Sap Ref 1

Download as pdf or txt
Download as pdf or txt
You are on page 1of 438

SAP2000

Integrated Finite Element Analysis and Design of Structures

ANALYSIS REFERENCE

COMPUTERS & STRUCTURES INC.

Computers and Structures, Inc. Berkeley, California, USA

Version 7.0 Revised October 1998

COPYRIGHT
The computer program SAP2000 and all associated documentation are proprietary and copyrighted products. Worldwide rights of ownership rest with Computers and Structures, Inc. Unlicensed use of the program or reproduction of the documentation in any form, without prior written authorization from Computers and Structures, Inc., is explicitly prohibited. Further information and copies of this documentation may be obtained from:

Computers and Structures, Inc. 1995 University Avenue Berkeley, California 94704 USA tel: (510) 845-2177 fax: (510) 845-4096 e-mail: [email protected] web: www.csiberkeley.com

Copyright Computers and Structures, Inc., 19781998. The CSI Logo is a registered trademark of Computers and Structures, Inc. SAP2000 is a registered trademark of Computers and Structures, Inc. Windows is a registered trademark of Microsoft Corporation.

DISCLAIMER
CONSIDERABLE TIME, EFFORT AND EXPENSE HAVE GONE INTO THE DEVELOPMENT AND DOCUMENTATION OF SAP2000. THE PROGRAM HAS BEEN THOROUGHLY TESTED AND USED. IN USING THE PROGRAM, HOWEVER, THE USER ACCEPTS AND UNDERSTANDS THAT NO WARRANTY IS EXPRESSED OR IMPLIED BY THE DEVELOPERS OR THE DISTRIBUTORS ON THE ACCURACY OR THE RELIABILITY OF THE PROGRAM. THE USER MUST EXPLICITLY UNDERSTAND THE ASSUMPTIONS OF THE PROGRAM AND MUST INDEPENDENTLY VERIFY THE RESULTS.

ACKNOWLEDGMENT
Thanks are due to all of the numerous structural engineers, who over the years have given valuable feedback that has contributed toward the enhancement of this product to its current state. Special recognition is due Dr. Edward L. Wilson, Professor Emeritus, University of California at Berkeley, who was responsible for the conception and development of the original SAP series of programs and whose continued originality has produced many unique concepts that have been implemented in this version.

Table of Contents
Chapter I Introduction
SAP2000 Analysis Features . . . . . . Structural Analysis and Design . . . . . About This Manual . . . . . . . . . . . Topics . . . . . . . . . . . . . . . . . . Typographical Conventions. . . . . . . Bold for Definitions . . . . . . . . Bold for Variable Data . . . . . . . Italics for Mathematical Variables . Italics for Emphasis . . . . . . . . All Capitals for Literal Data . . . . Capitalized Names . . . . . . . . . Bibliographic References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1
1 2 3 3 4 4 4 4 4 5 5 5

Chapter II

Labels, Arrays, and Generation

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Labels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Label Increments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Regular Arrays. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Single Label. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 One-dimensional Regular Arrays . . . . . . . . . . . . . . . . . . 11 Two-dimensional Regular Arrays . . . . . . . . . . . . . . . . . 12 Three-dimensional Regular Arrays . . . . . . . . . . . . . . . . . 13 Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Elements, Constraints, and Welds . . . . . . . . . . . . . . . . . 15

i
5

SAP2000 Analysis Reference


Deletion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Chapter III

Coordinate Systems
Overview. . . . . . . . . . . . . . . . . . . . . . Global Coordinate System. . . . . . . . . . . . . Upward and Horizontal Directions . . . . . . . . Defining Coordinate Systems . . . . . . . . . . . Vector Cross Product . . . . . . . . . . . . . Defining the Three Axes Using Two Vectors Local Coordinate Systems . . . . . . . . . . . . . Alternate Coordinate Systems . . . . . . . . . . . Cylindrical and Spherical Coordinates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

21
22 22 23 23 23 24 24 26 28

Chapter IV

Joint Coordinates
Overview. . . . . . . . . . . . . . . . . . . . . . Joint Definition . . . . . . . . . . . . . . . . . . One-dimensional Joint Generation . . . . . . . . One-dimensional Joint Array Specification . One-dimensional Joint Definition . . . . . . One-dimensional Linear Generation . . . . . One-dimensional Cylindrical Generation. . . Two-dimensional Joint Generation . . . . . . . . Two-dimensional Joint Array Specification . Two-dimensional Joint Definition . . . . . . Two-dimensional Linear Generation . . . . . Two-dimensional Frontal Generation. . . . . Two-dimensional Edge Generation. . . . . . Three-dimensional Joint Generation. . . . . . . . Three-dimensional Joint Array Specification. Three-dimensional Joint Definition. . . . . . Three-dimensional Linear Generation . . . . Three-dimensional Frontal Generation . . . . Three-dimensional Edge Generation . . . . . Variable Joint Spacing. . . . . . . . . . . . . . . One Dimension . . . . . . . . . . . . . . . . Two Dimensions . . . . . . . . . . . . . . . Three Dimensions. . . . . . . . . . . . . . . Joint Definition in Polar Coordinates . . . . . . . Cylindrical Coordinates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

31
32 32 33 33 33 34 34 36 37 37 38 38 40 43 43 44 45 46 46 47 48 48 49 50 50

ii
6

Table of Contents
Spherical Coordinates. . . . . . . . . . . . . . . . . . . . . . . . 52 Variable Joint Spacing in Polar Coordinates . . . . . . . . . . . . 53 Joint Definition in Alternate Coordinate Systems . . . . . . . . . . . . 54

Chapter V

Joint Patterns
Overview. . . . . . . . . . . . . . . . . . . Pattern Definition . . . . . . . . . . . . . . Gradient or Hydrostatic Specification . . . . One-dimensional Pattern-value Generation . Joint Array Specification . . . . . . . . One-dimensional Addition . . . . . . . One-dimensional Gradient Addition . . One-dimensional Linear Mapping . . . Two-dimensional Pattern-value Generation . Joint Array Specification . . . . . . . . Two-dimensional Addition . . . . . . . Two-dimensional Gradient Addition . . Two-dimensional Linear Mapping . . . Two-dimensional Frontal Mapping. . . Two-dimensional Edge Mapping . . . . Three-dimensional Pattern-value Generation Joint Array Specification . . . . . . . . Three-dimensional Addition . . . . . . Three-dimensional Gradient Addition . Three-dimensional Linear Mapping . . Three-dimensional Frontal Mapping . . Three-dimensional Edge Mapping . . . Variable Pattern-value Increments . . . . . One Dimension . . . . . . . . . . . . . Two Dimensions . . . . . . . . . . . . Three Dimensions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57
58 59 59 60 61 62 62 63 63 64 64 65 65 66 66 67 67 68 69 69 70 71 72 73 73 74

Chapter VI

Joints and Degrees of Freedom


Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . Modeling Considerations . . . . . . . . . . . . . . . . . . . Local Coordinate System . . . . . . . . . . . . . . . . . . . Advanced Local Coordinate System . . . . . . . . . . . . . Reference Vectors . . . . . . . . . . . . . . . . . . . . Defining the Axis Reference Vector . . . . . . . . . . . Defining the Plane Reference Vector . . . . . . . . . . . Determining the Local Axes from the Reference Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

75
76 77 78 79 79 80 80 81

iii
7

SAP2000 Analysis Reference


Joint Coordinate Angles . . . . . . . . . . . . . . . . . . . . . . 82 Degrees of Freedom . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 Available and Unavailable Degrees of Freedom . . . . . . . . . . 85 Restrained Degrees of Freedom . . . . . . . . . . . . . . . . . . 86 Constrained Degrees of Freedom . . . . . . . . . . . . . . . . . . 86 Active Degrees of Freedom. . . . . . . . . . . . . . . . . . . . . 86 Null Degrees of Freedom . . . . . . . . . . . . . . . . . . . . . . 88 Restraints and Reactions. . . . . . . . . . . . . . . . . . . . . . . . . 88 Springs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 Masses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 Force Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Restraint Displacement Load . . . . . . . . . . . . . . . . . . . . . . 93 Spring Displacement Load . . . . . . . . . . . . . . . . . . . . . . . 95 Degree of Freedom Output . . . . . . . . . . . . . . . . . . . . . . . 97 Joint Mass Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 Displacement and Acceleration Output . . . . . . . . . . . . . . . . 100 Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Applied Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Inertial Loads . . . . . . . . . . . . . . . . . . . . . . . . . . . 104 Stiffness Forces . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Spring Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Nllink Forces . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 Restraint Forces (Reactions). . . . . . . . . . . . . . . . . . . . 106 Constraint Forces . . . . . . . . . . . . . . . . . . . . . . . . . 107 P-Delta Moments . . . . . . . . . . . . . . . . . . . . . . . . . 107 Global Force Balance Output . . . . . . . . . . . . . . . . . . . . . 107

Chapter VII

Constraints and Welds


Overview . . . . . . . . . . . Body Constraint . . . . . . . Joint Connectivity . . . . Local Coordinate System Constraint Equations . . Plane Definition . . . . . . . Diaphragm Constraint . . . . Joint Connectivity . . . . Local Coordinate System Constraint Equations . . Plate Constraint. . . . . . . . Joint Connectivity . . . . Local Coordinate System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

113
114 115 115 115 115 116 117 117 117 118 119 119 119

iv
8

Table of Contents
Constraint Equations . . . . Axis Definition . . . . . . . . . . Rod Constraint . . . . . . . . . . Joint Connectivity . . . . . . Local Coordinate System . . Constraint Equations . . . . Beam Constraint . . . . . . . . . Joint Connectivity . . . . . . Local Coordinate System . . Constraint Equations . . . . Equal Constraint . . . . . . . . . Joint Connectivity . . . . . . Local Coordinate System . . Selected Degrees of Freedom Constraint Equations . . . . Local Constraint . . . . . . . . . Joint Connectivity . . . . . . No Local Coordinate System Selected Degrees of Freedom Constraint Equations . . . . Welds . . . . . . . . . . . . . . . Automatic Master Joints . . . . . Stiffness, Mass, and Loads . Local Coordinate Systems. . Constraint Output. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 120 120 121 122 122 122 123 123 123 124 124 125 125 125 127 127 128 128 128 131 132 132 133 133

Chapter VIII

Material Properties
Overview . . . . . . . . . . . . . . Local Coordinate System. . . . . . Stresses and Strains. . . . . . . . . Isotropic Materials . . . . . . . . . Orthotropic Materials. . . . . . . . Anisotropic Materials . . . . . . . Temperature-Dependent Properties Element Material Temperature . . . Mass Density . . . . . . . . . . . . Weight Density . . . . . . . . . . . Design-Type Indicator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

135
136 136 137 138 139 140 141 142 142 143 143

v
9

SAP2000 Analysis Reference Chapter IX The Frame Element


Overview . . . . . . . . . . . . . . . . . . . . . . Joint Connectivity . . . . . . . . . . . . . . . . . Degrees of Freedom . . . . . . . . . . . . . . . . Local Coordinate System. . . . . . . . . . . . . . Longitudinal Axis 1 . . . . . . . . . . . . . . Default Orientation . . . . . . . . . . . . . . Coordinate Angle . . . . . . . . . . . . . . . Advanced Local Coordinate System . . . . . . . . Reference Vector . . . . . . . . . . . . . . . Determining Transverse Axes 2 and 3 . . . . Section Properties . . . . . . . . . . . . . . . . . Local Coordinate System . . . . . . . . . . . Material Properties . . . . . . . . . . . . . . Geometric Properties and Section Stiffnesses. Shape Type . . . . . . . . . . . . . . . . . . Automatic Section Property Calculation . . . Section Property Database Files. . . . . . . . Additional Mass and Weight . . . . . . . . . Non-prismatic Sections . . . . . . . . . . . . End Offsets . . . . . . . . . . . . . . . . . . . . . Clear Length . . . . . . . . . . . . . . . . . . Rigid-end Factor. . . . . . . . . . . . . . . . Effect upon Non-prismatic Elements . . . . . Effect upon Internal Force Output. . . . . . . Effect upon End Releases . . . . . . . . . . . End Releases . . . . . . . . . . . . . . . . . . . . Unstable End Releases . . . . . . . . . . . . Effect of End Offsets . . . . . . . . . . . . . Effect upon Prestress Load . . . . . . . . . . Mass . . . . . . . . . . . . . . . . . . . . . . . . Self-Weight Load. . . . . . . . . . . . . . . . . . Gravity Load . . . . . . . . . . . . . . . . . . . . Concentrated Span Load . . . . . . . . . . . . . . Distributed Span Load . . . . . . . . . . . . . . . Loaded Length. . . . . . . . . . . . . . . . . Load Intensity . . . . . . . . . . . . . . . . . Projected Loads . . . . . . . . . . . . . . . . Temperature Load . . . . . . . . . . . . . . . . . Prestress Load . . . . . . . . . . . . . . . . . . . Prestressing Cables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

145
146 147 147 148 148 148 149 149 150 153 153 154 154 154 155 157 157 159 159 162 163 164 164 165 165 165 165 166 167 167 168 168 169 169 169 170 173 173 174 174

vi
10

Table of Contents
Prestress Load . . . . . . . . . . . Effect upon P-Delta Analysis . . . Internal Force Output. . . . . . . . . . Effect of End Offsets . . . . . . . Internal Forces in the Output File . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 174 175 176 177 179 179

Chapter X

The Shell Element


Overview . . . . . . . . . . . . . . . . . . Joint Connectivity . . . . . . . . . . . . . Degrees of Freedom . . . . . . . . . . . . Local Coordinate System. . . . . . . . . . Normal Axis 3 . . . . . . . . . . . . . Default Orientation . . . . . . . . . . Element Coordinate Angle . . . . . . Advanced Local Coordinate System . . . . Reference Vector . . . . . . . . . . . Determining Tangential Axes 1 and 2. Section Properties . . . . . . . . . . . . . Section Type. . . . . . . . . . . . . . Thickness Formulation . . . . . . . . Material Properties . . . . . . . . . . Material Angle. . . . . . . . . . . . . Thickness . . . . . . . . . . . . . . . Mass . . . . . . . . . . . . . . . . . . . . Self-Weight Load. . . . . . . . . . . . . . Gravity Load . . . . . . . . . . . . . . . . Uniform Load . . . . . . . . . . . . . . . Surface Pressure Load . . . . . . . . . . . Temperature Load . . . . . . . . . . . . . Internal Force and Stress Output . . . . . . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

181
182 183 185 186 186 187 187 187 189 190 191 191 191 192 193 194 194 195 195 195 196 197 198 203

Chapter XI

The Plane Element


Overview . . . . . . . . . Joint Connectivity . . . . Degrees of Freedom . . . Local Coordinate System. Stresses and Strains. . . . Material Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

205
206 206 208 210 211 211

vii
11

SAP2000 Analysis Reference


Material Angle . . . . Thickness . . . . . . . Mass . . . . . . . . . Self-Weight Load. . . Gravity Load . . . . . Surface Pressure Load Pore Pressure Load . . Temperature Load . . Stress Output . . . . . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212 212 213 213 214 214 214 215 215 216

Chapter XII

The Asolid Element


Overview . . . . . . . . . Joint Connectivity . . . . Degrees of Freedom . . . Local Coordinate System. Stresses and Strains. . . . Material Properties . . . . Material Angle . . . . . . Arc and Thickness . . . . Mass . . . . . . . . . . . Self-Weight Load. . . . . Gravity Load . . . . . . . Surface Pressure Load . . Pore Pressure Load . . . . Temperature Load . . . . Rotate Load. . . . . . . . Stress Output . . . . . . . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

219
220 221 224 224 224 225 226 227 227 228 228 229 229 229 230 230 231

Chapter XIII

The Solid Element


Overview . . . . . . . . . Joint Connectivity . . . . Degrees of Freedom . . . Local Coordinate System. Stresses and Strains. . . . Material Properties . . . . Material Angles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

233
234 234 236 236 236 236 237

viii
12

Table of Contents
Incompatible Bending Modes Mass . . . . . . . . . . . . . Self-Weight Load. . . . . . . Gravity Load . . . . . . . . . Surface Pressure Load . . . . Pore Pressure Load . . . . . . Temperature Load . . . . . . Stress Output . . . . . . . . . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 238 239 239 239 240 240 240 241

