slnkage (Fig. 3.11).

x 8.5kN:pi=0.8bar
a 6.5kN :/Oj=0.8bar
• 6.5kN:pj= 1.2bar

20 1.0 60 80 100 120mm HO

Sinkage
Fig. 3.11 I nf luence of sinkage on tyre deflectI on (Sltkel,
1969).

Max Imum allowed tyre deflectIon

The max Imum allowed tyre deflection ( £ / ^ ) m a x can be calculated
from man ufacturers' tables.
Sltkel (1969) found S/h = 0.15 at maximum load for cross ply
tractor rear tyres. From technical Information from Continental,
Krick (1 969) calculated that (<57ft)max lies between 0.13 and 0.17.
Terpstra (1978) based his calculations of (S/h)
from Eu ropean and American tyre manufacturers. He found max onvalues
Information

between 0.14 and 0.20. According to Inns and Kllgour (1978)

maxImum deflection of agricultural tractor and Implement tyres is
IImlted to about 18% to 20 % of the section height h In order to
prevent carcass damage.

DeflectI on models and geometry of the contact area

Bekker (1956) has provided amathematical tyre deflection model.
Furthermore, he assumes that the contact area Amay be determined
as an area of a rectangle reduced by 15 %. In a formula:

0.85/.B C3.7]

where, A = the area

/ = the length of the contact area
B = the width of the contact area.

42
Flg. 3.12. Tyre def/eet Ion model of Kriek.

K rick (1969) has provided a more simple model for deflection.

He a ssumes that the shape of a tyre Isa torus. This Implies that
the contact area Iselliptica Ion a firm surface (Fig. 3.12).
When deflection starts the arc s ABCand DBF, respectively In and
perp endlcular to the dlrectlo n of travel, are compressed to the
chor ds AECand DEF or the chor ds AECand DEF elongate sidewards.
The true lengths lie between the original chords and the arcs,
The differences between the original chords and the arcs are
rela 11vely small because of th e large diameters D and d. Thus the
arcs ABCand DBFare equalized to the chords AECand DEF respect-
Ivel y. The ellipse has the fol lowing half axes:

a =\[pf~ 2
b \fdS T "
The c o n t a c t a r e a A .TV^DS - <T2 . \ld<S - S2

c a n , because o f S2<<dS«D6 be w r i t t e n a s :

A =TT VDdJ 2 [3.8]

This calculated relation between deflection and contact area

correlates well with the measured relation. According to Painter
(1981) the chords AECand DEF can be approximated by the arcs ABC
and DBF for small deflections .The ellipse now has half axes:

a =D/2 arccosd -2<T/D)

b = d/2 arccosd -2<$7d)

43
and an area :
A = ( I T / 4 ) . D . d . a r c c o s d - 2&/D) ,arccos( 1 - 2S/d) [3.9]

Although the formulas 3.8 and 3.9 are both based on the same
deflection model, formula 3.8 iseasier to use.

3.2.1.2. TANGENTIAL CARCASS AND LUG DEFORMATIONS

Tangential ca rcass deformations are not only deformations in the

sense of stra Ins, but are relative displacements In relation toa
fIxed system of coordinates as well. Displacement depends on tyre
constructIon, load, inflation pressure, and torque. Fig. 3.13
shows that t angential carcass deformation is almost symmetrical
for a towed wheel (F - 0) and that the radial tyre shows dis-
placement aI Iover the circumference, while the cross ply tyre
has haIf acl rcumference free from relative displacement. Calcu-
latlons were made of the shortening of the carcass (see Table
3.6). The s hortenlng of the cross ply carcass Is about three
tImes as much as the one of the radial tyre carcass. The carcass
shortening of the radial tyre decreases when a torque is applied.

Table 3.6. Tangent Ial strains of tyre carcass (Stelner, 1979).

tyre inflation load carcass shortening
pressure (kN)
(bar) at r=0 Nm at T= 3000 Nm

13.6-28 0.8 10 32.3 mm 32.5 mm

13.6R28 0.8 10 10.1 mm 8.8 mm

True measurements of tangential lug deformations are unknown

and difficult to make. Relative displacement of lugs In the
mutual contact area consists of lug bending and carcass deforma-
tion. Measurements of lug movements In the mutual contact area
were made by Cegnar and FaustI (1961), Wann and Reed (1962), and
Stelner (1979). The most complicated method of determining lug
movement Is used by Stelner. He has drawn longitudinal tyre
sections by using castings of the contact areas. He found a
nearly linear relation between tangential lug displacement and
the lug's longitudinal position in the contact area. Cegnar and
FaustI used high-precision optical equipment, whereas Wann and
Reed used scratch plates and a "bar table". Both the optical
method and the bar table method showed a curved relation between
tangential lug displacement and the lug's longitudinal position
In the contact area (Fig. 3.14a). A comparison of the measured
and the theoretical curves shows that - the lugs, when entering the
contact area, follow the carcass with a certain delay due to lug-
to-surface friction. When the lugs leave the contact area, their
relative displacement Isgreater due to their elastic reaction.
This behaviour agrees with observations of the direction of lug
movement made by Anslow and Warrllow (1962). The optical and the

44
 30 90 150 270
Tyre rotation angle 9

Fig. 3.13. Tangential carcass déformât Ion of a cross ply and a

radial tyre (Stelner, 1979).

c (a) lb)
5
e mm
Ol measured — cross ply
o
o U theorie — radial
Q.
UI

200 mm 300
Position under the contact area

Fig. 3.14. Tangent lal lug displacement In the contact area of a

point 10 mm from the centre plane of a cross ply tyre (a), and
(b) total tangent lal lug displacement at different points of the
cross section of a cross ply and a radial tyre.

45
bar table method also measure thetangential lugdisplacement at
different points ofthecross section. Fig. 3.14bshows thepath
of thetotal tangential lug displacement versus thewidth ofthe
contact area. Thetotal displacement Isameasure fortheenergy
lost In friction perwheel revolution, and Is therefore a measure
for wear. SoFig. 3.14bpictures tyre wear across thewidth.The
area under thecurve of theradial tyre wasfound to be 26 %
smaller than theoneofthecross plytyre. Total lugdisplace-
ment washigher at low Inflation pressures than at high inflation
pressures inthescratch plate tests. These results agree with
the practical experience that radial tyres wear less than cross
ply tyres and that wear ishigher at low Inflation pressures.

3.2.1.3. INFLUENCE OF TYRE DEFORMATIONS ON SLIP

Only tyre deformations Inthemutual contact area can Influence
slip. Ona fIrm surface, slip Is influenced bycarcass and lug
deformation. Using his drawings of longitudinal tyre sections
Stelner measured slip values duetotyre deformations of7.54 %
and 5.15 %ofa cross plyanda radial tyre respectively (see
Table 3 . 7 ) . Slip is influenced more by lugdeformations than by
carcass deformations. Estimates of slip components ona deform-
able soil were also made. These show that the Influence of tyre
deformations on slip should notbeneglected and that radial
tyres have lower slip percentages dueto fewer tyre deformations.

Table 3.7. Measured slip components on a firm surface and esti-

mated slip components on a deformable soll (Stelner,
1979).

Slip caused by
carcass lug solI sIIppIng In totaI
Tyre defor-
défor- defor-
défor- defor-
défor- thecontact
mâtIon mâtion mâtIon area
13.6-28 1.86%%
1.86 5.79 % %
5.79 7.54%
13.6R28 0.55
0.55 %% 4.354.35
% % 5.15%
13.6-28 1.2 %%
1.2 4.34.3
% %8.0% 8.0 7.5 % % 21.0 %
13.6R28 00.4
.4 % 2 .2.6
6 % % 7 . 57.5
% % 7.5% 18.0%

3.2.2. MOVEMENTS OF A POINT ATTHECIRCUMFERENCE OF A ROLLER,

WHEEL, OR TYRE DURING MOTION INTHEMUTUAL CONTACT AREA
The trajectory ofa point attherimofa roller orwheel during
motion inthecontact area ona rigid surface canbedescribed by
formula 3.5. A rigid wheel with a radius rQ andmoving with n
revolutions per second ona rigid surface, has a theoretical
travelled distance s^n • 2ttr 0 n. Söhne (1952) neglects tangential

46
tyre deformationsandtherefore assumes that tyres only deformIn
the radial direction. Thetrajectory ofapoint atthecircumfer-
enceofatyreonadeformable surfacecannowbewrittenas:

x = r Q . ( 1 - Sfc) .<ut - r.slnÄ>t

y mr0 - r.cosat [3.10]

where, r =r0 - Sr
Sr =radial tyre deformation
S t =( s t n - s)/sth
When wetake tangential deformation 6± Into account, the for-
mulas change into:

x = r0.{^ - S^) .<»t - r.s\not +<ft.cosot

y = rQ - r.cosut - S^.s\nat [3.11]

The formulas forapoint attheInsideofthetyre carcass are:

x = r Q . ( 1 - Sf.) .ot - ( r + h).sin<yt + <ft.cos<ut

y = rQ - ( r + ri).cos<ut - <T t .sln»t [3.12]
where, h =thedistance between theInsideofthecarcassand the
lug face.

Stelner (1979) calculates theparametersofapoint atthe lug

faceasfollows: starting withapoint (X2.Y2) a t t n e Inside of
the carcass, heassumes that thelug Isperpendicular totheline
through thenearest measuring points (x^,y^)and (x3,y^) atthe
Inside of thecarcass. Thereare measuring pointsateach 2.5
degrees.Thedirection ofthelug Is:
tan eCL - ( y 3 - y^)/(xy - x3)
Sotheparametersofapoint atthelugface are:

X|_=Xg -r».slncC|_
/l_= y2 - rt.cos«|_ [3.13]
These calculations do nottake thebendingof the lugs Into
account.

3.2.3. TRAJECTORIESOFSOIL PARTICLES INTHECONTACT AREA BETWEEN

THE SOILAND
AROLLER, AWHEEL,ORATYRE

Poletayev (1964) used theangle between the direction ofthe

velocity vector and theradial atapoint ona wheel rim to
demarcate different zones in the wheel-soil interface.
Theoretical analysesoftangential displacement ofa soil par-
ticleattherimcircumference were madebyJanosI (1962), Onafe-
koandReece (1967), andWong andReece (1967a,1967b).
Wong (1967) determined thepathofparticles, onaclay soil,
Influenced by aroller (Fig. 3.15). Theparticles were first
pushed forward andupInastraight linebytheoncoming roller.
They were then shifted backwards Inacircle. After the roller

47
 ,u
negative slip
w %z>
negative slip positive slip
<Utc
positive slip
no rut left permanent rut norutleft permanent rut
Fig. 3.15. Movement of a soil material point relative to the
untouched soil, due to a roller passing from left to right.

mm*
•mmmmmmâÊÈmmM

mgm»ÈËËÊÊËÊmÊm

marl *$$$:;

Flg. 3.16. Movements of soil under the action of towed and driven
tyres (Söhne, 1952).

48
had passed by the particles were at the same vertical level as in
their Initial position because of the IncompressibIIIty of the
soli used. The final positions of soli surface particles were
ahead of their initial positions when a towed roller was used,
while the final positions of the particles were behind their
Initial positions when a driven roller was used.
Söhne (1952) showed the trajectory of soil between the lugs of
a tyre and of soli particles at the soli-lug Interface (See Fig.
3.16).
We determined the trajectories of soil surface particles be-
neath a towed tyre from our measurements of three-dimensional
soli deformation. Fig. 3.21 shows two such trajectories of soil
surface particles in the central longitudinal section of the ruts
formed by a towed tyre in Schinnen silt loam and in WagenIngen
sllty clay loam. These trajectories differ from the ones found by
Wong because we used deformable soils and a deflecting tyre.
Besides a vertical displacement Jv there isa resulting hori-
zontal displacement _/n In deformable soils. We placed small dots
at the soil surface In our soil bin Investigations. The dots were
placed in such places that after passing of the tyre, the dots
were In the central longitudinal section of the rut. The posi-
tions of the dots were measured, In horizontal and vertical
direction, before and after the first and second passing of a
towed tyre. In the test five different soils were used at differ-
ent moisture contents and bulk densities.
In Fig. 3.17 the resulting horizontal displacement 7 n has been
plotted versus the vertical displacement Jv of soil surface
particles In the central longitudinal section of the formed rut.
The vertical displacements are equal to the sinkage In the centre
of the rut. The 48 measuring points fit a power function rather
well (r=0.98). It Is remarkable that the ratio J n /J y Increases
with Increasing rut depth. This must be due to the presence of a
rigid bottom at 15cm depth. The soil under the tyre can move
less easily downwards at an increase of rut depth and therefore
will move more in a forward direction. Finally It isclear that
horizontal displacement can not be neglected, especially when
deep ruts are formed.

49
 horizontal displacement j.

o 20 i0 60 80 100 120 mm U0
O-jp
T.
• after f i r s t pass
o after second pass
20 o«

i.0
c
<D
E

!0 60

- y v =6 . 3 1 y h 0 5 8 ( r = 0 . 9 7 )

80

mm

100

Fig. 3.17. HorIzontal and vertical dIsplacement s of points at the

soil surface In the wheel-centre plane.

50
3.3. SUBSURFACE MOVEMENTS

The most simp e two-dimensional movement Is the one under InfI-

nIteIy wide roIIers because of the absence of a sideward move-
ment. Soll under a wheel wheel
or or a tyre can escape sideways This
sideward transport does not exist In the central longitudinal
section of the formed rut Isalso possible to show the two-
dimensional deformation In the cross section of the rut. This is
already a projection of a three-dimensional deformation on a flat
plane.

3.3.1. SOIL MOVEMENTS UNDER ROLLERS

When rollers are considered as Infinitely wide rollers we have a

true two-dimensional problem.

3.3.1.1 DISTRIBUTION OF SOIL VELOCITIES RELATIVE TO THE CENTRE

OF A ROLLER

Wong (1967) used the principle of continuity of flow for rollers

at Incompressible soils. When a roller ismoving and In equilib-
rium the soil surface, after passing of this roller, has to be at
the same level as the undisturbed soil In front of the roller.
Fig. 3.18a shows a horizontal distribution of soil velocities
under the centre of a slipping roller. The relative velocity of
the soli In front of the roller, where the soli is not Influenced
by the roller, Is uniformly distributed down the depth of the
soli and equal to the forward velocity v of the axle. The dis-
tribution of velocities beneath the axle requires sinkage of the
roller. The higher the slip of the roller the higher the sinkage
must be. Since the final level must be equal to the initial level
In front of the roller, It is necessary that the slipping roller
affects rut recovery.
The use of the principle of continuity of flow for a towed roller
gives a velocity distribution as shown in Fig. 3.18b. Such a
distribution requires no sinkage of the roller.

at positive slip at negative slip

"act "act
Fig. 3. 18a,b. Velocity profiles deduced from contInulty conslder-
at Ions.

51
3.3.1.2. FLOW ZONES UNDER ROLLERS
Flow patterns under rollers have been photographically determined
by Wong and Reece (1966), and Wong (1967). Their observations on
Incompressible soils show that there are normally two flow zones
beneath a towed or a driven roller: a forward zone a and a
backward zone b (Fig. 3.19a+b). The zones degenerate Into a
single backward flow zone at 100 % slip (Fig. 3.19c). It Is
Interesting that the locked roller Isconverted as a backward
raked blade to a vertIcal blade made of solI (Fig. 3.19d).
The boundaries of the flow zones are parts of logarithmic spirals
In sand. In clay, those boundaries are sections of circles.
Increasing the load of the towed roller leads to:
- Increased skid values
- growth of the flow zones a and b
- moving of the boundary between the two zones in the direc-
tion of travel
For a driven roller an Increasing load at equal slip results In:
- more sinkage of the roller and, therefore, more rut recov-
ery
- growth of the zones a and b
- moving of the boundary between these two zones In the
direction of travel.

3.3.2. SOIL MOVEMENTS UNDER WHEELS

A difference between a conventional wheel and the Infinitely wide
roller Is that It Is possible for the soil to flow sidewards
under a wheel, but not under a roller.
Wong (1967) studied the flow patterns under narrow wheels in
Incompressible sand. These experiments show that even under a
narrow wheel soil has two distinct flow zones In the longitudinal
section. These two zones have boundaries with similar character-
istics as those beneath rollers. Rut recovery under narrow wheels
grows with Increasing wheel width and slip. Longitudinal and
sideward flows occur together. The relative extent of the two
mechanisms depends on the degree of slip and on wheel width.
Comparisons of the profiles of ruts and heaves left by a driven
wheel and a narrow plate In an Incompressible soil show that:
- the Influence of the plate goes further sideways. The
reason is that all the soil under a plate flows sidewards,
while the soil under a wheel also flows forwards and
backwards.
- rut recovery is higher In the middle of the rut than on
the sides.
- the heaves on the border of the wheel rut are higher.
So the soil under a wheel does not flow In the same way as under
a plate.

52
 (a) (b)

Fig. 3.19. Flow zones In sand under (a) a towed wheel, (b)
driven roller, (c) a driven roller at 100 % slip, and (d)
locked wheel with 100 % skid (Wong, 1967).

3.3.3. SOIL MOVEMENTS UNDER A TYRE

Measurements of soil deformation due to the passing of a towed

tyre have been made by TijInk et al. (1988). From these measure-
ments we took the central longitudinal sections (Fig. 3.20).
These deformations agree with visualizations of the deformations
under a tyre, given by Söhne (1956). Wageningen sllty clay loam
compacts strongly under the action of the towed tyre, while
Schinnen silt loam shows more deformations and less compaction.
The measurements of Söhne (1952) Indicate that the horizontal
displacement under a tyre decreases with depth. Gliemeroth (1953)
measured photographically the trajectories of soil particles
under the action of a tractor In the field. He found that the
horizontal movements decrease more with depth than the vertical
movements.
The trajectories of soil particles at the surface and at 0.05
m and 0.10 m depth were determined from our longitudinal sections
of the ruts formed by a towed tyre (Fig. 3.20). The resulting
trajectories (Fig. 3.21) have a different shape for each of the
two types of soils. It isalso Interesting that the shape of the
trajectories changes with depth. The different shapes of the
trajectories must be due to the different shapes and extents of
the flow zones.

53
 original soil surface

g.N^..,..! ........v.,.! T ^,,..yt,.\ ^.t^V^iyt^....^. < r r , ^ , ... ..s, ..-..v.^. t ..\ 4 ]

bottom of soil bin

Fig. 3.20a,b. Déformât Ion of soil under the tyre centre plane.
The tyre moves from left to right.

soil surface

10cm

Wageningen Schinnen

Flg. 3.21. Movements of soil particles at the soil surface and at

5 cm and 10 cm depth due to a towed tyre passing from left to
right.

54
 CHAPTER 4
DYNAMIC ASPECTSOF LOAD-BEARING PROCESSES

4.1.FORCES,MOMENTS,ANDSTRESSESON ROLLERS,WHEELS,AND TYRES

There are three forcesand threemoments acting from the ground
on atyre.
F x isthe longitudinal, F y isthecornering (-lateral), and Fz
the normal component of the force from theground to the tyre.
The force F v is important for steered wheels, butwiIInot be
discussed inthischapter.
Theoverturning moment 7"x Isthemoment of forceabout theX-axis
acting on thetyre from theroad. Themoment about theY-axisis
called rolling resistancemoment 7 V . Aligning torque7"z is the
moment about theZ-axIsactingonthetyre. Adriving torque T
about the rotation axisof the tyreproducesa force that can
accelerate or decelerate thevehicle.

4.1.1.MECHANICAL EQUILIBRIUMOF ROLLERS,WHEELS,AND TYRES

The forces, torques, and stressesactingona roller,wheel,or
tyreon ahorizontal soil surfaceare illustrated InFig.4.1.

crn(9)
crn(9
Fig. 4.1. Forces, torques, and stresses on (a) a driven roller,
(b) a towed tyre, and (c) a driven tyre on deformatie soll.

55
Roller and rigid wheel
The equilibrium of a stationary moving roller or rigid wheel on
horizontal soil surface can be described by means of the follow-
ing equations:

W= 51vertical components ofoj + vertical components ofr tf [4.1]

P = Z horizontal ,, of <r. + horizontal ,, ofx0 [4.2]
T = £ r.Ti [4.3]

For a towed roller or wheel, Input torque T » 0 and, therefore,

shear stresses must occur In forward as well as backward direc-
tion. The rolling resistance of a towed roller or wheel can be
equalized to the towing force (equation4.2).
For a driven roller or wheel rolling resistance Ismore complex
and can be Illustrated by means of an energy balance:

Input =Output + Loss.

According to Corcoran (1979) energy components can be based on

unit wheel travel and unit ground travel.

Based on unit wheel travel the components of the energy balance

are:
Input = T/(r0.W)
output = P(1-S)/W
loss = rolling resistance energy = {T-P-re)IrQ.W

Based on unit ground travel the components are:

Input = r/[r 0 (1-S).W\

output = P/W
loss » rolling resistance energy = (T-P.re)/re.W

PneumatIc tyres
The equations for P and Ware the same as those for rollers and
wheels. Because of the tyre deflection the equation for T
changes. Because of this deflection the normal stresses <r con-
tribute to the torque T and the equation for the torque now
becomes:

T = Y.r.(components ofeg and T6 that are perpendicular tor ) .

Components of rolling res I stance

Rolling resistance defines the total amount of energy lost and Is
composed of several components:
- rut forming In the solI
- friction between the rolling device and the travelling sur-
face caused by sliding
- hysteresis in tyre materials due to deflections when the
tyre is roIIIng
- resistance due to air circulating Inside the tyre
- fan effect of the rotating wheel on the outside air
- friction of the wheel bearings.
The last three components are of secondary Importance and can be

56
neglected for agricultural purposes. On the road 90 to 95 % of
the energy losses are due to internal hysteresis of tyre m a -
terIaIs.

4.1.2. TRACTIVE PERFORMANCE

Tractive performance of tractors and single w h e e l s has been

measured by many researchers. The presentation of the results in
performance curves is sometimes confusing and, therefore, an
analysis will be given of the most common tyre performance p r é s -
entâtions.

4.1.2.1. ANALYSIS

The tractive performance of tractors have been measured by Bock

( 1 9 5 3 ) , Terpstra and van Maanen ( 1 9 7 2 ) , and o t h e r s . Their results
have been presented generally in puII-sIip c u r v e s . Their presen-
tations give information about the (drawbar) pull at different
slip v a l u e s , but no information about the efficiency of the
operation. Because of weight transfer during the m e a s u r e m e n t s It
is difficult to transfer the characteristics Into other tractors.
When tyres are tested with a single wheel tester It Is gen-
erally possible to m e a s u r e input torque, puII, and siIp. The
results of single wheel tests can be used for every tractor.
Traction performance presentations (Fig. 4.2) of such tests can
have the following c o e f f i c i e n t s :
- pull coefficient %= P/W
- tractive efficiency: ratio between output and input
[ -q= p.v/T.o]
- gross traction ratio M - U/W -,where U= T/rQ
- rolling resistance coefficient / = R/W
The pull coefficient versus slip combined with the tractive
efficiency versus slip m a k e it possible to chose a tyre that has
a high efficiency at the pulIs needed. Sometimes efficiency and
siIp are plotted versus pulI.
The rolling resistance coefficient and the gross traction ratio
are sometimes given in the presentations. This is rather c o n -
fusing because it suggests that gross traction and rolling resis-
tance of driven tyres can be measured. It also can give the
impression that there are two horizontal forces in the tyre-soil
contact area: gross traction (thrust) in a forward direction and
a rolling resistance force in the opposite direction. According
to Schuring (1968) these forces do not exist in the soil-tyre
Interface.
The most important aspects of tyre performance on farms are:
- for towed tyres: rolling resistance
- for driven tyres: pull, tractive efficiency, and slip.

57
 0.8

u 0.7

.ï ö
.y = 0.6H
«*- *•-
«ê- Qt
(V
o «
° ~ 0.5-
= o O.i.
=> P
o. iz° -
3.
o 0.3
o <u
t_
u
c
o "Si n?
^o o
o
o 01
r 0.1
V) o
in
o
L-
Ol
0
Dl 10 20 30 i.0 % 50
C
'S slip S

Flg. 4.2. Tract Ion performance of a tractor tyre.

4.1.2.2. TOWED TYRES

The most important aspect of a towed tyre is the force needed to

overcome motion resistance. This motion resistance can be equal-
ized to the towed force. In this dissertation the terms "rolling
resistance" and "motion resistance" are synonymous.
The rolling resistance depends on many parameters; the main ones
are: tyre construction, tyre dimensions. Inflation pressure,
wheel load, surface conditions, and travelling speed.

Tyre construction has a significant Influence on the rolling

resistance. It Is known of automobile tyres that radial tyres
have lower rolling resistance than cross ply tyres (Clark,
1971a). Radial tractor rear tyres also have lower rolling resis-
tance (Stelner and Söhne, 1979).
A larger number of carcass piles and thicker treads tends to
increase the rolling resistance because of higher hysteresis
losses. The type of rubber compound also affects rolling resis-
tance.

7"yre width and diameter influence rolling resistance as well.

Increasing the tyre width, at a constant tyre load and Inflation
pressure, results In a decrease of rolling resistance. On a rigid
road the Influence of the tyre diameter on rolling resistance Is
negligible. In field conditions the influence of the tyre diam-
eter can be considerable (Inns and Kllgour, 1978).

58
Inflation pressure affects tyre flexibility. Based on measure-
ments of Perdok (1978) fig. 4.3 was plotted. This figure shows
that rolling resistance on a concrete surface decreases with
Increasing inflation pressure. On deformable soil high inflation
pressure results In Increased ground penetration work and there-
fore higher rolling resistance. Conversely, lower Inflation
pressure, while decreasing ground penetration, Increases the
deflection of the tyre and hence hysteresis losses. So there must
be an optimum inflation pressure for a particular surface condi-
tion. Because of this Interaction It is possible that lowly
Inflated and excessively overloaded tyres have higher rolling
resistance on the road than in the field.

Wheel load affects tyre deflection and surface penetration. In-

creasing the wheel load results In an Increase In rolling resis-
tance (Fig.4.3).

Surface conditions have influence on rolling resistance. On a

hard, smooth, and dry surface rolling resistance Is considerably
lower than on a freshly ploughed field.

3800 1.0 bar

N
3400
ft:
01
o 3000
c
o
> field
VI 2600
01

c 2200

1800-

K00-

1000

600 concrete

200-
o-l—y^
12.170 15.990 19.820N
wheel load W

Fig. 4.3. Rolling reslstance of a 13-16 agricultural Implement

tyre on the road and In the field at different Inflation press-
ures and wheel loads.

59
Travel IIng speed can also affect rolling resistance.
On a rigid road rolling resistance Increases with Increasing
speed (Clark, 1971a). Stelner and Söhne (1979) found a nearly
constant rolling resistance for a tractor rear tyre on the road
at a speed ranging from 1to 3m/s.
Measurements In the field by Stelnkampf (1975) show Increasing
rolling resistance at increasing speed.

4.1.2.3. DRIVEN TYRES

Important aspects of the performance of driven tyres are pull

coefficient and tractive efficiency. These aspects depend on
parameters such as: slip, surface conditions, tyre load, Infla-
tion pressure, load ratio, tyre tread pattern, lug parameters,
tyre construction, tyre dimensions, driving speed, and direction
of rotation.
Pull coefficients and tractive efficiencies of driven tyres have
been published by Taylor and Williams (1959), Forrest et al.
(1962), Taylor (1973,1976), Dwyer et al. (1974), Stelnkampf
(1975,198la+b), Stelner (1978,1979), BII1er and Stelnkampf
(1978), and many others.

SI lp
Slip has great Influence on tractive performance. The pull coef-
ficient generally Increases with Increasing slip. Tractive effi-
ciency first Increases with slip until a maximum Is reached and
then decreases with further Increasing slip. The slip value at
which maximum efficiency occurs depends strongly on surface con-
d11ions.