Chapter XIV

The Nllink Element


Overview . . . . . . . . . . . . . . . . . . . Joint Connectivity . . . . . . . . . . . . . . Zero-Length Elements . . . . . . . . . . . . Degrees of Freedom . . . . . . . . . . . . . Local Coordinate System. . . . . . . . . . . Longitudinal Axis 1 . . . . . . . . . . . Default Orientation . . . . . . . . . . . Coordinate Angle . . . . . . . . . . . . Advanced Local Coordinate System . . . . . Axis Reference Vector . . . . . . . . . Plane Reference Vector . . . . . . . . . Determining Transverse Axes 2 and 3 . Internal Deformations . . . . . . . . . . . . Nlprop Properties. . . . . . . . . . . . . . . Local Coordinate System . . . . . . . . Internal Nonlinear Springs . . . . . . . Spring Force-Deformation Relationships Element Internal Forces . . . . . . . . . Linear Force-Deformation Relationships Linear Effective Stiffness . . . . . . . . Linear Effective Damping . . . . . . . . Nonlinear Properties. . . . . . . . . . . Nonlinear Deformation Loads . . . . . . . . Mass . . . . . . . . . . . . . . . . . . . . . Self-Weight Load. . . . . . . . . . . . . . . Gravity Load . . . . . . . . . . . . . . . . . Internal Force and Deformation Output . . . Joint Force Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

243
244 245 245 246 246 247 247 247 249 249 250 251 253 254 255 255 256 257 258 259 261 262 271 272 274 274 275 277

ix
13

SAP2000 Analysis Reference Chapter XV Load Cases


Overview . . . . . . . . . . . . . . . . . . Load Cases for Analysis . . . . . . . . . . Defining Load Cases . . . . . . . . . . . . Coordinate Systems and Load Components Force Load . . . . . . . . . . . . . . . . . Restraint Displacement Load. . . . . . . . Spring Displacement Load . . . . . . . . . Self-Weight Load. . . . . . . . . . . . . . Gravity Load . . . . . . . . . . . . . . . . Concentrated Span Load . . . . . . . . . . Distributed Span Load . . . . . . . . . . . Prestress Load . . . . . . . . . . . . . . . Uniform Load . . . . . . . . . . . . . . . Surface Pressure Load . . . . . . . . . . . Pore Pressure Load . . . . . . . . . . . . . Temperature Load . . . . . . . . . . . . . Reference Temperature. . . . . . . . . . . Rotate Load. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

279
280 281 281 282 282 283 283 283 284 284 285 285 285 286 286 287 288 289

Chapter XVI

Static and Dynamic Analysis


Overview . . . . . . . . . . . . . . . . . . . . . . Analysis Cases . . . . . . . . . . . . . . . . . . . Static Analysis . . . . . . . . . . . . . . . . . . . Harmonic Steady-State Analysis . . . . . . . . . . Acceleration Loads . . . . . . . . . . . . . . . . . Eigenvector Analysis . . . . . . . . . . . . . . . . Number of Modes . . . . . . . . . . . . . . . Frequency Range . . . . . . . . . . . . . . . Convergence Tolerance . . . . . . . . . . . . Ritz-vector Analysis . . . . . . . . . . . . . . . . Number of Modes . . . . . . . . . . . . . . . Starting Load Vectors . . . . . . . . . . . . . Number of Generation Cycles . . . . . . . . . Modal Analysis Output. . . . . . . . . . . . . . . Periods and Frequencies . . . . . . . . . . . . Participation Factors. . . . . . . . . . . . . . Participating Mass Ratios . . . . . . . . . . . Static and Dynamic Load Participation Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

291
292 292 293 294 295 296 297 297 298 299 300 301 302 303 303 304 304 306

x
14

Table of Contents
Functions . . . . . . . . . . . . . . . Response-Spectrum Analysis . . . . Local Coordinate System . . . . Response-Spectrum Curve . . . Modal Combination . . . . . . . Directional Combination . . . . Response-Spectrum Analysis Output Damping and Accelerations . . . Modal Amplitudes . . . . . . . . Modal Correlation Factors. . . . Base Reactions. . . . . . . . . . Time-History Analysis . . . . . . . . Loading . . . . . . . . . . . . . Mode Superposition . . . . . . . Modal Damping . . . . . . . . . Time Steps . . . . . . . . . . . . Initial Conditions . . . . . . . . Analysis Results . . . . . . . . . Nonlinear Time-History Analysis . . Nllink Effective Stiffness . . . . Mode Superposition . . . . . . . Modal Damping . . . . . . . . . Iterative Solution . . . . . . . . Static Period . . . . . . . . . . . Combos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309 310 311 311 313 315 317 317 317 319 319 319 320 323 324 325 325 327 328 328 329 330 330 333 333

Chapter XVII P-Delta Analysis


Overview . . . . . . . . . . . . . . . Geometric Nonlinearity . . . . . . . The P-Delta Effect . . . . . . . . . . Equilibrium Equations . . . . . . . . P-Delta Axial Forces . . . . . . . . . Directly Specified Axial Forces . P-Delta Load Combination . . . Iterative Analysis . . . . . . . . . . . Convergence Criterion. . . . . . Maximum Number of Iterations. Convergence Failure. . . . . . . Frame Element . . . . . . . . . . . . Small Deflections . . . . . . . . Cubic Deflected Shape . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

337
338 339 340 343 343 344 345 345 346 346 347 347 347 347

xi
15

SAP2000 Analysis Reference


Computed P-Delta Axial Forces. Prestress . . . . . . . . . . . . . Effect upon Other Analyses . . . . . Dynamic Analyses. . . . . . . . Harmonic Steady-State Analysis Bridge Moving-Load Analysis . Buckling . . . . . . . . . . . . . . . Detection of Buckling . . . . . . Estimating the Buckling Load . . Local Buckling . . . . . . . . . Practical Application . . . . . . . . . Preliminary Linear Analysis. . . Building Structures . . . . . . . Cable Structures . . . . . . . . . Guyed Towers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348 349 349 349 350 350 350 351 351 351 352 352 352 353 355

Chapter XVIII Bridge Analysis


Overview . . . . . . . . . . . . . . Modeling the Bridge Structure . . . Frame Elements . . . . . . . . Supports . . . . . . . . . . . . Bearings and Expansion Joints Other Element Types . . . . . Roadways and Lanes . . . . . . . . Roadways . . . . . . . . . . . Lanes. . . . . . . . . . . . . . Eccentricities . . . . . . . . . Modeling Guidelines . . . . . Examples . . . . . . . . . . . Spatial Resolution . . . . . . . . . Load and Output Points . . . . Resolution . . . . . . . . . . . Modeling Guidelines . . . . . Influence Lines . . . . . . . . . . . Vehicles . . . . . . . . . . . . . . Direction of Loads. . . . . . . Application of Loads . . . . . General Vehicle . . . . . . . . Standard Vehicles . . . . . . . Vehicle Classes. . . . . . . . . . . Moving Load Cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

357
358 359 359 360 361 361 363 363 363 364 364 365 367 367 369 369 370 372 372 372 373 377 383 383

xii
16

Table of Contents
Example 1 AASHTO HS Loading . . . . . . . Example 2 AASHTO HL Loading . . . . . . . Example 3 Caltrans Permit Loading . . . . . . Example 4 Restricted Caltrans Permit Loading Influence Line Tolerance. . . . . . . . . . . . . . . . Exact and Quick Response Calculation . . . . . . . . Moving Load Response Control . . . . . . . . . . . . Correspondence . . . . . . . . . . . . . . . . . . . . Computational Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384 386 387 389 391 391 392 392 393

Chapter XIX

The Output Files


Overview . . . . . . . . . . . . . . . . The Input Echo (.EKO) File . . . . . . The Analysis Log (.LOG) File . . . . . The Results Output (.OUT) File . . . . Element Joint Force Output . . . . . . Joint and Element Output Control . . . Joints. . . . . . . . . . . . . . . . Frame and Nllink Elements . . . . Shell Elements . . . . . . . . . . . Plane, Asolid, and Solid Elements Pagination Control . . . . . . . . . . . Pagination by Lines . . . . . . . . Pagination by Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

395
396 396 397 399 402 405 412 412 413 413 414 414 414

Chapter XX

References

415

xiii
17

18

Chapter I

Introduction
SAP2000 is the latest and most powerful version of the well-known SAP series of structural analysis programs. Basic Topics for All Users SAP2000 Analysis Features Structural Analysis and Design About This Manual Topics Typographical Conventions Bibliographic References

SAP2000 Analysis Features


The SAP2000 structural analysis program offers the following features: Static and dynamic analysis Linear and nonlinear analysis Dynamic seismic analysis and static pushover analysis SAP2000 Analysis Features
19

SAP2000 Analysis Reference Vehicle live-load analysis for bridges P-Delta analysis Frame and shell structural elements, including beam-column, truss, membrane, and plate behavior Two- and three-dimensional and axisymmetric solid elements Nonlinear link and spring elements Multiple coordinate systems Many types of constraints A wide variety of loading options Alpha-numeric labels Large capacity Highly efficient and stable solution algorithms These features, and many more, make SAP2000 the state-of-the-art in structural analysis programs.

Structural Analysis and Design


The following general steps are required to analyze and design a structure using SAP2000: 1. Create or modify a model that numerically defines the geometry, properties, loading, and analysis parameters for the structure 2. Perform an analysis of the model 3. Review the results of the analysis 4. Check the design of the structure This is usually an iterative process that may involve many cycles of the above sequence of steps. All of these steps can be performed seamlessly using the SAP2000 graphical user interface. A few advanced analysis options are not yet available through the SAP2000 graphical user interface. To access these features, you must edit an input data text file. This file can then be brought into the graphical user interface for analysis, display, and design. However, most users will find the SAP2000 graphical user interface sufficient for their needs.

Structural Analysis and Design


20

Chapter I Introduction

About This Manual


This manual describes the theoretical concepts behind the modeling and analysis features offered by the SAP2000 structural analysis program. The focus of this manual is on the analysis portion of the program. It is imperative that you read this manual and understand the assumptions and procedures used by the program before attempting to use the analysis features. The graphical user interface and the design modules are described in separate manuals. Static Pushover Analysis capabilities are descibed in the SAP2000 Detailed Tutorial including Pushover Analysis and in the on-line help feature of the graphical user interface. See the SAP2000 Getting Started manual for a listing of all the manuals supplied with the program.

Topics
Each chapter of this manual is divided into topics and subtopics. All chapters begin with a list of topics covered. These are divided into two groups: Basic topics recommended reading for all users Advanced topics for users with specialized needs, and for all users as they become more familiar with the program. Following the list of topics is an Overview which provides a summary of the chapter. Reading the Overview for every chapter will acquaint you with the full scope of the program. The SAP2000 Basic Analysis Reference is a condensation of the basic topics covered in the present manual.

About This Manual


21

SAP2000 Analysis Reference

Typographical Conventions
Throughout this manual the following typographic conventions are used.

Bold for Definitions


Bold roman type (e.g., example) is used whenever a new term or concept is defined. For example: The global coordinate system is a three-dimensional, right-handed, rectangular coordinate system. This sentence begins the definition of the global coordinate system.

Bold for Variable Data


Bold roman type (e.g., example) is used to represent variable data items for which you must specify values when defining a structural model and its analysis. For example: The Frame element coordinate angle, ang, is used to define element orientations that are different from the default orientation. Thus you will need to supply a numeric value for the variable ang if it is different from its default value of zero.

Italics for Mathematical Variables


Normal italic type (e.g., example) is used for scalar mathematical variables, and bold italic type (e.g., example) is used for vectors and matrices. If a variable data item is used in an equation, bold roman type is used as discussed above. For example: 0 da < db L Here da and db are variables that you specify, and L is a length calculated by the program.

Italics for Emphasis


Normal italic type (e.g., example) is used to emphasize an important point, or for the title of a book, manual, or journal.

Typographical Conventions
22

Chapter I Introduction

All Capitals for Literal Data


All capital type (e.g., EXAMPLE) is used to represent data that you type at the keyboard exactly as it is shown, except that you may actually type lower-case if you prefer. For example: SAP2000 indicates that you type SAP2000 or sap2000 at the keyboard.

Capitalized Names
Capitalized names (e.g., Example) are used for certain parts of the model and its analysis which have special meaning to SAP2000. Some examples: Frame element Diaphragm Constraint Frame Section Load Case Common entities, such as joint or element are not capitalized.

Bibliographic References
References are indicated throughout this manual by giving the name of the author(s) and the date of publication, using parentheses. For example: See Wilson and Tetsuji (1983). It has been demonstrated (Wilson, Yuan, and Dickens, 1982) that ... All bibliographic references are listed in alphabetical order in Chapter References (page 415).

Bibliographic References
23

SAP2000 Analysis Reference

24

C h a p t e r II

Labels, Arrays, and Generation


Labels are names that you assign to the various entities such as joints, elements, materials, and loads that define a structural model and its analysis. A regular array is group of labels that increment (differ from each other) in a regular fashion. Generation allows you to define large numbers of joints, elements, constraints, or welds using arrays and simple data specifications. Basic Topics for All Users Overview Labels Label Increments Regular Arrays Generation Deletion Assignment

7
25

SAP2000 Analysis Reference

Overview
Every item that you define as part of the structural model or its analysis has an alpha-numeric label. Certain types of entities, which are typically defined in large quantities, may be generated using simple data specifications. The generatable entities are: Joints Elements Constraints Welds The remaining types of labeled entities used by the program must each be explicitly defined and cannot be generated: Alternate Coordinate Systems Patterns Materials, Frame Sections, Shell Sections, and Nlprops Loads Functions Specs and Histories Lanes, Vehicles, Classes, and Moving Loads Combos A group of generatable entities may be referred to using a regular array, which is specified by giving the first and last label and the label increment for each of one, two, or three dimensions. Arrays may be used to generate and delete joints, elements, Constraints, and Welds; and to assign loads and properties to these entities.

Labels
Labels are names that you assign to the various entities that make up a structural model and its analysis. Such entities include joints, elements, Constraints, Materials, Loads, and analyses. A few entities, such as the Modes, are numbered automatically by the program. Labels are alpha-numeric according to the following rules:

Overview
26

Chapter II Labels, Arrays, and Generation They may contain from one to seven letters and/or numbers No spaces or other characters are permitted Uppercase and lowercase letters are treated the same, e.g., A3 and a3 are the same A label may not contain all zeroes Leading zeroes are ignored, e.g., 0A3 is the same as A3 Some examples of valid labels are:
1 2 A03 33B 123x567 Corner STEEL

Some examples of invalid labels are:


0 000 A_03 33% 123+567 -5001 ABC 123

The same label may be used for different types of entities without any conflict. For example, you may have a joint, a Frame element, a Material, and a Load Case, each with the label 1.

Label Increments
For the purpose of generation, it is convenient to define a group of entities having labels that increment in a regular fashion. Some examples are:
1, 2, 3, 4 ... A00, A05, A10, A15 ... 8TH, 9TH, 10TH, 11TH ... 1X1, 2X2, 3X3, 4X4 ... 9000, 8000, 7000, 6000 ...

Label Increments
27

SAP2000 Analysis Reference The label increment is the constant difference between successive labels in such a series. The following rules apply to label increments: Only the numeric parts of a label may increment; the alphabetic parts of the labels must be unchanged Leading blanks are considered to be numeric and may increment The increment is always specified as a number, with zeroes at each position where a letter appears Increments may be negative Thus the increments for the above examples are:
1 5 100 101 -1000

A series may continue to increment until one of the following occurs: A numeric part carries over into an alphabetic part A zero or negative number is generated Seven characters is exceeded Thus the limiting labels for the above examples are:
9999999 A95 99999TH 9X9 1000

Regular Arrays
The Regular Arrays described in this topic are used only in the input data text file. Skip this topic if you are preparing your model using the SAP2000 graphical user interface. A regular array, or simply an array, is set of labels that increment in a regular fashion. The labels in an array must correspond to entities of the same type, such as joints or Frame elements. Regular arrays are used to refer to large numbers of items for the purposes of generation, deletion, and assignment.

10

Regular Arrays
28

Chapter II Labels, Arrays, and Generation The labels in a regular array may increment separately in one, two, or three dimensions. The dimension of an array refers to the number of fixed increment values that are used to describe the set of labels. The main advantage of regular arrays is that they are easily described with just a few parameters. SAP2000 uses a simple, standardized format for specifying arrays that makes use of the following parameters: A starting label For each dimension, an ending label and a label increment These concepts will be made clearer in the following subtopics.

Single Label
For the sake of generality, a single label may be considered to be a zerodimensional regular array. It has a starting label, but no ending labels or increments.

One-dimensional Regular Arrays


A one-dimensional regular array is a set of labels that differ, one from the next, by a single increment value. For example, consider the following set of labels:
1 3 5 7 9 11 13

This set of labels can be specified by giving the starting label, 1, the ending label 13, and a label increment, 2. In general, the following information is required to specify a one-dimensional regular array: The label, a0, at the start of the array The label, a1, at the end of the array The label increment, ai1, between any pair of successive labels To say it in words, the array goes from a0 to a1 by ai1. The number of labels, n1, in the array is given by:
n1

a1 a0 +1 ai1

Regular Arrays
29

11

SAP2000 Analysis Reference You must choose the starting and ending labels and the label increment such that n1 is a whole, positive number. For example, an array that goes from 1 to 12 by 5 is not valid. If the ending label is less than the starting label, the increment must be negative.

Two-dimensional Regular Arrays


A two-dimensional regular array is a set of labels that differ, one from the next, by two different increment values in two different directions. For example, consider the following set of labels:
101 103 105 107 109 201 203 205 207 209 301 303 305 307 309

This set of labels increments by 2 in the horizontal direction and by 100 in the vertical direction. It can be specified by giving the starting label 101 in one corner, the ending labels 109 and 301 at the two adjacent corners, and the two increments. In general, the following information is required to specify a two-dimensional regular array: The label, a0, at the starting corner of the array The label, a1, at one ending corner of the array The label, a2, at the other ending corner of the array The label increment, ai1, between any pair of successive labels along side a0-a1 The label increment, ai2, between any pair of successive labels along side a0-a2 You may choose any corner to be the starting corner. The two ending corners must be adjacent to the starting corner. Two corners are adjacent if they are the two ends of one side of the array. To say it in words, the array goes from a0 to a1 by ai1 and to a2 by ai2. The number of labels, n1 and n2, along the two dimensions of the array are given by:
n1

a1 a0 +1 ai1

and

n2

a2 a0 +1 ai2

12

Regular Arrays
30

Chapter II Labels, Arrays, and Generation You must choose the starting and ending labels and the label increments such that n1 and n2 are a whole, positive numbers. The total number of labels in the array is given by n1 n2. The physical locations of the labeled items (joints, elements, etc.) do not have to correspond in any way to the layout of the array of the labels. For example, suppose six joints labeled JOINT1 to JOINT6 physically lie along a straight line. They may be identified by the following one-dimensional array from JOINT1 to JOINT6 by 1:
JOINT1 JOINT2 JOINT3 JOINT4 JOINT5 JOINT6

or by the following two-dimensional array from JOINT1 to JOINT3 by 1 and to JOINT4 by 3:


JOINT1 JOINT2 JOINT3 JOINT4 JOINT5 JOINT6

Three-dimensional Regular Arrays


A three-dimensional regular array is a set of labels that differ, one from the next, by three different increment values in three different directions. For example, consider the following set of labels:
111 112 121 122 131 132 211 212 221 222 231 232

311 312 321 322 331 332

411 412 421 422 431 432

This set of labels increments by 1 in the horizontal direction, by 10 in the vertical direction, and by 100 in the third out-of-plane direction. It can be specified by giving the starting label 111 in one corner, the ending labels 112, 131, and 411 at the three adjacent corners, and the three increments. In general, the following information is required to specify a three-dimensional regular array: The label, a0, at the starting corner of the array The label, a1, at one ending corner of the array The label, a2, at another ending corner of the array The label, a3, at the third ending corner of the array Regular Arrays
31

13

SAP2000 Analysis Reference The label increment, ai1, between any pair of successive labels along side a0-a1 The label increment, ai2, between any pair of successive labels along side a0-a2 The label increment, ai3, between any pair of successive labels along side a0-a3 You may choose any corner to be the starting corner. The three ending corners must be adjacent to the starting corner. Two corners are said to be adjacent if they are on the same side (edge) of the array. To say it in words, the array goes from a0 to a1 by ai1, to a2 by ai2, and to a3 by ai3. The number of labels, n1, n2, and n3, along the three dimensions of the array are given by:
n1

a1 a0 +1 , ai1

n2

a2 a0 +1 , ai2

and

n3

a3 a0 +1 ai3

You must choose the starting and ending labels and the label increments such that n1, n2, and n3 are a whole, positive numbers. The total number of labels in the array is given by n1 n2 n3.

Generation
Generation as described in this topic is used only in the input data text file. Skip this topic if you are preparing your model using the SAP2000 graphical user interface. Generation is used to define a regular array of joints, elements, Constraints, or Welds.

Joints
Two distinct methods are available for generating joints: All joints in an array are simultaneously defined from specified data Joints at the corners or along the edges of an array are first defined; the remaining joints are generated with respect to these previously-defined joints

14

Generation
32

Chapter II Labels, Arrays, and Generation Complete details about joint generation are given in Chapter Joint Coordinates (page 31).

Elements, Constraints, and Welds


Elements, Constraints, and Welds are identical for the purposes of generation. Each element, Constraint, or Weld is defined by: A label A set of connected joints Various other properties and parameters Elements, Constraints, and Welds will all be referred to as elements for the remainder of this discussion. Generation defines a regular array of elements based on the definition of the starting element. The starting element is the element having the starting label in the array. The starting element must have been previously defined or generated, and is unchanged by the generation. All other elements in the array are created if they do not exist, or are redefined if they already exist. Each generated element has the same properties and parameters as the starting element. Only the element label and the connected joints differ. The joint labels differ by joint increments that you specify. There is one joint increment for each dimension of the element array, except for the Frame and Nllink elements which permit two joint increments for each dimension of the array. This is described in more detail in the following. One-dimensional Generation The following information is required to specify one-dimensional element generation: The starting element label: e0 The ending element label: e1 The element label increment: ei1 The joint label increment: ji1 For the Frame and Nllink elements, the joint label increment: ii1

Generation
33

15

SAP2000 Analysis Reference Element e0 must already be defined. Suppose it is connected to joints j1, j2, j3, ..., jn. Then generated element e0+ei1 will be connected to joints j1+ji1, j2+ji1, j3+ji1, ..., jn+ji1, and so on for the rest of the generated elements. Thus all joint labels for a generated element differ from those of the starting element by the same amount. For a Frame or Nllink element, suppose that element e0 is connected to joints i and j. Then generated element e0+ei1 will be connected to joints i1+ii1 and j+ji1, and so on for the rest of the generated elements. The default for the joint label increment ji1 is the element label increment ei1. The default for the joint label increment ii1 is the joint label increment ji1. Thus it is often convenient to assign element labels that are consistent with the joint labels. Two-dimensional Generation The following information is required to specify two-dimensional element generation: The starting element label: e0 The ending element labels: e1 and e2 The element label increments: ei1 and ei2 The joint label increments: ji1 and ji2 For the Frame and Nllink elements, the joint label increments: ii1 and ii2 Element e0 must already be defined. Suppose it is connected to joints j1, j2, j3, ..., jn. Then generated element e0+ei1 will be connected to joints j1+ji1, j2+ji1, j3+ji1, ..., jn+ji1, generated element e0+ei2 will be connected to joints j1+ji2, j2+ji2, j3+ji2, ..., jn+ji2, and so on for the rest of the generated elements. Thus all joint labels for a generated element differ from those of the starting element by the same amount. See Figure 1 (page 17) and Figure 2 (page 17) for examples. For a Frame or Nllink element, suppose that element e0 is connected to joints i and j. Then generated element e0+ei1 will be connected to joints i1+ii1 and j+ji1, generated element e0+ei2 will be connected to joints i1+ii2 and j+ji2, and so on for the rest of the generated elements. The default for the joint label increments ji1 and ji2 are the element label increments ei1 and ei2, respectively. The default for the joint label increments ii1 and ii2 are the joint label increments ji1 and ji2, respectively. Thus it is often convenient to assign element labels that are consistent with the joint labels.

16

Generation
34

Chapter II Labels, Arrays, and Generation

Figure 1 Two-dimensional Generation of Shell elements

Figure 2 Two-dimensional Generation of Plane or Asolid Elements Generation


35

17

SAP2000 Analysis Reference Three-dimensional Generation The following information is required to specify three-dimensional element generation: The starting element label: e0 The ending element labels: e1, e2, and e3 The element label increments: ei1, ei2, and ei3 The joint label increments: ji1, ji2, and ji3 For the Frame and Nllink elements, the joint label increments: ii1, ii2, and ii3 Element e0 must already be defined. Suppose it is connected to joints j1, j2, j3, ..., jn. Then generated element e0+ei1 will be connected to joints j1+ji1, j2+ji1, j3+ji1, ..., jn+ji1, generated element e0+ei2 will be connected to joints j1+ji2, j2+ji2, j3+ji2, ..., jn+ji2, generated element e0+ei3 will be connected to joints j1+ji3, j2+ji3, j3+ji3, ..., jn+ji3, and so on for the rest of the generated elements. Thus all joint labels for a generated element differ from those of the starting element by the same amount. See Figure 3 (page 19) for an example. For a Frame or Nllink element, suppose that element e0 is connected to joints i and j. Then generated element e0+ei1 will be connected to joints i1+ii1 and j+ji1, generated element e0+ei2 will be connected to joints i1+ii2 and j+ji2, generated element e0+ei3 will be connected to joints i1+ii3 and j+ji3, and so on for the rest of the generated elements. The default for the joint label increments ji1, ji2, and ji3 are the element label increments ei1, ei2, and ei3, respectively. The default for the joint label increments ii1, ii2, and ii3 are the joint label increments ji1, ji2, and ji3, respectively. Thus it is often convenient to assign element labels that are consistent with the joint labels.

Deletion
Deletion as described in this topic is used only in the input data text file. Skip this topic if you are preparing your model using the SAP2000 graphical user interface. Deletion is used to eliminate a regular array of previously-defined elements, Constraints, or Welds from the model. You can use a combination of generation and deletion to efficiently model a structure that has gaps or holes.

18

Deletion
36

Chapter II Labels, Arrays, and Generation

Figure 3 Three-dimensional Generation of Solid Elements

Joints cannot be deleted once they have been defined. However, the program will automatically ignore any unloaded joint that is not connected to an element or Constraint.

Assignment
Assignment as described in this topic is used only in the .S2K input data text file. Skip this topic if you are preparing your model using the SAP2000 graphical user interface. Assignment is used to define loads and properties for regular arrays of joints or elements. The joints or elements must have been previously defined. Three types of assignment may be available, depending upon the load or property being assigned: Addition: The specified load or property values are added to the current values for each joint or element in the array

Assignment
37

19

SAP2000 Analysis Reference Replacement: The specified load or property values replace the current values for each joint or element in the array Removal: The specified type of load or property is removed from (set to zero for) each joint or element in the array Loads are applied to all joints and elements by assignment. All joint properties are defined by assignment. A few element properties are defined by assignment, but most properties are given when the elements are explicitly defined or generated.

20

Assignment
38

C h a p t e r III

Coordinate Systems
Each structure may use many different coordinate systems to describe the location of points and the directions of loads, displacement, internal forces, and stresses. Understanding these different coordinate systems is crucial to being able to properly define the model and interpret the results. Basic Topics for All Users Overview Global Coordinate System Upward and Horizontal Directions Defining Coordinate Systems Local Coordinate Systems Advanced Topics Alternate Coordinate Systems Cylindrical and Spherical Coordinates

21
39

SAP2000 Analysis Reference

Overview
Coordinate systems are used to locate different parts of the structural model and to define the directions of loads, displacements, internal forces, and stresses. All coordinate systems in the model are defined with respect to a single global coordinate system. Each part of the model (joint, element, or constraint) has its own local coordinate system. In addition, you may create alternate coordinate systems that are used to define locations and directions. All coordinate systems are three-dimensional, right-handed, rectangular (Cartesian) systems. Vector cross products are used to define the local and alternate coordinate systems with respect to the global system. SAP2000 always assumes that Z is the vertical axis, with +Z being upward. The upward direction is used to help define local coordinate systems, although local coordinate systems themselves do not have an upward direction. The locations of points in a coordinate system may be specified using rectangular, cylindrical, or spherical coordinates. Likewise, directions in a coordinate system may be specified using rectangular, cylindrical, or spherical coordinate directions at a point.

Global Coordinate System


The global coordinate system is a three-dimensional, right-handed, rectangular coordinate system. The three axes, denoted X, Y, and Z, are mutually perpendicular and satisfy the right-hand rule. The location and orientation of the global system are arbitrary. The Z direction is normally upward, but this is not required. Locations in the global coordinate system can be specified using the variables x, y, and z. A vector in the global coordinate system can be specified by giving the locations of two points, a pair of angles, or by specifying a coordinate direction. Coordinate directions are indicated using the values X, Y, and Z. For example, +X defines a vector parallel to and directed along the positive X axis. The sign is required. All other coordinate systems in the model are ultimately defined with respect to the global coordinate system, either directly or indirectly. Likewise, all joint coordinates are ultimately converted to global X, Y, and Z coordinates, regardless of how they were specified.

22

Overview
40

Chapter III Coordinate Systems

Upward and Horizontal Directions


SAP2000 always assumes that Z is the vertical axis, with +Z being upward. Local coordinate systems for joints, elements, and ground-acceleration loading are defined with respect to this upward direction. Self-weight loading always acts downward, in the Z direction. The X-Y plane is horizontal. The primary horizontal direction is +X. Angles in the horizontal plane are measured from the positive half of the X axis, with positive angles appearing counterclockwise when you are looking down at the X-Y plane. The upward and horizontal directions apply to the global coordinate system and all alternate coordinate systems.

Defining Coordinate Systems


Each coordinate system to be defined must have an origin and a set of three, mutually-perpendicular axes that satisfy the right-hand rule. The origin is defined by simply specifying three coordinates in the global coordinate system. The axes are defined as vectors using the concepts of vector algebra. A fundamental knowledge of the vector cross product operation is very helpful in clearly understanding how coordinate system axes are defined.

Vector Cross Product


A vector may be defined by two points. It has length, direction, and location in space. For the purposes of defining coordinate axes, only the direction is important. Hence any two vectors that are parallel and have the same sense (i.e., pointing the same way) may be considered to be the same vector. Any two vectors, Vi and Vj, that are not parallel to each other define a plane that is parallel to them both. The location of this plane is not important here, only its orientation. The cross product of Vi and Vj defines a third vector, Vk, that is perpendicular to them both, and hence normal to the plane. The cross product is written as: V k = Vi Vj

Upward and Horizontal Directions


41

23

SAP2000 Analysis Reference The length of Vk is not important here. The side of the Vi-Vj plane to which Vk points is determined by the right-hand rule: The vector Vk points toward you if the acute angle (less than 180) from Vi to Vj appears counterclockwise. Thus the sign of the cross product depends upon the order of the operands: Vj Vi = Vi Vj

Defining the Three Axes Using Two Vectors


A right-handed coordinate system R-S-T can be represented by the three mutuallyperpendicular vectors Vr, Vs, and Vt, respectively, that satisfy the relationship: Vt = Vr Vs This coordinate system can be defined by specifying two non-parallel vectors: An axis reference vector, Va, that is parallel to axis R A plane reference vector, Vp, that is parallel to plane R-S, and points toward the positive-S side of the R axis The axes are then defined as: Vr = Va Vt = V r V p Vs = V t V r Note that Vp can be any convenient vector parallel to the R-S plane; it does not have to be parallel to the S axis. This is illustrated in Figure 4 (page 25).

Local Coordinate Systems


Each part (joint, element, or constraint) of the structural model has its own local coordinate system used to define the properties, loads, and response for that part. The axes of the local coordinate systems are denoted 1, 2, and 3. In general, the local coordinate systems may vary from joint to joint, element to element, and constraint to constraint. There is no preferred upward direction for a local coordinate system. However, the upward +Z direction is used to define the default joint and element local coordinate systems with respect to the global or any alternate coordinate system.

24

Local Coordinate Systems


42

Chapter III Coordinate Systems

Figure 4 Determining an R-S-T Coordinate System from Reference Vectors Va and Vp

The joint local 1-2-3 coordinate system is normally the same as the global X-Y-Z coordinate system. However, you may define any arbitrary orientation for a joint local coordinate system by specifying two reference vectors and/or three angles of rotation. For the Frame, Shell, and Nllink elements, one of the element local axes is determined by the geometry of the individual element. You may define the orientation of the remaining two axes by specifying a single reference vector and/or a single angle of rotation. The element local coordinate systems for the Plane and Asolid elements are always aligned with the global coordinate axes. The definition varies according to which global plane is parallel to the element. The Solid element local 1-2-3 coordinate system is always the same as the global X-Y-Z coordinate system. The local coordinate system for a Body, Diaphragm, Plate, Beam, or Rod Constraint is normally determined automatically from the geometry or mass distribution of the constraint. Optionally, you may specify one local axis for any Dia-

Local Coordinate Systems


43

25

SAP2000 Analysis Reference phragm, Plate, Beam, or Rod Constraint (but not for the Body Constraint); the remaining two axes are determined automatically. The local coordinate system for an Equal Constraint may be arbitrarily specified; by default it is the global coordinate system. The Local Constraint does not have its own local coordinate system. For more information: See Topic Local Coordinate System (page 78) in Chapter Joints and Degrees of Freedom. See Topic Local Coordinate System (page 148) in Chapter The Frame Element. See Topic Local Coordinate System (page 186) in Chapter The Shell Element. See Topic Local Coordinate System (page 210) in Chapter The Plane Element. See Topic Local Coordinate System (page 224) in Chapter The Asolid Element. See Topic Local Coordinate System (page 236) in Chapter The Solid Element. See Topic Local Coordinate System (page 245) in Chapter The Nllink Element. See Chapter Constraints and Welds (page 113).