Surface condltIon
The soil Is the most important factor affecting the pull coeffi-
cient (Domler and WIIIams, 1978).
Field conditions are classified in the Handbook of Agricultural
Tyre Performance (Dwyer et al., 1974a) as follows: good, average,
poor, and bad. Under these conditions the Cone Index has a value
of 2000 kN/m 2 , 400 kN/m 2 ,250 kN/m 2 , and 150 kN/m 2 respectively.
Based on measurements of Dwyer et al. (1974), Steiner
(1978,1979), and Stelnkampf (1975,1981 a+b) we can give the
following values for slip and pull coefficient at maximum trac-
tive efficiency under different field conditions (Table4.1).

Table 4.1. Slip and pull coefficlent at maximum tractive effi-

ciency and at 20 % slip.
field max. effl- slip at %aty)m&x %&*•20%slip
condition clency(7?max)7? m a x

good 0.82 0.08 0.28 0.51

average 0.68 0.10 0.28 0.41
bad 0.53 0.16 0.21 0.23

60
Tractive efficiency is highest on rigid surfaces and decreases
with increasing deformabIIIty of the soil surface. The pull
coefficient also decreases with worsening surface conditions.

Tyre load
Under good and average conditions Increasing the wheel load
generally results In better tractive performance. Under bad con-
d11ions there Is no significant Influence of wheel load on trac-
tlve performance.

Tyre InfI at Ion pressure

Under good and average conditions decreasing the Inflation press-
ure creates a larger contact area. This results In better tyre
performance. At sInkage due to high slip and under bad conditions
the Influence of decreasing tyre Inflation is less because the
tyre becomes more rigid in relation to the soil.

Load ratio
The load ratio is the ratio of the wheel load to the maximum
allowed wheel load. The load ratio can be increased by decreasing
the tyre Inflation pressure or by Increasing the tyre load. Under
good and average conditions Increasing the load ratio results In
better tyre performance. This effect Ismore Influenced by tyre
Inflation than by wheel load (Fig. 4.4). Under worse conditions
the Influence of the load ratio Is less clear.

56

%
0.75barA
53 >>. load ratio
0.88b a r / y=i.3
12
1.05 b a r / \
/ 11
a« 50- W=UkN 16kN
\
Z__—^-
/1.0 ^rr?^^
IS.OKIN/
\ 0.9/ ^-pj=1.2bar=cons
o
o 0.8 /U5bar
i.7-
E /-W=18kN=constant
•D
XI >.=1.75bar
E
U
6 7 8 kN 10
medium pull P

Flg. A.A. Median tractive performance of a 16.9-30 tractor rear

tyre at several tyre load ratios. Load ratio (If) was changed by
changing wheel load W (p/ = constant) or by changing Inf lat Ion
pressure p/ (W« constant) (Stelner, 1978).

61
Tread pattern
The main agrlcul tural tread patterns are Regu Iar AgrleuIturaI
(Rl), High-Lugged ( R 2 ) , Sportsfleld (R3),and Industrial Lugged
(Ft4)•
Measurements by Te rpstra (1973), Taylor (1976), and Stelnkampf
(1981) lead to the following conclusions. Under good and average
condi11ons the R 1 generally performs best, es pecIaIIy at hIgh
slip values. The R2 Is superior on wet, cohesI ve solIs. Under
good conditions R2 performs lesswell than R-|.T he performance of
the R3 and R4 typ es can nearly equal the one of '1 when oper-
atIng under good conditions at low slip values. When hIgh puIIs
are needed R3 and R 4 are less useful. R 2 and R] have the highest
tyre wear. R3 Is intended for use on gra ss fleIds, whIleR 4 Is
used when almost a Itraffic takes place on the road.

Lug parameters
Main lug parameters are lug height, lug angle, and lug spacing.
Tractor rear tyres generally have a lug angle of about 45 de-
grees. One manufacturer supplies tyres with a lug angle of 67
degrees. This difference In lug angle Is not likely to have a
significant Influence on tyre tractive performance (Taylor,
1973).

(a) tyre 16.9-30 (b) tyre 16.9-30

wheel load W=15887N wheel load W= 15887N
inflation pressure:p- =1.1bar inflation p r e s s u r e ^ =1.4 bar
0.8 0.8-1

.1 0.7 .Ï 0.7
u o

8 0.6 o
o 0.6H

o. 0.5 i . 0.5

0.4 0.4-

c«
0.3 >, 0.3
u
c
<u
ö 0.2 '.Ü 0.2-1
01 a»
>o °-1 > aH
u
• o
~ 0 - 0
10 20 30 40 50 % 60 0 10 20 30 40%50
slip S slip S
Flg. 4.5. Tractive performances of radial, cross ply, and high
lugged tractor rear tyres under good (a) and poor (b) tract Ion
condltIons (Stelnkampf, 1974).

62
Under good conditions the number of lugs (lug spacing) has no
significant Influence on ty re performance either (Taylor, 1974).
However, under bad condItI ons the self-cleaning ability of tyres
with small lug spaces dec reases, resulting In worse tractive
performance.
According to Gee-Clough e t al. (1977a)maximum tractive effl-
clency Is independent of lu g height. Furthermore, they found that
the pull coefficient deer eased when the lug height increased
beyond 20 mm. Gill and va n den Berg (1967) state that high lugs
decrease efficiency, but that they may be required under poor
traction conditions in orde r to increase maximum pull,
Terpstra (1973) and Stelnk ampf (1974,1975,1981) found that high-
lugged tyres (R2) result in better tractive performance than
regular agricultural tyres, when operating on wet, cohesive soils
(Fig4.5).

Radial and cross ply construct Ion

Radial tyres generally are characterized by better tractive per-
formance than cross plies when the conditions are relatively
good. Under bad conditions the differences In tractive efficiency
are small, but the radial tyre still has a higher pull coeffi-
cient. Steinkampf (1981a+b) found great differences between tyres

tyre 18.4-34
wheel load W= 14600N
0.8
inflation pressure :p j =1.1 bar

S 0.6-I
u
o. 0.5-

0.4

o
c 1.62 km/h
<v
;Ö 0.2- 3.99 "
"oi 6.05 "
I 0.
o
o
£ 0
0 10 20 30 40 % 50
slip S
Flg. 4.6. Tractive performance of a 18.4-34 tractor rear tyre at
different driving speeds.

63
in the groups of radiais and cross piles as well. The tractive
performance of a good radial tyre can be 25 %better than the one
of a bad cross ply; on the other hand, the tractive performance
of a good cross ply can be as much as 14 %better than the one of
a bad radial one.

Tyre diameter and width

Increasing the diameter and Increasing the width both result Ina
small Improvement of tractive performance. The effect of width Is
more pronounced than the effect of diameter.

Driving speed
Under poor driving conditions Steinkampf (1975) measured the
performance of a 18.4-34 tractor rear tyre at driving speeds of
1.62, 3.99 and 6.05 km/h. An Increase In speed resulted In a
decrease of pull coefficient and tractive efficiency (Fig. 4.6).
These results only show a tendency and they need to be proved for
tyres driven at higher speeds and under better field conditions.

DlrectIon of rotation
Sometimes the direction of the rotation of front tyres of four-
wheel drive tractors Is reversed Inorder to decrease tyre wear.
Reversed rotation direction results In a longer life of the front
tyres when much road travel Isdone (Blller, 1983).
B M Ier and Stelnkampf (1978) found that the direction of rotation
hardly Influenced tractive performance under good driving condi-
tions. Under less Ideal conditions the performance of the re-
versed tyres decreased.
The choice between low front tyre wear (reversed mounting) and
best tractive performance under bad conditions (normal mounting)
will depend on the use of the tractor.

Tyre wear
Tyre wear results In a decrease of lug height and an Increase In
the area of the lug faces. Therefore, on dry roads tyre perform-
ance Increases with Increasing tyre wear. In the field starting
tyre wear has no pronounced Influence on tractive performance.
When driving conditions are poor tyre wear can result In a worse
performance.

4.1.2.4. TRACTOR TRACTIVE PERFORMANCE

Tractive performance depends on the tractor type, the traction

properties of the tyres, the weight, and the surface conditions.

Tractor type
The tractor type Includes the number of driven axles and weight
distribution. There are 4 tractor types available:
- TWD : a two-wheel drive tractor with driven rear tyres
and towed front tyres
- FWDa : a four-wheel drive tractor with driven front
tyres that are smaller In diameter and width com-
pared to the driven rear tyres

64
 - FWDb : a four-wheel drive tractor with driven front
tyres that have equal width and a smaller diameter
than the driven rear tyres
- FWDc : a four-wheel drive tractor with driven front
tyres that are equal to the driven rear tyres

The rear tyres of a FWD perform better than the front tyres.
Dwyer et al. (1977) found the pull for a 13.6-38 tyre to be
approximately 7 %higher when the tractor was passing In the same
track for the second time. This multi-pass behaviour also re-
sulted in about 5 % higher maximum efficiency.
Dwyer and Pearson (1976) compared a TWD and a FWD In the field.
The difference In performance between the two tractors was
greater than was expected from the multi-pass effect. The reason
must be the zero-slip condition used, and the fact that the FWD
weighed more than the TWD. To make more realistic comparisons, we
have made recalculations, using Dwyer and Pearson's results. The
performances of the tractor types compared with respect to maxi-
mum drawbar power per unit tractor weight, show that the FWD with
unequal wheels performs on the average 2.1 % better than the TWD,
while the FWD with equal wheels performs on the average 8.7 %
better than the TWD.
Weight distribution also affects tractor tractive performance. A
TWD generally has a static weight distribution of about 35:65.
This means that 35 % of the tractor weight acts on the front axle
and 65 % on the rear axle. At a dynamic weight distribution of
15 : 85 a TWD performs somewhat better than at a distribution of
35 :65 (Steinkampf, 1977).
Söhne and Bolling (1981) compared three FWD types at different
dynamic weight distributions. They show weight distributions for
optimum tractor tractive performance (Table 4.2).Type 2 performs
better than type 1, and almost equally to type 3.

Table A.Z. Dynamic weight dlstrI but Ion for optimum tractor
performance (Sohne and Boiling, 1981).

Tractor type weight distribution

1: FWD a 32 68
2: FWD b 45 55
3: FWD c 50 50

Single and dual wheels

At a given slip and dynamic load, the pull of a single tyre Is
practically the same as the one of a dual system (Bailey and
Burt, 1981). Because duals have greater load-carrying capacity
than a single tyre they can develop greater pull If their greater
load-carrying capacity Isused. The tractive efficiency of a
single tyre is slightly higher than the one of the dual system.
On sand, pull coefficient and tractive efficiency of duals de-
crease with increasing spacing between duals (Melzer and Knight,
1973).

65
Weight
Many researchers give formulas fortheoptimum ratio ofweight to
available power (Table4.3).
Table 4.3. Weight to power ratio for optimum tractor tractive
performance.

tractor type formula references

TWD W/Pe . 1 17/v Reece (1970)
FWD W/Pe =0 82/v Reece (1970)
FWD W&/P -1 .50/v BrIxlus andZoz (1976)
TWDand FWD W&/P -1 .33/v Hofman (1977)
TWDand FWD W&/P -1 .79/v Dwyer (1978)
TWD W/Pe =1 .2/v Melborg andPerdok (1979)
P « power available fora driven axle(kW)
engine power (kW) v mdriving speed(m/s)
W tractor weight (kN) W . -weight ondriven axle(kN)
We canseefrom these formulasth at atthesame weight to power
ratio aTWDneeds a higher velocl ty than aFWDforoptimum per-
formance. The drawback of allt hese formulas Isthat they are
valid under oneparticular condI tlon only. Neither do those
formulas take Into account theInf luenceof slip and field condI-
tlons ontheweight needed foropt imum performance. The Influence
of slip ontheweight topowerrat lohasbeen shown by Stelnkampf
(1975).
To calculate theoptimum weigh t topower ratio for different
field conditions, velocities, an d tractor types It Isnecessary
to useamore detailed formula:

+ Pr + P, [4.4]
where, effective engine power (kW)
*t = transmission power [P t - (1-7?t)Pe]
3t - transmission efficiency
Pf = front axle power (kW)
Pr - rear axle power (kW)

For aTWDthis formula changes into:

7.wr.v
p + W V + [4.5]
n -7t> e F- f-
where, ƒ> - rolling resistance coefficient
Wf -dynamic front axle load(kN)
Wr - dynamic rear axle load(kN)
v - velocity
% = pull coefficient
n - tractive efficiency

66
For a F W D f o r m u l a 4 . 4 .c h a n g e s into:

1
?1 ?2
where, X-\ = pull c o e f f i c i e n t 1stp a s s
22. = puII c o e f f i c i e n t 2 n dp a s s
7?1 = t r a c t i v e e f f i c i e n c y 1stp a s s
7j2 = tractive efficiency 2nd pass

Calculations of optimum tractor - tillage Implement combina-

tions have been presented by Jahns andStelnkampf (1982), and
Schäfer(1983).
For optimum performance theratio ofweight on driving axleto
engine power generally lies under 900N/kW.Tractors availableIn
the Netherlands generally have a tractor weight toengine power
ratio between 500and700N/kW. About twothirds oftheweightof
a TWDacts ontherear axle. Attheoften used ploughing veloc-
ities of5to7km/hthestandard available weight topower ratio
Isnotoptimal. Therefore, thetransfer aswell astheaddingof
weight are important foroptimum tractive performance during
ploughing operations. The ratio W&/Pe canalso beenlarged by
ballasting weights and filling thetyres with water.
When thetractor isdriven Inthefurrow when theploughing Is
done with amouldboard plough, the largest allowed tyre width Is
13.6, 16.9, or 18.4Inches depending on ploughing width. There-
fore, the load-carrying capacity oftherear tyres determinesup
to which engine power TWDcanbeused andwhen FWD Is needed for
optimum performance. Table 4.4shows themaximum TWDengine power
that canbeused fordifferent tyre sizes andatcertain weight
to power ratios. Themaximum power ata load topower ratio of
W a / P e = 900N/kWisalso theheaviest TWDthat canbe loaded to
900N/kW.

Table 4.4. Maximum allowed engine power (kW) of TWD at different

levels of weight to power ratio for 3 tyre sizes.
InfI at Ion pressure 1.0 bar; load ratio 120 %.
weight topower ratio Wa/Pe N/kW
Tyre Size 600 700 800 900
13.6-38 51 44 38 34
16.9-30 66 56 49 44
16.9-38 74 63 55 49
18.4-30 80 69 60 53
18.4-38 90 77 68 60

67
4.1.2.5. PLOUGHING CAPACITY

One of the main traction aspects of a tractor is the steady pull

needed for ploughing. To know more about ploughing capacity on
farms we measured ploughing capacity during our slip measurements
(see also 3.1.1.3).
Ploughing capacity was calculated with the following formula:

^000.v.b.d
C=

where, C • ploughing capacity (m 3 /kWh)

v - ploughing velocity (km/h)
b - ploughing width (m)
d - ploughing depth (m)
P e - available effective engine power of the tractor (kW)

The results of our measurements have been plotted in Fig. 4.7.

This figure shows that the ploughing capacity was generally
higher on light soils than on heavy soils. It Isamazing that the
lowest and the highest capacity were equal on both soil types.
The ploughing capacities varied from 11.4 to 54.6 m 3 /kWh. The
median value on light soils was 30.0 m 3 /kWh. On heavy soils this
was 24.0 m 3 /kWh. The percentage measurements at a ploughing
capacity below 18.0 m 3 /kWh was 13 and 15 for light and heavy
soils respectively. On the light soils the ploughing capacity of
35 % of the participating farmers was above 36.0 m 3 /kWh. On heavy
soils this was true for only 15 %of the farmers.
The median ploughing depth was 25 cm on both light and heavy
soils. At a ploughing resistance of 25 kN/m 2 for light soils and
75 kN/m 2 for heavy slolIs, the energy needed for ploughing has
been calculated. The median available tractor energy has been
calculated from the median values of ploughing capacity and
ploughing depth. A comparison of the needed and available energy
(Table 4.5) shows that there Is twice as much tractor energy
available than is needed for ploughing.
The main reason for the sometimes low ploughing capacity Is the
Inefficient use of the available tractor engine power. This Is
generally due to several of the following aspects: wheel slip,
tractor load and weight transfer, traction conditions, tyre as-
pects, tractor-plough combination, ploughing width, ploughing
depth, velocity, and adjustment of the plough.
A good tractor engine load at heavy ploughing Includes an engine
speed of 90 %of the rated engine speed and a torque of 90 % of
torque at rated engine speed. This means that the tractor engine
load (A )equals 81 % of the power at rated engine speed. During
ploughing on farms the values were generally lower due to non
optimal settings of tractor, plough, and tractor-plough combina-
tion.
In West Germany a common figure for calculating Is A - 83 %
(Renlus, 1985). Rutherford (1973) found A values between 44 %
and 87 % when ploughing was done on farms.
We may conclude that there Isa great potential for Improve-
ments of ploughing capacity.

68
Table 4.5. Energy during ploughing on farms.

calculated median energy medIan avalIabIe

needed for ploughing (kWh/ha) energy (kWh/ha)

IIght soIIs 17.4 83.3

heavy solIs 52. 1 104.2

100
%

vi
heavy soils ight soils
£ 80
E
ai
t-
D
(A
O
ai
E
60

40-

20

0 10 20 30 i.0 50 60m3/kWh
ploughing capacity

Fig. 4.7. Ploughing capacity on Dutch farms In 1980.

69
4.2. STRESSES IN THE CONTACT AREA

Radial (normal) stresses and tangential (shear) stresses gen-

eralIy occur in the contact area between the solIand a towed or
drIven devIce.
Stresses can be measured by making use of pressure transducers In
the soil or by pressure cells embedded In the wheel periphery.
The disadvantage of using pressure transducers in the soil Is
that they can move during the compacting process and, therefore,
can provide erroneous information. All researchers referred to In
section 4.2 have used pressure cells embedded In the wheel peri-
phery.

4.2.1 STRESSES IN THE CONTACT AREA BETWEEN SOIL AND RIGID WHEEL
OR ROLLER

Hegedus (1965) measured normal stress distribution in the soII-

wheel Interface and the influence of wheel load and slip on this
distribution (Fig. 4.8). The distribution of both normal and
shear stresses In the soil-wheel Interface has been measured by
Onafeko and Reece (1967) and Krlck (1969). Wong and Reece
(1967a,1967b) and Wong (1967,1978) relate stress distribution to
flow zones beneath rollers and wheels.

(b)

S=100%
IV=227N
W=45£N
W=680N
W=907N

Fig. 4.8. Normal pressure dlstrI but Ions under a wheel on dry
sand.
(a) non-slip with load as parameter
(b) constant load with slip as parameter.

70
4.2.1.1. RADIAL STRESS DISTRIBUTION

Maximum radial stress does not occur at the point beneath the
axle as suggested by the plate slnkage analogy (Bekker, 1956). It
actually occurs In front of the wheel axle and shifts forward at
Increasing slip. Maximum radial stress occurs at the point where
the two soil failure zones beneath the wheel Join each other.
Increasing the wheel load results In an Increase In maximum
radial stress. At the edge of the wheel rim there Isa concentra-
tion of stress which decreases at Increasing slip (Table4.6).

Table 4.6. Ratio of maximum radial stress OR at the edge of the

wheel rim to maximum radial stress Vy In the middle of
the wheel rim (Krlck, 1969).

towed silp=12 % Silp=25 % Sllp=40 %

V^M 1.75 1.6 1.5 1.4

4.2.1.2. TANGENTIAL STRESS DISTRIBUTION

At the transition point (point A In Fig 3 .19a) of a towed wheel

shear stress changes Its direction; having occurred In the dIréc-
tlon opposite to wheel rotation (positive) it now occurs In the
same direction Inwhich the wheel rotates (negatIve).
In zone a (Fig. 3.19a+b) shear stress occu rs In forward direction
(positive). Beneath a towed wheel (Fig. 3.19a) the soil moves
forward slowly between A and Ewhile the wheel moves forward
relatively fast. Therefore, shear stress occurs In the direction
of wheel rotation (negative). Shear st ress In the two zones
occurs In opposite directions. These fi ndlngs agree with the
requirement that the sum of the moments eq uaIs zero.
Beneath a driven wheel both soil and whee rlm move backward In
zone b (Fig. 3.19c). The wheel moves fas ter than the soil, and
therefore, shear stress In zone b occurs in forward direction
(pos11ive).

4.2.2. STRESSES IN THE TYRE CONTACT AREA

The essential difference between a rigid wheel and a pneumatic

tyre Is the flexibility of the tyre. When the tyre moves on soft
soil there generaIlywiIIbe an equilibrium between the tyre-
deflecting forces and the rut-forming forces.

71
4.2.2.1. AVERAGE CONTACT PRESSURE

Average contact pressure Isdefined as:p m = W/A [4.7]

wh e r e ,Pm= average contact pressure

W = vertical wheel load
A = area of the tyre-road Interface
In terramechanics this average contact pressure Isoften given
as sum of tyre Inflation pressure p\ and a pressure p c for car-
cass stIffness:

p m - p, + p c [4.8]

Average contact pressure on a rigid surface

This pressure depends on Inflation pressure, wheel load, and tyre
parameters.
An Increase In Inflation pressure (at a constant wheel load)
results In a decrease of carcass stiffness pressure p c . An In-
crease in wheel load (at a constant P|) results In a small in-
crease in p c (Söhne, 1952).
The dynamic contact area of a tyre Is larger than the static
contact area (Van den Berg and Gill, 1962).
Tractor tyres with a closed centre tread have a higher average
contact pressure than tyres with an open centre tread (Söhne,
1952).
Kriek (1969) tested smooth tractor rear tyres, at nominal wheel
load, over an Inflation pressure range from 0.6 to 2.0 bar. He
found a constant carcass stiffness pressure p c of about 0.45 bar.
Steiner and Söhne (1979) tested several lugged tractor rear tyres
of different sizes and construction. They found the following
empirical equations for the average contact pressure:

Cross ply tractor rear tyres:

p m = 1.128 + 0 . 6 6 5 p | + 0.0091? - 0.004D [4.9]

Radial tractor rear tyres:

p m = 2 . 6 7 7 + 0 . 5 7 5 p | + 0.0111V - 0.016D [4.10]

where, p m = average contact pressure (bar)

P| = tyre inflation pressure (bar)
W =wheel load (kN)
D = tyre diameter (cm).

Average contact pressure on deformable soil

Contact pressure Is generally smaller on deformable soli than on
a rigid surface because the contact area Is enlarged when ruts
are formed.
When a tyre passes In a track for the second time wheel slnkage
Is generally lower than when It does so In the first pass. When
the tyre passes for the second time the contact area Is smaller
and the average contact pressure is therefore higher than when
the tyre passes for the first time.
Average contact pressure on deformable soil can be remarkably
lower than Inflation pressure (Baganz and Kunath, 1963; Söhne
1952 and 1963). Formula 4.8. is therefore not very useful for

72
field traffle.

4.2.2.2. RADIAL STRESS DISTRIBUTION

Smooth tyre on a rigid surface

Normal stress distributions In longitudinal and lateral direc-
tions have been given by Van den Berg and Gill (1962), Clarck
(1971b), and Van EIdIk Thleme and Pacejka (1971). Shear peaks
occur at the edges (Fig. 4.9).Stress Increases fast on the front
side of the tyre, but decreases also fast at the end of the
contact area.

Smooth tyre on deformable soil

On soft soil normal stress distribution Ismore uniform In both
longitudinal and lateral directions (Fig. 4.9). At high slip
values there are an initial and a terminal peak In the longitudi-
nal stress distribution measured In the centre line of the tyre
(Fig. 4.10)

Lugged tyre on deformable soil

Trabblc et al. (1959) and Liang and Yung (1966) measured normal
pressure distribution across the lug face, on the undertread, and
on the leading and trailing lug face. An increase In inflation
pressure resulted in an increase In normal pressures In the
middle of the tyre and a decrease In normal pressures at the
edges of the tyre circumference. Normal pressure on the leading
lug face and the undertread Increased at higher pull, while
normal pressure on the lug face and the trailing side decreased.

4.2.2.3. SHEAR STRESS DISTRIBUTION

Measurements of shear stress In the tyre contact area have been

made by Liang and Yung (1966), Krlck (1969), Van Eldik Thleme and
Pacejka (1971), and Clarck (1971b).

Smooth tyre on a rigid surface

Shear stresses generally show peaks on the edges of the contact
area. An Increase in Inflation pressure Increases the maximum
shear stress (Clark, 1971b).
Shear stress becomes positive by means of the action of a driving
torque. Maximum shear stress moves backward with Increasing driv-
ing torque. The stress peak becomes negative as a result of the
action of a breaking torque.

Smooth tyre on deformable soil

In longitudinal as well as lateral direction stress distribution
for tyres Ismore uniform than for rigid wheels (KIrck,1969).
There Is still a stress Increase towards the edge of the tyre at
small slip values, but at high slip this disappears. Stress peaks
have been observed In the middle of the tyre at high wheel slip
(Fig. 4.10).

73
Lugged tyre on deformable soils
Shear stress on the lug face, the undertread, the leading lug
side, and the trailing iug side have been measured by Liang and
Yung (1968) for a 1 1 - 2 8 tractor rear tyre used on sand.
Stresses were measured at two pulls. An increase In pull resulted
In a decrease in shear stress on the lug face and the lug sides,
while shear stress on the undertread Increased.

(a) (b)
stress directly under axle stress under centre
bar bar
firm sand
2.1- A firm
U A

Jr
1.4-

0.7 0.7
•--„''- N.."
0
soft clay
>°"
10 5 0 5 10 cm 30 15 0 15 cm
lateral distance from centre front rear
length

Fig. 4.9.a.b. Lateral and longltudal stress dlstrIbutIon under a

11-38 towed smooth tyre at an InfI at Ion pressure of 1 bar
(VandenBerg and Gill, 1962).

tangential stress

radial stress

Fig. 4.10. Radial and tangent Ial stress dlstrIbutIon under

smooth 11.5-15 tyre at 40 % slip (Krlck, 1969).

74
4.3. STRESSES INTHESOIL
Soll compaction during agricultural field traffic Isa result of
stresses under tyres, wheels, and rollers.
To gain a better understanding of this problem, especially In
order topredict soil compaction duetoagricultural field traf-
fic, It Is Important toknow more about:
- stress distribution under thetyres
- theparameters that affect this stress distribution
- the resulting compaction ofthesoil andthe parameters
that affect this.

4.3.1. STRESS DISTRIBUTION INTHESOIL UNDER TYRES, WHEELS, AND

ROLLERS
Many theories about stress dlstlbutlon under tyres arebased on
the equations of Bousslnesq that describe stress distribution
within Isotropic andelastic materials such assteel. Theprinci-
pal stress (<Tr) Ina volume element, Induced bya point load (p)
acting perpendicularly ontheflat boundary ofa semI-InfInIte
mass, canbecalculated with:
3p
<r r = cos 6 [4.11]
2TTr 2

where, r =thedistance between thevolume element andthepoint

load
8 -theangle between thevertical line andthedirection
of thepoint load.
Civil engineers found that stress distribution Insoil deviated
from theone Ina homogeneous Isotropic mass. Themore plastic
the soil Is,the larger these deviations willbe.
Fröhlich (1934) Introduced theconcentration factor T) In for-
mula 4.11 toaccount forthe Influence of soil plasticity on
stress distribution:

IP N
Crr » (COS 9 ) V " 2 [4.12]
2Kr2
For V = 3thestress distribution Is Identical to Bousslnesq's
formula foranelastic Isotropic mass. A concentration factorfor
•v= 4to ~i= 6 Is common for soils.
During agricultural field traffic load transfer does notoccurat
one point, but ina finite contact area. Assuming that a tyre-
soil system equals a uniformly loaded circle, Fröhlich (1934)
calculated stresses crz:

°z -%i< 1 -cos'3« ) [4.13]

75
where,oc - half top angle of a cone which has Its top in the
point In the soil In question and the area of the
circle as Its base.
pm =mean pressure In the contact area.

Söhne (1953a) gives a numer c summation procedure. He divides the

elliptical tyre-soil contac t area Into 25 elements of equal size,
The load on each element isconsidered to be a vertical point
load. The stress In a poi nt A In the soil can be estimated as
foIlows:
for aII the elements the stress in point A is calculated
with the point-load met hod. By summoning all these contrl-
butions to the stress InA, an estimation of the principal
stress In A Is found.
In this way Söhne calcul ated many pressure distributions of
agricultural tyres under di fferent soil conditions. The results
of these calculations can be shown In pressure bulbs (curves of
equal principal stress), Fig. 4.11 shows calculated pressure
bulbs under a tractor tyre for different soil conditions,
Actually pressures measur ed by Reaves and Cooper (1960), and
Grecenko (1967) are very si mllar to the computed distributions of
Söhne.

t y r e : 13.6R28
inflation pressure
press : p , = 0.9 bar
wheel load:W =10kN

firm soil ( >> = 4 ) normal soil (»1 =5) soft soil (V =6 )

Fig. 4.11. CalcuIated curves of equal pressure aj (pressure bulbs)
under a tractor rear tyre for different soil condltIons.

Bekker (1956) assumes that the tyre-soil system equals a strip

load acting on a semi-infInIte elastic medium. The width of the
strip load Is the tyre width. The stress V on the strip load Is
the mean normal stress In the tyre-soil interface.

76
The major principal stress follows:
V
0\ = (d+ sin««')
TT

where, / » the angle (In radians) between the lines that connect
the point considered with the sides of the strip load
In a vertical plane.
From this equation Bekker assimilated the following rule of
thumb: at a depth equal to the width of the loaded strip, the
stress under the center of the strip load Is about half of the
stress on the strip load. Compared with the method of Söhne, the
approach of Bekker results In values which are too high.

4.3.2. PARAMETERS THAT AFFECT THE STRESS DISTRIBUTION UNDER

AGRICULTURAL DRIVING EQUIPMENT
Not only the soil but also the technical concept of the machinery
affects stress distribution.
The concept of agricultural traction and transport devices with
pneumatic tyres has changed drastically during the last 30 years.
The mean engine power of the tractors Increased by more than 300
%. The ratio tractor weight to engine power decreased with a
factor of about 1.5. Therefore the driving speed of tractors at
work Increased. The mean weight of tractors more than doubled.
By Increasing tractor power the tractor type changed from two-
wheel drive (TWD) to four-wheel drive (FWD). By Increasing trac-
tor power the load-carrying capacity of the transport devices
Increased.
Numerical data of these developments have been given by Danfors
(1980), Söhne (1980), Söhne and Bolling (1981), Perdok and Terp-
stra (1983), and Renlus (1986).

4.3.2.1. SOIL PARAMETERS

Soil properties Influence stress distribution In the soil under-
neath the tyres. In the theories based on the Bousslnesq equa-
tions, soil properties are characterized by the concentration
factor V .For agricultural soils i) has values between 4 and 6 .
This Implies that soils are supposed to be Isotropic.
In agricultural soils two or more layers can be distinguished:
top soil, ploughing layer, plough sole, and subsoil. Most likely
this layered structure of soil will influence the actual stress
dIstrIbutIon.

77
4.3.2.2. TRAILER TYRE CONCEPT

Tyre parameters like inflation pressure ( P | ) , wheel load (W),

tyre dimensions, and single/dual mounting Influence pressure
dIstr ibution.
Fig. 4.12 shows pressure bulbs under three contact areas accord-
Ing to Boiling and Söhne (1982). From this figure we can see
that:
- the tyre with the lowest load has the smallest volume
inside the pressure bulbs.
- Increasing the wheel load, at constant contact surface
pressure, results In an Increase In volume inside the
pressure bulbs.
- equal wheel loads result In almost equal reach (In depth)
of the pressure bulbs.

Figure 4.13 shows three tyre concepts at a constant wheel loadW.

A greater tyre width or double mounting results In less sInkage
of the wheel and therefore less compaction near the surface. The
pressure-bulbs of the three concepts reach nearly at equal depth
Into the subsoII.
Modern tyre agricultural transport devices have, despite their
modern concept, pressure bulbs that reach deeply into the soil.
The reason is that much heavier loads are applied.

contact area 1000 cm

inflation pressure 1 bar
wheel load 10kN

Fig. 4.12. Pressure bulbs under three contact areas

78
4.3.2.3. TRACTOR TYPE

Increasing tractor power generally Implies that tractor weight

Increases. To support this heavier weight, the tractor needs
tyres with larger dimensions.
A limiting factor In Increasing tractor tyre width Is the width
of the open furrow during ln-furrow ploughing. Therefore at
greater tractor power the front axle load Increases. To prevent
the front wheel rolling resistance from getting too high and to
Improve the weight-pull ratio, driven front wheels may be used.
For these four-wheel drive tractors (FWD-tractors) different
types are available. The pull-slip behaviour of these types has
been discussed In section 4.1.2.4.
Boiling and Söhne (1982) present calculated pressure bulbs under
one TWD and two FWD tractors (Fig. 4.14). We can see that, des-
pite larger tyre dimensions, the pressure bulbs reach more deeply
when tractor power Is Increased. This Ismainly due to the In-
crease of tractor weight at Increasing tractor power. The back-
ward bend In the pressure bulbs Iscaused by the pulling force.
To calculate these distributions advanced computing programmes
and extensive calculation facilities are needed.

170-20AW 10-18AW 170-20AW 10-18AW

Q.
XI
'5

Fig. 4.13. Caleu Iated pressure bulbs under tyres with 7.0 kN axle
loads. From left to right: narrow tyres, wide tyres, and dual
trailer tyres at Inf lat Ion pressures of 3.0, 1.5 and 0.74 bar
respectIvely (Söhne, 1953a).

79
 W=3600 N S=-2% W=10810 N S=2 3 %
7.5-16 U.9-30

"Vf -
.">îo\ plough sole
.'Xv.v

W=9320N S=17% W=17170N S = 1 7 %

U.9-24 16.9-38

(b)
'•^v. plough sole

IV=12850 N S=19% W=27880 N S=19%

16.9-26 18.4-38

(c)
^ plough sole

Fig. 4.14.a-c. Pressure bulbs under three tractors,

(a) 30 kW Two-wheel drive tractor
(b) 60 kw Four-wheel drive tractor
(c) 90 kW Four-wheel drive tractor.

80
 CHAPTER 5
SOILCHARACTERISTICSCONCERNINGLOAD-BEARING PROCESSES

5.1.RELATIONSHIPSBETWEEN "TREATMENT"AND "BEHAVIOUR"OFSOIL

The compaction behaviour of asoli dependsonprior processesto
which thesoil hasbeen subjected. Iftheseprior processes are
known they can beused asabasis for prediction of subsequent
behaviour (Kooien, 1977).
The treatment of thesoilmay comprise aseriesof tillage oper-
ations,changes Intemperature,changes Inmoisture content,etc.
In a uniaxial compression test thebehaviour of Lexkesveer
sandy loam was tested for repeated compressionswhile the soli
was subjected todifferent treatments between the compressions.
Cylinders (81 mm indiameter andwith aheight of 50 mm) were
filled with about 300grammesof soil. Thesamples were com-
pressed with ahydraulicpressataspeed of 2mm/s up to press-
uresof 1.5 and 4.0 bar.
The testconsisted of 4series. Each serieshad amoisturecon-
tent range between 10and 21 percentwater content byweight. To
get the samestarting position for all cylinders, they were
precompressed at0.5 bar.
Series 1:repeated compressionsweremadewith negligible In-
tervals between thecompressions. Series 1awas loaded
15timesto 1.5 bar;series 1b 15times to4.0bar.
Series 2:the soil sampleswerecompressed to4.0 bar and there
wasa rest-period of 3.5 days. Series 2awas loaded5
times to4.0 barwith a rest-period of 3.5 days be-
tween compressions; series 2bwas loaded 15timesto
4.0 barwith arest-period of 3.5 daysafter every 3
compressIons.
Series 3:the sampleswerecompressed to4.0 bar followed by a
rest-period of 7days.
Series 3awas loaded 5times to4.0 barwith a rest-
period of 7daysafter every compression; series 3b
was loaded 15times to4.0 barwith a rest-period of7
daysafter every 3compressions.
Series 4: like3, butduring the rest-period thesoil sample
was subjected toadrying/wetting cycle. We used the
following procedure during a drying/wetting cycle:
after compression thesamplesweredried inanovenat
105°Cduring 0.5 hour. Thesamplesweredried after-
wards during oneday at room temperature. After one

81
 day the samples were rewetted with crushed Ice to the
original moisture content.

The results of these tests are given in Fig. 5.1. From these
results we can see that In the dry area (under 16% m.c.) soils
do not behave differently as a result of the different treat-
ments. In the moisture content range above 16 % soil behaviour
depends on the treatment.
From the series 2 and 3we can see that a rest-period between
compressions results In a higher compressibility of the soil.
This can be explained by the phenomenon of entrapped air. When a
soil iscompressed air pressure builds up In this entrapped air
before the density of saturation Is reached. This phenomenon of
"entrapped air" has been mentioned by Faure (1980). When the
rest-period between compressions is long enough the air pressure
In the sample will return to normal values. During the next
compression this soli can be compressed further than when a rest-
period between the compressions Is lacking. There is no differ-
ence In soli behaviour when the rest-period lasts 3.5 or 7 days.
Series 4 shows less compaction of the samples (in the range >16 %
m.c.) when they are subjected to a drying/wetting cycle In the
rest-period. Because of this drying/wetting cycle this soil part-
ly regains its resistance to compaction. A clear explanation of
this phenomenon Isdifficult to give. Swelling and shrinking
processes are not likely to be the cause because this soli hardly
has a swelling and shrinking capacity.

In the field soli Isexposed to natural forces like freezing

and thawing, and shrinking and swelling as a consequence of
drying and wetting respectively.
The results of studies on the persistence of compaction after
freezing and thawing are contradictory. Krumbach and White (1964)
reported decreased bulk densities on a freezing surface soli of
CelIna loam inMichigan. Wittsell and Hobbs (1965) found a de-
crease of bulk density after 1year of freezing and thawing In
the 2.5 to 10cm and the 18 to 25 cm layer of a compacted silt
loam.
More recent research results tend to disagree with the above
results. In Sweden wheeI-Induced soli compaction persisted under
freezing and thawing (Danfors, 1974). Mech et al.(1967) found
that on a compacted silt loam InWashington crop yield problems
persisted over a four-year period, even though extreme frost
heaving took place. NJds (1976) found on loam and clay-loam soils
an effect of autumn compaction on the yield of cereals In Norway.
Compactions made In spring resulted In significant yield reduc-
tions even after two years. Blake et al. (1976) noted no change
Inbulk density below the plough layer over a nine-year period In
Nicollet clay loam InMinnesota. Neither crop nor the soil water
content in the frozen period affected the compacted subsoil.
From these studies we can conclude that in the top soil, freezing
and thawing can reduce soil compaction because of lateral soil
displacements and a small overburden. Compaction in the subsoil
(below the plough layer) persists even after freezing.

82
 1.80 Saturation line-
10 vol. %air

•1-4*-
1.70 ^r=-t*
g/cm3
£
1.60
vi
c
Ol
•o

3
1.50

1.40-

u 8 10 12 U 16 18 % 20
moisture content
F/g. 5.1. Behaviour of Lexkesveer loam In uniaxial compression
tests during different treatments. Series 1 ( • ) , Series 2 (à),
series 3 ( a ) , and series 4 (o) after 15 loadings to 4.0 bar.

Shrinking and swelling, as a result of changes In the moisture

content, depend mainly on the type and proportion of clay In the
soII. Upon sweIIing there Isan Increase in soiIpor oslty. After
excessive shrinkage, man y cracks are formed in the soiI. These
cracks faciIItäte water In filtration and root pénétra tIon. SoIIs
with a high sand content have a small capacity for shrInking.
Wetting and drying do no t seem to aIlevIate solI compaction.
Porosities hardly change d in the compacted zone of a silt loam
with a naturally compacted layer (Mech et al, 1967). Van Ouwer-
kerk (1968) observed the e ffeet of time and wetting a nd drying on
soil poros-ity In a sandy loam. The surface porosit y graduaIIy
decreased due to natural f orces. Where the subsoiIha d an artIfI-
clally compacted layer, porosity did not Increase with tIme.
VomocIIand Flocher (1965) found on Yolo loam (20 per cent clay of
which 75 percent montmo rilIonIte) that had been artIfIcially
compacted, a persistance of the compaction effect d espIte many
periods of wetting and dry Ing.

83
5.2. CHARACTERIZING PROPERTIES USED IN SOIL SCIENCE

Sol I texture
The compactIbIIIty of a soil depends on Its particle size dis-
tribution (Bodman and Constantin, 1965; van der Watt, 1969;
Chancellor, 1976). Static compression tests show that the highest
compaction occurs In soils with almost equal amounts of the
various particle size classes. An explanation for this has been
given by Bodman and Constantin (1965). High soil compaction can
only be obtained ifcoarse sand Is present in fair amounts.
Porosities and range of pore spaces tend to be higher In clay
soils than in sandy soils. Kooien and Kuipers (1983) give credit
for this to "the stronger tendency of clay to form aggregates".

Mo/sture content
The Influence of moisture content on soli compaction Is shown In
Fig. 5.2a. We can see from these unIaxIaI-compressIon curves that
for dry soils porosity continues to decrease at increasing press-
ure. Within the limits of 10 to 19 percent moisture content there
Is a strong increase incompactIbIIity when moisture content
Increases. When the moisture content exceeds 19 percent the
curves have a different shape because the compressed soli comes
near to the saturation line. Fig. 5.2b shows the relation between
porosity and water content at different pressure levels. The
curve AB of the figure shows the water-fIIled part of the pore
space. The soil can be compacted to this limit only by short
compression actions. If compression to this limit lasts for some
time water will be squeezed out of the solI.
Amir et al. (1976) used these data of Söhne for calculations. In
the range of 10 to 20 percent moisture content they calculated a
highly significant relationship between porosity, pressure, and
soil moisture content:

N - A n - B n Inp -C n Ino

where N = porosity(%)
A n , B n and C n are constants
p = pressure
© =moisture content (%by volume)

Air content
Fig. 5.1 and 5.2 show that the compaction curves do not reach the
saturation line; this Iscaused by entrapped air. During compres-
sion of a soIIsample air flows out. Not all air can escape
freely. Entrapped air can be compressed. After unloading the
compressed air may cause a rebound of the soil.
The Influence of soil air on rebound will be discussed In section
7.1.2.
When it becomes possible for the air to flow out completely the
compaction curve touches the saturation line (Faure, 1980). Dur-
ing compression tests there Isa normal amount of entrapped air
of about 5 percent.

84
 (a) (b)

60 i

% moisture content
by weightI%)
O", (bar)

0 12 10 bar 0 U 8 12 16 20 2U 28 32%
moisture content

Fig. 5.2.a,b. Soil compactibI 11ty as affected by moisture content

(Söhne, 1958).

Lime content
Proctor tests carried out by Kezdi (1969) and Sommer (1972)
showed a decrease of compactIbiIity ifmore limewas added. The
moisture content at which highest compactions occurred also in-
creased. For a clay loam Sommer found that maximum compaction and
the moisture content at which this Isobtained changed by liming
the soil with about 2 percent by volume and 2 percent by weight
respectively. According to Kezdi these phenomena can be explained
by the change in soil structure caused by liming: water is needed
for hydration and themobility of coagulated particles decreases.

Organic matter
Organic matter increases soil strength under wet conditions. At a
low soil moisture content resistance to compaction is higher In
soils containing a normal amount of organic matter than in soils
containing a large amount of organic matter. So, during drying,
compaction resistance increases most in soils low in organic
matter (Kuipers, 1959).

Water suction (pF)

Soli water suction incombination with saturation degree Is an
important soil strength factor. The pF-curve Is sensitive to soil
structure. During compression aggregates flatten at the contact
points and the size of pores between the aggregates decreases.
Fig. 5.3 shows pF-curves of samples from the plough sole and from
the subsoil of Winnigstedt silt loam (Moreno et al, 1974). The
curves at p=0 bar represent undisturbed soil samples. The p=14

85
curves represent the same samples after static compaction at 14
bar and successive equilibration to the different suctions. These
pF-curves show that, below pF2.4, the water content of the undis-
turbed samples from the plough sole Is lower than the water con-
tent of those from the subsoil; above pF2.4 the opposite Is
true.
The curves for p-14 bar are almost equal for the plough sole and
the subsoil. Below pF2.4 the static compaction results In a
decrease of the water content; while an Increase has been found
between pF 2.4 and 3.7. Above pF3.7 static compaction has no
effect on the water content.
Kuipers (1966) also found a decrease of water content (by weight)
at pF2.0 after compaction, but after loosening the compressed
soil, the water content at pF2.0 increased to higher values than
before compaction.

( Iplough sole
( (subsoil

Fig. 5.3. pF-curves (averaged) Fig. 5.4. A loaded soil volume

for soil samples taken from the element
plough sole and from the sub-
sol I of "Wlnnlgstedf silt
loam.

86
5.3. MECHANICAL PROPERTIES
The soil mechanical properties comprise the geometrical changes
(compaction, deformation, and failure) that a loaded soil volume
element (Fig. 5.4) will show for all possible values of CT-\ , CT2
and0"3.
The most Important mechanical properties of load-bearing process-
es are compactibiIIty, deformabMIty, and resistance against
shear faIIure.

5.3.1. COMPACTIBILITY
The loading of soil during soil tillage and agricultural field
traffic Isa relatively quick type of transient loading. There-
fore, we can safely assume that the developed pore-water press-
ures will not have sufficient time to dissipate and that the role
of water In the solIcan be Iimited to its effect on solI
strength.

5.3.1.1. MEASURES FOR COMPACTION

Compaction can be expressed In volumetric properties of solid,
liquid, and gas phases:
Porosity of a soiIvolume element isthe ratio of volume of pores
(water + air) to the total volume of the element. It is expressed
as a percentage or as a fraction. Porosity has the advantage of
being dimensIonIess and independent of particle density. Other
names for porosity are pore space and total pore space.
Void ratio (e) is the total volume of pores divided by the total
volume of solids. According to Karaflath and Nowatzkl (1978) void
ratio iseasier to handle than porosity when the volume of the
solids remains essentially constant for a given Initial total
volume of soil solids whereas with porosity both the volume of
voids and the total volume change.
Dry bulk density (y d ) is the weight of solids in a unit of
volume. It is usually expressed in g/cm^ or kg/m 3 . For the inter-
pretation of dry bulk density we need to know the specific den-
sity of the soIIds.
Bulk weight volume (BWV) is 1/tfd (cm 3 /g).
The relationships between these quantities are given in Fig. 5.5.
Besides changes In volumetric quantities compaction is often
associated with Increased penetration resistance and changes in
pore structure properties such as reduction of conductivity, dif-
fusiveness and permeability by water and air.

87
 BWV I Xà I

2.00

25 30 35 40 45 50 55 60 65 70 % 75
pore space P *•
Fig. 5.5. InterreI atlonshIps between pore space (P), void ratio
(e), bulk weight volume (BWV) and dry bulk density (fa).

5.3.1.2. MEASURING COMPACTIBILITY

The process of soil compaction Is generally expressed In terms of

stress-strain relationships:

state of strain f (state of stress)

The stresses 0^, Ojj, O3, and their directions change during load-
ing:

state of strain - f ( G\(t) • 02(t) •0"3(t)>•

After some simplifications (Kooien and Kuipers,1983) this formula
may be written as:

(Bwv) t -f (cr 1(t) ,çr 3(t ) )•

This function can be measured In a trlaxlal test. In a triaxial
apparatus a cylindrical soli sample is enclosed by a cylindrical
rubber membrane and rigid top and bottom plates (Fig. 5.6). The
enclosed sample Is placed Ina rigid transparant cylinder and Is
surrounded by water. The pressure (O3) In this water can be
measured and regulated. Through a loading ram on the top-plate0~i
can be changed and measured. By varying 0"iand0*3every desired
combination of stresses can be examined. Trlaxlal tests show that
at any moment BWV Isonly determined by (Ji and<T3. So the formula

88
can be written as:

BWV = f (0"i.0*3).

( -!
By definition <Tm - 1/3 CT +0*2+03) andT m a x - 1/2(0i -°3>
Substitution results In an other function:

BWV F «rmm>
,T.
"-max ).
For a particular soil at a particular moisture content both fand
F can be determined.

In the (0"m, T m a x > B W V ) space the compaction surface can be

represented as BWV • F (0~m, T m a x ) - F can be determined from a
series of triaxlal tests with different paths of (7^ and CT3.
During such measurements the compressive stresses can be separ-
ated from shear stresses. When 0~i Iskept equal to0~3then T m a x -
0; Increase of a\ and decrease ofO 3 can be regulated In such a
way that 0"m stays constant. At a certain moment it appears that
0"idoes not increase any further with further ram movement, and
at the same time sample volume and, therefore, BWV do not change
anymore. This situation iscalled "critical state". In critical
state distortion continues at constant <Tm, T m a x and BWV. Along
the critical state line, "^max^m l s constant. This behaviour Is
described In the "critical state soil mechanics" (CSSM). CSSM has
been developed at Cambridge by Roscoe and colleagues and has been
described by Schofleld and Wroth (1968). There Isevidence that
CSSM can be applied to unsaturated agricultural soils (Kurtay and
Reece (1970); Hettlaratchl and O'Callaghan (1980)).
Triaxlal measurements of critical states can be found In Bailey
and VandenBerg (1968), and Kumar and Weber (1974).

topplate

rubber membrane

— water under
pressure applying 0"^

Fig. 5.6. Triaxlal apparatus.

89
 In the (BWV, cTi,03) s pace the compaction surface can be repre-
sented as BWV - f(0~i,O 3 ).
Compaction caused by ag rlcultural activities will be along the
surface f between the c rltical state line and the line for corn-
paction at a constant diameter of the sample. For a number of
constant values of0~3/0~i fwas calculated from the functions F
published by Bailey and Vandenberg, and Dunlap and Weber. From
these calculations, made by Kooien and Kuipers (1983), and Kooien
and Vaandrager (1984), we can conclude that0*3only has IIttle
Influence on the major principal stress-pore space relationship
(see Fig.5 . 7 ) .
In agricultural operatlo nsO3/O1 generally Is0.5 (Kooien, 1974).
So soil compactlbIIIty can be represented by a relation between
Q\ and degree of compact Ion. For this no trlaxal test is needed;
uniaxial compression Is suffielent.

£ Congaree
• silt loam

,# Lloyd clay

9 Hiwassee
+
sandyloam

0.8 1.2 1.6 2.0 2Mbar

Fig. 5.7. O) -pore space relat lonshlps as affected by O3/CTj

ratlo.

90
Uniaxial compression test
During a uniaxial compression test (see Fig. 5.8) a soil sample
in a rigid cylinder Iscompacted under a downward moving piston.
The stress on the piston and the sample volume are recorded
continuously. Generally the piston moves gradually downward at a
constant speed. The ratio of sample height to sample diameter
should be rather small to keep the friction between the cylinder
wall and the soil negligible. A suitable ratio can be found from
the formula (Kooien, 1974):

<Jj^ (D/h) - 2Ktan S

0~t (D/h) + 2Ktan£

where: o>
CTt =mean normal stress on top of the sample
CTb mean normal stress at the bottom of the sample
D inner diameter of the cylinder,
h actual height of the soli In the cylinder
tanS coefficient of friction between the cylinder wall
and the solI
= <T3/0\ In "ideal" uniaxial compression, wlthout
wall friction. In general K m0.5.

u
1
Fig. 5.8. Uniaxial compress Ion Fig. 5.9. Loads Inkage relation-
test. ship In an uniaxial compress Ion
test with repeated loadings and
un loadings (Söhne, 1952).

91
5.3.1.3. THE INFLUENCE OF REPEATED LOADING, LOADING SPEED, AND
VIBRATIONS ON COMPACTIBILITY

Repeated loading
Agricultural field traffic isa process of loading, and several
unloadlngs and reloadings.
In uniaxial compression tests the path of unloading and re-
loading, in a pressure - sInkage relationship, shows a hysteresis
effect (see Fig. 5.9). Such hysteresis effects were found by
Söhne (1951,1952,1956), and Yong and Warkentin (1966,1975). This
hysteresis effect occurs inmany materials; NadaI (1950) shows
the same effect inmetals.
After reloading the sample behaves In the same way as it would
have done If the last (re)loading was continued without un-
loading.
A uniaxial compression test with several cycles of unloading and
reloading shows that compaction continues. Every reloading re-
sults in extra compaction. This extra compaction decreases with
number of loadings. Söhne (1956) found that further compressibil-
ity was still possible after 10 reloadings with no rest-period
between the unloading and the next reloading (see Fig. 5.10).
In the field there isgenerally a rest-period between successive
reloadings. During this rest-period the soil can regenerate. If
there is entrapped air In the soil during loading, the air press-
ure can equalize in the rest-period. During the next reloading
the soil can be further compressed than when the rest-period is
negligllble (see Section5.1).
If there is a drying-wetting cycle during the rest-period the
compressibility of the soil Is lower (see Section 5.1). During
our measurements we found still additional compaction after 15
re-loadings.

Rebound
A loaded solIsamplewill rebound when the sample isunloaded. A
rebound has two components: elastic recovery and creep. Elastic
recovery occurs Immediately at unloading. Creep Increases with
11me (see Fig. 5.11).
Elastic recovery depends on soil type, soil density, and moisture
content. In soil samples where air was entrapped during loading
we found an Increase In elastic recovery (see Section 7.1.2).
Creep velocity decreases with time. Creep values can well be
higher than elastic recovery.
We found that rebound can decrease bulk density up to 0.1 g/cm 3 .

Loading speed
In uniaxial compression tests using Lexkesveer loam, we found no
difference incompactIbIIIty between loading speeds of 0.25 mm/s
and 3.0 mm/s. Söhne (1953a) found a difference in pore space of
about 0.5 % between loading speeds of 20mm/s and 120mm/s. These
speed effects, measured in uniaxial compression tests, are much
lower than those found in the field under rollers and tyres; see
Baganz (1963/1964), Abaoba (1969), Sltkei and Fekete (1975),
Steinkamp (1975), and Karczewskl (1978).

92
 123 10

5mm

Flg. 5.W. Uniaxial compress Ion with ten repeated loadings,

0.10n

gr/cm 3

0.08

*'
0.06
0)
Ol
c
o
X
o 0.04
U)
c
(V
X)
0.02
XI

nAr-n-, -l 1—|—i

0.1 0.5 5 10 mm. 50 100

time
Fig. 5.11. Rebound behavlour of a peaty soil after uniaxial com-
pression to CTj» 4.0 bar.
a - rebound (elastic recovery)
b - creep (slow elastic recovery).

93
VIbrat Ions
Vlbrations can causecompaction intwoways: bymeansof stress
changesand acceleration effects (Wu, 1971).
Stress changes duetovibrations, cancompact soiIinthe same
way as repeated loadingsdo. Thereare hardlymeasurements known
about thecompactingeffect of stresschangesdueto vibrations
undermachinerywith pneumatictyres.
Compaction byacceleration canoccur when theupward acceleration
of soil particles,caused byvibrations,exceeds theacceleration
of gravity; soil particles have theopportunity to fall freely
and becomevery densely packed. Thiseffectwill onlyoccur in
soilswith negligiblecohesion. Theacceleration effect Is used
to advantage inhighway construction and underwater foundation
constructions likethe "OosterscheIde-dam" IntheNetherlands.
We know thatcrawler-tractorscan highly compact sandy soils.
Further research Isneeded, beforewecanmake statements on
vibrating effectsof agricultural machinery with pneumatic tyres
oncompaction.