Alternate Coordinate Systems


You may define alternate coordinate systems that can be used for locating the joints; for defining local coordinate systems for joints, elements, and constraints; and as a reference for defining other properties and loads. The axes of the alternate coordinate systems are denoted X, Y, and Z. The global coordinate system and all alternate systems are called fixed coordinate systems, since they apply to the whole structural model, not just to individual parts as do the local coordinate systems. Each fixed coordinate system may be used in rectangular, cylindrical or spherical form. The definition of the upward and horizontal directions for each alternate coordinate system is the same as for the global coordinate system.

26

Alternate Coordinate Systems


44

Chapter III Coordinate Systems

Figure 5 Definition of Alternate Coordinate System Using Three Points

Each alternate coordinate system is defined by specifying the location of three points in the global coordinate system: Point P0 at the origin of the new system Point P3 anywhere on the +Z half of the new Z axis Point P1 anywhere on the +X half of the new Z-X plane An axis reference vector, Va, is defined from point P0 to point P3, and a plane reference vector Vp is defined from point P0 to point P1. The new, positive X, Y and Z axes then have the directions of V1, V2, and V3, respectively, defined as: V3 = Va V2 = V 3 Vp V1 = V 2 V3 This is illustrated in Figure 5 (page 27).

Alternate Coordinate Systems


45

27

SAP2000 Analysis Reference

Cylindrical and Spherical Coordinates


The location of points in the global or an alternate coordinate system may be specified using polar coordinates instead of rectangular X-Y-Z coordinates. Polar coordinates include cylindrical CR-CA-CZ coordinates and spherical SB-SA-SR coordinates. See Figure 6 (page 29) for the definition of the polar coordinate systems. Polar coordinate systems are always defined with respect to a rectangular X-Y-Z system. The coordinates CR, CZ, and SR are lineal and are specified in length units. The coordinates CA, SB, and SA are angular and are specified in degrees. Locations are specified in cylindrical coordinates using the variables cr, ca, and cz. These are related to the rectangular coordinates as: cr = x 2 ca = tan -1 cz = z Locations are specified in spherical coordinates using the variables sb, sa, and sr. These are related to the rectangular coordinates as: sb = tan
-1

+y

y x

x2

+y

z y x
+y
2

sa = tan -1 sr = x 2

+z

A vector in a fixed coordinate system can be specified by giving the locations of two points or by specifying a coordinate direction at a single point P. Coordinate directions are tangential to the coordinate curves at point P. A positive coordinate direction indicates the direction of increasing coordinate value at that point. Cylindrical coordinate directions are indicated using the values CR, CA, and CZ. Spherical coordinate directions are indicated using the values SB, SA, and SR. The sign is required. See Figure 6 (page 29).

28

Cylindrical and Spherical Coordinates


46

Chapter III Coordinate Systems

Figure 6 Cylindrical and Spherical Coordinates and Coordinate Directions Cylindrical and Spherical Coordinates
47

29

SAP2000 Analysis Reference The cylindrical and spherical coordinate directions are not constant but vary with angular position. The coordinate directions do not change with the lineal coordinates. For example, +SR defines a vector directed from the origin to point P. Note that the coordinates Z and CZ are identical, as are the corresponding coordinate directions. Similarly, the coordinates CA and SA and their corresponding coordinate directions are identical.

30

Cylindrical and Spherical Coordinates


48

C h a p t e r IV

Joint Coordinates
This chapter describes the definition of the joints and their use in defining the geometry of the structure. This chapter is only of interest if you are preparing your model using the input data text file. See Chapter Joints and Degrees of Freedom (page 75) for more general information about the joints. Basic Topics for All Users Overview Advanced Topics Joint Definition One-dimensional Joint Generation Two-dimensional Joint Generation Three-dimensional Joint Generation Variable Joint Spacing Joint Definition in Polar Coordinates Joint Definition in Alternate Coordinate Systems

31
49

SAP2000 Analysis Reference

Overview
Joints, also known as nodal points or nodes, are a fundamental part of every structural model. Joints perform a variety of functions, which are discussed in Chapter Joints and Degrees of Freedom (page 75). The method used to define the structural model affects how joints are created: Using the SAP2000 graphical interface joints are automatically created at the ends of each Frame or Nllink element and at the corners of each Shell element; additional joints may also be defined independently of any element Using the input data text file joint locations must be explicitly defined in order to describe the geometry of the structure; these joints are then connected by elements to build the structure The use of joints to define the geometry of the structure is discussed in this chapter. If you are using the SAP2000 graphical user interface, you may skip the rest of this chapter. A variety of methods are available in the input data text file to define the layout of the joints, and hence the geometry of the structure. Joints may be located in arbitrary coordinate systems using rectangular, cylindrical, or spherical coordinates. Large arrays of joints may be generated in one, two, and three dimensions after defining a smaller number of joints at the corners or along the edges of the region.

Joint Definition
Joint definition as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. A joint is defined by specifying its label, j, and three spatial coordinates, x, y, z, that locate the joint in space. You may define a joint individually, or use a generation operation that defines many joints on a line (or curve), a surface, or throughout a three-dimensional region. You may define extraneous joints in the system, for geometric reference, that do not attach to any elements. The program will automatically restrain them.

32

Overview
50

Chapter IV Joint Coordinates

One-dimensional Joint Generation


Joint generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. It is often useful to define a large number of equally-spaced joints along a straight line or curve. This is easily done with the generation options in SAP2000 if the labels of the joints increment in a uniform fashion along the line or curve. Such a set of joints is called a one-dimensional joint array. Each generation option requires the specification of the labels for the joint array. This is described in Subtopic One-dimensional Joint Array Specification below. The additional information that is required to define the location of the joints is described in the subtopics that follow. See Subtopic One-dimensional Regular Arrays (page 11) in Chapter Labels, Arrays, and Generation for more information.

One-dimensional Joint Array Specification


The following information is required to specify a one-dimensional array of joints: The label, j0, of the joint at the start of the array The label, j1, of the joint at the end of the array The label increment (constant difference), ji1, between any pair of successive joints along the line The number of spaces, n1, between the joints along the line is then given by:
n1

j1 j0 ji1

The labels and label increment must be chosen such that n1 is a whole, positive number. The total number of joints in the array, including the ends, is n1+1.

One-dimensional Joint Definition


A set of equally-spaced joints along a straight line can be directly defined by specifying:

One-dimensional Joint Generation


51

33

SAP2000 Analysis Reference The information in Subtopic One-dimensional Joint Array Specification above The three spatial coordinates, (x0, y0, z0), for joint j0 The three spatial coordinates, (x1, y1, z1), for joint j1 The spatial coordinates of the intermediate joints are determined by linear interpolation such that the constant coordinate differences, x 1 , y 1 , and z 1 , between any pair of successive joints are given by: x 1 = x1 x0
n1

y 1 =

y1 y0
n1

and

z 1 =

z 1 z0
n1

See Figure 7 (page 35) for an example of one-dimensional joint definition.

One-dimensional Linear Generation


One-dimensional linear generation is the same as one-dimensional joint array definition, except that the joints at the two ends must have previously been defined: 1. Joint j0. The spatial coordinates, (x0, y0, z0), for this joint are known 2. Joint j1. The spatial coordinates, (x1, y1, z1), for this joint are known It is then only necessary to provide the information in Subtopic One-dimensional Joint Array Specification above. The spatial coordinates of the intermediate joints are determined by linear interpolation such that the constant coordinate differences, x 1 , y 1 , and z 1 , between any pair of successive joints are given by: x 1 = x1 x0
n1

y 1 =

y1 y0
n1

and

z 1 =

z1 z0
n1

One-dimensional Cylindrical Generation


One-dimensional cylindrical generation can be used to generate joints along circular, helical, and spiral curves. Before using cylindrical generation, the starting joint j0 must have been previously defined; the spatial coordinates, (x0, y0, z0), for this joint are known. The following additional information must be specified: 1. The information in Subtopic One-dimensional Joint Array Specification above.

34

One-dimensional Joint Generation


52

Chapter IV Joint Coordinates

Figure 7 One-dimensional Joint Array Definition One-dimensional Joint Generation


53

35

SAP2000 Analysis Reference 2. A pair of previously-defined joints, axveca and axvecb, that define the axis about which the generation is performed. The positive direction of this axis is from axveca to axvecb. Joint j0 must not lie on this axis. 3. The angle increment, da, between successive joints in the array. The positive direction of da is given by the right-hand rule with respect to the axis of generation. 4. For spiral generation, the radius increment, dr, between successive joints in the array. The positive direction of dr is away from the axis of generation. 5. For helical generation, the axial increment, dz, between successive joints in the array. The positive direction of dz is along the positive direction of the axis of generation. The spatial coordinates of each generated joint are obtained from the previous joint in the array by moving a distance dz parallel to the axis of generation, moving a distance dr away from the axis, and moving through an angle da about the axis. A circular arc is defined if dr and dz are both zero. A spiral is defined if dr is nonzero and dz is zero. A helix is defined if dz is nonzero; it is a cylindrical helix if dr is zero, and a conical helix otherwise. If da is zero, a straight line is defined.

Two-dimensional Joint Generation


Joint generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Two-dimensional joint generation can be used to define large numbers of joints over four-sided regions on flat or curved surfaces. All that is required is that the labels of the joints increment in a uniform fashion in two separate directions. Such a set of joints is called a two-dimensional joint array. A variety of generation options is available, differing primarily in the amount of information that must be specified and therefore the complexity of the shapes that can be described. Each generation option requires the specification of the labels for the joint array. This is described in Subtopic Two-dimensional Joint Array Specification below. The additional information that is required to define the location of the joints is described in the subtopics that follow.

36

Two-dimensional Joint Generation


54

Chapter IV Joint Coordinates See Subtopic Two-dimensional Regular Arrays (page 12) in Chapter Labels, Arrays, and Generation for more information.

Two-dimensional Joint Array Specification


The following information is required to specify a two-dimensional array of joints: The label, j0, of the joint at one corner The label, j1, of the joint at the end of one side that contains joint j0 The label, j2, of the joint at the end of the other side that contains joint j0 The label increment (constant difference), ji1, between any pair of successive joints along side j0-j1 The label increment (constant difference), ji2, between any pair of successive joints along side j0-j2 The numbers of spaces, n1 and n2, between the joints along directions parallel to sides j0-j1 and j0-j2, respectively, are then given by:
n1

j1 j0 ji1

and

n2

j2 j0 ji2

The labels and label increments must be chosen such that n1 and n2 are whole, positive numbers. The total number of joints in the array is ( n1 + 1) ( n 2 + 1).

Two-dimensional Joint Definition


A set of equally-spaced joints on a rectangular or parallelogram-shaped region of a flat surface can be defined by specifying: The information in Subtopic Two-dimensional Joint Array Specification above The three spatial coordinates, (x0, y0, z0), for joint j0 The three spatial coordinates, (x1, y1, z1), for joint j1 The three spatial coordinates, (x2, y2, z2), for joint j2 The spatial coordinates of the remaining joints are determined by linear interpolation and extrapolation. The constant coordinate differences, x 1 , y 1 , and z 1 , between any pair of joints whose labels differ by ji1 are given by:

Two-dimensional Joint Generation


55

37

SAP2000 Analysis Reference x 1 = x1 x0


n1

y 1 =

y1 y0
n1

and

z 1 =

z1 z0
n1

Similarly, the constant coordinate differences, x 2 , y 2 , and z 2 , between any pair of joints whose labels differ by ji2 are given by: x 2 = x2 x0
n2

y 2 =

y2 y0
n2

and

z 2 =

z2 z0
n2

Two-dimensional Linear Generation


Two-dimensional linear generation creates a set of joints on a straight-sided quadrilateral surface determined by four previously-defined joints at the corners. If the four corner joints do not lie in a plane, the surface is called a hyperbolic-paraboloid, commonly known as a hypar (Timoshenko and Woinowsky-Krieger, 1959). The generated joints are equally spaced along each of the four sides of the surface. Before using two-dimensional linear generation, the four corner joints must have already been defined: 1. Joint j0 2. Joint j1 3. Joint j2 4. Joint j1+j2j0 at the corner opposite joint j0 It is then only necessary to provide the information in Subtopic Two-dimensional Joint Array Specification above. The spatial coordinates of the remaining joints are determined by standard bilinear interpolation (Cook, Malkus, and Plesha, 1989). Any subset of the generated joints having labels that increment uniformly by ji1 or ji2 will lie on a straight line and will be equally spaced. See Figure 8 (page 39) for an example.

Two-dimensional Frontal Generation


Two-dimensional frontal generation creates a set of joints on a four-sided surface determined by previously-defined joints along two adjacent sides. These

38

Two-dimensional Joint Generation


56

Chapter IV Joint Coordinates

Figure 8 Two-dimensional Linear Generation

Two-dimensional Joint Generation


57

39

SAP2000 Analysis Reference previously-defined sides may have any shape, and the joint spacing may be uniform or nonuniform. The generated joints lie along lines or curves that are parallel to the two defining sides. The spacing between the lines or curves is determined by the joints on the two sides. Before using two-dimensional frontal generation, all joints on the two adjacent sides must have already been defined: 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 It is then only necessary to provide the information in Subtopic Two-dimensional Joint Array Specification above. See Figure 9 (page 41) for an example of frontal generation used for a planar building frame.

Two-dimensional Edge Generation


Two-dimensional edge generation creates a set of joints over a complex, four-sided surface determined by previously-defined joints along all four sides. The previously-defined sides may have any shape, and the joint spacing may be uniform or nonuniform. The joints in the interior of the surface are generated using serendipity polynomial interpolation (Cook, Malkus, and Plesha, 1989). Before using two-dimensional edge generation, all joints on the four sides must have already been defined: 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 3. Joints j2 to j1+j2j0 by increment ji1 4. Joints j1 to j1+j2j0 by increment ji2 It is then only necessary to provide the information in Subtopic Two-dimensional Joint Array Specification above. See Figure 10 (page 42) for an example of two-dimensional edge generation.

40

Two-dimensional Joint Generation


58

Chapter IV Joint Coordinates

Figure 9 Two-dimensional Frontal Generation

Two-dimensional Joint Generation


59

41

SAP2000 Analysis Reference

Figure 10 Two-dimensional Edge Generation

42

Two-dimensional Joint Generation


60

Chapter IV Joint Coordinates

Three-dimensional Joint Generation


Joint generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Three-dimensional joint generation can be used to define large numbers of joints over three-dimensional regions enclosed by hexahedra, i.e., having six faces. All that is required is that the labels of the joints increment in a uniform fashion in three separate directions. Such a set of joints is called a three-dimensional joint array. A variety of generation options is available, differing primarily in the amount of information that must be specified and therefore the complexity of the shapes that can be described. Each generation option requires the specification of the labels for the joint array. This is described in Subtopic Three-dimensional Joint Array Specification below. The additional information that is required to define the location of the joints is described in the subtopics that follow. Three-dimensional joint generation is a generalization of two-dimensional joint generation. For more information: See Topic Two-dimensional Joint Generation (page 36) in this chapter. See Subtopic Three-dimensional Regular Arrays (page 13) in Chapter Labels, Arrays, and Generation.

Three-dimensional Joint Array Specification


The following information is required to specify a three-dimensional array of joints: The label, j0, of the joint at one corner The label, j1, of the joint at the end of one side that contains joint j0 The label, j2, of the joint at the end of another side that contains joint j0 The label, j3, of the joint at the end of the third side that contains joint j0 The label increment (constant difference), ji1, between any pair of successive joints along side j0-j1 The label increment (constant difference), ji2, between any pair of successive joints along side j0-j2 Three-dimensional Joint Generation
61

43

SAP2000 Analysis Reference The label increment (constant difference), ji3, between any pair of successive joints along side j0-j3 The numbers of spaces, n1, n2, and n3, between the joints along directions parallel to sides j0-j1, j0-j2, and j0-j3, respectively, are then given by:
n1

j1 j0 ji1

n2

j2 j0 ji2

and

n3

j3 j0 ji3

The labels and label increments must be chosen such that n1, n2, and n3 are whole, positive numbers. The total number of joints in the array is ( n1 + 1) ( n 2 + 1) ( n 3 + 1).