5.3.2.DEFORMABILITY
The loading of soli by tyres, wheels, and rollers Isgenerally
accompanied by soil deformation. Thedegreeof deformation de-
pendson the load, moisturecontent, soil density, and loading
speed.
Soli deformation occurs Inavertical planeA through the tyre
center In the direction of travel. Thechanging position of
points at thesoli surface Inthisplanewereobserved In soil
bin investigations. During theseteststhevehicle parameters
(tyre, load, speed, Inflation pressure)were keptconstant.The
soli conditionswere varied inabroad range. Fig. 3.17 shows
horizontal displacement of pointsat thesoil surface Inplane A
versus theaccompanying vertical displacement (sinkage). After
two runswith atowed tyrewecould seethat largesinkages were
accompanied by largehorizontal displacementsat thesoil sur-
face.
For driven tyres thedisplacementsarebackwards becauseof tyre
siIp. At normal pulI there is15-20%siip. This means a
backwardmovement ofabout 0.25 m for thepresent tractor tyres.
Thesemovements decreasewithdepth.
In another vertical planeB, through thetyrecenter and Its
wheel axle,there isalsosoil deformation. Inwet or densesoils
deformation wlII take placeatconstant volumeand solI will
bulge be%Ide thetyre. Under drier or looser conditions the
deformation will beaccompanied bycompactionand therewill be
littleor nobulging beside thetyre (Kuipers, 1970).
Fig. 5.12 shows twocrosssectionsofour soil bin tests. Fig.
5.12a shows planeB foradensesoil after one and two tyre
passes. There Is nobulging. Fig. 5.12b showsbulging In a
softer soiI.
An elementary treatment of deformation hasbeen given by Kooien
and Kuipers (1983).

94
 (a) Lexkesveer
moisture content :17.06%
pore space: 45.9% first pass

original surface second pass

-2.5cm

(b) Schinnen
Vts'K
moisture content :26.53 %
pore space :47.1%
first pass
original surface

Fig. 5.12. Cross sections of ruts formed by a 7-12 Implement

tyre with an axle load of 4874 N. The area above the original
surface Is bulged because of the action of the tyre.

5.3.3. RESISTANCE AGAINST SHEAR

According to the well-known law of Coulomb, shear stress at

failure (Tf) Iscomposed of a cohesion and a friction part:

T f « c + Ctan tf

where, Tf • shear stress at failure

c » cohesion
<T - normal stress on the failure plane
é - angle of Internal friction.

For tyre-soil studies not only shear strength Xf Is Important,

but also the whole shear deformation process.
Shear deformation proceeds as follows: with Increasing shear
stress elastic and plastic deformation occurs first until maximum

95
plastic deformation has been achieved and then shear failure
planes are formed. At this point shear stress has Its maximum
value: shear strength Tf.After this point shear stress decreases
and reaches a residual level due to soll-to-soil friction.
Fig. 5.13a+c show schematic shear deformation curves for (cohe-
sionless) sand and cohesive natural soil. Fig. 5.13b+d show
Coulomb's law for these examples In graphs. The shape of the
stress deformation curves depends on soil type, bulk density,
(a)

<T1

shear deformation normal stress a "

shear deformation normal stress CT

•dense,wet
•loose,dry

O~=constant

shear deformation shear deformation

Fig. 5.13. Shear stress - shear déformât Ion re IatlonshIps for

pure sand (a), for cohesive soil (c), for cohesive soil under
different condltIons (e), and at different shear areas (f).
Coulomb's lines for pure sand (b) and for cohesive soil (b).
(Söhne and Stubenböck, 1978)

96
molsture content, and the area of the shear failure plane. Fig.
5.13e shows schematically shear deformation curves for dry, loose
and dense, cohesive soil with equal moisture content and for dry
and wet cohesive soils of equal density. For dense, cohesive soil
with a moderate moisture content shear deformation curves are
given at two different areas of shear planes (Fig. 5.13f).

5.4. CHARACTERIZATION PROCESSES

During passing of a wheel the state of stress and strain varies

continuously. This makes Itvery difficult to relate the process-
es to specific mechanical properties. Research workers therefore
often use simpler but comparable processes Induced by measuring
devices as a method of characterization. A problem with this
method of characterization Is that there are many different tests
incommon use. In section 5.4.2 results of comparisons of differ-
ent methods are shown and discussed. It Isclear that if two
tests are highly correlated it isenough to use one of them to
characterize soil behaviour.

Cone pénétrât Ion tests

With a cone penetration test resistance to penetration, as a
composite soil property, can be measured and described rapidly
and in sItu.
This test gives no specific soli values such as angle of fric-
tion, cohesion, or coefficient of soil-metal friction. Because
the results are Influenced by the cone type and method used, a
standardization of cone type and method is needed for comparing
measurements.
A widely accepted cone penetrometer Is the one standardized by
ASAE (Fig. 5.14). ASAE Standard S313.1 (Hahn et al., 1984) spec-
ifies two standard 30 degrees cone sizes (20.27 and 12.83 mm base
diameter) with a smooth finish (63 microlnches maximum) and a
penetration velocity of 30.5 mm/s.
Cone resistance Isusually expressed as the quotient of penetra-
tion force and the base area, called cone Index (CI). Cone base
size should always be stated.
CI strongly depends on many factors other than bulk density. The
most Important of the factors are soil moisture content, type of
soil, and pore space. Chancellor (1976) gives examples of rela-
tionships between CI and dry bulk density obtained by several
Investigators. .In these examples the relationships vary from
almost linear to strongly curved. Chessners et al. (1972) show
that even when the same soil and the same moisture content are
used the relationship between CI en bulk density can be different
for solIcompacted In the laboratory and solIcompacted In the
field.
Despite its limitations CI Is frequently used for comparative
Indications of soil conditions because of the ease and rapidity
with which numerous measurements can be made. With the newly
developed electronic recording hand-held penetrometer which can
be linked to a programmable calculator, CI can be measured more
easM y stIII.

97
 15.9

9.5

20.27
V-° r: use cone
sizes ( A ) or (B)
(B) 12.83

Flg. 5.14. ASAE standard cone penetrometer (Hahn et al 1984).

Torsional shear tests

These tests may be grouped according to the type of shearing
device, namely vane, annulus, cylinder, or plate with grousers
(Fig. 5.15).
The vane shear test Isa quick test that can be carried out below
the soil surface. The height to diameter ratio Is normally be-
tween 4:1 and 2:1 (Karaflath and Nowatzki, 1978). An annulus with
grousers on the bottom (Bekker, 1969) can be pressed into the
soil surface, loaded vertically and rotated. A miniature annulus
shear test Instrument Is the "Torvane" made by Solltest Inc. (NN
1971).

(a) (d)

<*>

Fig. 5.15.a-d. Torsional shear devices.

98
A shear annulus of two concentric cylinders with radial grousers
has been shown by Söhne (1953b), and Dwyer et al. (1974b).
In all torsional shear measuring devices for field use there is
uncertainty as to the actual plane of shearing.

Fa IIIng weight
To measure the response of soil to vertical impact we used a
falling weight. The soil measurement Is the depth of a hole
formed by a cylindrical iron weight (10 kg, 99,1 mm diameter),
which was allowed to fall freely onto the soil surface from a
height of 1.0 m. To prevent the weight from tilting as it col-
lided with the soil surface, Itwas dropped through a perspex
pipe having an inside diameter of 99.8 mm, which was placed on
the soil surface. The pipe wall was perforated to prevent a
bul'ld-up of air pressure In front of the falling weight.

5.4.1. COMPARISON OF DIFFERENT CONES

Important aspects of a cone are the base area, the cone surface
roughness, and the tip angle.
Two base areas have been standardized: ASAE(A) with a 3.23 sq.cm
and ASAE(B) with a 1.39 sq.cm base area. The area of the A-type
Is a factor 2.5 higher than the one of the B-type. In soil bins
we measured penetration resistance with both cones, from 5 to 15
cm depth. Each soil bin was prepared In such a way that a very

200

200 250 300 N 350

penetration force ASAE (A)
Fig. 5.16. Relatlonshlp between pénétrât Ion force measured with
ASAE(A) cone and ASAE(B) cone.

99
uniform soil condition was achieved over the whole bin. We exam-
ined five soil types (see Table 5.1) whose water content and
porosity we had varied. The measurements show (Fig. 5.16) that
the vertical force of the ASAE(A) cone was a factor 1.9 higher
than the one of the ASAE(B)-cone. So CI decreases with increasing
cone base area. Schothorst (1974) found the same effect for cones
with a 60° tip angle. To avoid confusion it Is desirable to state
the size of the cone used.
Gill (1968) shows that teflon coated cones have lower CI values
than steel cones. Therefore, it Isadvisable to use standard
cone-roughness and material: AISI416 stainless steel, machined
to a smooth finish (63 mlcroinches maximum).
The use of a 30° cone tip angle Iswidely accepted, OccasIonaM y ,
however, cone measurements with other angles are reported. To
give an indication of the Influence of the cone tip angle on the
results of cone index measurements and to promote the use of a
30° angle In The Netherlands, TIJInk and Vaandrager (1983) meas-
ured CI, averaged over the layer between 5 cm and 15cm depth,
with tip angles of 15°, 30°, 60°,90°, and 180° (all with the same
base area), on 67 different agricultural fields. Each cone index
value was the mean of five measurements. The measurements showed
significant differences In CI values at different tip angles. The
CI values at a 30° tip angle were significantly the lowest. For
each field the mean of all cone index values, irrespective of tip
angle, was calculated and called the overall cone Index. The
results have been presented In a graph giving C//(overaI\CI) as a
function of the tip angle to which CI applies (Fig. 5.17). The
standard deviation of CI/(overaI\CI) at a given tip angle was, on
the average 15% of the mean value at that tip angle, and was
minimal (9.7%) at 30°. This variability Ismainly due to the
heterogeneity of each field: standard deviations within each set
of five single measurements ranged from 7 % to 20 % of the mean
of the five measurements.

o
| 1
>
o

15 30 60 90 180
cone tip angle,degrees

Fig. 5.17. Relatlonshlp between cone tip angle and Cl/overal I CI

(TIJInk and Vaandrager, 1983).

100
Kooien and Vaandrager(1984) replaced the Cl/(overaI IC/) scale by
CI/CI at 30°. With this curve (Fig. 5.18) It Is possible to
"translate" cone measurements at unusual tip angles Into 30° tip
angle measurements.
Despite the possibility of converting cone Index measurements at
an unusual base area tip angle Into standard CI measurements, It
Is better to use a standard cone.
For Interpretation of CI measurements the cone type (the base
area and the tip angle) and the units Inwhich CI Is expressed
should always be mentioned.

1.83
o 1.56.
o 1.26.
1.06 1.00
o !•

0 15 30 60 90 180
cone tip angle,degrees
Fig. 5.18. Graph to transform CI at cone angles other than 30°
Into CI at 30° cone angle (Kooien and Vaandrager, 1984).

5.4.2. COMPARISON OF CONE PENETROMETER, MICRO-PENETROMETER, VANE,

AND TORVANE

To investigate the relationships between characterization

methods, based on different processes, four different tests were
carried out In 15 differently prepared soil bins. in each bin
soil bulk density, water content, and pore space were very homo-
geneous. We used five soil types (see Table 5.1) whose water
content and porosity we had varied as Indicated in Table 5.2.
Cone penetration resistance was measured with a spring-type hand-
operated penetrometer which recorded cone resistance as a func-
tion of depth. The cone used was a ASAE standard cone A with a
tip angle of 30° and a base area of 3.2 cm 2 . The mean cone index
(C/) was calculated as the quotient of the mean resistance over
the 5 to 15 cm depth layer and the base area of the cone. In each
bin ten single cone penetration measurements were made. Each
value mentioned In table 5.2. Is the mean of ten single CI
measurements. Micro-penetrometer tests were carried out with a
"Solltest" pocket penetrometer (N.N. 1971). in each bin tests
were carried out ten times. Torsional shear tests were carried
out with a "PIlcon"dIrect-readIng hand vane tester (Serota and
Jangle,1972) with a 33 mm diameter shear vane and with a "Soll-
test Torvane" micro-annuIus shear test Instrument (N.N. 1971).
Both shear tests were carried out ten times In each soil bin.
Table 5.2. shows the data of the measurements.

101
Table 5.1. Soll descrlptIon.

Moisture
Soll %ofminerals %ofsoll P H KCI content
atpF2
(100cm
<2/u 2-16/* ^6-50JU Humus CaC03 suction),
%

WagenIngen
sIIty clay loam, 36 27 21 2.3 3.3 7.4 27.3
Lexkesveer loam, 15 12 17 1.6 10.4 7.2 20.2
SehInnen
sI11 Ioam, 17 14 57 2.1 2.9 7.2 27.2
Ede sand, 3.5 2.5 4.5 3.6 - 4.4 17.4
Almkerk
siIty clay. 40 24 24 1.2 0.1 6.8 26.9

Table 5.2. Measurements of soll character 1st les.

Bin Soll*MC P CI CIM Si S2 T

NR. % vol.% (kPa) (kPa) (kPa) (kPa) (kPa)

1 L 16.96 43.1 782 1.28 3.4 4 8 22 4

2 L 17.16 40.4 933 1.69 4.3 6 1 25 8
3 L 20.18 39.7 497 0.84 2.7 3 7 26 7
4 A 24.79 53.4 539 0.76 3.3 4 6 22 7
5 A 24.47 48.1 761 1.17 5.4 6 9 40 0
6 A 26.66 46.1 532 0.75 3.9 4 8 30 5
7 W 22.85 52.4 700 1.18 4.0 5 7 25 6
8 W 23.12 46.5 870 1.83 6.6 9 5 44 9
9 W 27.58 46.2 436 0.58 3.1 3 8 25 3
10 S 19.76 47.5 779 1.13 3.4 4 6 18 0
11 S 20.05 42.8 1122 2.01 5.0 6 9 26 3
12 S 26.83 42.7 237 0.23 1.1 1 6 12 4
13 E 16.47 46.3 821 1.32 3.0 4 1 12 6
14 E 16.24 44.1 996 1.64 3.6 5 7 13 2
15 E 19.24 45.9 696 0.93 2.2 3 8 12 3

*: soll type (seeTable5.1)

MC =moisture content byweight
P = porosity
CI =cone Index measured with ASAE typeA
CIM =cone Index measured with MIcro-penetrometer
S-) =shear strength Inthe0-4cm layer
S2 =shear strength inthe3-7cm layer
F =Torvane shear strength.

102
By means of linear regression analysis, the co rrelation coeffi-
d e n t s (r)were determined between the soil cha racterIstles:cone
Index (CI), micro-penetrometer cone Index ( CIM), vane shear
strength in the 0 - 4 cm layer (Si), vane shear strength over the
3 - 7 cm layer (S2), and Torvane strength (7"). See Table 5.3.
High correlation between two tests means thatI t is enough to use
one of the two tests to characterize the solI conditions. At a
low correlation between two tests the second test gives extra
Information about the soli condition. We can conclude that for
laboratory samples It Isenough to measure CI and T, or CIM and
T. In the field the best measurements are CI a nd S' Because of
the enormous variations In the 0 to 1cm top la yer CIM, T, and S 1
are Iess usefuI.

Table 5.3. Correlation coeffle lents (r) showing the linear rela-
tionship between soil character IstIcs .

CI - CIM r » 0.96 s
1 -s2 0.98
Cl - Si r = 0.63 0.83
CI - S2 r - 0.71
s,-T 0.75
CI - T r a 0 . 14 S2 - T
CIM - S-] r - 0.73
CIM - S2 r - 0.81
CIM - T r = 0.30

5.4.3. RELATIONSHIPS BETWEEN CONE-PENETROMETER, SHEAR VANE, AND

FALLING WEIGHT

Inour soil bin Investigations, aimed at predicting the rolling

resistance of a wheel, the depth of the rut, and soil compaction,
we used the following soli characterization tests: cone Index
(CI), vane shear strength (S2) , depth of hole formed by a falling
weight (D), and cone Index measured In the hole (Clf). See Fig.
5.19. To predict the behaviour of the wheel-soil system at the
second wheel pass we also measured CI, D, and S2 In the first
rut.
Between CI, Clf, D, S2 we found the linear correlation coeffi-
cients shown In Table 5.4.

Fig. 5.19. Soil measurements : a) cone res Istance; b) shear by

vane; c) depth of hole formed by a falling weight; d) cone resis-
tance below the hole formed by a falling weight.

103
The combination ofCI, Clf, and Dproved tobe suitable for
predicting rolling resistance, soil compaction, andruth depthin
the first and second passofawheel (see chapter 6.2.2).

Table 5.4. CorrelatIon coeffIclents (r) showing the relatlonshlp

between the different soil character Istles used.

CI - Clf r m 0.96
CI - D r--0.78
CI - S 2 r»0.96
Clf- D r--0.75
Clf- S 2 r=0.95
D - S2 r--0.73

104
 CHAPTER 6

RELATIONSHIPS BETWEEN SOIL CHARACTERISTICS AND PROCESS ASPECTS

AND THEIR SUITABILITY TO PREDICT PROCESS ASPECTS

It is useful to predict certain aspects of the processes Involved

In agricultural field traffic. It Is particularly useful to
predict required pull (rolling resistance), compaction, and rut
depth.
To predict a process, we need data on the characteristics of the
wheel, the behaviour of the wheel, and the soil. Data on tyre
characteristics and the kinematic and dynamic aspects of tyre
behaviour are discussed In sections 2 through 4. Soil charac-
teristics concerning load-bearing processes are discussed In
section 5.
Two general catagorles of prediction methods can be distinguished
(Kooien, 1977):
a) methods that are mainly based on observations of relationships
between characteristics and process aspects and
b) methods that are mainly based on knowledge of the mechanism of
the process under consideration.
The first category Is the most Important and may be divided Into:
comparative methods and methods using empirical formulas or
graphs.
A method from the second category Iscalled exact when this
method uses a hypothetical mechanism almost equal to the mechan-
ism that Is simulated. When this Is not the case, the method Is
called approximate.

6.1. COMPARATIVE METHODS

These methods start with the assumption that a tyre-process

aspect y depends on characteristics x-j ,x n of soil and tyre
which can be expressed by the following formula:

y = f (xi x n ),

Where, f Is the way y depends on Xi xn.

Even when f Is unknown these methods can be used because a once

observed value y at certain values of x-| x n , will be found
again when the same values of x-| x n occur again. This means
that soils which have equal soli mechanical properties will show
equal behaviour in the soII-wheeI process.

105
Rolling resistance as process characteristic depends strongly on
the soil mechanical property known as cone Index (CI)- Table 6.1
shows examples of the same rolling resistance at equal cone index
values. These examples are from the Investigations reported In
6.2.2.1.

Table 6.1. Equal behaviour at equal soil mechanical properties.

Sol1 pore space moIsture cone Index «1 «2
(%) content(%) (MPa) (N) (N)

Sch1nnen 47.9 17.97 0.95 225 221

Wagen1nge n 42.2 16.98 0.90 223 221

R-\= Rol1 Ing resistance In the first pass

R2 = Ro11 1ng resistance In the second pass

6.2. EMPIRICAL METHODS

These methods use empirical relationships, formulas, or graphs.

It Isassumed that a process aspect y is dependent on character-
istics xi x:
n

y = f(x 1 , ,x n )

Where, f represents the way inwhich y depends on x-| x

n•
The prediction function f is known to such an extent that y can
be calculated for any arb Itrary values of x-| x n (within
certain IImIts).
Methods that use relations hips between soil characteristics and
process aspects are notI ntended to be dimenslonaIIy correct.
These methods only try tof Ind relationships that can be used for
prediction purposes. Sectio n 6.2.2. deals with these methods,
Empirical methods based on dimensional analysis use relationships
between soli characterIstI cs and process aspects as well. The
main difference, compared to the methods from 6.2.2, is that
attempts are made to formu late dimenslonaIly correct prediction
functIons.

6.2.1. EMPIRICAL GRAPHS

Examples of empirical graphs are presented by Zoz (1972), and

Witney and OskouI (1982).
Zoz (1972) presents a chart that may be used to determine the
expected drawbar horse power, travel speed and travel reduction
of any two-wheel drive tractor under various soil conditions.
This graphical solution Is based on average tyre performance for
single tyres on four surfaces. Three weight transfer coefficients
are used for each soil type. This predictor does not require
actual tractor specifications such as tyre sizes, tyre Inflation

106
pressures and front-axle load. Only a reasonable relationship
between tractor weight and tyre size Is necessary. Therefore, the
accuracy of prediction will generally not be very high.
At the Edinburgh School of Agriculture a ploughing performance
predictor nomograph has been devised by Witney and OskouI (1982).
This chart, one of the results of a five-year research programme
on tractor power selection for ploughing, eliminates laborious
calculations to find the most efficient tractor/plough combina-
tion for any specific conditions. The predictor uses three plough
body types, three compaction levels of the top layer, six soil
types, three traction efficiency levels, and several possibil-
ities of depth of cut, width of cut, and drawbar power.
The chart can be used In several ways. It shows the work rate
obtainable and the sensitivity of the power requirements to
changes In the various characteristics of a particular plough.
Specific Iterns such as the best plough type, travel speed and
best working depth can be selected by trial and error. In agri-
cultural field practice this chart can be used to examine the
existing ploughing procedure. This predictor is most accurate
when the soil moisture content isclose to field capacity and the
compaction level agrees with the one of medIurn loam soils.

6.2.2. EMPIRICAL METHODS USING RELATIONSHIPS BETWEEN SOIL CHARAC-

TERISTICS AND PROCESS ASPECTS

The main characteristics of a towed tyre-soli system are rolling

resistance of the wheel, depth of the rut, and soli compaction
caused by the wheel. The numerical value of these characteristics
depends not only on tyre parameters, such as size, vertical load
exerted by the wheel on the soil, Inflation pressure, and forward
speed, but also on properties of the soil. A wide range of me-
thods to predict the value of these characteristics from tyre
parameters and soli properties are mentioned In literature.
Usually, these methods are Intended for a range of tyre sizes,
vertical loads and Inflation pressures. Very often, verification
of predictions Is limited to one or two typical soil types whose
porosity and moisture content were almost always constant. More-
over only a single characteristic to express soil properties
(e.g., cone Index)or a combination of very few soli character-
istics selected a priori are used.
In the Investigation presented In this paragraph, tyre parameters
were kept constant, and soli types and conditions varied con-
siderably. Different measurements of soli mechanical properties
were carried out to predict rolling resistance, rut depth, and
compaction under a wide range of soli conditions for the particu-
lar tyre parameters tested.
In these tests, the soil was moving instead of the wheel axis.
The wheel was suspended In a rigid frame that had been built over
a set of fixed rails. Soli bins had been prepared in a separate
building and were transported to a carriage which could be pulled
along the rails by an electrically-powered cable winch. The
linear dimensions of the tyre tested were about half of those of
current agricultural trailer tyres. The depth of the soli In the

107
/?1 - 2 0 4 . C / - 1 - 4 2 with r-0.98 (for all bins) [6.3]

«1 = 2 0 6 . C / - 1 - 4 2 with r=0.99 (for all bins except Ede Sands).

[6.4]

The two measuring programmes together result In:

«1 = 2 0 3 . C / - 1 - 4 3 with r-0.97 [6.5]

This analysis shows a high degree of correlation between CI and
R-\ In five soils and under a variety of soil conditions. Under
all forty-two soil conditions the wheel parameters were kept
constant. The slightly different behaviour of Ede Sand Is In line
with the fact that cone Index readings for these bins differed
from the readings for the other bins (see TIJInk and Kooien,
1985).

Tyre rolling resistance In the second pass (R2) correlates well

with CI, Clf,and C/ r :

For Almkerk, Lexkesveer, Schinnen, and WagenIngen soils:

R2 = 1 6 1 . C / - 1 - 2 0 with r-0.92 [6.6]

R2 = 1 8 1 . C / f - 1 - 3 2 with r-0.96 [6.7]

-1 23
R2 = 2 0 8 . C / r - with r-0.97 [6.8]

For all examined soils:

R2 = 157.C/~ 1 - 2 1 with r=0.92 [6.9]

R2 = 1 7 5 . C / f - 1 - 3 4 with r-O.95 [6.10]

R2 = 1 9 9 . C / r - 1 - 2 3 with r-0.94 [6.11]

Where, R2 *measured rolling resistance in the second pass (N)

CI »measured cone Index prior to the first pass (MPa)
Cl f - cone Index measured in the hole formed by a falling
weight (MPa); see chapter 5.4.
C/ r - cone index measured In the rut prior to the sec-
ond pass (MPa).

For all soils the highest correlation was found between R2 and
Cl f (Flg. 6.3). In attempts to find linear relationships between
the wheel system and soil characteristics, the following derived
values were tried:

1/C/2

1/C/ f 2

CI f.D2.

110
Moreover, multiple regression was applied.
In the first pass the best linear relationships for rolllnç
resistance were found with 1/C/2 (Fig.6.4):

/?! - 109 + 96.1 0 1 2 (1/C/ 2 ) with r-0.98 [6.12]

In the second pass 1/C/f2 correlated slightly better with Rz

than 1/C/2:

98 + 8 2 . 1 0 1 2 (1/C/ f 2 ) with r=0.96 [6.13]

89 + 6 5 . 1 0 1 2 (1/C/ 2 ) with r=0.93 [6.14]

Fig. 6.5 shows the relationship between 1/C/f2 and R^. •

We can see from this Investigation that In a wide range of
soils and soil conditions, there Isa high correlation between
cone Index and tyre rolling resistance. We may conclude that
rolling resistance can be predicted accurately In uniform soils.

u all dots:fl 2 =175-CJ f - 1 - 3 4 (r=0.95)

MPa

closed dots only:/? 2 =181 CJ f " 1 ' 3 2 (r =0.96)

1.2

o 1.0

0.8
S*
0.6

0M

0.2
-1 r-
0 100 500 N 1000
rolling resistance in the second pass,/? 2

Fig. 6.3. Rolling rest stance In the second pass related to Clf.

111
 u
12-

10

e
» 8

fM

fl.|=:109*96-1012-CJ~2 (r =0.98)

+ a n d o :Tijink (1979)
• and o :Tijink andKooien (1985)
2-

0 100 500 1000 N 1500

rolling resistance in the first pass,/?^

Flg. 6 A. Relatlonshlp between measured values of rolling resls-

tance In first passes and 1/CI2, where CI Is the measured cone
Index.

12

10

8-

o"" 6- R2 =98+82.4-10 1 2 -(I/C/ f 2 ) (r=0.96)

U-

2 y
0
P
0 100 500 N1000
rolling resistance in thesecond pass,/?2

Fig. 6.5. Rolling res Istance In the second pass related to

1/Clf2.

112
6.2.2.2. RELATIONSHIPS BETWEEN SOIL CHARACTERISTICS ANDSOIL COM-
PACTION DUETOWHEEL ACTION
In the Investigations discussed above, soil compaction due to
wheel action wasalso measured. Dimensions oftherut left bythe
tyre and changes Insurface height near therut were measured
with a profile gauge consisting ofa rowof51 vertical needles
each set 10mm apart. Therowof needles could beplaced above
the soil perpendicularly tothedirection of wheel travel, so
that theshape ofthesurface under therowcould bemeasured by
lowering theneedles until they touched thesoil andbymeasuring
the sInkage of all needles. Themeasurements were taken at a
fixed place Inthebinbefore thefirst andafter each tyre pass,
and were presented graphically. Wedetermined thedifferencebe-
tween the Initial surface area ofavertical cross section ofthe
bin and thesurface area ofthecross section after each tyre
pass from these graphs. Thedifference isequal tothearea lost
Incompaction duetowheel action (seeFig.6.6).
Data on soil compaction andsoil properties were analysed.
High correlations were found between soil compaction andthesoil
characteristics CI.D 2 and Clf.D2-.

AA1 - -6.9 +0.16C/.D' r-0.94 [6.15]

AA1 - -5.3 + 0.13C/ f .D 2 r-0.97 [6.16]
AA1+2" -8.2 +0.21C/.D 2 r=0.95 [6.17]
AA1+2- -4.8 + 0.16C/ f .0' r=0.95 [6.18]
where, AA1 =soil compaction inthefirst pass expressed as
the cross-sectional area lost (cm 2 )
so
A A1+2 " '' compaction intwopasses expressed as
cross-sectional area lost(cm 2 )
CI =measured cone Index(MPa)
Cl f =cone index measured under holes formed bya
falIing weight(MPa)
D =depth ofthehole formed bythefalling weight
prior tothefirst pass(mm).
With equations 6.15 and6.17soil compaction was calculated.
Figures 6.7and6.8 show therelationships between measured and

Fig. 6.6. Decrease of soil area In a vertical cross section of

the rut, due to compact Ion of soil by the wheel. This decrease
equals the dotted area minus the sum of the hatched areas.

113
 70-1
cm2
60-
c
o
o 50-
CL
E
o
Ü
40-
£
o
D
O 30-
O
O

20-
(r=0.97)

10 H

I 1 -I 1 1 1 1 1

0 10 20 30 CO 50 cm260 70
measured compaction,A^
Flg. 6.7. ReIatlonshlp between measured and predicted compact Ion
In the first pass. Compact Ion Is expressed In cross-sectlonal
area lost due to tyre action.

i.0 50 60 70cm280 90
measured compaction,A A 1 + 2

Fig. 6.8. ReIatlonshlp between measured and predicted compact Ion

due to the first and the second wheel passes.

114
calculated soll compaction. From this Itseems that soil compac-
tion duetowheel action canbepredicted satisfactorily fromthe
soil measurements CI, C/f,andD.
Sometimes It Is thought that rutdepth Isa measure for soil
compaction (Boels, 1978). An analysis of data onrutdepthsand
soil compactions showed a poor relationship between these charac-
teristics (seeFig. 6.9). This poor correlation Isdueto the
bulging ofsolIbeside therut. Forthewheel passes with negII-
glble bulging, high correlations (r-0.99) between rutdepthand
soil compaction were found.

100

mm
CM
"O 80 H
T3
C
O

° 60
Q.

Z i.0

••
20

uc20 i.0 60 80 cm2 100

compaction (A A - and A . . _ )

Fig. 6.9. RelatlonshIp between soil compact Ion In cross-sectlonal

area lost due to wheel action In first pass, AAJ, and In first
and second passes, hAj+2' ancl rut depth after first pass, d j , and
after second pass, d2.

115
 ÖU-
y
mm • X

60-
.c
CL
"
TD
40- •
-)
t_
/ /
•o *•
<u /
o 20 ( r =0.98)
•D
Ol
Q.
* /
n » '
j

0 20 40 60 mm 80
measured rut depth,d^

Flg. 6.10. Predicted and measured values of rut depth after the
first pass.

100-
mm'
80-

Q.
"° 60

ai
y 40
T>
<V

20
1 °*y' (r=0.99)

? • •
0 20 40 60 80 100

measured r u t depth, d ^

Fig. 6.11. Predicted and measured values of rut depth after the
second wheel pass.

116
6.2.2.3. RELATIONSHIPS BETWEEN SOIL CHARACTERISTICS AND RUT
DEPTH

Accurate predictions of rut depth are not known from literature.

In our Investigations rut depth was measured In the centre of the
rut as the difference between the initial soil height and final
solIhelght.
The use of relatively Independent soil characteristics CI and D
resulted In good predictions of rut depth:

'1 -0.85 + 2.65C/ - 2 + 0.028D 2 r-0.98 [6.19]

-0.01 + 5.01C/ - 2 + 0.018D 2 r-0.99 [6.20]

where, tf^ « rut depth after the first pass (mm)

d2 - rut depth after the second pass (mm)
CI - cone Index prior to first pass (MPa)
D » depth of hole formed by the falling weight prior
to the first pass (mm).

Fig. 6.10 and 6.11 show the relationships between measured and
calculated rut depths after the first and the second pass of the
wheeI.

6.2.3. EMPIRICAL FORMULAS BASED ON DIMENSIONAL ANALYSES

Empirical formulas for tyre performance predictions often are the

result of dimensional analyses. This method evolves from con-
sideration of the variables of the problem and the dimensions of
these variables. To apply dimensional analysis it Is necessary to
determine the variables and their basic dimensions. Application
of the Buckingham Pl-theorem leads to a set of dimensionless
parameters. The number of parameters generally isdecreased by
reducing the domain of the problem. In tests the relationships
between characteristics of the tyre and the soil reactions can be
evaluated In terms of dImensionless characteristics. The rela-
tionships obtained can be used to predict tyre and soil behav-
iour.

6.2.3.1. PREDICTING OFF THE ROAD TYRE ROLLING RESISTANCE

Empirical predicting formulas for tyre rolling resistance have

been reviewed and examined with data presented in this disserta-
tion and with data published by Perdok (1978) and Lumkes and
Perdok (1981).

117
REV IEW:

F r e i t a g ( 1 9 6 5 ) u s e d t h em e t h o d o f d i m e n s i o n a l a n a l y s i s t o m o d e l
t h e t y r e - s o i l s y s t e m . T y r e - s o i l s y s t e m v a r i a b l e s i d e n t i f i e d by
F r e i t a g a r ep r e s e n t e d in T a b l e 6 . 2 . T h r e e b a s i c d i m e n s i o n s a r e
i n d i c a t e d a n d a c c o r d i n g t o t h eB u c k i n g h a m P i - t h e o r e m , t h e n u m b e r
of v a r i a b l e s c a nb e r e d u c e d b y t h r e e . T h eP l - t e r m s w h i c h w e r e
d e r i v e d a r ep r e s e n t e d in T a b l e6 . 3 .
T h e f u n c t i o n a l r e l a t i o n s h i p b e t w e e n t h eP l - t e r m s is:

P/W ,R/W ,T/d.W ,zld m f(ft,f(,S,b/d,h/d,S/h,c.d2/W,Xd3/W,g.d/V2)

T e r m s t h a t i n c l u d e f r i c t i o n a n g l e , c o h e s i o n a n ds p e c i f i c w e i g h t
c a n b e o m i t t e d w h e n t h ec o n e index ( C / )a n dt h eg r a d i e n t of c o n e
index ( G ) a r eu s e d t o c h a r a c t e r i z e c o h e s i v e s o i l s ( c l a y ) a n d
f r i c t l o n a l soli ( s a n d ) r e s p e c t i v e l y . T h e i n f l u e n c e of d r i v i n g
s p e e d a n d (h/d) h a sb e e n c o n s i d e r e d I n s i g n i f i c a n t . If, In a d d i -
t i o n , a c o n s t a n t d e g r e e o f s l i p is s u p p o s e d t o o c c u r , t h e f u n c -
t i o n a l r e l a t i o n Is r e d u c e d t o :

P/W,R/W,T/d.W,z/d = f(Cl.d2/W,b/d,6/h) for clay and

P/W,R/W,T/d.W,z/d = f ' { G . d 2 / W , b / d , S/h) for sand.

Table 6.2. Tyre-sol I system var Iab les used by Freitag.

Variable Symbol Dimensions

INDEPENDENT VARIABLES

SolI
Internal friction angle •
CohesIon c FL-2
SpecifleweIght y FL-3
Penetration resistance CI FL-2

Tyre
DIameter d L
SectIonwidth b L
Section height h L
DefIectIon S L

System
Load W
TransiationaI speed V LT-1
Slip s
Tyre-soII frletion f*
Acceleration due to gravity g LT-2

DEPENDENT VARIABLES
PuI I p F
Rol IIng r e sis t a n c e R F
Torque T FL
SInkage z L

118
Table 6.3. PI terms used by Freitag.

TT1=P/W T C 2-R/W TT 3-T/d.W TT 4 - Z/d TT.5 =/*

Ttg=/ TT7=
S Tts- b/d TTg•h/d ^10°° S/h
Ttn = cd2/VI Tt12- Xd*/W K 13 - CI.d2/W Mit - g.d/V2

Freitag developed twodimensIonless prediction terms (mobility

numbers) fortreadless, two-ply tyres with circular-shaped cross
sectIons:
C l a y m o b i l i t y number Nc = (CI.b.d/W).(S/h)°'5

Sand mobility number N s=(G(fc.d) 3/2 /^).( S/h)

Relationships between mobility numbers andtyre performance were
presented.
Turnage (1972) extended theclay mobility number (Nc) , proposed
by Freitag, with an additional factor (1/(1+b/2c0)totakethe
cross-sectional shape (circular or rectangular) into account:
M - (CI.b.d/W).(S/h)0-5.(1/(1+b/2c0 )

The shape of therelationships between prediction termsand

rolling resistance coefficient arepresented inFig. 6.12.a+b.
Wlsmer andLuth (1973) used an approach similar totheone used
by Freitag andTurnage anddevelopedadImensionIess wheel nume-
ric (C n ) foratyre moving In cohesive-frletlonaI soli:
Cn = CI.b.d/W

This wheel numeric wasrelated torolling resistance by:

R/W-(1.2/C n )+0.04 [6.21]
The shape oftherelationship Is given In Flg. 6.12.C. TriIs
formula hasbeen adopted bytheAmerican Society of Agricultural
Engineers: ASAE Data 230.4 (Hahn et al.,1984), andcanbeapplied
to cohesive frletlonaI soils ofmoderate compactlbIIity andto
tyres operating at inflation pressures that produce radial tyre
deflections of approximately 20percent oftheundeflected tyre
section height. Typical tyre dimensions areasection widthto
diameter ratio (b/d) of0.3andarolIing radius to diameter
ratio of 0.475.
Values ofC n fortypical surfacesare:
50-hard soIIs
30-fIrm solIs
20-11IledsolIs
15-soft, sandy soils

119
 0.30-

\ 0.20 • clay
Q:

0.10-

0-
5 10 15 20 25 30
CIbd £ 1
W h 1+b/2d

0.301

^ 0.20 i

0.10-

0
10 20 30 £0 50 60 70 80
32
Gibd) ' {
W h

k 0.60
(c)
Q:

"c.
<u
V
iE o.io
o

u
c
a
« 0.20 cohesive frictional soil
U)
ai
L-
Dl
C

10 20 30 i.0 50 60 70
r . CIbd
W

Flg. 6.12. Rolling resistance coefficient of a towed tyre related

to clay mobility number (a), sand mobility number (b), and wheel
numeric (c).

120
Dwyer et al. (1975) and Gee-Clough et al. (1978) Investigated the
field performance of tractor rear tyres ranging In size from
12.4-36 to 18.4-38. The following prediction function for rolling
resistance was proposed:

R/W = (0.287/M) + 0.049 [6.22]

To Include the effect of tyre construction, Me AIIIster (1983)

proposed the following relationships:

for cross ply tyres: R/W = (0.323/M) +0.054 [6.23]

and for radial tyres R/W= (0.321/M) + 0.037 [6.24]

ANALYSIS:

In the Investigations, described In section 6.2.2, tyre and sys-

tem parameters were kept constant. So, the parameters b, d, Wand
S/h are constants. Mobility numbers change into:

Nc = & 1 . C /

M = a 2 .C/
are
where, a-)and az constants.

In Fig. 6.1 the relationship between cone Index (C/) and rolling
resistance Is presented. We may conclude that mobility numbers,
used at constant tyre and system parameters, accurately predict
tyre rolling resistance.
When tyre deflection measured In a test bin (£ a ) was used almost
equal correlations between mobility number M and rolling resis-
tance were found. The use of the actual tyre deflection (ó"a)did
not Improve the accuracy of prediction.

The drawback of mobility numbers (Nc and M) and wheel numeric

(C n ) Is that they have no linear relationship with rolling resis-
tance (see Fig. 6.12). Neither do they Include the effect of tyre
construction (cross ply or radial) and tread pattern. To get
linear relationships It can be useful to use C / - 2 Instead of CI
(see Fig.6.4).

Equations 6.21, 6.22, and 6.23 were examined using data from our
soil bin experiments (see TIJInk and Kooien (1985) for soil
conditions, soil characteristics and results of wheel-soil system
measurements). Using equation 6.21 the accuracy of prediction was
poor (Fig. 6.13) although tyre dimensions agreed with the spec-
ificationsmentioned by Wismer and Luth. In Fig. 6.14 calculated
rolling resistance, using equation 6.22, Is plotted versus meas-
ured rolling resistance. The same poor result was obtained. Mc
AIlister's prediction formula for cross ply tyres (equation 6.23)
also resulted In poor predictions of rolling resistance (see Fig.
6.15).

121
The differences between these prediction equations and the poor
accuracy they achieved In predicting rolling resistance In test
bins demonstrate that every equation has a limited field of
application and that those equation cannot be extrapolated with
confidence. The different equations were developed from Investi-
gations with different tyres. Freitag used 4 different tyres with
diameters ranging from 0.36 m to 0.72 m. These were all tread-
less, thln-walled, cross ply tyres with two-ply rating. Equation
6.22 was based on experiments with tractor rear tyres ranging
from 1.45 m to 1.75 m diameter. This expression was valid for
stubble, ploughed and cultivated fields only (r-0.73). Dry grass
fields and fields with a loose, wet, top layer showed signifi-
cantly different results. InMe AIlister's experiments tyre diam-
eter ranged from 0.8 m to 0.95 m. Measured rolling resistance
values In Fig. 6.13 to 6.15 have been determined using an Imple-
ment tyre with a diameter of 0.67 m.

1600 i

<D
O
c
o

Ol
c
o 1000

200 400 600 800 1000 1200 U00 1600N1800

measured rolling resistance, R^

Flg. 6.13. Relation between calculated rolling resistance, calcu-

lated using equ. 6.21 proposed by Wlsmer and Luth, and measured
rolling rest stance (Rj) of a 7.00-12 Implement tyre.

122
 1600

O
c
o

Ï 1200 •
at
c

•o
01
'S 800

Flg. 6 14. Relation

between calculated 400
rol I1ng reslstance,
calcuIated us Ing
equ. 6.22 proposed
by Gee-Clough, and
measured rolling
reslstance (Rj) of a 400 800 1200 N 1600
7.00-12 Implement
tyre. measured rolling resistance,/?^

1600

ai
o
o
Vi
1200
O)
c

~ 600-

Flg. 6.15. Rela-

tion between calcu- 400
lated rolling resis-
tance, calculated
using equ. 6.23 pro-
posed by McAlI Ister,
and measured rolling
res Istance (Rj) of a 400 1200 1400 N 1600
7.00-12 Implement
tyre. measured rolling resistance,/?^

123
Predicting terms can be Improved by using empirical methods In
strongly reduced domains. Some examples are given below, using
experimental data presented In section 6 2.2 and data published
by Perdok (1978).
The domain of the problem can be reduced strongly by the use of
the same tyre-system settings In the same fleid at two dIfferent
times. On both days rolIing resistance was measured at four
Inflation pressure levels. Only soli conditions differed on these
days. Linear relationships between rolling resistance measured
on two days are presented In Table 6.4.

Table 6.4. Coeff/clent of linear correlation between rolling

res Istance on 2 different days In the same field and
with the same tyre-system sett Ings.
tyre R
day 1 <day 2
13-16 Implement r=0.99
12-38 Treadless tractor rear r-0.95
16-15.5 Implement r-0.99
16-20 Implement r=1.00
20-22.5 Implement r-0.94

Another example of a very reduced domain Isa test carried out

under one field condition with only variable levels of tyre
inflation pressure. This resulted In high linear correlations
between R/W and p\ (Table6.5).

Table 6.5. Linear correI at Ions between rolling res Istance coeffi-
cient (R/W) and tyre Inflation pressure(P|).

tyre wheel load

day 1 day 2
wheel load allowed
at p\ used

13-16 Implement 0.99 0.98 1.09 - 1.57

12-38 TreadIess 0.99 0.89 1.12 - 1.78
Tractor Rear
16-15.5 ImpIement 0.98 0.99 0.99 - 1.78
16-20 Implement 0.99 0.98 0.79 - 1.28
20-22.5 ImpIement 0.99 0.91 0.64 - 1.68

High linear correlations between rolling resistance Rand W 2 were

also found In our analyses. The variables Inflation pressure p\
and wheel load were combined to p\.W2. Fig. 6.16 shows the rela-
tionship between P\.WZ and R for a 13-16 implement tyre in a
particular field. A high degree of correlation (r=0.99) between
P\.WZ and Rwas found. For the other tyres similar correlations
were found.
We may conclude that the term p\.Wz can be applied to make accu-
rate predictions of rolling resistance in a restricted domain.
If rolling resistance A Is a function of P\.WZ , then the coeffI-

124
 t y r e : 13-16 Implement

uo-
1012N3/m2
120-

100-
CM

|- 80

60

R=0.043-10" 1 2 -Pj^ 2 -30.9 (r=0.99)

40

20-

oV 1000 2000 3000 N 4000

rolling resistance,/?
2
Flg. 6.16. Relation between p/W and rolling resistance measured In
a part leu Iar field.

clent of rolling resistance is a function of P\.W.

To achieve dImensionless prediction numbers, we developed
terms IIke:

(p,.W)/(CI2.b.d), (ps.W)/(CI2.h.d), (p,.W)/(CI2.h2),

(p,.W./7)/(C/ 2 .to 3 ), (P| .W.b)/(CI2.h2.d) , (p\.W .b)/(.CI2.h.d2) ,

(Pl.W.b)/(CI2.h3), (P|.W.b)/{CI2.b2.d), etc.

In order to decrease the number of variables, the domain was

reduced by choosing only one soil condition. For the tyres tested
the dimensIonIess term ( P | . W ) / ( C I 2 . b . d ) achieved the highest
accuracy In predicting the rolling resistance coefficient (see
Fig. 6.17).
Despite the high accuracy achieved In Fig. 6.17 Itmust be clear
that more tyre variables are necessary when an universal predic-
tion term Iswanted. It Is not enough to characterize a tyre with
b, d, h, Wand S/h or with b, d, h, Wand p ( . Additional terms
should be used (for tyre construction, tyre flexibility, and
tread pattern) or the application of the prediction term Is
limited to a particular domain.

125
 20
tyre: • =13-16
%
+ =20-22.5
x =16-20
o =16-15.5

i= 15

10

5 R/W=2.60*695lpiW/CI2bd) (r=0.98)

o£
5000 10.000 15.000 20.000
p\NlCIlbd
Fig. 6.17. Relation between rolling rest stance coeffIclent and
the dI mens Ion less term (p/W)/Cl^bd for Implement tyres In a
particular field.

6.2.3.2. PREDICTING TYREROLLINGRESISTANCEONAHARD SURFACE

Stelner and Söhne (1979)measured tyre rolling resistance of
towed tractor rear tyresonaconcrete surface. The test tyres
weresized from 11-28to 16.9-38and 18.4-26. Bothcrossplyand
radial tyreswere examined. Itwas found that the Influence of
driving speed can beneglected inthe range from 0.8 to 3 m/s.
All rolling resistancemeasurementsweremadeatadriving speed
of 0.8 m/s. Thisproved that tyre rolling resistance ona con-
crete surface Isa functionof tyre inflation pressure (P|),
wheel load (W), tyredimensions, tyreconstruction, and ply
ratIng.
For equal ply rating and construction: R/W= f(p|,W,b,d,h)

Several dImensIonIesstermsweredetermined. The dimensionless

term bW/h3pi correlated highestwithmeasured valuesof rolling
resistancecoefficients. For six-ply agricultural rear tyresthe
following equationswere found:
Cross ply tyres: R/W= 1.27 +0.65(bW/h3p|) (%) with r=0.85
[6.25]
Radial tyres: R/W = 1.33 +0.40(bW//i3P|) (%) with r=0.57
[6.26]

126
Perdok (1978) and Lumkesand Perdok (1981) published data on
rolling resistanceof somecrossply Implementtyres.An analysis
of thesedata Ispresented here.
As expected from section 6.2.3.1, theequations 6.25 and 6.26
could not beextrapolated satisfactorily to Implement tyres.How-
ever, a good prediction of rolling resistance of the tested
implement tyrewas possible,using thedimensionIess term bW/h^p\
(seeFig. 6.18). Theprediction function found can beexpressed
as follows:
R/W=0.46 +0.83(bW/h3p,) (%)with r=0.97 C6.27]
where, R =rolling resistance (N)
W =vertical wheelload (N)
to=tyresectionwidth (m)
h =tyresection height (m)
P\ =tyre Inflation pressure (N/m2)
We m may conclude that for tyreswithequal
with equal construction and ply
rati ng the dimensIonIessterm bW/h3p\ is"* useful for predicting
tyre•• rolling resistanceonahard surface.

5 R/W=0X6+0.83 lbW/h3p-.) (r=0.97)

^ %

<u

0)
o 3-
o
ai
o
c
-
o 2
tyre: • =13-16
+ =20-22.5
O)
c x =16-20
o =16-15.5

0 1 2 3 4 5
2
bWlh pi

Flg. 6.18. Relation between rolling reslstance coeffIclents and

predict Ing term' bW/h3Pi using four Implement tyres on a concrete
surface. Lines A and B are relatlonshlps found by Stelner and
Sohne (1979) for six-ply tractor rear tyres.

127
6.3. APPROXIMATE METHODS

Approximate methods to predict off-the-road wheel performance are

generally based on pressure-sInkage relationships.
Bernstein (1913) stated that rolling resistance of a wheel is due
to the vertical work done In rut forming. Predictions have been
based on the pressure-sinkage relationships of a rectangular
piate:

p« K.zn

where, p » pressure on the plate

k mmodulus of soil deformation
z - sInkage
n - exponent of deformation.

Bernstein found n = 0.5 for agricultural soils. The parameter k

proved to be dependent on plate dimensions.

Bekker (1956, 1960, 1969) used the Bernstein theory to predict

rolling resistance. To get independent parameters Bekker proposed
that the modulus of soli deformation has two components; kc Is
the cohesive component and k^the frlctlonal component:

k = kc/b + k$

Bekker's pressure-sInkage r e l a t i o n s h i p of a r e c t a n g u l a r plate

states that:

p = (kc/b + kó)zn

To o b t a i n v a l u e s for kc and k^, p e n e t r a t i o n t e s t s w i t h d i f f e r e n t

plate width bare needed.
A s s u m i n g that r o l l i n g r e s i s t a n c e Is equal to v e r t i c a l w o r k d o n e
In rut forming and that soil r e a c t i o n , at a point In t h e c o n t a c t
a r e a , Is a l w a y s radial to t h ewheel and equal t o t h e p r e s s u r e
under a v e r t i c a l p l a t e at t h es a m e d e p t h , B e k k e r f o r m u l a t e d t h e
following rolling resistance expression:

b(kc/b + k^) ZW (2n+2/2n+1)

n +1 b(kc/b + *0)c*o-5(3 n)

To determine ground pressure (P g ) under a tyre Bekker proposed to

use the following expression:

Pg - P\ +P c -
where, p\ =•Inflation pressure of the tyre
pc - carcass stiffness.
+ 1 / n
[£>(P| + p c ) ] C )
Rolling resistance now becomes: R
(kc + ö * Ä ) 1 / n ( n + 1)

128
Uffelmann (1961) also assumed that the soil reaction Is radial to
the wheel and that the rolling resistance Isequal to vertical
work done. The pressure beneath a track Is expressed as:

p = 5.7c

where, c - cohesion of the soli.

Uffelmann's rolling resistance equation for towed rigid wheels

Is:

R = W2/5.7cbd

In fact this Isa special case of Bekker's theory with n-0 and
p=kc/b + k^.

Wills et aI.(1965) Investigated the validity of the Bekker the-

ories. They found that:
- rolling resistance Is not only due to vertical work done
- the assumption of a pure radial soil reaction Is not cor-
rect
- n Isa soM parameter
- kc, k$ and k are dependent on plate dimensions.

In his theory Gee-Clough (1976,1979) Included the effects of slip

and deep slnkage on rolling resistance of rigid wheels In sand.
He proposed to determine soil slnkage parameters from measure-
ments with a wheel of known dimensions. This principle Is the
basis of the "Wheel Bevameter". It Is less difficult to determine
the Bekker soil parameters with the "Wheel Bevameter". The soil
reaction under a "Wheel Bevameter" Ismore similar to the one
under awheel than to the one under a strip footing.

To make the Bekker theory applicable to pneumatic tyres, Perdok

(1978) assumed that a flexible tyre of a defined size behaves
like a rigid wheel of a larger diameter (see Fig. 6.19). A defor-
mation factor C was introduced:

D* = CD

where, D* = alternative rigid wheel diameter

D mtyre dlameter.

The deformation factor depends on Inflation pressure, wheel load,

and soil conditions. Full-sized rigid wheels were used as "Wheel
Bevameter".
The "Wheel Bevameter" gives better predictions than the plate
slnkage tests. It also makes the pressure-sInkage method more
empIrIcal.

Kooien (1976) gives an approximate method for soil compaction

under a towed tyre. His theory is based on the hypothesis that
the compaction of a volume element of soil under a certain tyre
for a certain soil and wheel load can be predicted by uniaxial
compression of a sample of that soil (Fig. 6.20). In this unl-

129
axial compression test the piston pressure Isequal to the Infla-
tion pressure plus some corrections that should be applied to
obtain the same compaction as the volume element of soil under
the wheel.
Corrections are to be made for:
- carcass stiffness of the tyre
- unequal pressure distribution In the contact area
- shear stress In the contact area
- decrease of pressure with depth
- area of the contact surface
- bulging of soli beside the tyre
- wall friction In the compression test.

Stienstra (1976) investigated the sum of corrections to be made

for a solIsample at 5 cm depth In the longitudinal centre plane
of a rut formed by a tyre with an inflation pressure of 3.3 bar
and a wheel load of 30 kN. In the uniaxial compression test a
pressure of 7.1 bar was needed to achieve the same compaction.
So, in this case, the sum of the corrections had a value of 3.8
bar.

Fig equivalent rigid wheel

tyre inflation pressure

+ corrections

•A"\".-V

Äs« » vc^
Fig. 6.20. Approximate compaction model

130
6.4. EXACT METHODS

Exact methods try to model soil deformation in a mathematical

way. Two approaches to predict soil-wheel behaviour have been
adopted: the sIIp-1Ine method and the finite element method.

6.4.1. SLIP-LINE METHODS

In the longitudinal plane of awheel-soIIsystem two fields of

soil failure were found by Wong and Reece (1966,1967). A math-
ematical model based on limit equilibrium was used to predict the
stress distribution under two-dimensional wheels. Good agreement
between measured and predicted values was achieved.
The model assumes an Incompressible medIurnand is therefore not
adequate for predicting soil compaction due to wheel action.
Karaflath and Nowatzkl (1978) showed similar failure zones. Their
solutions assumed that a rut was formed. However, this can not be
combined with plain strain and an incompressible medium.

6.4.2. FINITE ELEMENT METHODS

Perumpral et al.(1971) adopted the finite element method (FEM) to

predict wheel-soil behaviour. The FEM assumes linear stress-
strain relationships. Soils do not satisfy this assumption.
Therefore, a variable elastic modulus for the soil was assumed.
So, each finite element was given such different elastic values
that the non-linear stress-strain curve was approximated. Strict-
ly speaking this Is incorrect. Yong and Fattah (1976) used a
similar method. Wheel performance was predicted In terms of
energy components as a function of wheel slip. Investigations
with tyres operating on an undeformable surface were made by Yong
et al.(1978) to Include tyre deformation in the prediction model.
Although attemps have been made to apply the FEM to flexible
wheels, the FEM-approach Is two-dimensional and therefore less
useful In predicting wheel and tyre performance In field oper-
ations, where bulging occurs beside the wheel.

6.5. CLOSING REMARKS ON PREDICTION METHODS

With empirical methods wheel-soil characteristics can be predict-

ed accurately in uniform soils. The relationships discussed In
section 6.2.2 show that easily obtainable soil characteristics
can be used for these predictions.
In agricultural field practice soli conditions usually vary with
depth. Therefore, further studies should be made to account for
solIvarlatIons.
Under uniform soil conditions dImensionaIly correct empirical
methods can also achieve a high degree of accuracy In predictions
(see section 6.2.3).

131
 Approximate methods can Improve the understanding of the mech-
anism. The complexity of such methods Increases enormously the
more similar the model Is to the practical situation. Accuracy Is
higher when the model becomes more empirical.
In practice exact methods are difficult to apply because
appropriate ways of describing the mechanical properties of soil
are not available. Further Investigations into the mechanical
behaviour of soil are needed before computer-based solutions can
be expected to become adequate.
The attraction of exact methods Is their ability to predict the
state of stress In the entire soil during the passing of awheel.
Comparative and empirical methods proved to be the easiest and
most realistic approaches to predict wheel-soIIbehaviour under
practical conditions. These methods do not Improve the under-
standing of the wheel-soIImechanism.
Further development of approximate and exact methods is needed
for a better understanding of the processes. This, In turn, will
make It possible to choose better process aspects which can then
be used inempirical prediction methods.