Three-dimensional Joint Definition


A set of equally-spaced joints within a parallelepiped (the six faces are parallelograms) can be defined by specifying: The information in Subtopic Three-dimensional Joint Array Specification above The three spatial coordinates, (x0, y0, z0), for joint j0 The three spatial coordinates, (x1, y1, z1), for joint j1 The three spatial coordinates, (x2, y2, z2), for joint j2 The three spatial coordinates, (x3, y3, z3), for joint j3 The spatial coordinates of the remaining joints are determined by linear interpolation and extrapolation. The constant coordinate differences, x 1 , y 1 , and z 1 , between any pair of joints whose labels differ by ji1 are given by: x 1 = x1 x0
n1

y 1 =

y1 y0
n1

and

z 1 =

z1 z0
n1

Similarly, the constant coordinate differences, x 2 , y 2 , and z 2 , between any pair of joints whose labels differ by ji2 are given by: x 2 = x2 x0
n2

y 2 =

y2 y0
n2

and

z 2 =

z2 z0
n2

Finally, the constant coordinate differences, x 3 , y 3 , and z 3 , between any pair of joints whose labels differ by ji3 are given by:

44

Three-dimensional Joint Generation


62

Chapter IV Joint Coordinates x 3 = x3 x0


n3

y 3 =

y3 y0
n3

and

z 3 =

z3 z0
n3

Three-dimensional Linear Generation


Three-dimensional linear generation creates a set of joints over a three-dimensional region enclosed by six quadrilateral faces. Each of the twelve edges is a straight line. The shape of the region is determined by eight previously-defined joints at the corners. If the four corner joints of a face do not lie in a plane, the face is a surface called a hyperbolic-paraboloid, commonly known as a hypar (Timoshenko and Woinowsky-Krieger, 1959). The generated joints are equally spaced along each of the twelve edges of the region. Before using three-dimensional linear generation, the eight corner joints must have already been defined: 1. Joint j0 2. Joint j1 3. Joint j2 4. Joint j3 5. Joint j1+j2j0 at the corner opposite joint j3 6. Joint j1+j3j0 at the corner opposite joint j2 7. Joint j2+j3j0 at the corner opposite joint j1 8. Joint j1+j2+j3j0j0 at the corner opposite joint j0 It is then only necessary to provide the information in Subtopic Three-dimensional Joint Array Specification above. The spatial coordinates of the remaining joints are determined by standard trilinear interpolation (Cook, Malkus, and Plesha, 1989). Any subset of the generated joints having labels that increment uniformly by ji1, ji2, or ji3 will lie on a straight line and will be equally spaced.

Three-dimensional Joint Generation


63

45

SAP2000 Analysis Reference

Three-dimensional Frontal Generation


Three-dimensional frontal generation creates a set of joints over a threedimensional region determined by previously-defined joints along three adjacent sides. These previously-defined sides may have any shape, and the joint spacing may be uniform or nonuniform. The generated joints lie along lines or curves that are parallel to the three defining sides. The spacing between the lines or curves is determined by the joints on the three sides. Before using three-dimensional frontal generation, all joints on the three adjacent sides must have already been defined: 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 3. Joints j0 to j3 by increment ji3 It is then only necessary to provide the information in Subtopic Three-dimensional Joint Array Specification above.

Three-dimensional Edge Generation


Three-dimensional edge generation creates a set of joints over a complex, six-faced volume determined by previously-defined joints along all twelve edges. The previously-defined edges may have any shape, and the joint spacing may be uniform or nonuniform. The joints in the interior of the volume and the faces are generated using serendipity polynomial interpolation (Zienkiewicz and Taylor, 1989). Before using three-dimensional frontal generation, all joints on the twelve edges must have already been defined: 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 3. Joints j0 to j3 by increment ji3 4. Joints j1 to j1+j2j0 by increment ji2 5. Joints j1 to j1+j3j0 by increment ji3

46

Three-dimensional Joint Generation


64

Chapter IV Joint Coordinates 6. Joints j2 to j1+j2j0 by increment ji1 7. Joints j2 to j2+j3j0 by increment ji3 8. Joints j3 to j1+j3j0 by increment ji1 9. Joints j3 to j2+j3j0 by increment ji2 10. Joints j1+j2j0 to j1+j2+j3j0j0 by increment ji3 11. Joints j1+j3j0 to j1+j2+j3j0j0 by increment ji2 12. Joints j2+j3j0 to j1+j2+j3j0j0 by increment ji1 It is then only necessary to provide the information in Subtopic Two-dimensional Joint Array Specification above.

Variable Joint Spacing


Variable joint spacing as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Joints created by array definition or linear generation normally have uniform spacing along any straight line, provided that the joint labels differ by one of the specified label increments. Sometimes, however, it is desirable to have the spacing between the joints vary gradually along these straight lines. This is often the case in finite element meshes where smaller elements are needed near a region of interest, such as at a stress concentration, and larger elements are to be used farther away for computational efficiency. Such a mesh is called a graded mesh. An option is available to specify how the joint spacing is to vary along each generation dimension. The joint spacing changes by arithmetic progression so that the difference between successive joint spacings is constant. This option is only available for array definition and linear generation. For frontal generation and edge generation, joint spacing is determined by the locations of the joints on the previously-defined edges. This option may also be used for array definition in polar coordinates. In this case, it is the polar-coordinate differences, rather than the absolute joint spacing, that varies along the generation dimensions. Variable Joint Spacing
65

47

SAP2000 Analysis Reference For more information: See Subtopic One-dimensional Joint Definition (page 33) in this chapter. See Subtopic One-dimensional Linear Generation (page 34) in this chapter. See Subtopic Two-dimensional Joint Definition (page 37) in this chapter. See Subtopic Two-dimensional Linear Generation (page 38) in this chapter. See Subtopic Three-dimensional Joint Definition (page 44) in this chapter. See Subtopic Three-dimensional Linear Generation (page 45) in this chapter. See Topic Joint Definition in Polar Coordinates (page 50) in this chapter.

One Dimension
The parameter ratio1 can be used to specify how the joint spacing varies along the line of generation. It gives the ratio of the last spacing, d 1n , between joints j1ji1 and j1, and the first spacing, d 11 , between joints j0 and j0+ji1: ratio1 =
d1

d1

If the value of ratio1 is greater than unity, then the distance between joints increases along the length. Connected elements would be smaller at the start of the line, and larger at the end of the line. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing. See Figure 7 (page 35) for an example.

Two Dimensions
The two parameters ratio1 and ratio2 can be used to specify how the joint spacing varies along two generation dimensions. Parameter ratio1 is defined above. Parameter ratio2 is defined similarly: it gives the ratio of the last spacing, d 2n , between joints j2ji2 and j2, and the first spacing, d 21 , between joints j0 and j0+ji2: ratio2 =
d2

d2

If the value of ratio1 is greater than unity, then the distance between joints increases along any straight line where the joint labels increment by ji1. Connected

48

Variable Joint Spacing


66

Chapter IV Joint Coordinates elements would be smaller toward side j0-j2. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing. Similarly, if the value of ratio2 is greater than unity, then the distance between joints increases along any straight line where the joint labels increment by ji2. Connected elements would be smaller toward side j0-j1. If the value of ratio2 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing.

Three Dimensions
The three parameters ratio1, ratio2 and ratio3 can be used to specify how the joint spacing varies along three generation dimensions. Parameters ratio1 and ratio2 are defined above. Parameter ratio3 is defined similarly: it gives the ratio of the last spacing, d 3 n , between joints j3ji3 and j3, and the first spacing, d 31 , between joints j0 and j0+ji3: ratio3 =
d3 d3

n 1

If the value of ratio1 is greater than unity, then the distance between joints increases along any straight line where the joint labels increment by ji1. Connected elements would be smaller toward face j0-j2-j3. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing. Similarly, if the value of ratio2 is greater than unity, then the distance between joints increases along any straight line where the joint labels increment by ji2. Connected elements would be smaller toward face j0-j1-j3. If the value of ratio2 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing. Finally, if the value of ratio3 is greater than unity, then the distance between joints increases along any straight line where the joint labels increment by ji3. Connected elements would be smaller toward face j0-j1-j2. If the value of ratio3 is less than unity, then the reverse would be true. The default value of unity indicates uniform joint spacing.

Variable Joint Spacing


67

49

SAP2000 Analysis Reference

Joint Definition in Polar Coordinates


Joint definition in polar coordinates as described in this topic is used only in the input data text file. Cylindrical coordinates are available in the SAP2000 graphical user interface through the use of cylindrical grids. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. The location of the joints may be specified using polar coordinates instead of rectangular X-Y-Z coordinates. Polar coordinates include cylindrical CR-CA-CZ coordinates and spherical SB-SA-SR coordinates. See Figure 11 (page 51) for the definition of the polar coordinate systems. Polar coordinate systems are always defined with respect to a rectangular X-Y-Z system. Polar coordinates may be used for the definition of a single joint or for the definition of joint arrays in one, two, or three dimensions. When defining a joint array, all specified coordinates must be of the same type: rectangular, cylindrical, or spherical. Linear generation, frontal generation, and edge generation are always performed in rectangular coordinates, even if polar coordinates were used to locate the previously-defined joints. The location of each joint is converted to rectangular coordinates in the global coordinate system for printing in the .EKO output file. For more information: See Chapter Coordinate Systems (page 21). See Topic Joint Definition (page 32) in this chapter. See Subtopic One-dimensional Joint Definition (page 33) in this chapter. See Subtopic Two-dimensional Joint Definition (page 37) in this chapter. See Subtopic Three-dimensional Joint Definition (page 44) in this chapter. See Topic Joint Definition in Alternate Coordinate Systems (page 54) in this chapter.

Cylindrical Coordinates
The following coordinate values must be specified when defining joints in cylindrical coordinates: For a single joint, j, the coordinates (cr, ca, cz)

50

Joint Definition in Polar Coordinates


68

Chapter IV Joint Coordinates

Figure 11 Locating a Joint Using Cylindrical or Spherical Coordinates Joint Definition in Polar Coordinates
69

51

SAP2000 Analysis Reference For one-, two-, or three-dimensional joint definition, the coordinates (cr0, ca0, cz0) of joint j0 and coordinates (cr1, ca1, cz1) of joint j1 For two- or three-dimensional joint definition, the coordinates (cr2, ca2, cz2) of joint j2 For three-dimensional joint definition, the coordinates (cr3, ca3, cz3) of joint j3 The spatial coordinates of the remaining joints are determined by linear interpolation and extrapolation of the cylindrical coordinate values. The constant coordinate differences, cr1 , ca1 , and cz 1 , between any pair of joints whose labels differ by ji1 are given by: cr1 = cr1 cr0
n1

ca1 =

ca1 ca0
n1

and

cz 1 =

cz1 cz0
n1

Similar relationships exist for the constant coordinate differences between pairs of joints whose labels differ by ji2 or ji3 for two- and three-dimensional joint arrays. See Figure 12 (page 53) for an example of one-dimensional joint array definition.

Spherical Coordinates
The following coordinate values must be specified when defining joints in spherical coordinates: For a single joint, j, the coordinates (sb, sa, sr) For one-, two-, or three-dimensional joint definition, the coordinates (sb0, sa0, sr0) of joint j0 and coordinates (sb1, sa1, sr1) of joint j1 For two- or three-dimensional joint definition, the coordinates (sb2, sa2, sr2) of joint j2 For three-dimensional joint definition, the coordinates (sb3, sa3, sr3) of joint j3 The spatial coordinates of the remaining joints are determined by linear interpolation and extrapolation of the spherical coordinate values. The constant coordinate differences, sb1 , sa1 , and sr1 , between any pair of joints whose labels differ by ji1 are given by: sb1 = sb1 sb0
n1

sa1 =

sa1 sa0
n1

and

sr1 =

sr1 sr0
n1

52

Joint Definition in Polar Coordinates


70

Chapter IV Joint Coordinates

Figure 12 One-dimensional Joint Array Definition in Cylindrical Coordinates

Similar relationships exist for the constant coordinate differences between pairs of joints whose labels differ by ji2 or ji3 for two- and three-dimensional joint arrays.

Variable Joint Spacing in Polar Coordinates


The coordinate differences are normally constant between joints generated using one-, two-, three-dimensional joint array definition in polar coordinates. More specifically, equal cylindrical or spherical coordinate differences exist between the joints along any given path where the joint labels increment by ji1, ji2, or ji3. This does not correspond to constant spacing between the joints along this path unless only a single coordinate value is changing. It is sometimes desirable to create a graded mesh where smaller elements are defined in one portion of the region. The parameters ratio1, ratio2, and ratio3 can be specified for this purpose when using one-, two-, or three-dimensional joint array definition in polar coordinates.

Joint Definition in Polar Coordinates


71

53

SAP2000 Analysis Reference In cylindrical coordinates, parameter ratio1 gives the ratio of the coordinate differences, cr1n , ca1n , and cz 1n , between joints j1ji1 and j1, and the coordinate differences, cr11 , ca11 , and cz 11 , between joints j0 and j0+ji1: ratio1 = cr1n cr11 = ca1n ca11 = cz 1n cz 11

In spherical coordinates the coordinate differences are sb1n , sa1n , and sr1n and sb11 , sa11 , and sr11 , respectively, and: ratio1 = sb1n sb11 = sa1n sa11 = sr1n sr11

Similar definitions apply for ratio2 and ratio3 for joints whose labels increment by ji2 and ji3, respectively, in two- and three-dimensional joint arrays. The coordinate differences in each case change by arithmetic progression so that the difference between successive coordinate differences is constant. See Topic Variable Joint Spacing (page 47) in this chapter for more information.

Joint Definition in Alternate Coordinate Systems


Joint coordinates are normally given in the global coordinate system. However, it is sometimes convenient to define different parts of a structure in separate coordinate systems. Each coordinate system has its own origin and orientation of the axes. For example, you might use a separate coordinate system for each wing of a building or each bent of a bridge. The parameter csys is used to specify which coordinate system is being used to define joint coordinates: If csys=0, the global coordinate system is used; this is the default Otherwise, csys is the label of an Alternate Coordinate System A different value of csys may be used for each joint or set of joints. The coordinates of a joint may be specified in rectangular, cylindrical, or spherical coordinates with respect to coordinate system csys. The choice of the coordinate system affects only the definition of individual joints and joint arrays in one, two, or three dimensions. Linear generation, frontal genera-

54

Joint Definition in Alternate Coordinate Systems


72

Chapter IV Joint Coordinates tion, and edge generation are always performed in rectangular coordinates in the global coordinate system. The location of each joint is converted to rectangular coordinates in the global coordinate system for printing in the .EKO output file. It is important to understand that the use of an Alternate Coordinate System to locate a joint has no effect upon the local coordinate system defined for that joint. For more information: See Chapter Coordinate Systems (page 21). See Topic Joint Definition (page 32) in this chapter. See Subtopic One-dimensional Joint Definition (page 33) in this chapter. See Subtopic Two-dimensional Joint Definition (page 37) in this chapter. See Subtopic Three-dimensional Joint Definition (page 44) in this chapter. See Topic Joint Definition in Polar Coordinates (page 50) in this chapter.

Joint Definition in Alternate Coordinate Systems


73

55

SAP2000 Analysis Reference

56

Joint Definition in Alternate Coordinate Systems


74

Chapter V

Joint Patterns
Joint Patterns can be used to describe complicated spatial distributions of loads and properties over the structure. Patterns are optional and are not required for simple problems. This chapter describes the definition of the joint Patterns. Their use is described in Chapter Joints and Degrees of Freedom (page 75) and Chapter Load Cases (page 279). Advanced Topics Overview Pattern Definition Gradient or Hydrostatic Specification One-dimensional Pattern-value Generation Two-dimensional Pattern-value Generation Three-dimensional Pattern-value Generation Variable Pattern-value Increments

57
75

SAP2000 Analysis Reference

Overview
A joint Pattern consists of a set of scalar numeric values, one value for each joint of the structure. A Pattern can be used to describe how certain loads and properties vary over the structure: Joint spring and mass properties Joint force and displacement loads Pressure-type loads Temperature-type loads Patterns are most effective for describing complicated spatial distributions of numeric values. Their use is optional and is not required for simple problems. Joint Patterns do not affect the location of the joints, or have any other effect upon the structure except as utilized in the specification of loads and properties. Pattern values are assigned to previously-defined joints. The specification of Pattern values is very similar to the definition of the joint coordinates, except that only a single value, v, is assigned to a joint in a Pattern, whereas three coordinate values, (x, y, z), are assigned during joint definition. You may define as many Patterns as you need, each having a single value at every joint. The generation options available for specifying joint Pattern values correspond exactly to the generation options for joint coordinates. If the given Pattern values for a generation specification are proportional to the x, y, or z coordinates of the joints, then all generated values will also be proportional to x, y, or z, respectively. This is guaranteed by the properties of isoparametric interpolation (Cook, Malkus, and Plesha, 1989). Pattern values may also be specified that vary linearly in a given gradient direction from zero at a given datum point. An option is available to permit only positive or only negative values to be defined. This is useful for defining hydrostatic pressure distributions. Since Pattern values are scalar quantities, they are independent of any coordinate system. For more information: See Chapter Joints and Degrees of Freedom (page 75). See Chapter Load Cases (page 279).