132
 CHAPTER 7

SOIL PHYSICAL ASPECTS OF LOAD-BEARING PROCESSES

The physical properties of a soil can change due to load-bearing

processes. On the other hand soil physical properties influence
the mechanical behaviour of a soil.
For a given soil with specific Intrinsic properties (partIcIe
size distribution, types of minerals, organic matter content,
etc.) Important physical factors on the micro level are:
- pore space and the distribution of pore sizes
- the distrIbutIon of soiIpartleleswithin the soII
- soil water content and the distribution of water within the
solI
- soli air content and the distribution of air within the soil
- points of contact: number of contacts, bonds, and distributions
of these bonds.
These properties, so-called micro-factors, can be changed not
only by tillage processes but also by natural forces (freezing,
thawing, etc). Tillage processes Influence the mechanical prop-
erties of the soIIas welI.
In general farmers are not interested inmicro-factors, but In
soIIquaI111es such as: pF-curves, water and gas conductivities,
thermal properties, penetration resistance, erodiblllty, till-
abIIIty, etc. From thls point of view soli mechanical properties
can be seen as solI quaIitles as welI.
The micro-factors can be seen as the Interface between tillage
processes and sol quaI11les.
Kooien (1986) deals with the relationships between micro-factors,
soil mechanical properties, mechanical processes, and process
effects.
In this chapter we discuss the Influence of some physical prop-
erties on mechanical properties and the Influence of mechanical
treatment on physical properties.

7.1. THE INFLUENCE OF PHYSICAL PROPERTIES ON MECHANICAL

PROPERTIES

Soil mechanical properties often are divided Into compactIbIIIty,

deformablIity, and breakablIIty.
Kooien (1986) who uses a more fundamental approach distinguishes
between stable and unstable behaviour. The behaviour of a process

133
 Is called stable if the process tends to occur throughout the
entire soil. An example of stable behaviour Is the compaction of
loose soil. Unstable behaviour occurs when the process has a
tendency to concentrate on a spot. Unstable behaviour occurs on
the spots where soil breaking starts.
In practice physical soil degradation generally Is not only
associated with the effects of Increased bulk density but also
with those of large soil deformations that occur when the soil
volume remains almost constant. CompactIbIIIty and deformabIIIty
are therefore Important mechanical properties In relation to soil
physical degradation.
A simple test to measure compactibIIIty Is the uniaxial compres-
sion test. This test and the test facility parameters that in-
fluence this test have been discussed In section 5.3.
Soils used in uniaxial compression tests generally have been
sieved. Depending on the screens used the soil sample can have a
small or a wide aggregate diameter range. For soil samples this
may result In different bulk densities during filling and/or a
different number of contact points between aggregates. A differ-
ent number of contact points between aggregates may result In a
different compactIbIIIty. To Investigate a possible effect of
aggregate diameter on compactIbIIIty, uniaxial compression tests
were carried out for different aggregate diameter classes and
mixtures of aggregate diameter classes. Section 7.1.1 deals with
these tests.
In uniaxial compression tests on a soil with specific Intrinsic
properties (particle size distribution, type of minerals, organic
matter content, etc) the moisture content Is probably the major
strength-determining factor for compactIbIIIty. Kooien (1986)
distinguishes between dry compaction and wet compaction. After
dry compaction the initial structure Isat least locally pre-
served in the compacted soil. During wet compaction of soil, the
bulk density Increase is limited by partial water saturation. In
this final phase of wet compaction air may be entrapped. In this
entrapped air pressure may build up. When the soil Is unloaded
this air pressure may contribute to rebound. Uniaxial compression
tests have been carried out investigate the Influence of soil air
on rebound. These tests are discussed In 7.1.2.

7.1.1. THE INFLUENCE OF AGGREGATE DIAMETER ON UNIAXIAL COMPRES-

SIBILITY.

The uniaxial compression behaviour of soil samples with different

aggregate diameters has been examined for two soils as follows:
WagenIngen sIIty clay loam and Lexkesveer loam were passed
through a 10mm screen. After having been dried at room tempe-
rature to a moisture content below the hygroscopic point, each
soil was sieved into 3 aggregate diameter classes:
3.0 - 3.4 mm
4.0 - 4.8 mm
6.8 - 8.0 mm.
Aggregate portions of 1kg were placed on soil water suction bins
at a value of pF2.0. One day later suctions were decreased to

134
pF1.0, pF1.5, pF1.7, andpF2.0. After 5days theaggregates were
taken from thesuction bins andmeasurements of soil moisture
content were made. Bymixing soils from those different suction
bins a range ofmoisture contents was made. Five days were
aIlowed toequIIIbrate. CyIInders (50mm high and81mm in diam-
eter) were carefully filled with a spoon. Ontopof the soil
sample a 152.81 gram piston wasplaced. Uniaxial compressions
were carried outata piston speed of3mm/sec until a pressure
of 5 barwasreached. Pressure slnkage curves were recorded.
After testing thesamples were oven dried during five days (at
105°C)tomake soil moisture content determinations.
Both soils were tested inthemoisture content range from the
hygroscopic point topF2.0. Forboth soils thefollowing aggre-
gate size classes andmixtures were tested:
Class 3.0 -3.4mm In diameter
Class l 4.0 -4.8mm In diameter
Class II 6.8 -8.0mm in diameter
Mlxture 50% (byweight) 3.0-3.4mmand50%4.0-4.8mm
Mixture I 50% (byweight) 3.0-3.4mmand50%6.8-8.0mm
Mixture I I 33.3% (byweight) 3.0-3.4mm,33.3%4.0-4.8mm,and
33.3% 6.8-8.0mm.
Aggregates from thethree classes have anoutside area ratioof
1 :1.9 :5.4.Aggregate volume hasa ratio of 1 :2.6 : 12.4.
Before theuniaxial compression tests were carried out deter-
minations were made of pore space of aggregates from thediffer-
ent classes. Forboth soils andeach aggregate diameter class,
aggregate pore spacewasdetermined (10fold)onairdry aggre-
gates with the"petroleum method". Thepetroleum method Is de-
scribed InMcIntIreandStirk (1954). Table 7.1. shows the re-
sults of these determinations.

Table 7.1. Pore space of air dry aggregates from three diameter
classes of Wagenlngen and Lexkesveer soils.
soil diameter pore space standard
(mm) (%) deviation
WagenIngen 3 0-3 4 33.91 0.52
WagenIngen 4 0-4 8 33.91 0.38
Wageningen 6 8-8 0 33.91 0.34
Lexkesveer 3 0-3 4 34.56 0.39
Lexkesveer 4 0-4 8 34.65 0.49
Lexkesveer 6 8-8 0 34.78 0.67

We may conclude that there Isnodifference In aggregate pore

space between theaggregate size classes tested. Sothediffer-
ences found intheuniaxial compression tests are not due to
differences Inaggregate pore space.
From theuniaxial compression curves sample heights at <Tivalues
o f O , 0.5, 1.0, 2.5, and5.0barwere obtained. For theseo"i
values therelationships between void ratio (e)andsoil moisture
content will bediscussed below. Fig. 7.1and7.2show these

135
relationships for two classes and one mixture.
For both the soils void ratio during fM i l ng increases with
Increasing moisture content. This effect Is due to clod forming
as the cylinders are filled with a spoon. Int he moisture content
ranges tested, the void ratio Increased from 1.74 to 2.19 and
from 1.99 to 3.10 for Lexkesveer and WagenIng en soil respectIve-
ly. The void ratio Increase during filling wa s highest for Wage-
nIngen soil because this soil has a stronger tendency to form
clods. At the highest moisture content tested aggregate class I
showed a stronger cIoddIness resulting in higher void ratio
values. Compared with aggregate classes theml xtures showed lower

Lexkesveer
2.251
• = 3 . 0 - 3 .i. mm
+ = 6 . 8 - 8 . 0 mm
01
0=3.0-3.4/6.8-8.0 en =0
o
P 2.00
•\ +
+
+
+ o

1.75 \ *
00
o

1.50
r
8<x>
oo
0+ +
1.25 o o
o +„
\%ï +
$< +
%'
8 +
1.00 * fi" 'oo
"O + %• »
+ 0„
*+ 8+#+
M o»» o° <Xj =0.5bar
0.75 +
*»o*
+ o ** '<P Q] =1.0 bar
-t/°o+t fa &l =2.5bar
•yoo+ + *o <r^ =5.0bar
0.50

O.i.0

0 6 10 15 20 % 23
moisture content

Fig. 7.1. Uniaxial compression behaviour of two aggregate classes

and one mixture of Lexkesveer loam.

136
 Wageningen
• =3.0 - 3.4 mm
3.00 <J;=O
+ =6.8 - 8 . 0 mm
o =3.0-3.4/6.8-8.0
*0

2.75 +
+o
.• %

2.50 • oo
+
+

2.25- <o

2.00
O o

°°*
+
1.75- * '<s> +<è +
% +
j * 9> o OD* +
OD o ++ o +
+
° "o » .. * / +
1.50 • +
<$> • + °ooV
h O
O (DJ + 0
> £+
1.25- t o CTÏ=0.5bar
° t ++

»
o + o»,
1.00 o« * t <b CTj=1.0bar
* •+
•°oo«t
V
0.75 K+ • • 0 ° 3 * : 9 > q =2.5bar
*a>
** dfoOoW 1 * CT,=5.0bar

0.50

0i/- 12 15 20 25 % 28
moisture content

Flg. 7.2. Uniaxial compress Ion behaviour of two aggregate classes

and one mixture of Wageningen sI Ity clay loam.

137
void ratio values during filling. When the moisture content
Increases the difference disappeared gradually due to clod
forming. The lower void ratio of dry mixtures could have been
expected because It Is known that a denser packing can be
achieved with different diameters.

The compaction of a relatively dry sample starts as dry com-

paction but with Increasing pressure compaction It can change
Into wet compaction. The line that represents 10 volume percent
air content was chosen more or less arbitrarily as transition
between the dry and wet compaction areas.
In the wet compaction area there is no difference In behaviour
between aggregate classes and/or mixtures. In this area aggre-
gates have been damaged to such an extent that they almost disap-
pear.
In the dry compaction area It seems sometimes that a certain
aggregate class has more resistance to compaction than another
one. For example, in the moisture content range between 7 %and
15 % Lexkesveer class Ishows for the pressures applied a some-
what higher pore ratio level compared to class III. The values
for class II M e between those of the other classes. These small
differences, however, are not due to aggregate size but to dif-
ferences In packing during filling. Samples with denser packing
during filling have somewhat higher void ratio values during
compaction. When the moisture content Increases these differences
disappear both during filling and during compaction. The WagenIn-
gen soil class III has In the moisture content range between 12 %
and 20 % a somewhat denser packing during filling and therefore
somewhat higher void ratio values during compaction than the
other classes. Also these small differences disappear when the
moisture content Increases In this soil. At the lower moisture
contents the mixtures have lower void ratio values during com-
paction, but these differences also disappear when the moisture
content Increases.
From these tests we may conclude that aggregate size has no
Influence on uniaxial compactiblIIty. This can be explained by
the stable behaviour of the compaction process as described by
Kooien (1986). The weakest spot Iscompacted first and gets
stronger. Compaction stops at this spot and starts at the second
weakest spot. In this way compaction will occur throughout the
entIre soII.
For both soils the relationship between critical pressure and
moisture content has been calculated from the uniaxial compres-
sions (Fig. 7.3). Critical pressure Is the value of the applied
pressure at the moment that the critical pore space is reached.
Kuipers (1959) defines critical pore space as the pore space at
which there Isan air content of 10 %at the pF2 moisture con-
tent. Critical pore space is reached at void ratio values of 0.71
and 0.92 for Lexkesveer and WagenIngen soil respectively. The
difference between the two lines in Fig. 7.3 iscomparable to the
differences Inmoisture content at pF2. These comparable equal
shapes of the curves of these soils agree with the results of
uniaxial compression tests carried out by Kooien (1974) for the
same soils. At pF2 both soils have a critical pressure of 1.2
bar. For field traffic Fig. 7.3 means that the critical pore

138
space will be reached at the same tyre settlngs( Inflation pres-
sure, wheel load, speed, etc.) for both soils. Fig. 7.3 shows as
well that soil moisture content Is the most Important soil physi-
cal aspect of compactIblIIty.

PF2 Wageningen

c
"c
o PF2 Lexkesveer
u
Wageningen

Lexkesveer

2 3 U _5 bar 6
critical pressure

Fig. 7.3. Critical pressure-mo Isture content relatlonshlp for

"Lexkesveer" and "WagenIngen" In uniaxial compression tests.

7.1.2. THE INFLUENCE OF SOIL AIR ON REBOUND

Söhne (1952,1956) observed rebound (elastic recovery) In repeated

uniaxial compression tests of soil samples. Yong and Warkentln
(1966) show a model for soil behaviour during loading and unload-
ing (Fig. 7.4 and 7.5). In this dissertation the elastic recovery
during unloading (a in Fig. 7.4) will be called rebound and the
relaxation (b In Fig. 7.4) will be called creep.
A hysteresis effect during the unloading and rel loadIng of soiI
was found not only by Söhne but also by Yong and Warkentln.
Similar effects for metals have been presented by Nadal (1950).
Stone and Larson (1980) measured rebound on 60 gramme solI
samples with a diameter of 5.65 cm. A static load was applied
until sample volume and soil water suction were constant. This
took about half an hour. The total recovery was measured by
reducing the load and holding the new load until the sample
height did not change anymore. In these tests the total recovery
usually was less than 0.05 gr/cm 3 dry bulk density change. No
relationship between total recovery and moisture content was
found. The tests proved that the clay content was an Important
factor of total recovery. Because Stone and Larson's experiments
used relatively long lasting static loadings, there Is no evi-
dence of a possible influence of entrapped air on the recovery.

139
 Loading

Elastic recovery

Unload
c
'o
Reload'

Constant stressö*released here time log of pressure

Fig. 7.4. Course of strain In Fig. 7.5. Behavlour of soil

time of a volume element of soil at loading, un loading, and
at constant stress and after reload!ng.
re leas Ing stress.

pressure CHj

Fig. 7.6. Typical uniaxial pressures Inkage curve.

140
Neither Söhne's nor Yong and Warkentin's experiments mention the
possible influence of soil air on rebound.
To become better Informed about rebound and the influence of
soil air on this phenomenon, several series of uniaxial compres-
sion tests were carried out with three soil types. Rebound was
measured at different loading speeds, soil moisture contents, and
In soil samples of varying dimensions.
During uniaxial compression a pressure-sInkage curve was rec-
orded. An example of a pressure-sInkage curve Isgiven In Fig.
7.6. Distance a in the pressure-sInkage relationship Is rebound.
Distance b can be seen as creep during relaxation of the weak
spring that was placed between piston and pressure transducer.
To measure creep the soil sample was placed on a micrometer
stand.
Some samples were vacuum saturated and loaded to 0.5 bar with a
piston with a diameter that was much smaller than the sample
dIameter.
The soils tested were Lexkesveer loam, Almkerk sIIty clay, and
Boskoop peat. Lexkesveer and Almkerk were passed through a 3mm
screen; for the Boskoop soll a 5mm screen was used. A moisture
content range was made to correspond to the most common field
moisture contents of these soils.
Samples with diameters of 8 cm, 18cm, 30 cm, and 40 cm were
tested at a compression speed of 3mm/s. All samples had a diam-
eter to height ratio between 1.6 and 1.0. No significant differ-
ences In compaction and rebound were found for the different soil
sample dimensions. Therefore, in the other tests only sample
cylinders with a height of 50 mm and an inside diameter of 81 mm
were used.
PrecompressIng to 0.5 bar and subsequent reloading to 4.0 bar
resulted in rebounds almost equal to those occurIng when com-
pression was Increased to 4.0 bar without precompactIon.
Fig. 7.7 shows the rebound for the Lexkesveer and Boskoop soil at
two pressure levels and the different air contents of the soils.
Soil samples were loaded to 0.5 bar, unloaded, and reloaded to
4.0 bar. The compression speed was 0.25 mm/s. In both soils the
rebound Increased when the pressure was Increased. For Lexkesveer
soil the rebound Increased when the air content was decreased. At
a pressure of 0.5 bar this increase In rebound was less pro-
nounced than at 4.0 bar. When the air content exceeded 20 % the
Boskoop soil showed an Increase In rebound when the air content
decreased. Below an content of 20 %rebound decreased when the
air content decreased. The vacuum-saturated samples showed a
rebound level that was almost equal to the one of samples with a
high air content. Reloading to 4.0 bar was Impossible for the two
samples from both soils with the lowest air content at 0.5 bar.
In these samples water was squeezed out at a pressure below 1
bar. This resulted In such a fast pressure Increase In time that
loading could not be stopped at exactly 4.0 bar. For one peat
sample In Fig. 7.7. a negative air content was calculated. This
was due to water that was squeezed out temporarily during com-
pression and that was absorbed again by the soil sample during
creep.
No difference in behaviour was found between compression speeds
of 0.25 mm/s and 3mm/s for the Lexkesveer soil. At an air

141
content above about 5 %by vo Iume the Boskoop soll showed higher
rebound values after the faste r compression (Fig. 7.8). This Is
normal behaviour of fibrous ma terlals. At a low air content the
soil particles can move easily because of the high water content,
Fig. 7.9 shows the rebound- air content relationships for the
three soils tested. All sampI es were precompacted at 0.5 bar and
subsequently loaded to 4.0 bar . The compression speed was 3mm/s.
It seems that the Boskoop soll behaves very differently from the
Lexkesveer and Almkerk soil.
To understand this behavlou r which occurs when the soil air
content Is low and to find the influence of entrapped air on
rebound, rebound was expres sed as the difference In soil air
content between loading to 4 0 bar and after rebound. For the
three soils calculated reboun d (expressed in volume percent air)

Boskoopsoil
O loadedto0.5bar;ovacuumsaturated
andloadedto0.5bar
• loadedto0.5bar,unloaded,andreloadedto£..0bar

Lexkesveersoil
+ loadedto0.5bar;$vacuumsaturated
andloadedto0.5bar
A loadedto0.5bar,unloaded,andreloadedtoA.Obar

Fig. 7.7. Rebound-alr content relationshlp for "Boskoop" and

"Lexkesveer" soils at two stress levels.
Compress Ion speed «= 0.25 mm/s; cylinder diameter = 81 mm.

142
 0.05-

g/crn^

0.04-

)? 003

TD
C
O

ï 0.02

0.01

—1 1

16 24 32 £0 48
vol.% air

Flg. 7.8. Rebound-alr content relatlonshlp for "Boskoop" soil at

two compress Ion speeds. PrecompactIon stress = 0.5 bar; com-
paction stress = 4.0 bar-, cylinder diameter = 81 mm.

0.05T
g/cm3

0.04 Boskoop
<3

c 0.03
o
_o
(V
1_

0.02

Lexkesveer
0.01

10 20 30 vol.% air 40
after unloading
Flg. 7.9. Rebound after loading to 4.0 bar for three soils.
Compress Ion speed - 3 mm/s.

143
was plotted against measured air content after unloading (Fig.
7.10). The straight lines represent volume changes expected at
the Indicated pressure Increases of the entrapped air. An Inter-
section of a rebound curve with the line p-2 bar means that the
volume of air after unloading had to be compressed to a pressure
which is 2 bar higher during compression to account for the
volume decrease. We can see that for Almkerk and Lexkesveer
soils higher pressures In smaller quantities of entrapped air can
account for the Increased rebound. The Boskoop soll showed the
same effect of an Increased rebound and an Increasing pressure of
the entrapped air when the air content was higher than about 10
%. When the air content was lower rebound decreased. This de-
crease must be explained as follows: when air pressure Increases
soli water will be squeezed out more easily. For Boskoop soil
this means that a further increase In air pressure does not seem
to occur when the air pressure exceeds 2 bar because water Is
squeezed out. Fig. 7.10 seems to suggest that the pressure In-
crease within the entrapped air was higher than 4.0 bar. However,
this Is not possible because such pressures were not applied. At
4.0 bar loading water was squeezed out of the samples that had
been positioned on the left side of the 4 bar air pressure In-
crease line. A part of thiswater was squeezed out temporarily

p air =4bar
1
p air =2bar
p air =1bar
p air =0.5bar

p a i r =0.1bar

30 35 i.0 45
vol.% air after unloading

Fig. 7.10. Rebound (expressed as change In air content) after

loading to 4.0 bar for three soils. Samples are uni axially pre-
compacted at 0.5 bar and subsequently loaded to 4.0 bar. Compres-
sion speed - 3 mm/s.

144
during compression and absorbed again during rebound and creep.
Because of this sponge-Iike behaviour we can not make exact
calculations of maximum pressures built up within the entrapped
air.
We can conclude that a build up of air pressure can occur
which can influence rebound considerably when air Is entrapped
during the final part of a compression process.
Total recovery (rebound +creep) ismuch higher than rebound.
Fig. 7.11 shows rebound and total recovery of the Boskoop soil
after 1hour for different soil air contents.

0.10 n

g/cm J
0.09

0.08-
x \ x
x
*» 0.07i
<3 rebound+creep
•D after 1 hour
C
I 0.06i
01

0.05

0.04-1

0.03- rebound

0.02

0.01

16 24 32 40
vol.% air

Fig. 7.11. Rebound and total recovery after one hour for "Bos-
koop" soll. Samples are untax Ially precompacted at 0.5 bar and
subsequently loaded to 4.0 bar. Compression speed - 3 mm/s.

145
7.2. THE INFLUENCE OF MECHANICAL TREATMENT ON PHYSICAL PROPERTIES

Mechanical treatment of the soil by wheels results In compaction

and/or deformation. Because of this treatment micro-factors and
soil qualities can change. Changes on the micro level Influence
soil qualities such as pF-curves and conductivities. A review of
the Influence of mechanical treatment on micro-factors and soil
qualities Is given In 7.2.1.
In arable farming It Is Important to know the Influence of soli
compaction on subsequent tillage operations. The major soil mech-
anical quality that can change due to compaction is tillability.
To Investigate the Influence of soli compaction on tillability a
test Is needed that can characterize this influence. Tillability,
tillability tests, and attempts to use the unconflned compression
test to characterize tillability are discussed in 7.2.2.

7.2.1. THE INFLUENCE OF MECHANICAL TREATMENT ON MICRO-FACTORS AND

SOIL QUALITIES

On the macro level compaction reduces pore space. On the micro-

level this pore space reduction Is at the expense of the largest
pores (Sommer et al., 1972; Sommer et al.,1975; MaIdI and FIsch-
beck,1985). This effect is strongest during wet compaction. Table
7.2 shows the Influence of autumn ploughing on pore size dis-
tributions under dry and wet soil conditions.

Table 7.2. The Influence of ploughing condltIons on pore size

dlstrI but Ions (after Maldl and FIschbeck. 1985)
Depth top soil plough sole
(0 - 35 cm) (35 - 40 cm)
Ploughing Adry Awet Adry Awet

Vol.% pores>10 jj. 11 2 6 5 7 9 6 0

Vol. % pores 10 -0.2/* 14 0 13 0 11 3 11 8
Vol. %pores <0.2/U 17 5 21 3 22 . 2 22 2

Adry = autumn ploughed under dry soil conditions

Awet = autumn ploughed under wet soil conditions

Pores >10/U are Important for air conductivity. After wet plough-
ing the content of these pores was lower than after dry plough-
ing; in the top soil this amounted to a reduction to 58 % and in
the plough pan to 76 %of the values measured after dry plough-
ing. This reduced air conductivity.
The pores 10 - 0.2/u. determine the amount of water available for
plants. Compared to dry conditions the amount of available water
decreased In the top soil after ploughing under wet conditions.
Water In pores <0.2^< Is not available for plants. The content of
pores < 0.2/u. Increased strongly in the top soil after ploughing
under wet conditions as well.
A wet compacted soil has a higher content of unavailable water

146
and a lower content of pores for air conductivity. This means
that such a soII iswetter and colder than a dry compacted soII.
Maidl and FIschbeck (1985) also examined the influence of
these differences in pore size distributions, due to ploughing
under wet and dry conditions, on plant growth. Their investiga-
tion proved that the decrease of yield after wet compaction was
mainly due to an insufficient supply of nitrogen. The decrease of
N supply after wet compaction indicates a lower N mineralization
due to lower soil temperature, and higher gaseous N losses (denl-
trification) due to lower air conductivity.
Inter-partleIe bonds not arising from water suction can be
destroyed at large deformations. The strength of these bonds
influences the water suction -moisture content relationship.
Dawidowskl and Kooien (1987) applied triaxial tests to field
samples of WagenIngen silty clay loam. The sample volume was
almost constant during deformation. Samples were deformed to 50 %
of the original sample heights. Many bonds were destroyed by the
deforming and smearing action of the test, resulting In a strong
increase inwater suction. Compared with undeformed samples, the
samples tested showed higher values for volumetric shrinkage and
dry tensile strength.
Influences of compaction on pF-curves have been reported by
Warkentln (1971), Sommer et al. (1972), and Moreno et al. (1974).
Kooien (1978) found that the precompactIon moisture content
greatly Influences moisture contents at pF2.0 and pF2.7.

7.2.2. TILLABILITY

TlIIabiIity Isdefined as the degree of ease with which a soiI

may be manipulated for a specific purpose (ASAE Engeneerlng
Practice ASAE EP291.1, Hahn et al. 1984).
In agricultural field practice farmers decide on how to till from
experience. However, their experience does not suffice when they
need to choose new Implements and when they find themselves In
unknown situations. Also, field experience is not easy to convey.
To obtain an objective approach to tillability, a measurable par-
ameter of tillability, and therefore a tillability test Is
needed.

7.2.2.1. TILLABILITY TEST

To develop a tillability test It Is necessary to take the follow-

ing factors into account:
- the Initial soil condition. This condition depends on soil
physical properties and the history of the soil.
- the type of tillage operation. Primary tillage and field traf-
fic have other demands than secondary tillage operations.
In this dissertation we concentrate tillability on seed-bed
preparation in spring.
- the processes that occur. In tillage operations we should
distinguish between Intended processes and accidental pro-

147
 cesses. In seed-bed preparation the major Intended process is
"crumbling". This process Isaccompanied by the accidental
process of "compaction". Tillability tests should concentrate
on the major intended process, but must also take into account
the major accidental process.
the energy needed for the tillage operatI on. Energy needed for
tillage is associated with energy needed for tillage processes
and for locomotion of the tractor on the field. Gross energy
consumption Is the amount of energy needed for both tillage
processes and tractor locomotion. Net energy consumption only
deals with the energy needed for the tillage process (for
Instance draft).
In the Netherlands soil conditions generally are too wet In the
period before seed-bed preparation. Therefore, tillabtIty re-
search has concentrated on the upper tillage limits.
It isadvisable to increase the ratio net energy consumption to
gross energy consumption. Ultimately tillability depends not
only on the results of the tillage operations but also on the
energy required. An optimum ratio of tillage results to energy
consumption Is pursued.
Energy Is not yet a limiting factor. Therefore, the Interest in
tiIIabIIty nowadays is limited to a tillabiIty test that indi-
cates whether or not the soil can be tilled inorder to achieve
the Intended results. In times of energy scarcity the tiIlabII-
Ity test should also pay much attention to the energy aspects.
the Intended tillage results. The Dutch farmer Is generally
mostly Interested In the results of the tillage operations.
Therefore criteria for the results of tillage operations are
required Inorder to determine tillability. Tillage criteria
have been published In "Tillage Advice" for cereals (Andrlnga
et al, 1979), sugar beets (Perdok et al, 1974a), and potatoes
(van Glist et al, 1975).
the tillage Implement to be used. The tillage implement should
also be considered because of the energy consumption and the
kind of the work expected. The capacity of the tillage imple-
ment can also play an Important part in farm management decI-
sions.

7.2.2.2. TILLABILITY TESTS BASED ON PROCESSES OTHER THAN

CRUMBLING

Boekel (1977) tried to relate soil tillability to deformabiIIty

measured with a plasticity meter. The pressure that is needed to
force the soil through a small nozzle Is measured with this
Instrument. The required pressure is lower for a wet plastic soil
than for a dry crumbly soil. It is proposed that this Is a
measure for crumbling. This test is not suitable for field
samples and does not account for the effect of pore space on
11IlablIity.
Havlnga and Perdok (1969) developed a tillablIIty test that
relates tlIlablIIty to mechanical properties of solis. The com-
presslblllty of loose soil ischosen as major process, The ap-
piled pressure Isameasure for the energy consumption. In thls

148
test air permeability ismeasured as a function of soil moisture
content and applied pressure. Perdok et al. (1974b) discovered
that 20 fields could be tilled for potato seed-bed preparation at
a working depth of 8 cm, when their soils still had an air
permeability of at least 100 c m 2 x l 0 - 1 0 after having been loaded
to 4 bar. Aggregates ranging from 2.8 - 4.0 mm were used for
these tests . This means that the soil structure was not taken
Into account. Because this test uses compressibility as a major
process It can also be used to predict field traffIccabiIIty
(Perdok, 1976).
Koenigs (1976) determined the upper tillage limit with a micro
tillage test. To carry out this test a small saucer Is filled
with a thin layer of soil aggregates (2-4 mm fraction), with a
hammer a sliding action (micro tillage) isapplied to the soli in
the saucer. After the micro tlIlage the pulIthat is needed for
moving a spatula through the soil Ismeasured. In a range with
increasing moisture contents the draft for the spatula abruptly
increased from a certain moisture content on. Koenigs considers
this moisture content minus one percent as the upper tillage
limit for spring seed-bed preparation.
After some preliminary tests Wageningen sllty clay loam was uni-
axialIy compressed to three pore space levels over a wide range
of moisture contents. Having been dried these samples were

1000-
N/mm

750
P=35%
o
•D P=4 0 %
O
Q.
U)
P=4 5 %
4.
O
«•— 500
II
o
i_
•o

250-

-
0-ty«^—r^x*~*t—
% 30
moisture content
Fig. 7.12. Behavlour of precompacted "WagenIngen" silty clay loam
at three pore space levels In a micro tillage test.

149
crushed and sieved. For each sample the 3.4-4.8mm fraction was
rewetted to the moisture content before compression. After one
day of equIIIbratlon these aggregates were used in a micro tlII-
age test. The relationships between spatula draft and moisture
content are shown In Fig. 7.12. For the three pore space levels
tested the draft for the spatula suddenly Increased. It is re-
markable that for each Initial pore space level this sudden
increase occurs at the same moisture content. This means that
this upper tillage limit is independent of pore space at precom-
pactIon.
Neither does this test use field samples. The micro tlIlage with
the hammer and the draft measurement with the spatula depend on
the person who does the test. Some other drawbacks of this test
are that soil sometimes sticks to the hammer and that soil may
curl up when wet cohesive.

7.2.2.3. TILLABILITY TESTS BASED ON CRUMBLING PROCESSES

These tests try to determine the intensity of crumbling. Further-

more, the energy needed for this crumbling can be measured. In
field practice the most interesting aspects of spring seed-bed
preparation are Indeed crumbling and the energy needed for the
tillage operation. Therefore, it Isattractive to use these
methods to characterize field processes.
Kooien (1977) used a pendulum-type impact machine to measure
tillage process aspects on soil blocks in the laboratory. For
each test he measured the volume and the mean aggregate size of
the loosened soil, and the energy consumption. Although his test
showed good results for soil blocks prepared in the laboratory it
isdifficult to use It for undisturbed field samples.
For our examination the unconfined compression test has been
chosen to characterize the crumbling process. in an unconfined
compression test a soil sample ismade to fail between two plates
(Fig. 7.13). At the moment of failure, the sample shortening
expressed as a fraction of Initial height (£f )and the prevai-
ling failure stress ( C T m a x ) are measured. The failure strain £f
represents the soli deformation that occurs before failure. This
means that £f gives an indication of breakabillty and therefore
of crumbling as well. The unconfined compressive strength (0"max)
Isa measure of the energy consumption.
Kooien (1973 and 1978) successfully used the unconfined compres-
sion test to characterize soil mechanical properties.
When laboratory samples are used the Influence of bulk density on
the crumbling process can be examined by adjusting different
levels of pore space. The unconfined compression test can also be
applied to field samples.

150
Flg. 7.13. Unconflned compression test.

TESTS OF LABORATORY SAMPLES

At two pore space levels unconflned compression tests were

carried out notonly todetermine therelationship between Sf and
moisture content, but alsotheonebetween<Tmax and moisture
content. A jawcrusherwas used todeterminecrumbling at these
pore spacesandmoisture contents. Inboth experimentsWagenIngen
siItyclay loamand Lexkesveer loamwere used. ThesolImoisture
content rangewaschosen betweenmoisture content atpF2andhalf
of thismoisturecontent.Twoporespace levelsweremadecompar-
ableto field porespace levelsthatcan be found inspring.
SoiIsampleswere prepared as follows: thedry soiIswere passed
through a 3mm screen and subsequently wetted to the desired
moisture content ranges. After equilibration thesoil was com-
pressed in cylinders withan insidediameter of 50 mm and a
height of 100mm. The Insidewallsof thecylinderweregreased
with siIIcongrease. Thecompressionsweremade in layersof 20
mm. Thediameter toheight ratioof thesampleswas 0.5 inorder
to minimize the influenceof friction between platesand sample
during unconfined compression (Kooien, 1978). Beforeunconflned
compressionsweremade thesamples had twodays toequilibrate.
From theunconflned compression tests, made atapiston speedof
10mm/min, £ f andc m a x were determined. After thetests thesoil
moisture contentwasmeasured.
The tests with the Jawcrusher weremadeas follows: the two
soilsand thesoil preparationmethod usedwere thesameasthose
used for theunconflned compression tests.Thecylindersused had
an insidediameter and aheight of 50mm. Having beencompressed
to the desired porespace thesampleshad two days to equi-
librate. The jawcrusher (Fig. 7.14) hasabinwith onemovable
walI (d). Ahandle isused tooseIIlatethemovable waII. The

151
 200-/.00 0-105

Fig. 7.14. Jaw crusher.

outlet (f) Isadjustable between O and 200 mm. The oscillating

movement of the movable wall can be adjusted by the amplitude of
the oseiIIatIon and by the angleotbetween the movable walI and
the opposite wall. After three test series the following settings
were used:
- a width of the outlet of 11.5 mm
- an angle oc ranging from 21 to 59 degrees for the
Lexkesveer soiI
- during the first five oscillations of the WagenIngen
solIthe angle ocwas 32 to 41 degrees (to avoid too
high pressures on the soli samples); the next oscil-
lations were made at an angle ranging from 21 to 59
degrees.
The aggregates were sieved directly after thhe tests.
From the unconfined compression tests curved relationships
have been found between £ f and moisture content (Fig. 7.15 and
7.16). At increasing moisture content soil deformation before
breaking increases. The lowest pore space level shows higher f f
values for both soils. This isdue to the lower amount of weak
zones in the most compacted soiI.
FaIlure stress 'max shows a IInear decrease with increasing
moisture content. So under wet conditions these soIIs need less
energy to break than under dry conditions.
In the moisture content ranges tested, the Wageningen soil showed
higher levels of £f and0"max compared to Lexkesveer soli So the
Wageningen soil is harder to break and more energy is needed for
this breaking. This agrees with the practical experience that
this heavier soil Is less easy to till.

152
 17 18 19 20 21 22 23 2U 25 % 26
moisture content

Flg. 7.15. UnconfI ned compression behaviour of "Wageningen" sllty

clay loam.

153
 The measuremen ts with the Jaw c rusher showed a sudden decrease
of the amount o f soIIreleased f rom the crusher outlet at In-
creasing moisture content (see Fig 7.17 and 7.18). At the same
moisture content the Mass Median Diameter (MMD) of the soil that
passed through th e Jaw crusher str ongIy Increased. The moisture
content at which these sudden cha nges occured can be considered
the upper tiIlage IImlt. For Wagen Ingen sIIty clay loam the upper
11Ilage limit II es for both pore spaces at amoisture content of
about 21 %. This agrees with the tests caried out with the micro
tillage test acco rdIng to Koenigs (Fig. 7.12). The upper tillage
limit for Lexkesv eer Iles between a moisture content of 18 % and
19 %. Because of the low number of measurements made at these
moisture content s Itwas not pos sible to make a more precise
estImatIon.

x
•
E
in
ui
a>
i_
In
0»

'5 -oo c P - p = 4 9 8 2o/o

0.06
Lexkesveer F=45.12%
to
. 0.05
c
'o
Tn P=49.62%
* 0.04
_2
'5
0.03-

0.02

0.01?—•

oV- 11 12 13 14 15 16 17 18 19 %20
moisture content

Fig. 7.16. UnconfI ned compress Ion behaviour of "Lexkesveer" loam.

154
 From the tests applied to laboratory samples of WagenIngen
sIIty clay loam and Lexkesveer loam w e can c o n c l u d e that:
- pore space Influences crumbling In a solI-breakIng p r o -
c e s s . A loose soil crumbles m o r e strongly than a dense soil.
- an unconflned compression test can be used to Investigate the
Influence of soil compaction on breakabIIIty.
- failure strain £f and failure s t r e s s < T m a x do not give an Indi-
cation of the upper tillage limit. €f gives an indication of
the breaking sensitivity of a soil and < T m - v Indicates the
max
amount of energy needed for breaking.
- i n a soiI-breakIng process crumbling decreases at Increasing
m o i s t u r e content.
the m o i s t u r e content at the upper tillage limit is almost equal
for the pore spaces tested, although the results of the crum-
bllng tillage operation differ at these pore s p a c e s .

100
%

,_ 80
JZ

60 Wageningen
o
-C i.0
o • = P =46.13%
Ol mm
40
O • x=P=50.99%
•30
-o o
O)
U) 20-
U)
o •20
Q.
n-r

J&ï" 10

0 16 18 20 22 24 26 %
moisture content

Fig. 7.17. Behavlour of "WagenIngen" soil at two pore space le-

vels In a Jaw crusher.

155
 100 Mt<*«0<-X4a»-X—X—«•••(-X
%
<u
in
3 80-
Lexkesveer
S
sz o • =P=45.88%
60 30
3 o x =P=50.78%
O mm

T3
(U
i.0- 20
UI
UI o
O
Q.
S 20 10
**--&;:=&!&&&=&
^i-ä^D-D-°°
o-l-A-
10 12 U 16 18 20 %
moisture content

F/g. 7.18. Behavlour of 'Lexkesveer" soll at two pore space le-

vel s In a Jaw crusher.

TESTS OF FIELD SAMPLES

To Include soil field structure In tillability tests preliminary

Investigations were made on the possIbI111es to use field samples
for the unconfined compression test.
The first three series of 80 samples each were used to determine:
- a procedure for taking the field samples
- a procedure for the unconfined compression test with field
samples.
The second six series were used to investigate the reaction of
the unconfined compression values to changes In weather condi-
tions. The field samples were taken during the months of March
and April 1980 on an autumn-ploughed field. The soil can be
classified between the WagenIngen and the Lexkesveer soil.
The cylinders used have an Inside diameter and a height of 50 mm.
Before samples were taken the cylinder Inside walls were oiled.
For each series samples were taken from 8 places in the field
from the furrow slice and from the Junction area between furrow
slices. The soil for seed-bed preparation comes for the greater
part from the furrow slices. In the first two series half of the
samples were taken from the Junction area to get an Idea of the
differences between furrow slices and the Junction area between
the furrow slices. All the other samples were taken from the
furrow sIices.
For sampling the following rules and procedures were used:
- cracks In the soli were avoided.
- cylinders were not allowed to be twisted during sampling,
because twisting can create breaking surfaces resulting In

156
 smaller £f values. When the cylinder was pushed far enough Into
the soll Itwas lifted with a spade, and trimmed with a knife.
- on sampling the soil was compressed. Therefore only samples
with a decrease In height of less than 0.5 cm were accepted.
- the thin dry top layer of about 1cm In thickness was removed
before samples were taken.

The mean moisture content in furrow slices Is higher than In the

Junction area. This difference Is Just below significance for the
first series and significant (95 % probability) for the second
series. The mean difference has a value of 1- 1.5 percent by
weight. The average pore space was about 5 percent higher In the
Junction area. This could have been expected because this soil
was loosened from the furrow slices by weathering.
To obtain a height to diameter ratio of 2 In the unconflned
compression test two samples were been put on top of each other
according to Kooien (1978). Before testing the samples were
weighed. The sample pairs were composed of two samples with
almost equal weight. This was done to avoid that samples with
very different pore space were used as a sample pair In the
unconflned compression test.
In the field the moisture content Increases with depth. There-
fore, soil samples topsldes are drier and more crumbly than
bottomsIdes. So there are three ways to combine the sample pairs:
- loose/loose: the upside of the first sample against the upside
of the second sample.
- loose/firm: the upside of the first sample against the under-
side of the second sample.
- firm/firm: the underside of the first sample against the under-
side of the second sample.
In the sample height values which exist at the beginning of the
unconflned compression tests, the decrease in sample height which
occurs when the samples are pushed out of the cylinders is ac-
counted for.
In the different combinations no significant differences were
found in moisture content and pore space. So the differences
found as a result of unconflned compression are due to the method
used. For each series the loose/loose method resulted In the
lowest values for _mean failure strain £f and mean unconflned
compressive strengthO m a x . The firm/firm method showed the highest
values while the loose/firm method resulted in Intermediate
values. The differences between loose/loose and firm/firm are
significant (95 % probability) but no significant differences
were found between loose/loose and loose/firm and between
loose/firm and firm/firm. Compared with the firm/firm method the
standard deviation of Zf Is a factor 2.5 lower In the
loose/loose method. For <r max standard deviations are almost equal
for loose/loose and firm/firm. It is remarkable that the standard
deviation values for Zf are almost constant for all the series
within a soil. Standard deviation Increases for Sr m a x when the
soil becomes drier. The standard deviation of if was almost equal
for a series of field samples (loose/loosemethod) and a series
of laboratory samples. Forön,^ the laboratory samples had a
factor 3 lower standard deviation than the field samples
(loose/loose). Fig. 7.19 shows thecrmav -moisture content rela-

157
tlonshIp for a loose/loose s e r i e s . This c l e a r l y shows the d i f f l -
cuItlest hat a r i s e w h e n field samples a r e used Instead of labora-
tory samp Ies (Fig. 7.15 and 7 . 1 6 ) .
The crack formation differed for the different c o m b i n a t i o n s . In
the loos e / l o o s e method crack formation a l w a y s occurred In the
mIddIeof the sample p a i r s . This could have been expected because
this part Is driest and h a sm o r e weak z o n e s . C r a c k formation for
the fIrm /firm m e t h o d a l w a y s started on topor at the bottom of
the samp e p a i r s . Inthe loose/firm m e t h o d crack formation could
start e v e rywhere.
The if v a l u e s for furrow s l i c e s and the Junction area a r e not
sIgnIflea ntly different for the three m e t h o d s . T h e unconflned
c o m p r e s s ve stength v a l u e s a r em u c h lower for the Junction area.
This c o u Id be expected because this w e a t h e r e d soil is easy to
crumble. Further sampling c o n c e n t r a t e d on furrow s l i c e s because
of ther eported d i f f e r e n c e s and the k n o w l e d g e that t h e soil for
the seed bed c o m e s m o s t l y from the furrow s l i c e s .

Six s e r i e s of forty samp les each w e r e taken from the furrow

s l i c e s on six d i f f e r e n t d a t e s . T h ec o u r s e of t h e unconflned
c o m p r e s s i o n v a l u e s (loose /loose m e t h o d ) Is compared with the
c l i m a t e diagram. The c l l m a te diagram (Fig. 7 . 2 0 )shows that 1980
had a w e t s p r i n g . There w a s a dry period between 10and 18 April.
In this period a seed-bed could dry enough in order to be p r e -
pared for p o t a t o e s . Fig. 7 20 shows thec o u r s e of ff and0 " m a x In
time a s w e l l . W e can s e et hat the unconflned c o m p r e s s i o n v a l u e s
tend to follow the drying of the field top layer. Even when the
c o n d i t i o n s a r e known at whI eh this soil r e a c h e s the upper tillage
limit, w e can draw no c o n eI u s i o n s about t h ed a t e in the spring of
1980 on w h i c h a seed-bed f or p o t a t o e s could h a v e been p r e p a r e d .

1.00 -i
xbar
o
b 0.80
in
IA
a>
i-
"
«0.60
_2
a
0.40 i
-y=1.89-0.05x
0.20 (r=-0.82)

0V
0 20 22 24 26 28 30 %32
moisturecontent

Fig. 7.19. UnconfIned compressive strength of field samples.

158
First, the following questions are to be answered:
- are the sampling depth and the thickness of the top layer that
is needed for the seed-bed equal?
- has the dry layer that was removed been taken Into account?
- have cracks been avoided?

The thickness of the top layer that Isused to prepare a seed-bed

has two components: the layer that Is used for surface levelling
and the thickness of the loose layer that Is needed for the seed-
bed. The thickness of the levelling layer can be calculated using
a rellefmeter (Kuipers, 19S7). For a level field a loose top
layer of 8 cm Is needed for potatoes (Kouwenhoven, 1974). For our
test field the necessary sampling depth Isestimated at 10.8 cm
(2.8 cm of levelling layer and 8 cm for the seed-bed). Our sam-
pling depth was 9 cm: 1.5cm for the removed top layer, 2 cm for

|CT 0.080

ai 0.060
(_ -•„
_2 •»»

'5
H—
c
o
<u 0.040-
lbC
E « 0.50
<u bar

U 18
March

E3 = rain
— =potential evaporation

Fig. 7.20. Course of climate and the unconfIned compress Ion cha-
racteristics of field samples.

159
trimming, 0.5 cm compression during sampling, and 5 cm cylinder
height. This means that the height and diameter of our cylinders
were 1.8cm too smaII.

From the tests with field samples we can conclude that:

- field samples can be used for unconflned compressions tests.
- for high accuracies a standard sampling method Is necessary.
- sampling depth depends on roughness of the field, the needed
thickness of the seed-bed, the compressibility of the soil, and
the thickness of the removed dry top layer.
- the highest accuracy in unconflned compression tests Is achiev-
ed when two samples are combined with the upsides against each
other (loose/loose method).
- the course offf and (f max tend to follow the drying of a field
solI.

160
 CHAPTER 8

SUMMARY

In soil dynamics we distinguish between loosening and load-

bearing processes. This dissertation deals with load-bearing
processes under agricultural rollers, wheels, and tyres.
Agricultural rollers, wheels, and tyres have been classified. In
addition some general aspects of these devices are presented.
This report presents the fundamentals of load-bearing processes,
I.e. kinematic, dynamic, and soil physical aspects. Kinematic and
dynamic aspects are discussed on three levels: device (roller,
wheel, or tyre), contact area, and soil under the device. Physi-
cal properties of a soil can change due to load-bearing pro-
cesses. On the other hand the soil physical properties influence
the mechanical behaviour of a soil. In the section on soil physi-
cal aspects we have paid special attention to the influence of
aggregate diameter and soil air on compression behaviour of soil
and to 11IlabiIIty.
Soil characteristics concerning load-bearing processes are pre-
sented. Different tests that were used to characterize soli
mechanical behaviour are compared. Relationships between soil
characteristics and process aspects and their suitability to
predict process aspects are discussed. Contributions are made to
the prediction of some process aspects of a towed tyre under
different soil conditions and in different soli types. Under
laboratory conditions the use of characterizing processes (cone,
vane, and falling weight) and empirical prediction methods re-
sulted In accurate predictions of rolling resistance, rut depth,
and compaction caused by a towed tyre.

161
 CHAPTER 9

SAMENVATTING

In de gronddynamlca onderscheidt men losmakende en afsteunende

processen. BIJ het rijden over grond treden afsteunende processen
op onder rollen, wielen en banden. Dit proefschrift behandelt
afsteunende processen onder rollen, wielen en banden in de land-
bouw.
In deze dissertatie worden een indeling en enkele algemene aspec-
ten gepresenteerd van In de landbouw voorkomende rollen, wielen
en banden.
De grondslagen van afsteunende processen worden behandeld: kine-
matische, dynamische en bodemfysische aspecten. De bespreking van
kinematische en dynamische aspecten vindt plaats op drie ni-
veau's:werkend deel (rol, wiel of band), contactvlak en de grond
onder het werkend deel.
Bodemfyslche eigenschappen kunnen veranderen als gevolg van af-
steunende processen. Anderzijds hebben bodemfysische eigenschap-
pen invloed op het mechanische gedrag van grond. In het hoofdstuk
over bodemfysische aspecten krijgen met name aandacht: bewerk-
baarheid van grond en de Invloed van aggregaat diameter en bodem-
lucht op het verdichtingsgedrag van grond.
Bodemkarakteristieken, die verband houden met afsteunende proces-
sen worden gepresenteerd. Verschillende voor het karakteriseren
van het mechanisch gedrag van grond gebruikte testen worden
vergeleken. De betrekkingen tussen bodemkarakteristieken en pro-
cesaspecten worden behandeld. Evenals de mogelijkheid om deze
betrekkingen te gebruiken In voorspellingsmethoden. Er worden
nieuwe bijdragen geleverd aan de voorspelling van enkele proces-
aspecten van een getrokken band. Nauwkeurige voorspellingen van
rolweerstand, InsporIng en verdichting waren mogelijk met behulp
van karakteriserende processen (conus, vane en valgewicht) en
empIrIsehe voorspeIIIngsmethoden.

162
 REFERENCES

Aboaba, F.O., 1969. Effects of time on compaction of soils by

rollers. Trans. ASAE 12(1):302.
Amir, I., G.S.V. Raghavan, E.Mckyes and R.S. Broughton, 1976.
Soil compaction as a function of contact pressure and soil
moisture content. Canadian Agrlc. Eng. 18(1):54-57.
Andrlnga, J.T. et al., 1979.GrondbewerkIngsadvIes voor granen.
BedrIjfsontwlkkelIng 10(5):495-501.
Anslow, B.J. and R.J. Warrllow, 1962. Tractor tyres. I.Mech. E.
Referred In Inns and Kllgour (1978).
Baganz, K. und L. Kunath, 1963. Einige Spannungs- und Verdich-
tungsmessungen unter SchIepperIaufwerken.
Agrartechnik 13:180-182.
Baganz, K., 1963/1964. Spannungs- und Verdichtungsmessungen Im
boden bel verschiedenen Fahrgeschwindigkeiten.
Archiv für Landtechnik 4(1):35-45.
Bailey, A.C. and G.E. VandenBerg, 1968. Yielding by compaction
and shear In unsaturated soil. Trans. ASAE 11:307-311,317.
Bailey, A.C., E.C. Burt and R.L. Schäfer, 1974.
Digital calculation of rolling radius for wheels and tracks.
Trans. ASAE 17(5):871-873.
Bailey, A.C. and E.C. Burt, 1981. Performance of tandem, dual and
single tires. Trans. ASAE 24(5):1103-1107.
Bekker, M.G., 1956. Theory of land locomotion.
University of Michigan Press, Ann Arbor.
Bekker, M.G., 1960.Off-the-road locomotion.
University of Michigan Press, Ann Arbor.
Bekker, M.G., 1969. Introduction to terrain-vehicle systems.
University of Michigan Press, Ann Arbor.
VandenBerg, G.E. and W.R. Gill, 1962. Pressure distribution be-
tween a smooth tire and the soli. Trans. ASAE 5:105-107.
Bernackl, H., J. Haman and Cz. Kanafojsk, 1972. Agricultural
machines, theory and construction. Vol. I, Warsaw.
Bernstein, R., 1913. Probleme zur experimentellen Motorpflug-
mechanik. Der Motorwagen, Vol. 16.
B N 1er, R.H. and H. Steinkampf, 1978. Elnfluss der ProfIIrIchtung
von AS-Relfen auf die Zugkraftübertragung des Schleppers.
Landtechnik 33:57-60.
Bil1er, R.H., 1983. RichtIgeMontage von Fronttriebreifen hilft
sparen, ohne Zugkraft einzubüssen. dlz 34(11):1444-1446.
Blake, G.R., W.W. Nelson and R.R. AIImaras, 1976. Persistance of
subsoIIcompactIon In aMoIIIsoI.
Soil ScI.Soc. Am. J. 40(6):943-948.

163
Bock, G., 1952. Feldversuche über die Zugfählgkelt von Acker-
schlepperreifen. Grundl. Landtechnik 3:88-100.
Bock, G., 1953. Beobachtungen bei Feldversuchen über die Zug-
fählgkelt von Ackerschleppern. Grundl. Landtechnik 5:42-48.
Bodman, G.B. and G.K. Constantin, 1965. The Influence of particle
size distribution In soil compaction.
Hllgardia 36(15):567-591.
Boekel, P., 1977. De werkbaarheid van de grond In het voorjaar In
verband met de ontwatering en de samenstelling van de grond.
Symposium werkbaarheid en bedrijf, WagenIngen.
Boels, D., 1978. Kanttekeningen bij bestekvoorwaarden ter beper-
king van bandspanning van dumpers en trekkers.
Nota 1081, ICW-Wageningen.
Boll, E. und E. Isensée, 1987. Eignung von Geschwindigkeits-
gebern. Landtechnik 42(2):57-59.
Boiling, I. und W. Söhne, 1982. Der Bodendruck schwerer
Ackerschlepper und Fahrzeuge. Landtechnik 37(2):54, 56-57.
Brlxlus, W.W. and F.M. Zoz, 1976. Tires and tracks In agricul-
ture. Am. Soc. Automotive Eng., Paper 760653, Milwaukee.
Buckingham, F., 1976. Fundamentels of machine operations.
Tillage. Deere and Company, MolIne.
Cegnar, A. and F. Fausti, 1961.Movements under the contact area
of radial and conventional tires. Trans. ASAE 4:224-225.
Chancellor, W.J., 1976. Compaction of soil. Bull. 1881, Dlv.
Agrlc. Sei. University of California, Davis.
Chessness, J.L., E.E. Ruiz, and C. Cobb, 1972.
Quantitative description of soil compaction in peach orchards
using a portable penetrometer.
Trans. ASAE 15(2):217-219.
Clark, S.K. (Ed.), 1971a. Mechanics of pneumatic tires. National
Bureau of Standards Monograph 122. U.S. Government Prltlng
Office, Washington USA.
Clark, S.K., 1971b. The contact between tire and roadway.
Chapter 6 inMechanics of pneumatic tires. National Bureau
of standards Monograph 122. U.S. Government Prltlng Office,
Washington :445-500.
Corcoran, P.T., 1979. Analysis of tractive performance. ASAE-
paper, No. 79-1550, New Orleans, L.A.
Danfors, B., 1974. Packning IaIven. SpeciaImeddelande S24,
Jordbruksteknlska Institutet, Ultuna, Uppsala, Sweden.
Danfors, B., 1980. Dack for traktorer och redskap.
Meddelande 386, JordbruksteknIska InstItutet,UppsaIa, Sweden.
Davies, D.B., D.J. Eagle and J.B. Finney, 1972. Soli management.
Farming Press, Ipswich.
Dawidowskl, J.B. and A.J. Kooien, 1987.
Changes of soli suction, conductivity, and dry strength during
deformation of wet undisturbed samples.
Soil and Tillage Research 9(2):169-180.
Domler, K.W. and A.E.Willams, 1978. Tractive efficiency -
maximum or optimum? Trans. ASAE 21(4):650-653.
Dwyer, M.J., D.R. Comely and D.W. Evernden, 1974a. Handbook of
Agricultural tyre performance. NIAE, Sllsoe.
Dwyer, M.J., D.R. Comely and D.W. Evernden, 1974b. The field
perfomance of some tractor tyres related to soil mechanical
properties. Journ. Agr. Eng. Res. 19(1):35-50.