58

Overview
76

Chapter V Joint Patterns

Pattern Definition
For each Pattern to be defined, you must assign a label that will be referenced whenever you later use the Pattern to describe the variation of a load or property. Each Pattern starts with a numeric value of zero at every joint. You then define the numeric values at the joints using a combination of addition, resetting, and mapping specifications as needed: Each addition specification adds numeric values to one or more joints. The assigned values are algebraically added to the values that are already present at the joints. Each resetting specification resets the numeric values to zero at one or more joints. Each mapping specification interpolates or extrapolates numeric values from certain joints in an array to the remaining joints in the array. Mapping is not additive, but overwrites previous numeric values at the joints. These specifications are applied separately to each Pattern being defined. Addition and resetting specifications may be applied to individual joints or to arrays of joints. Mapping specifications apply only to joint arrays. The application of these specifications to arrays of one, two, and three dimensions is described in the remainder of this chapter.

Gradient or Hydrostatic Specification


You may define Pattern values for a given array of joints that vary linearly along a specified gradient direction from zero value at a specified datum point. You have the option of allowing only negative values, only positive values, or both to be generated this way. For example, only positive values would be desired when defining a Pattern for hydrostatic pressure. Any Pattern values defined by a gradient specification are algebraically additive to the values already present at the joints of the given array. The following parameters are needed for a gradient specification: A fixed coordinate system, csys (the default is zero, indicating the global coordinate system) The components of the gradient, vx, vy, and vz, in coordinate system csys Pattern Definition
77

59

SAP2000 Analysis Reference The coordinates of the zero datum point, x, y, and z, in coordinate system csys The parameter setzero which can take on values: NEG, indicating to set negative values to zero POS, indicating to set positive values to zero NO, indicating to set no values to zero (this is the default) The component vx indicates, for example, how much the Pattern value changes per unit of distance parallel to the X axis in coordinate system csys. The Pattern value, vj, defined for a joint j that has coordinates (xj, yj, zj) is given by: vj = vx (xj x) + vy (yj y) + vz (zj z) if vj < 0 and setzero = NEG, then vj = 0 if vj > 0 and setzero = POS, then vj = 0 In most cases, the gradient will be parallel to one of the coordinate axes, and only one term in the above equation is needed. For example, consider a hydrostatic pressure distribution caused by water impounded behind a dam as shown in Figure 13 (page 61). The Z direction is up in the global coordinate system. The pressure gradient is simply given by the fluid weight density acting in the downward direction. Therefore, vx = 0, vy = 0 , and vz = 62.4 3 3 lb/ft or 9810 N/m . The zero-pressure datum can be any point on the free surface of the water. Thus z should be set to the elevation of the free surface in feet or meters, as appropriate. For hydrostatic pressure, the parameter setzero should be specified as NEG so that any joints above the free surface will be assigned a zero value for pressure. (Eqn. 1)

One-dimensional Pattern-value Generation


Pattern-value generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Pattern values can be generated for a one-dimensional array of joints, i.e., a set of joints whose labels increment uniformly along a line or curve. The generation options include addition, resetting, and linear mapping.

60

One-dimensional Pattern-value Generation


78

Chapter V Joint Patterns

Figure 13 Example of a Hydrostatic Pressure Pattern

Each generation option requires the specification of a joint array for which the Pattern values are to be generated, as described below. No other information is required for a resetting specification: the Pattern values for all joints in the array are set to zero. The additional information that is required for addition and linearmapping specifications is described in the subtopics below. See Subtopic One-dimensional Regular Arrays (page 11) in Chapter Labels, Arrays, and Generation for more information.

Joint Array Specification


Every one-dimensional generation specification requires the following information to describe the array of joints to which Pattern values are being assigned: The label, j0, of the joint at the start of the line or curve The label, j1, of the joint at the end of the line or curve The label increment (constant difference), ji1, between any pair of successive joints along the line or curve The number of spaces, n1, between the joints along the line or curve is then given by:

One-dimensional Pattern-value Generation


79

61

SAP2000 Analysis Reference = j1 j0 ji1

n1

The labels and label increment must be chosen such that n1 is a whole, positive number. The total number of joints in the array, including the ends, is n1+1.

One-dimensional Addition
A set of uniformly incrementing Pattern values can be added to the joints along a line or curve by specifying: The information in Subtopic Joint Array Specification above The Pattern value, v0, to be added to joint j0 at the start of the line or curve The Pattern value, v1, to be added to joint j1 at the end of the line or curve The Pattern values added to the intermediate joints are determined by linear interpolation from these values such that the constant difference, v 1 , between the values added to any pair of successive joints is given by: v 1 = v1 v0
n1

The computed values are algebraically added to the values that are already present at the joints.

One-dimensional Gradient Addition


A set of Pattern values determined from a constant gradient can be added to the joints along a line or curve by specifying: The information in Subtopic Joint Array Specification above The information in Topic Gradient or Hydrostatic Specification The Pattern values added to the joints are determined by Equation 1 in Topic Gradient or Hydrostatic Specification (page 59) in this chapter. The computed values are algebraically added to the values that are already present at the joints.

62

One-dimensional Pattern-value Generation


80

Chapter V Joint Patterns

One-dimensional Linear Mapping


One-dimensional linear mapping interpolates (maps) Pattern values from existing values at the ends of a line or curve. Before using this option, the Pattern values at the following joints must have already been defined (or allowed to default to zero): 1. Joint j0 at the start of the line or curve 2. Joint j1 at the end of the line or curve It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the intermediate joints are determined by linear interpolation such that the constant value difference, v 1 , between any pair of successive joints is given by: v 1 = v1 v0
n1

The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Two-dimensional Pattern-value Generation


Pattern-value generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Pattern values can be generated for a two-dimensional array of joints, i.e., a set of joints whose labels increment in a uniform fashion in two separate directions. This will typically correspond to a four-sided region on a flat or curved surface. The generation options include addition, resetting, and several types of mapping. Each generation option requires the specification of a joint array for which the Pattern values are to be generated. This is described in Subtopic Joint Array Specification below. No other information is required for a resetting specification: the Pattern values for all joints in the array are set to zero. The additional information that is required for addition and mapping specifications is described in the remaining subtopics below. See Subtopic Two-dimensional Regular Arrays (page 12) in Chapter Labels, Arrays, and Generation for more information. Two-dimensional Pattern-value Generation
81

63

SAP2000 Analysis Reference

Joint Array Specification


Each two-dimensional generation specification requires the following information to describe the array of joints to which Pattern values are being assigned: The label, j0, of the joint at one corner The label, j1, of the joint at the end of one side that contains joint j0 The label, j2, of the joint at the end of the other side that contains joint j0 The label increment (constant difference), ji1, between any pair of successive joints along side j0-j1 The label increment (constant difference), ji2, between any pair of successive joints along side j0-j2 The numbers of spaces, n1 and n2, between the joints along directions parallel to sides j0-j1 and j0-j2, respectively, are then given by:
n1

j1 j0 ji1

and

n2

j2 j0 ji2

The labels and label increments must be chosen such that n1 and n2 are whole, positive numbers. The total number of joints in the array is ( n1 + 1)( n 2 + 1).

Two-dimensional Addition
A set of uniformly incrementing Pattern values can be added to the joints in a twodimensional array by specifying: The information in Subtopic Joint Array Specification above The Pattern value, v0, to be added to joint j0 The Pattern value, v1, to be added to joint j1 The Pattern value, v2, to be added to joint j2 The Pattern values added to the remaining joints are determined by linear interpolation and extrapolation from these values. The constant difference, v 1 , between the values added to any pair of joints whose labels differ by ji1 is given by: v 1 = v1 v0
n1

64

Two-dimensional Pattern-value Generation


82

Chapter V Joint Patterns Similarly, the constant difference, v 2 , between the values added to any pair of joints whose labels differ by ji2 is given by: v 2 = v2 v0
n2

The computed values are algebraically added to the values that are already present at the joints.

Two-dimensional Gradient Addition


A set of Pattern values determined from a constant gradient can be added to the joints in a two-dimensional array by specifying: The information in Subtopic Joint Array Specification above The information in Topic Gradient or Hydrostatic Specification The Pattern values added to the joints are determined by Equation 1 in Topic Gradient or Hydrostatic Specification (page 59) in this chapter. The computed values are algebraically added to the values that are already present at the joints.

Two-dimensional Linear Mapping


Two-dimensional linear mapping interpolates (maps) a set of Pattern values over a four-sided region from existing values at the four corners. Before using this option, the Pattern values at the corner joints must have already been defined (or allowed to default to zero): 1. Joint j0 2. Joint j1 3. Joint j2 4. Joint j1+j2j0 at the corner opposite joint j0 It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined by standard bilinear interpolation. Any subset of the joints having labels that increment uniformly by ji1 or ji2 will have a constant difference between the generated Pattern values. Two-dimensional Pattern-value Generation
83

65

SAP2000 Analysis Reference The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Two-dimensional Frontal Mapping


Two-dimensional frontal mapping extrapolates (maps) a set of Pattern values over a four-sided region from existing values along two adjacent sides. Before using this option, the Pattern values at all joints on the two adjacent sides must have already been defined (or allowed to default to zero): 1. Joints j0 to j1 by increment ji1 along one side 2. Joints j0 to j2 by increment ji2 along the adjacent side It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined so that: Any subset of the joints having labels that increment uniformly by ji1 will have differences between the generated Pattern values that are the same as for the corresponding joints along side j0-j1 Any subset of the joints having labels that increment uniformly by ji2 will have differences between the generated Pattern values that are the same as for the corresponding joints along side j0-j2 The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Two-dimensional Edge Mapping


Two-dimensional edge mapping interpolates (maps) a set of Pattern values over a four-sided region from existing values along all four sides. Before using this option, the Pattern values at all joints on the four sides must have already been defined (or allowed to default to zero): 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 3. Joints j2 to j1+j2j0 by increment ji1 4. Joints j1 to j1+j2j0 by increment ji2

66

Two-dimensional Pattern-value Generation


84

Chapter V Joint Patterns It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined by serendipity polynomial interpolation (Cook, Malkus, and Plesha, 1989). The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Three-dimensional Pattern-value Generation


Pattern-value generation as described in this topic is used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Pattern values can be generated for a three-dimensional array of joints, i.e., a set of joints whose labels increment in a uniform fashion in three separate directions. This will typically correspond to a three-dimensional region enclosed by a hexahedron, i.e., having six faces. The generation options include addition, resetting, and several types of mapping. Each generation option requires the specification of a joint array for which the Pattern values are to be generated. This is described in Subtopic Joint Array Specification below. No other information is required for a resetting specification: the Pattern values for all joints in the array are set to zero. The additional information that is required for addition and mapping specifications is described in the remaining subtopics below. Three-dimensional Pattern-value generation is a generalization of two-dimensional Pattern-value generation. For more information: See Topic Two-dimensional Pattern-value Generation (page 63) in this chapter. See Subtopic Three-dimensional Regular Arrays (page 13) in Chapter Labels, Arrays, and Generation.

Joint Array Specification


Each three-dimensional generation specification requires the following information to describe the array of joints to which Pattern values are being assigned: Three-dimensional Pattern-value Generation
85

67

SAP2000 Analysis Reference The label, j0, of the joint at one corner The label, j1, of the joint at the end of one side that contains joint j0 The label, j2, of the joint at the end of another side that contains joint j0 The label, j3, of the joint at the end of the third side that contains joint j0 The label increment (constant difference), ji1, between any pair of successive joints along side j0-j1 The label increment (constant difference), ji2, between any pair of successive joints along side j0-j2 The label increment (constant difference), ji3, between any pair of successive joints along side j0-j3 The numbers of spaces, n1, n2, and n3, between the joints along directions parallel to sides j0-j1, j0-j2, and j0-j3, respectively, are then given by:
n1

j1 j0 ji1

n2

j2 j0 ji2

and

n3

j3 j0 ji3

The labels and label increments must be chosen such that n1, n2, and n3 are whole, positive numbers. The total number of joints in the array is ( n1 + 1)( n 2 + 1)( n 3 + 1).

Three-dimensional Addition
A set of uniformly incrementing Pattern values can be added to the joints in a three-dimensional array by specifying: The information in Subtopic Joint Array Specification above The Pattern value, v0, to be added to joint j0 The Pattern value, v1, to be added to joint j1 The Pattern value, v2, to be added to joint j2 The Pattern value, v3, to be added to joint j3 The Pattern values added to the remaining joints are determined by linear interpolation and extrapolation from these values. The constant difference, v 1 , between the values added to any pair of joints whose labels differ by ji1 is given by: v 1 = v1 v0
n1

68

Three-dimensional Pattern-value Generation


86

Chapter V Joint Patterns Similarly, the constant difference, v 2 , between the values added to any pair of joints whose labels differ by ji2 is given by: v 2 = v2 v0
n2

Finally, the constant difference, v 3 , between the values added to any pair of joints whose labels differ by ji3 is given by: v 3 = v3 v0
n3

The computed values are algebraically added to the values that are already present at the joints.

Three-dimensional Gradient Addition


A set of Pattern values determined from a constant gradient can be added to the joints in a three-dimensional array by specifying: The information in Subtopic Joint Array Specification above The information in Topic Gradient or Hydrostatic Specification The Pattern values added to the joints are determined by Equation 1 in Topic Gradient or Hydrostatic Specification (page 59) in this chapter. The computed values are algebraically added to the values that are already present at the joints.

Three-dimensional Linear Mapping


Three-dimensional linear mapping interpolates (maps) a set of Pattern values over a hexahedral (six-faced) region from existing values at the eight corners. Before using this option, the Pattern values at the corner joints must have already been defined (or allowed to default to zero): 1. Joint j0 2. Joint j1 3. Joint j2 4. Joint j3 Three-dimensional Pattern-value Generation
87

69

SAP2000 Analysis Reference 5. Joint j1+j2j0 at the corner opposite joint j3 6. Joint j1+j3j0 at the corner opposite joint j2 7. Joint j2+j3j0 at the corner opposite joint j1 8. Joint j1+j2+j3j0j0 at the corner opposite joint j0 It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined by standard trilinear interpolation. Any subset of the joints having labels that increment uniformly by ji1, ji2, or ji3 will have a constant difference between the generated Pattern values. The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Three-dimensional Frontal Mapping


Three-dimensional frontal mapping extrapolates (maps) a set of Pattern values over a hexahedral (six-faced) region from existing values along three adjacent edges. Before using this option, the Pattern values at all joints on the three adjacent edges must have already been defined (or allowed to default to zero): 1. Joints j0 to j1 by increment ji1 along one edge 2. Joints j0 to j2 by increment ji2 along an adjacent edge 3. Joints j0 to j3 by increment ji3 along the other adjacent edge It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined so that: Any subset of the joints having labels that increment uniformly by ji1 will have differences between the generated Pattern values that are the same as for the corresponding joints along edge j0-j1 Any subset of the joints having labels that increment uniformly by ji2 will have differences between the generated Pattern values that are the same as for the corresponding joints along edge j0-j2

70

Three-dimensional Pattern-value Generation


88

Chapter V Joint Patterns Any subset of the joints having labels that increment uniformly by ji3 will have differences between the generated Pattern values that are the same as for the corresponding joints along edge j0-j3 The assigned values are not additive, but instead they replace (overwrite) the values present at the joints.

Three-dimensional Edge Mapping


Three-dimensional edge mapping interpolates (maps) a set of Pattern values over a hexahedral (six-faced) region from existing values along all twelve edges. Before using this option, the Pattern values at all joints on the twelve edges must have already been defined (or allowed to default to zero): 1. Joints j0 to j1 by increment ji1 2. Joints j0 to j2 by increment ji2 3. Joints j0 to j3 by increment ji3 4. Joints j1 to j1+j2j0 by increment ji2 5. Joints j1 to j1+j3j0 by increment ji3 6. Joints j2 to j1+j2j0 by increment ji1 7. Joints j2 to j2+j3j0 by increment ji3 8. Joints j3 to j1+j3j0 by increment ji1 9. Joints j3 to j2+j3j0 by increment ji2 10. Joints j1+j2j0 to j1+j2+j3j0j0 by increment ji3 11. Joints j1+j3j0 to j1+j2+j3j0j0 by increment ji2 12. Joints j2+j3j0 to j1+j2+j3j0j0 by increment ji1 It is then only necessary to provide the information in Subtopic Joint Array Specification above. The Pattern values assigned to the remaining joints are determined by serendipity polynomial interpolation (Zienkiewicz and Taylor, 1989). The assigned values are not additive, but instead they replace (overwrite) the values present at the joints. Three-dimensional Pattern-value Generation
89

71

SAP2000 Analysis Reference

Variable Pattern-value Increments


Variable pattern-value increments as described in this topic are used only in the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. Pattern values generated by an addition specification or a linear-mapping specification normally have a constant increment (difference) between successive joints along any line or curve in a joint array, provided that the joint labels differ by one of the specified label increments. Sometimes, however, it is desirable to have the increment in Pattern values vary gradually along these lines or curves. The principal situation where this is useful is when you want the Pattern values to be proportional to the x, y, or z coordinates of the joints and the joint spacing is not constant, as in a graded mesh. The addition and linear-mapping specifications for Pattern values permit the use of the parameters ratio1, ratio2, and ratio3 to control the variation of generated Pattern values. The effect of these parameters is analogous to their effect upon joint spacing in joint array definition and joint linear mapping. To generate Pattern values proportional to the x, y, or z coordinates of joints that were created using joint array definition: Use an addition specification for the same joint array Specify pattern values that are proportional to the x, y, or z coordinates of the joints, as desired Specify the same values of the parameters ratio1, ratio2, and ratio3 as used for the joint array definition To generate Pattern values proportional to the x, y, or z coordinates of joints that were created using joint linear generation: Use a linear-mapping specification for the same joint array Make sure that the Pattern values at the corners have already been defined to be proportional to the x, y, or z coordinates of the joints, as desired Specify the same values of the parameters ratio1, ratio2, and ratio3 as used for the joint linear generation The increments in Pattern values change by arithmetic progression so that the difference between successive increments is constant.

72

Variable Pattern-value Increments


90

Chapter V Joint Patterns This option is only available for addition and linear-mapping specifications. For frontal-mapping and edge-mapping specifications, the increments in Pattern values is determined from the values at the joints along the previously-defined edges. The ratio parameters are described in more detail in the subtopics below. For more information: See Topic Variable Joint Spacing (page 47) in Chapter Joint Coordinates. See Subtopic One-dimensional Addition (page 62) in this chapter. See Subtopic One-dimensional Linear Mapping (page 63) in this chapter. See Subtopic Two-dimensional Addition (page 64) in this chapter. See Subtopic Two-dimensional Linear Mapping (page 65) in this chapter. See Subtopic Three-dimensional Addition (page 68) in this chapter. See Subtopic Three-dimensional Linear Mapping (page 69) in this chapter.

One Dimension
The parameter ratio1 can be used to specify how the Pattern-value increment varies along the line of generation. It gives the ratio of the last increment, v 1n , between the values at joints j1ji1 and j1, and the first increment, v 11 , between the values at joints j0 and j0+ji1: ratio1 = v 1n v 11

If the value of ratio1 is greater than unity, then the increment increases along the length. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment.

Two Dimensions
The two parameters ratio1 and ratio2 can be used to specify how the Pattern-value increment varies along two generation dimensions. Parameter ratio1 is defined above. Parameter ratio2 is defined similarly: it gives the ratio of the last increment, v 2n , between the values at joints j2ji2 and j2, and the first increment, v 21 , between the values at joints j0 and j0+ji2:

Variable Pattern-value Increments


91

73

SAP2000 Analysis Reference ratio2 = v 2n v 21

If the value of ratio1 is greater than unity, then the increment increases away from side j0-j2 along any line or curve where the joint labels increment by ji1. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment. Similarly, if the value of ratio2 is greater than unity, then the increment increases away from side j0-j1 along any line or curve where the joint labels increment by ji2. If the value of ratio2 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment.

Three Dimensions
The three parameters ratio1, ratio2 and ratio3 can be used to specify how the Pattern-value increment varies along three generation dimensions. Parameters ratio1 and ratio2 are defined above. Parameter ratio3 is defined similarly: it gives the ratio of the last increment, v 3 n , between the values at joints j3ji3 and j3, and the first increment, v 31 , between the values at joints j0 and j0+ji3: ratio3 = v 3 n v 31

If the value of ratio1 is greater than unity, then the increment increases away from face j0-j2-j3 along any line or curve where the joint labels increment by ji1. If the value of ratio1 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment. Similarly, if the value of ratio2 is greater than unity, then the increment increases away from face j0-j1-j3 along any line or curve where the joint labels increment by ji2. If the value of ratio2 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment. Finally, if the value of ratio3 is greater than unity, then the increment increases away from face j0-j1-j2 along any line or curve where the joint labels increment by ji3. If the value of ratio3 is less than unity, then the reverse would be true. The default value of unity indicates a uniform Pattern-value increment.

74

Variable Pattern-value Increments


92

C h a p t e r VI

Joints and Degrees of Freedom


The joints play a fundamental role in the analysis of any structure. Joints are the points of connection between the elements, and they are the primary locations in the structure at which the displacements are known or are to be determined. The displacement components (translations and rotations) at the joints are called the degrees of freedom. This chapter describes joint properties, degrees of freedom, loads, and output. Chapter Joint Coordinates (page 31) describes the use of joints to define the geometry of the structure. Additional information about joints and degrees of freedom is given in Chapter Constraints and Welds (page 113). Basic Topics for All Users Overview Modeling Considerations Local Coordinate System Degrees of Freedom Restraints and Reactions Springs Masses

75
93

SAP2000 Analysis Reference Force Load Restraint Displacement Load Spring Displacement Load Degree of Freedom Output Joint Mass Output Displacement and Acceleration Output Force Output Global Force Balance Output Advanced Topics Advanced Local Coordinate System

Overview
Joints, also known as nodal points or nodes, are a fundamental part of every structural model. Joints perform a variety of functions: All elements are connected to the structure (and hence to each other) at the joints The structure is supported at the joints using Restraints and/or Springs Rigid-body behavior and symmetry conditions can be specified using Constraints that apply to the joints Concentrated loads may be applied at the joints Lumped (concentrated) masses and rotational inertia may be placed at the joints All loads and masses applied to the elements are actually transferred to the joints Joints are the primary locations in the structure at which the displacements are known (the supports) or are to be determined All of these functions are discussed in this chapter except for the Constraints, which are described in Chapter Constraints and Welds (page 113). The method used to define the structural model affects how joints are created:

76

Overview
94

Chapter VI Joints and Degrees of Freedom Using the SAP2000 graphical interface joints are automatically created at the ends of each Frame or Nllink element and at the corners of each Shell element; additional joints may also be defined independently of any element Using the input data text file joint locations must be explicitly defined in order to describe the geometry of the structure; these joints are then connected by elements to build the structure The use of joints to define the geometry of the structure is discussed in Chapter Joint Coordinates (page 31). Joints act independently of each other unless connected by elements. Frame and Nllink elements connect to the joints at the element ends. The two- and threedimensional elements connect to the joints at the element corners, and possibly at other points on the element. The joints may themselves be considered as elements with no connectivity. Each joint may have its own local coordinate system for defining the degrees of freedom, Restraints, joint properties, and loads; and for interpreting joint output. In most cases, however, the global X-Y-Z coordinate system is used as the local coordinate system for all joints in the model. There are six displacement degrees of freedom at every joint three translations and three rotations. These displacement components are aligned along the local coordinate system of each joint. Joints may be loaded directly by concentrated loads or indirectly by ground displacements acting though Restraints or spring supports. Displacements (translations and rotations) are produced at every joint. The external and internal forces and moments acting on each joint are also produced. For more information: See Chapter Joint Coordinates (page 31). See Chapter Constraints and Welds (page 113).

Modeling Considerations
The location of the joints and elements is critical in determining the accuracy of the structural model. Some of the factors that you need to consider when defining the elements, and hence the joints, for the structure are:

Modeling Considerations
95

77

SAP2000 Analysis Reference The number of elements should be sufficient to describe the geometry of the structure. For straight lines and edges, one element is adequate. For curves and curved surfaces, one element should be used for every arc of 15 or less. Element boundaries, and hence joints, should be located at points, lines, and surfaces of discontinuity: Structural boundaries, e.g., corners and edges Changes in material properties Changes in thickness and other geometric properties Support points (Restraints and Springs) Points of application of concentrated loads, except that Frame elements may have concentrated loads applied within their spans In regions having large stress gradients, i.e., where the stresses are changing rapidly, a Shell element mesh should be refined using small elements and closely-spaced joints. This may require changing the mesh after one or more preliminary analyses. More that one element should be used to model the length of any span for which dynamic behavior is important. This is required because the mass is always lumped at the joints, even if it is contributed by the elements.

Local Coordinate System


Each joint has its own joint local coordinate system used to define the degrees of freedom, Restraints, properties, and loads at the joint; and for interpreting joint output. The axes of the joint local coordinate system are denoted 1, 2, and 3. By default these axes are identical to the global X, Y, and Z axes, respectively. Both systems are right-handed coordinate systems. The default local coordinate system is adequate for most situations. However, for certain modeling purposes it may be useful to use different local coordinate systems at some or all of the joints. This is described in the next topic. For more information: See Topic Upward and Horizontal Directions (page 23) in Chapter Coordinate Systems. See Topic Advanced Local Coordinate System (page 79) in this chapter.

78

Local Coordinate System


96

Chapter VI Joints and Degrees of Freedom

Advanced Local Coordinate System


By default, the joint local 1-2-3 coordinate system is identical to the global X-Y-Z coordinate system, as described in the previous topic. However, it may be necessary to use different local coordinate systems at some or all joints in the following cases: Skewed Restraints (supports) are present Constraints are used to impose rotational symmetry Constraints are used to impose symmetry about a plane that is not parallel to a global coordinate plane The principal axes for the joint mass (translational or rotational) are not aligned with the global axes Joint displacement and force output is desired in another coordinate system Joint local coordinate systems need only be defined for the affected joints. The global system is used for all joints for which no local coordinate system is explicitly specified. A variety of methods are available to define a joint local coordinate system. These may be used separately or together. Local coordinate axes may be defined to be parallel to arbitrary coordinate directions in an arbitrary coordinate system or to vectors between pairs of joints. In addition, the joint local coordinate system may be specified by a set of three joint coordinate angles. These methods are described in the subtopics that follow. For more information: See Chapter Coordinate Systems (page 21). See Topic Local Coordinate System (page 78) in this chapter.

Reference Vectors
To define a joint local coordinate system you must specify two reference vectors that are parallel to one of the joint local coordinate planes. The axis reference vector, Va , must be parallel to one of the local axes (i = 1, 2, or 3) in this plane and have a positive projection upon that axis. The plane reference vector, V p , must have a positive projection upon the other local axis (j = 1, 2, or 3, but i j) in this plane, but need not be parallel to that axis. Having a positive projection means that the posi-

Advanced Local Coordinate System


97

79

SAP2000 Analysis Reference tive direction of the reference vector must make an angle of less than 90 with the positive direction of the local axis. Together, the two reference vectors define a local axis, i, and a local plane, i-j. From this, the program can determine the third local axis, k, using vector algebra. For example, you could choose the axis reference vector parallel to local axis 1 and the plane reference vector parallel to the local 1-2 plane (i = 1, j = 2). Alternatively, you could choose the axis reference vector parallel to local axis 3 and the plane reference vector parallel to the local 3-2 plane (i = 3, j = 2). You may choose the plane that is most convenient to define using the parameter local, which may take on the values 12, 13, 21, 23, 31, or 32. The two digits correspond to i and j, respectively. The default is value is 31.

Defining the Axis Reference Vector


To define the axis reference vector for joint j, you must first specify or use the default values for: A coordinate direction axdir (the default is +Z) A fixed coordinate system csys (the default is zero, indicating the global coordinate system) You may optionally specify: A pair of joints, axveca and axvecb (the default for each is zero, indicating joint j itself). If both are zero, this option is not used. For each joint, the axis reference vector is determined as follows: 1. A vector is found from joint axveca to joint axvecb. If this vector is of finite length, it is used as the reference vector Va 2. Otherwise, the coordinate direction axdir is evaluated at joint j in fixed coordinate system csys, and is used as the reference vector Va

Defining the Plane Reference Vector


To define the plane reference vector for joint j, you must first specify or use the default values for: A primary coordinate direction pldirp (the default is +X)

80

Advanced Local Coordinate System


98

Chapter VI Joints and Degrees of Freedom A secondary coordinate direction pldirs (the default is +Y). Directions pldirs and pldirp should not be parallel to each other unless you are sure that they are not parallel to local axis 1 A fixed coordinate system csys (the default is zero, indicating the global coordinate system). This will be the same coordinate system that was used to define the axis reference vector, as described above You may optionally specify: A pair of joints, plveca and plvecb (the default for each is zero, indicating joint j itself). If both are zero, this option is not used. For each joint, the plane reference vector is determined as follows: 1. A vector is found from joint plveca to joint plvecb. If this vector is of finite length and is not parallel to local axis i, it is used as the reference vector V p 2. Otherwise, the primary coordinate direction pldirp is evaluated at joint j in fixed coordinate system csys. If this direction is not parallel to local axis i, it is used as the reference vector V p 3. Otherwise, the secondary coordinate direction pldirs is evaluated at joint j in fixed coordinate system csys. If this direction is not parallel to local axis i, it is used as the reference vector V p 4. Otherwise, the method fails and the analysis terminates. This will never happen if pldirp is not parallel to pldirs A vector is considered to be parallel to local axis i if the sine of the angle between -3 them is less than 10 .

Determining the Local Axes from the Reference Vectors


The program uses vector cross products to determine the local axes from the reference vectors. The three axes are represented by the three unit vectors V1 , V2 and V3 , respectively. The vectors satisfy the cross-product relationship: V1 = V2 V3 The local axis Vi is given by the vector Va after it has been normalized to unit length. The remaining two axes, V j and Vk , are defined as follows: Advanced Local Coordinate System
99

81

SAP2000 Analysis Reference

Figure 14 Example of the Determination of the Joint Local Coordinate System Using Reference Vectors for local=31

If i and j permute in a positive sense, i.e., local = 12, 23, or 31, then: Vk = Vi V p and V j = Vk Vi If i and j permute in a negative sense, i.e., local = 21, 32, or 13, then: Vk = V p Vi and V j = Vi Vk An example showing the determination of the joint local coordinate system using reference vectors is given in Figure 14 (page 82).

Joint Coordinate Angles


The joint local coordinate axes determined from the reference vectors may be further modified by the use of three joint coordinate angles, denoted a, b, and c. In the case where the default reference vectors are used, the joint coordinate angles define the orientation of the joint local coordinate system with respect to the global axes.

82

Advanced Local Coordinate System


100

Chapter VI Joints and Degrees of Freedom The joint coordinate angles specify rotations of the local coordinate system about its own current axes. The resulting orientation of the joint local coordinate system is obtained according to the following procedure: 1. The local system is first rotated about its +3 axis by angle a 2. The local system is next rotated about its resulting +2 axis by angle b 3. The local system is lastly rotated about its resulting +1 axis by angle c The order in which the rotations are performed is important. The use of coordinate angles to orient the joint local coordinate system with respect to the global system is shown in Figure 15 (page 84).

Degrees of Freedom
The deflection of the structural model is governed by the displacements of the joints. Every joint of the structural model may have up to six displacement components: The joint may translate along its three local axes. These translations are denoted U1, U2, and U3. The joint may rotate about its three local axes. These rotations are denoted R1, R2, and R3. These six displacement components are known as the degrees of freedom of the joint. In the usual case where the joint local coordinate system is parallel to the global system, the degrees of freedom may also be identified as UX, UY, UZ, RX, RY and RZ, according to which global axes are parallel to which local axes. The joint local degrees of freedom are illustrated in Figure 16 (page 85). In addition to the regular joints that you explicitly define as part of your structural model, the program automatically creates master joints that govern the behavior of any Constraints and Welds that you may have defined. Each master joint has the same six degrees of freedom as do the regular joints. See Chapter Constraints and Welds (page 113) for more information. Each degree of freedom in the structural model must be one of the following types: Active the displacement is computed during the analysis Restrained the displacement is specified, and the corresponding reaction is computed during the analysis

Degrees of Freedom
101

83

SAP2000 Analysis Reference

Figure 15 Use of Joint Coordinate Angles to Orient the Joint Local Coordinate System

84

Degrees of Freedom
102

Chapter VI Joints and Degrees of Freedom

Figure 16 The Six Displacement Degrees of Freedom in the Joint Local Coordinate System

Constrained the displacement is determined from the displacements at other degrees of freedom Null the displacement does not affect the structure and is ignored by the analysis Unavailable the displacement has been explicitly excluded from the analysis These different types of degrees of freedom are described in the following subtopics.

Available and Unavailable Degrees of Freedom


You may explicitly specify the global degrees of freedom that are available to every joint in the structural model. By default, all six degrees of freedom are available to every joint. This default should generally be used for all three-dimensional structures. For certain planar structures, however, you may wish to restrict the available degrees of freedom. For example, in the X-Y plane: a planar truss needs only UX and UY; a planar frame needs only UX, UY, and RZ; and a planar grid or flat plate needs only UZ, RX, and RY. The degrees of freedom that are not specified as being available are called unavailable degrees of freedom. Any stiffness, loads, mass, Restraints, or Constraints that are applied to the unavailable degrees of freedom are ignored by the analysis. Degrees of Freedom
103

85

SAP2000 Analysis Reference The available degrees of freedom are always referred to the global coordinate system, and they are the same for every joint in the model. If any joint local coordinate systems are used, they must not couple available degrees of freedom with the unavailable degrees of freedom at any joint. For example, if the available degrees of freedom are UX, UY, and RZ, then all joint local coordinate systems must have one local axis parallel to the global Z axis.