164
Dwyer, M.J., D.W. Evernden and M. McAllister, 1975. The develop-
ment of a handbook of agricultural tyre performance.
Proc. 5th Int. Conf. ISTVS, Detroit, USA.
Dwyer, M.J. and G. Pearson, 1976. A field comparison of the
tractive performance of two- and four-wheel drive tractors.
Journ. Agr. Eng. Res. 21:77-85.
Dwyer, M.J., M. McAllister and D.W. Evernden, 1977. Comparison of
the tractive performance of a tractor driving wheel during Its
first and second passes In the same track.
J. Terramech. 14(1):1-10.
Dwyer, M.J., 1978.Maximising Agricultural Tractor Performance by
Matching Weight, Tyre Size and Speed to the Power Available.
Proc. 6th Int. Conf. ISTVS, Vienna, Vol. 1:479-499.
Eldlk Thleme, H.C.A. van, and H.B. Pacejka, 1971. The tire as a
vehicle component InClark (Ed):Mechanics of Pneumatic Tires.
National Bureau of Standard Monograph 122,WashIngton,545-839.
Estler, M., H. Knlttel und E. Zeltner, 1984. Bodenbearbeitung
aktuel. DLG-Verlag Frankfurt (Main).
Faure, A., 1980/1981. A new conception of the plastic and liquid
limits of clay. Soil and Tillage Research 1:97-105.
Forrest, P.J., I.F. Reed and G.V. ConstantakIs, 1962. Tractive
characteristics of radial-ply tires. Trans. ASAE 5:108,115.
Freitag, D.R, 1965. A dimensional analysis of the performance of
pneumatic tyres on soft soils. Tech. Rep. 3-688. U.S. Army
Engineers Waterways Experiment Station, Vlcksburg.
Freitag, D.R., 1979. History of wheels for off-road transport.
J. Terramech. 16(2):49~68.
Fröhlich, O.K., 1934. Druckverteilung Im Baugrunde.
Springer-Verlag, Wien.
Gee-Clough, D., 1976. The Bekker theory of rolling resistance
amended to take account of skid and deep slnkage.
J. Terramech. 13(2)-.87-105.
Gee-Clough, D.,M. McAllister and D.W. Evernden, 1977a. Tractive
performance of tractor drive tyres. I. The effect of lug
height. Journ. Agr. Eng. Res. 22(4):373-384.
Gee-Clough, D.,M. McAllister and D.W. Evernden, 1977b. Tractive
performance of tractor drive tyres. II. A comparison of radial
and cross-ply carcass construction.
Journ. Agr. Eng. Res. 22(4):385-395.
Gee-Clough, D.,M. McAllister and D.W. Evernden, 1977c. Tractive
performance of tractor drive tyres. III. Running In the furrow
bottom. Journ. Agr. Eng. Res. 22.
Gee-Clough, D. et al., 1978. The empirical prediction of tractor-
implement field performance. J. Terramech. 15(2):81-94.
Gee-Clough, D., 1979. The effect of wheel width on the rolling
resistance of rigid wheels In sand.
J. Terramech. 15(4):161-194.
Gee-Clough, D., G. Pearson and M. McAllister, 1982. Ballasting
wheeled tractors to achieve maximum power output In
frictlonaI-cohesIve soils. Journ. Agr. Eng. Res. 27(1):1-19.
Gill, W.R. and G.E. VandenBerg, 1967. Soli dynamics In tillage
and traction. Agricultural handbook 316.
U.S. Dep. of Agr., Washington.
Gill, W.R., 1968. Influence of Compaction Hardening of soil on
penetration resistance. Trans. ASAE 11:741-745.

165
Gl1st, K.B. van, J.K. Kouwenhoven en L.M. Lumkes, 1975.
GrondbewerkIngsadvles voor aardappelen.
BedrIJfsontwlkkelIng 6(10):819-824.
Gllemeroth, G., 1953. Untersuchungen über VerfestIgungs- und
Verlagerungsvorgange Im Ackerboden unter Rad- und Raupen-
fahrzeugen. Zeltschrift für Acker- und Pflanzenbau 96:219-234.
Grecenko, A., 1967.Vllv nekollka alternativ podvozku koloveho
traktoru na napetI v pude.
Zemedelska Technika 13(7):361-378.
Hahn R.H., M.A. Purschwitz and E.E. Rosentreter (editors), 1984.
ASAE Standards 1984. Standards, Engineering Practices and Data
adopted by the American Society of Agricultural Engineers.
American Society of Agricultural Engineers, St. Joseph, Ml.
Havlnga, L. en U.D. Perdok, 1969.Methode ter karakterisering van
het mechanisch gedrag van grond. Nota 534, ICW Wageningen.
Hegedus, E., 1965. Pressure distribution under rigid wheels.
Trans. ASAE 8:305-308,311.
Hesse, H., 1986. Schlupfregeln - Sprit sparen.
Lohnunternehmen 41(1):39-40.
HettlaratchI, D.R.P. and J.R. O'Callaghan, 1980.Mechanical be-
haviour of agricultural soils.
Journ. Agr. Eng. Res. 25:239-259.
Hofferberth, W. and R. Relnhold, 1969. Probleme der Reifen-
entwicklung für die derzeitige Schlepper und Landmachinen.
Proc. 3th Int. Conf. ISTVS, Essen, Vol. II1:44-67.
Hofman, V.L., 1977.Model tractor demonstration on ballasting 2WD
und 4WD tractors for efficient use of horsepower.
ASAE Paper No.77-1519, Chlchago.
Holm, J.C., 1969. Das Verhalten von Reifen beim mehrmaligen
überfahren einer Spur.
Proc. 3th Int. Conf. ISTVS, Essen, Vol. 2:96-123.
Inns, F.M. and J. Kllgour, 1978. Agricultural tyres.
DunIop, London.
Jahns, G. und H. Steinkampf, 1982. Einflussgrössen auf Flächen-
leistung und Energieaufwand beim SchIeppereInsatz.
Grundl. Landtechnik 32(1):20-27.
Janosl, Z., 1962. Theoretical analysis of the performance of
tracks and wheels operating on deformable soils.
Trans. ASAE 5:133,134,146.
Karaflath, L.L. and E.A. Nowatzki, 1978. Soil mechanics for off-
road vehicle engineering.
Trans. Tech. PubI., Clausthal, Germany.
Karczewskl, T.,1978. The Influence of passing speed of agricul-
tural machinery wheels on soil compaction. Zeszyty Problemowe
Postepow Nauk, Rolnlczych.
Kellermann, F., 1985. 1 Impuls » 1 Zentimeter.
Die Formel für exacte Geschwindigkeitsmessung.
Agrartechnik 64(11):46.
KezdI, A., 1969. Handbuch der bodenmechanIk.
Band II, V.E.B. Verlag, Berlin.
Knlght, S.J. and A.J. Green, 1962. Deflection of a moving tire on
firm to soft surfaces. Trans. ASAE 5:116-120.
Koenigs, F.F.R., 1976. Determination of the Upper Tillage Limit
for spring tillage by lab. test.
Proc. 7th Conf. ISTRO, Sweden 19.1-19.7.

166
Kooien, A.J., 1973. Failure patterns In different soils produced
by a curved blade with a small angle of approach. NIAE Subject
Day on the mechanical behaviour of agricultural soil. Sllsoe.
Kooien, A.J., 1974. A method for soil compactabIIIty determina-
tion. Journ. Agr. Eng. Res. 19:271-278.
Kooien, A.J., 1976. De Invloed van de mechanisatie op de bodem-
struktuur. 100 Jaar onderwijs, voorlichting en onderzoek In de
landbouw. WagenIngen.
Kooien, A.J., 1977. Soil loosening processes In tillage. Analysis
systematics and predictability.
Mededelingen Landbouwhogeschool WagenIngen,77-17.
Kooien, A.J., 1978. The Influence of a soil compaction process on
subsequent soil tillage processes. A new research method.
Neth Journ. Agr. ScI. 26:191-199.
Kooien, A.J. and H. Kuipers, 1983. Agricultural soil mechanics.
Springer-Verlag, Berlin etc.
Kooien, A.J. and P. Vaandrager, 1984. Relationships between soil
mechanical properties. Journ. Agr. Eng. Res. 29:313-319.
Kooien, A.J., 1986. Behaviour of elemental soil volumes In
compaction and effects on properties of significance In soli
use. Transactions XIII Congress of the Int. Soc. of Soil
Science, Hamburg, Vol. IV:1368-1371.
Kouwenhoven, J.K., 1974. Ruggenteelt: teeItmaatregeI en groelre-
gulator. Landbouwmechanisatie 25:677-682.
KrIck, G., 1969. Druck- und SchubverteIIung unter Rädern und
Reifen auf nachgiebigem Boden unter Berücksichtigung der
RelfendeformatIon.
Proc. 3th Int. Conf. ISTVS, Essen, Vol. 11:50-75.
Krumbach, A.W. and D.P. White, 1964.Moisture pore space, and
bulk density changes In frozen soil.
Soll Sel. Soc. Am. Proc. 28:422-425.
Kuipers, H., 1957. A reliefmeter for soll cultivation studies.
Neth. Journ. Agr. Sei. 5(4):255-262.
Kuipers, H., 1959. Confined compression tests on soil aggregate
samples. Meded. Landbouwhogeschool en Opzoekingsstations Staat
Gent 24(1):349-357.
Kuipers, H., 1966. Die Bearbeitung schwerer Böden in den Nieder-
landen. Tagungsberichte 82, Teil II.
Deutsche Akad. der Landw. zu Berlln:233-240.
Kuipers, H., 1970. Entgegengesetzte Effekte von Reifen und Boden-
bearbeitungen. Ber. Int. Symp. Warna, Sofia :59-65.
Kumar, L. and J.A. Weber, 1974. compaction of unsaturated soil by
different stress paths. Trans. ASAE 17:1064-1069,1072.
Kurtay, T. and A.R. Reece, 1970. Plasticity theory and critical
state soil mechanics. J. Terramech. 7:23-56.
Liang, T. and C. Yung, 1966. A microscopic study of tractive
performance of a lugged tire operating on sand.
Trans. ASAE 9:513-515.
Lumkes, L.M. en U.D. Perdok, 1981. Volgteelt van stamslabonen na
doperwten. PubI. 17, PAGV Lelystad.
MaidI, F.x. und G. Fischbeck, 1985.Wenn dem Boden die Luft
ausgeht. DLG-Mlttellungen 100(13):1354-1356.
McAllister, M., 1983. Reducing the rolling resistance of tyres
for trailed agricultural machinery.
Journ. Agr. Eng. Res. 28:127-137.

167
Mclntlre, D.S. and G.B. Stlrk, 1954. A method for determination
of apparent density of soil aggregates.
Austr. J. Agrlc. Research 5:291-296.
McKlbben, E.G. and J.B. Davidson, 1940. Transport wheels for
agricultural machines. Agr. Eng. 21(2):57-58.
Mech, S.J., G.M. Horner, L.M.Cox and E.E. Cary, 1967. Soil
profile modification by backhoe mixing and deep plowing.
Trans. ASAE 10(6):775-779.
Melborg, R.H. en U.D. Perdok, 1979. Grondbewerking en vermogens-
behoefte. Rapport 11, IMAG WagenIngen.
Melzer, K.J. and S.J. Knight, 1973. Dual-wheel performance In
sand. Trans. ASAE 16:204-207.
Melzer, K.J., 1976. Power requirements for wheels operating in
sand. J. Terramech. 13(2):75-85.
Moreno, F., C. Sommer und W. Czeratzki, 1974. Einige bodenphysi-
kallsche Untersuchungen an der Schleppersohle einer degradier-
ten Schwarzerde. Landbauforschung Völkenrode 24(2):123-132.
Nadai, A., 1950. Theory of flow and fracture of solids. Vol. I.
New York.
NJds, A., 1976. Long term effects of tractor traffic
In two field experiments In Norway. Report 45, Division of
Soil Management of the Agr. Coll. of Sweden, Uppsala.
N.N., 1971. Solltest Inc. Gen. Catalog, vol 2a. Evanston.
N.N., 1980. Two-wheel-drIve tractor slip down the field.
Farmers weekly sept. 26th.
N.N., 1981a. There's danger In use too much tyre ballast.
Farmers weekly septh. 18th.
N.N., 1981b. Landbouwbanden. Vlugschrift voor de landbouw 335,
WagenIngen.
Onafeko, O. and A.R. Reece, 1967. Soil stresses and deformations
beneath rigid wheels. J. Terramech. 4:59-80.
Ouwerkerk, C. van, 1968. Two model experiments on the
durability of subsoil compaction.
Neth. Journ. Agr. ScI.16:204-210.
Painter, D.J., 1981. A simple deflection model for agricultural
tyres. Journ. Agr. Eng. Res. 26:9-20.
Perdok, U.D., P. Boekel en J. Jorrltsma, 1974a.
Het grondbewerkIngsadvles voor suikerbieten.
BedrIJfsontwlkkelIng 5(10):861-865.
Perdok, U.D., J.J. Klooster en M.C. Sprong, 1974b. De bewerkbaar-
heid van de grond tijdens de voorJaarswerkzaamheden.
Rapport 249, lLRWagenIngen.
Perdok, U.D., 1976. BewerkbaarheId en berijdbaarheid van grond.
Landbouwkundig Tijdschrift 88(6):173-178.
Perdok, U.D., 1978. A prediction model for the selection of tyres
for towed vehicles on tilled soil.
Journ. Agr. Eng. Res. 23:369-383.
Perdok, U.D. en J. Terpstra, 1983. BerijdbaarheId van landbouw-
grond. Bandspanning en grondverdichtIng.
LandbouwmechanIsatle 34(4):363-366.
Perumpral, J.V., J.B. Llljedahl a n d W . H . Perloff, 1971.
The finite element method for predicting stress distribution
and soil deformation under a tractive device.
Trans. ASAE 14(6):1184-1188.

168
Poletayev, A.F., 1964. The compaction of soil under a rolling
wheel. J. Terramech. 1(3):7-17.
Reaves, C.A. and A.W. Cooper, 1960. Stress distribution In soils
under tractor loads. Agrlc. Engineering 41(1):20-21, 31.
Reece, A.R., 1970. The shape of the farm tractor. Paper presented
at Inst.Mech. Eng. Conf. "Agricultural and Allied Industrial
Tractors", London.
Renlus, K.T., 1985. Traktoren. Technik und Ihre Anwendung.
BLV Verlag, München.
Renlus, K.T., 1986. Zur Entwicklung des Traktors nach 1945.
Landtechnik 41(10):420-425.
Riemer, G., P.W. Bakker Arkema en L.H. Hulsman, 1957.
Handboek voor IandbouwwerktuIgen en trekkers.
Tjeenk WiIIInk, ZwolIe.
Rutherford, l., 1973. Wheeled and tracklaylng tractors - utiliz-
ation, performance and tyre and track costs.
Proc. Int. Conf. "Perspectives of Agricultural Tractor Devel-
opment", Warsawa:115-156.
Schäfer, W., 1983. Theoretische Untersuchungen zur optimalen
Kombination von Allradschleppern und gezogenen Geräten zur
Bodenbearbeitung. Dissertation Univ. Hohenhelm, Stuttgart.
Scheltema, W., 1974. Puddling against dry plowing for lowland
rice culture In Surinam. Dissertation Agricultural University
WagenIngen.
Schilling, E., 1962. Landmaschinen. Schilling, Köln.
Schmitt, H., 1986. Einsatz von Radarmessgeräten in der Landwirt-
schaft. Landtechnik 41(10):434-435.
Schofleld, A. and P. Wroth, 1968.Critical State Soll Mechanics.
McGraw Hill, London.
Schothorst, C.J., 1974. De meting van IndrIngIngsweerstanden In
het terrein. Intern rapport, ICW WagenIngen.
Schuring, D., 1968. Zur Theorie des Geländerads.
Forsch. Ing.-Wes 34:165-200 und 35:7-12.
Serota, S. and A. Jangle, 1972. A direct-reading pocket shear
vane. Civil Engineering - ASCE 42(1):73-74.
Sltkel, G., 1969. Die Kennzahlen von AS-Reifen und die Probleme
der BereIfung.
Proc. 3th Int. Conf. ISTVS, Essen, Vol. II1:23-43.
Sitkei, G. and A. Fekete, 1975. Design of Off-the-road vehicles
for minimum soil compaction.
Proc. 5th Int Conf. ISTVS, DetroIt,534-545.
Söhne, W., 1951. Das mechanische Verhalten des Ackerbodens
bei Belastungen, unter rollenden Rädern sowie bei der
Bodenbearbeitung. Grundl. Landtechnik 1:87-94.
Söhne, W., 1952. Die Kraftübertragung zwischen Schlepperreifen
und Ackerboden. Grundl. Landtechnik 3:75-87.
Söhne, W., 1953a. Druckverteilung im Boden und Bodenverformung
unter SehIepperreIfen. Grundl. Landtechnik 5:49-63.
Söhne, W., 1953b. Reibung und kohäsion bei Ackerboden.
Grundl. Landtechnik 5:64-80.
Söhne, W., 1956. Einige Grundlagen für eine Landtechnische Soden-
mechanlk. Grundl. Landtechnik 7:11-27.
Söhne, W., 1958. Fundamentals of pressure distribution and soll
compaction under tractor tires. Agr. Eng. 39:276-281,290.

169
Söhne, W., 1963. Beitrag zur Mechanik des Systems Fahrzeug-Boden
unter besonderer Berücksichtigung der Ackerschlepper.
Grundl. Landtechnik 17:5-16.
Söhne, W., 1969. St. Christopher Lecture. Agricultural engineer-
ing and terramechanles. Third Int. Conf. ISTVS, Essen.
Söhne, W. und H. Stubenböck, 1978. Theoretische Grundlagen der
Mechanik der Bodenbearbeitung.
Berichte über Landwirtschaft 56:390-414.
Söhne, W., 1980. Entwicklungstendenzen und -mögIIchkeIten bel
Allradschleppern. Landtechnik 35(4):156-161.
Söhne, W. und I.Bolling, 1981. Der Elnfluss der LastvertelIung
auf die TrIebkraft-SchIupf-Kurve von Ackerschleppern.
Grundl. Landtechnik 31(3).81-85.
Sommer, C , 1972.Modellversuche über den Elnfluss hoher Kalkga-
ben auf die Verformung und VerdIchtbarkelt eines Bodens. Mit-
teilungen Deutschen BodenkundlIchen Gesellschaft 15:65-72.
Sommer, C , K. Stolnev und H.J. Altemüller, 1972. Das verhalten
vier verschlendener Modellböden unter vertikaler Belastung.
Landbauforschung VöIkenrode 22:45-56.
Sommer, C , H. Steinkampf, M. Zach und W. Czeratzkl, 1975.
Ein Beitrag zum Problem der Bodenverdichtung beim einsatz
leistungsstarker Schlepper.
Landbauforschung Völkenrode 25(2):69-74.
Sonnen, F.J., 1970. über den Elnfluss von Form und Länge der
Aufstandsflache auf die Zugfähigkeit und den Rollwiderstand
von AS-Relfen. Landbauforsch. Völkenrode, Sonderheft 3.
Stelner, M., 1978.Messungen für Triebkraft-Schlupf-Kurven ver-
schiedener Ackerschlepperreifen In der Bodenrinne.
Grundl. Landtechnik 28(5):169-208.
Stelner, M., 1979. Analyse, Synthese und Berechnungsmethoden der
Triebkraft-Schlupf-Kurve von Luftreifen auf nachgiebigen
Boden. Dissertation Technische Universität München.
Stelner, M. und W. Söhne, 1979. Berechnung der Tragfähigkeit von
Ackerschlepperrelfen sowie des Kontaktflächen mltteIdruckes
und des Rollwiderstandes auf starrer Fahrbahn.
Grundl. Landtechnik 29(5):145-152.
Steinkampf, H., 1971. Zur Methodik der Rollradien- und Rad-
schlupfmessung. Grundl. Landtechnik 21:40-44.
Steinkampf, H., 1974. Probleme der effizienten Umwandlung der
Motorleistung In Zugleistung bei Leistungsstarken Schleppern.
Grundl. Landtechnik 24(1):14-20.
Steinkampf, H., 1975. Ermittlung von ReIfenkennIInlen und Gerä-
tezugleistungen für Ackerschlepper. Dissertation Techn. Uni-
versität Braunschweig.
Steinkampf, H., 1977. Problematik der Leistungsumwandlung über
die Triebräder bei Leistungsstarken Schleppern.
Grundl. Landtechnik 27(5):168-172.
Steinkampf, H., 1981a. Zur Entwicklung der Schlepperreifen.
Landtechnik 4:178-181.
Steinkampf, H., 1981b. Ergebnisse aus RelfenvergIeIchsversuchen.
Landtechnik 10:479-482.
Stlenstra, J., 1976. De voorspelbaarheid van verdichting onder
banden. Unpublished thesis, Tillage Laboratory, Agricultural
University, WagenIngen.

170
Stone, J.A. and W.F. Larson, 1980. Rebound of five one-dImensIon-
ally compressed unsaturated granular soils.
Soil Sel. Soc. Am. Journ. 44:819-822.
Taylor, P.A. and N.Y. Williams, 1959. Traction characteristics of
11-36 agricultural tractor tyres on hard surfaces.
Journ. Agr. Eng. Res. 4:3-8.
Taylor, J.H., 1973. Lug angle effect on traction performance of
pneumatic tractor tires. Trans. ASAE 16:16-18.
Taylor, J.H., 1974. Lug spacing effect on traction of pneumatic
tractor tires. Trans. ASAE 17(2):195-197.
Taylor, J.H., 1976. Comparative traction performance of R-1, R-3,
and R-4 tractor tires. Trans. ASAE 19(1):14-16.
Terpstra, J. en J. van Maanen, 1972. Aspecten van het trekkerban-
denonderzoek. PubI. 160, ILR Wageningen.
Terpstra, J., 1973. Performance characteristics of deep lug
tires. Beitrage der Int. Konferenz "Entwicklungsperspektive
landwirtschaftlicher Schlepper", Warschau, Vol 1:233-258.
Terpstra, J., 1978.Wetenswaardigheden over banden.
Publ. 108, IMAG Wageningen.
Thansandote, A. et al., 1977. A new slip monitor for traction
equipment. Trans. ASAE 20:851-856.
TijInk, F.G.J., 1979. Mechanische eIgenschapsmetIngen en band-
proeven gericht op het ontwikkelen van voorspeIIIngsmethodes.
Unpublished thesis, Tillage Laboratory, Agr. University Wage-
nIngen.
Tijlnk, F.G.J, en W.P. den Haan, 1981.Wlelsllp tijdens ploegen.
LandbouwmecnanIsatie 32(10):961-964.
Tijlnk, F.G.J, en P. Vaandrager, 1983. De conusvorm bij penetro-
meterwaarnemIngen. CuItuurtechn. TIJdschr. 22(6):377-383.
Tijlnk, F.G.J, and A.J. Kooien, 1985. Prediction of tire rolling
resistance and soil compaction, using cone, shear vane, and a
falling weight. Proc. Int. Conf. on Soil Dynamics, Auburn,
Alabama USA, Vol. 4:800-813.
Tijlnk, F.G.J., P. LerInk and A.J. Kooien, 1988. Summation of
shear deformation In stream tubes In soil under amoving tyre.
To be published In Soil and Tillage Research.
Trabbic, G.W., K.V. Lask and W.F. Buchele, 1959. Measurements of
soil-tire Interface pressures. Agr. Eng. 40:678-681.
Traulsen, H. und W. Spingies, 1978. Schlepper mussten auf schwe-
rem Boden ihr Können beweisen. Top Agrar:68-71.
Turnage, G.W., 1972. Tire selection and performance prediction
for off-road wheeled-vehicle operations.
Proc. 4th Int. Conf. ISTVS, Stockholm, Vol. 1:61-81.
Uffelmann, F.L., 1961. The performance of rigid wheels on clay-
soils. Proc. 1st Int. Conf. SoIIVehIcIeMechanles, TurIn.
VomocII, J.A. and W.J. Flocker, 1965. Degradation of stucture of
Yolo Loam by compaction. Soll Sei. Soc. Am. Proc. 29(1):7-12.
Wann, R.L. and I.F. Reed, 1962. Studies of tractor tire-tread
movement. Trans. ASAE 5:130-132.
Warkentin, B.P., 1971. Effects of compaction on content and
transmission of water In soils. Chapter 5 InCompaction of
Agricultural Soils. ASAE St. Jozeph, Ml.:126-153.
Watt, H.v.H. van der, 1969. Influence of particle size distribu-
tion on soil compactabIIIty. Agrochemophyslea 1:79-86.

171
Werkhoven, C , 1975.Ontwikkeling van een enkelwleItester voor
het bandenonderzoek. Publ. 17, IMAG WagenIngen.
Wills, B.M.D., E.M. Barret and G.J. Shaw, 1965. An Investigation
Into rolling resistance theories for towed rigid wheels.
J. Terramech. 1(1):24-53.
Wlsmer, R.D. and H.J. Luth, 1973.Off-road traction prediction
for wheeled vehicles. J. Terramech. 10(2):49-61.
Witney, B.D. and K.E. OskouI, 1982. Ploughing performance pre-
dictor. Power farming 61(12):46-49.
Wlttsell, L.E. and J.A. Hobbs, 1965. Soil compaction effects on
field plant growth. Agron. J. 57:534-537.
Wong, J.Y. and A.R. Reece, 1966. Soil failure beneath rigid
wheels. Proc. 2nd Int. Conf. ISTVS, Quebec, Canada.
Wong, J.Y., 1967. Behaviour of soil beneath rigid wheels.
Journ. Agr. Eng. Res. 12(4):257-269.
Wong, J.Y. and A.R. Reece, 1967a. Prediction of rigid wheel
performance based on the analysis of soil-wheel stresses.
Part I, Performance of driven rigid wheels.
J. Terramech. 4(1):81-98.
Wong, J.Y. and A.R. Reece, 1987b. Prediction of rigid wheel
performance based on the analysis of soil-wheel stresses.
Part II, Performance of towed rigid wheels.
J. Terramech. 4(2):7-25.
Wong, J.Y., 1978. Theory of ground vehicles. John Wiley & Sons,
New York.
Wu, T.H., 1971. Soil Dynamics. Allyn and Bacon, Boston.
Yong, R.N. and B.P. Warkentln, 1966. Introduction to soil behav-
iour. Mac MlIIan Co, New York.
Yong, R.N. and B.P. Warkentln, 1975. Soil properties and behav-
iour. Elsevier, Amsterdam.
Yong, R.N. and E.A. Fattah, 1976. Prediction of wheel-soil Inter-
action and performance using the finite element method.
J. Terramech. 13(4):227-240.
Yong, R.N., E.A. Fattah and P. Boonslnsuk, 1978. Analysis and
prediction of tyre-soil Interaction and performance using
finite elements. J. Terramech. 15(l):43-63.
Zoz, F.M., 1972. Predicting tractor field performance.
Trans. ASAE 15:249-255.

172
 CURRICULUM VITAE

Franciscus Gerhardus Jozeph TIJInkwerd geboren op 11 september

1955 te Denekamp.
Na het behalen van het diploma HBS-b in 1973 aan het Twents
Carmellyceum teOldenzaal begon hij zijn studie aan de Landbouw-
hogeschool te WagenIngen. Het doctoraalexamen in de studierich-
ting Landbouwtechniek legde hij af in september 1979. Naast de
hoofdvakken GrondbewerkIng en Landbouwbedrijfsgebouwen omvatte de
doctoraalstudie het vak LandbouwwerktuIgkunde.
In aansluiting daarop trad hij In dienst als promotie-assistent
bij de Landbouwhogeschool, waar hij tot 1augustus 1982 bij de
vakgroep GrondbewerkIng het onderzoek naar afsteunende processen
inwlel-grond systemen heeft uitgevoerd. Vervolgens was hij als
docent verbonden aan de Agrarische Hogeschool te Dordrecht. In
deze periode heeft hij dit proefschrift voltooid.
Per 1september 1987 Is hij in dienst getreden als hoofd van de
afdeling Tractie en GrondbewerkIng bij het Instituut voor Mecha-
nisatie, Arbeid en Gebouwen (IMAG) te WagenIngen.

173