Restrained Degrees of Freedom


If the displacement of a joint along any one of its available degrees of freedom is known, such as at a support point, that degree of freedom is restrained. The known value of the displacement may be zero or non-zero, and may be different in different Load Cases. The force along the restrained degree of freedom that is required to impose the specified restraint displacement is called the reaction, and is determined by the analysis. Unavailable degrees of freedom are essentially restrained. However, they are excluded from the analysis and no reactions are computed, even if they are non-zero. See Topic Restraints and Reactions (page 88) in this chapter for more information.

Constrained Degrees of Freedom


Any joint that is part of a Constraint or Weld may have one or more of its available degrees of freedom constrained. The program automatically creates a master joint to govern the behavior of each Constraint, and a master joint to govern the behavior of each set of joints that are connected together by a Weld. The displacement of a constrained degree of freedom is then computed as a linear combination of the displacements along the degrees of freedom at the corresponding master joint. A degree of freedom may not be both constrained and restrained. See Chapter Constraints and Welds (page 113) for more information.

Active Degrees of Freedom


All available degrees of freedom that are neither constrained nor restrained must be either active or null. The program will automatically determine the active degrees of freedom as follows:

86

Degrees of Freedom
104

Chapter VI Joints and Degrees of Freedom If any load or stiffness is applied along any translational degree of freedom at a joint, then all available translational degrees of freedom at that joint are made active unless they are constrained or restrained. If any load or stiffness is applied along any rotational degree of freedom at a joint, then all available rotational degrees of freedom at that joint are made active unless they are constrained or restrained. All degrees of freedom at a master joint that govern constrained degrees of freedom are made active. A joint that is connected to any element or to a translational spring will have all of its translational degrees of freedom activated. A joint that is connected to a Frame, Shell, or Nllink element, or to any rotational spring will have all of its rotational degrees of freedom activated. An exception is a Frame element with only truss-type stiffness, which will not activate rotational degrees of freedom. Every active degree of freedom has an associated equation to be solved. If there are N active degrees of freedom in the structure, there are N equations in the system, and the structural stiffness matrix is said to be of order N. The amount of computational effort required to perform the analysis increases with N. The load acting along each active degree of freedom is known (it may be zero). The corresponding displacement will be determined by the analysis. If there are active degrees of freedom in the system at which the stiffness is known to be zero, such as the out-of-plane translation in a planar-frame, these must either be restrained or made unavailable. Otherwise, the structure is unstable and the solution of the static equations will fail. For more information: See Topic Springs (page 89) in this chapter. See Topic Degrees of Freedom (page 147) in Chapter The Frame Element. See Topic Degrees of Freedom (page 185) in Chapter The Shell Element. See Topic Degrees of Freedom (page 208) in Chapter The Plane Element. See Topic Degrees of Freedom (page 224) in Chapter The Asolid Element. See Topic Degrees of Freedom (page 236) in Chapter The Solid Element. See Topic Degrees of Freedom (page 245) in Chapter The Nllink Element.

Degrees of Freedom
105

87

SAP2000 Analysis Reference

Null Degrees of Freedom


The available degrees of freedom that are not restrained, constrained, or active, are called the null degrees of freedom. Because they have no load or stiffness, their displacements and reactions are zero, and they have no effect on the rest of the structure. The program automatically excludes them from the analysis. Joints that have no elements connected to them typically have all six degrees of freedom null. Joints that have only solid-type elements (Plane, Asolid, and Solid) connected to them typically have the three rotational degrees of freedom null.

Restraints and Reactions


If the displacement of a joint along any of its available degrees of freedom has a known value, either zero (e.g., at support points) or non-zero (e.g., due to support settlement), a Restraint must be applied to that degree of freedom. The known value of the displacement may differ from one Load Case to the next, but the degree of freedom is restrained for all Load Cases. In other words, it is not possible to have the displacement known in one Load Case and unknown (unrestrained) in another Load Case. Restraints should also be applied to available degrees of freedom in the system at which the stiffness is known to be zero, such as the out-of-plane translation and inplane rotations of a planar-frame. Otherwise, the structure is unstable and the solution of the static equations will fail. Restraints are always applied to the joint local degrees of freedom U1, U2, U3, R1, R2, and R3. However, it is permissible to specify restraints in terms of the global degrees of freedom UX, UY, UZ, RX, RY, and RZ. If a local degree of freedom is found at the joint that is parallel to the specified global degree of freedom, the restraint is applied to that local degree of freedom. If no parallel local degree of freedom is found, then the specified restraint is not applied. The force or moment along the degree of freedom that is required to enforce the restraint is called the reaction, and it is determined by the analysis. The reaction may differ from one Load Case to the next. The reaction includes the forces (or moments) from all elements and springs connected to the restrained degree of freedom, as well as all loads applied to the degree of freedom. A restrained degree of freedom may not be constrained. If a restraint is applied to an unavailable degree of freedom, it is ignored and no reaction is computed.

88

Restraints and Reactions


106

Chapter VI Joints and Degrees of Freedom Examples of Restraints are shown in Figure 17 (page 90). For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Topic Restraint Displacement Load (page 93) in this chapter.

Springs
Any of the six degrees of freedom at any of the joints in the structure can have translational or rotational spring support conditions. These springs elastically connect the joint to the ground. Spring supports along restrained degrees of freedom do not contribute to the stiffness of the structure. Springs may be specified that couple the degrees of freedom at a joint. The spring forces that act on a joint are related to the displacements of that joint by a 6x6 symmetric matrix of spring stiffness coefficients. These forces tend to oppose the displacements. Spring stiffness coefficients may be specified in the global coordinate system, an Alternate Coordinate System, or the joint local coordinate system. In a joint local coordinate system, the spring forces and moments F1, F2, F3, M1, M2 and M3 at a joint are given by: u1 u1u2 u1u3 u1r1 u1r2 u1r3 u1 F1 F u2 u2u3 u2r1 u2r2 u2r3 u 2 2 u3 u3r1 u3r2 u3r3 u 3 F3 = r1 r1r2 r1r3 r1 M1 sym. r2 r2r3 r2 M 2 M r3 r3 3 (Eqn. 1)

where u1, u2, u3, r1, r2 and r3 are the joint displacements and rotations, and the terms u1, u1u2, u2, ... are the specified spring stiffness coefficients. In any fixed coordinate system, the spring forces and moments Fx, Fy, Fz, Mx, My and Mz at a joint are given by:

Springs
107

89

SAP2000 Analysis Reference

Figure 17 Examples of Restraints

90

Springs
108

Chapter VI Joints and Degrees of Freedom ux Fx F y Fz = Mx M y M z uxuy uxuz uxrx uy uyuz uyrx uz uzrx rx sym. uxry uxrz u x uyry uyrz u y uzry uzrz u z rxry rxrz rx ry ryrz r y rz r z

where ux, uy, uz, rx, ry and rz are the joint displacements and rotations, and the terms ux, uxuy, uy, ... are the specified spring stiffness coefficients. For springs that do not couple the degrees of freedom in a particular coordinate system, only the six diagonal terms need to be specified since the off-diagonal terms are all zero. When coupling is present, all 21 coefficients in the upper triangle of the matrix must be given; the other 15 terms are then known by symmetry. If the springs at a joint are specified in more than one coordinate system, standard coordinate transformation techniques are used to convert the 6x6 spring stiffness matrices to the joint local coordinate system, and the resulting stiffness matrices are then added together on a term-by-term basis. The final spring stiffness matrix at each joint in the structure should have a determinant that is zero or positive. Otherwise the springs may cause the structure to be unstable. The displacement of the grounded end of the spring may be specified to be zero or non-zero (e.g., due to support settlement). This spring displacement may vary from one Load Case to the next. For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Topic Spring Displacement Load (page 95) in this chapter.

Masses
In a dynamic analysis, the mass of the structure is used to compute inertial forces. Normally, the mass is obtained from the elements using the mass density of the material and the volume of the element. This automatically produces lumped (uncoupled) masses at the joints. The element mass values are equal for each of the three translational degrees of freedom. No mass moments of inertia are produced for the rotational degrees of freedom. This approach is adequate for most analyses.

Masses
109

91

SAP2000 Analysis Reference It is often necessary to place additional concentrated masses and/or mass moments of inertia at the joints. These can be applied to any of the six degrees of freedom at any of the joints in the structure. For computational efficiency and solution accuracy, SAP2000 always uses lumped masses. This means that there is no mass coupling between degrees of freedom at a joint or between different joints. These uncoupled masses are always referred to the local coordinate system of each joint. Mass values along restrained degrees of freedom are ignored. Inertial forces acting on the joints are related to the accelerations at the joints by a 6x6 matrix of mass values. These forces tend to oppose the accelerations. In a joint local coordinate system, the inertia forces and moments F1, F2, F3, M1, M2 and M3 at a joint are given by: && 0 0 0 0 0 u1 u1 F1 F && u2 0 0 0 0 u2 2 && u3 0 0 0 u 3 F3 = && r1 0 0 r1 M1 sym. r2 0 && r2 M 2 M && r3 r3 3 && && && r1 r2 where u1 , u 2 , u 3 , && , && and && are the translational and rotational accelerations at r3 the joint, and the terms u1, u2, u3, r1, r2, and r3 are the specified mass values. Uncoupled joint masses may instead be specified in the global coordinate system, in which case they are transformed to the joint local coordinate system. Coupling terms will be generated during this transformation in the following situation: The joint local coordinate system directions are not parallel to global coordinate directions, and The three translational masses or the three rotational mass moments of inertia are not equal at a joint. These coupling terms will be discarded by the program, resulting in some loss of accuracy. For this reason, it is recommended that you choose joint local coordinate systems that are aligned with the principal directions of translational or rotational mass at a joint, and then specify mass values in these joint local coordinates. Mass values must be given in consistent mass units (W/g) and mass moments of in2 ertia must be in WL /g units. Here W is weight, L is length, and g is the acceleration

92

Masses
110

Chapter VI Joints and Degrees of Freedom due to gravity. The net mass values at each joint in the structure should be zero or positive. See Figure 18 (page 94) for mass moment of inertia formulations for various planar configurations. For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Chapter Static and Dynamic Analysis (page 291).

Force Load
The Force Load is used to apply concentrated forces and moments at the joints. Values may be specified in a fixed coordinate system (global or alternate coordinates) or the joint local coordinate system. All forces and moments at a joint are transformed to the joint local coordinate system and added together. The specified values are shown in Figure 19 (page 95). Forces and moments applied along restrained degrees of freedom add to the corresponding reaction, but do not otherwise affect the structure. For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Chapter Load Cases (page 279).

Restraint Displacement Load


The Restraint Displacement Load is used to apply specified displacements (translations and rotations) at the grounded end of joint restraints. Restraint displacements may be specified in a fixed coordinate system (global or alternate coordinates) or the joint local coordinate system. The specified values are shown in Figure 19 (page 95). All displacements at a joint are transformed to the joint local coordinate system and added together. Restraints may be considered as rigid connections between the joint degrees of freedom and the ground, and are always defined in the joint local coordinate system. If a particular joint degree of freedom is restrained, the displacement of the joint is equal to the restraint displacement along that local degree of freedom. Com-

Force Load
111

93

SAP2000 Analysis Reference

Figure 18 Formulae for Mass Moments of Inertia

94

Restraint Displacement Load


112

Chapter VI Joints and Degrees of Freedom

Figure 19 Specified Values for Force Load, Restraint Displacement Load, and Spring Displacement Load

ponents of restraint displacement that are not along restrained degrees of freedom do not load the structure. An example of this is illustrated in Figure 20 (page 96). For more information: See Topic Restraints and Reactions (page 88) in this chapter. See Chapter Load Cases (page 279).

Spring Displacement Load


The Spring Displacement Load is used to apply specified displacements (translations and rotations) at the grounded end of joint spring supports. Spring displacements may be specified in a fixed coordinate system (global or alternate coordinates) or the joint local coordinate system. The specified values are shown in Figure 19 (page 95). All displacements at a joint are transformed to the joint local coordinate system and added together. The resulting displacements are multiplied by the spring stiffness Spring Displacement Load
113

95

SAP2000 Analysis Reference

Figure 20 Example of Restraint Displacement Not Aligned with Local Degrees of Freedom

coefficients to obtain effective forces and moments that are applied to the joint. Spring displacements applied in a direction with no spring stiffness result in zero applied load. In a joint local coordinate system, the forces and moments F1, F2, F3, M1, M2 and M3 at a joint due to spring displacements are given by: F1 u1 u1u2 u1u3 u1r1 u1r2 u1r3 u s1 F u2 u2u3 u2r1 u2r2 u2r3 u s2 2 u3 u3r1 u3r2 u3r3 u s3 F3 = r1 r1r2 r1r3 rs1 M1 sym. r2 r2r3 rs2 M 2 M r3 rs3 3 (Eqn. 2)

where u s1 , u s2 , u s3 , rs1 , rs2 and rs3 are the spring displacements and rotations, and the terms u1, u1u2, u2, ... are the specified spring stiffness coefficients. The net spring forces and moments acting on the joint are the sum of the forces and moments given in Equations (1) and (2); note that these are of opposite sign. At a restrained degree of freedom, a restraint displacement and a spring displacement may both be specified. The restraint displacement will control the displacement at the joint, and the net spring forces and moments acting on the joint are still

96

Spring Displacement Load


114

Chapter VI Joints and Degrees of Freedom

Figure 21 Interaction Between Restraint Displacements and Spring Displacements

given by the sum of the forces and moments of Equations (1) and (2). An example of this is illustrated in Figure 21 (page 97). For more information: See Topic Springs (page 89) in this chapter. See Chapter Load Cases (page 279).

Degree of Freedom Output


A table of the types of degrees of freedom present at every joint in the model is printed in the output file under the heading:
DISPLACEMENT DEGREES OF FREEDOM

The degrees of freedom are listed for all of the regular joints, as well as for the master joints created automatically by the program. For Constraints, the master joints are identified by the labels of their corresponding Constraints. For Welds, the master joint for each set of joints that are welded together is identified by the label of one of the welded joints. Joints are printed in alpha-numeric order of the labels.

Degree of Freedom Output


115

97

SAP2000 Analysis Reference The type of each of the six degrees of freedom at a joint is identified by the following symbols: (A) (-) (+) ( ) Active degree of freedom Restrained degree of freedom Constrained degree of freedom Null or unavailable degree of freedom

The degrees of freedom are always referred to the local axes of the joint. They are identified in the output as U1, U2, U3, R1, R2, and R3 for all joints. However, if all regular joints use the global coordinate system as the local system (the usual situation), then the degrees of freedom for the regular joints are identified as UX, UY, UZ, RX, RY, and RZ. The types of degrees of freedom are a property of the structure and are independent of the Analysis Cases. Typical degree-of-freedom output is shown in Figure 22 (page 99). For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Chapter The Output Files (page 395).

Joint Mass Output


A table of the assembled masses at every joint in the model is printed in the output file under the heading:
ASSEMBLED JOINT MASSES

The mass at a given joint includes the mass assigned directly to that joint as well as a portion of the mass from each element connected to that joint. Mass at restrained degrees of freedom is set to zero. All mass assigned to the elements is apportioned to the connected joints, so that this table represents the total unrestrained mass of the structure. Joints are printed in alpha-numeric order of the labels. Joints that have no unrestrained mass are not printed. The masses are always referred to the local axes of the joint. They are identified in the output as U1, U2, U3, R1, R2, and R3. However, if all joints use the global coor-

98

Joint Mass Output


116

Chapter VI Joints and Degrees of Freedom


D I S P L A C E M E N T (A) (-) (+) ( ) JOINTS 1 3 4 5 6 8 9 10 11 13 14 15 16 18 19 20 21 23 24 25 26 28 29 30 31 35 = = = = D E G R E E S O F F R E E D O M

Active DOF, equilibrium equation Restrained DOF, reaction computed Constrained DOF Null DOF UX + + + + A + + + + + + A + + + U1 UY + + + + + + + + + + + + + + + + + + + + + + + + + U2 A A A A A A A A A A UZ + + + + + + + + + + + + + + + + + + + + + + + + + U3 A A A A A A A A A A A A A A RX RY RZ

TO

+ +

TO

+ +

TO

12

+ +

TO

17

+ +

TO

22

+ +

TO

27

TO TO

34 36

+ + + R1 A

+ R2 A

+ + R3

CONSTRAINTS PLATE SYM01 TO SYM03 SYM04 SYM05 SYM06 TO SYM08 SYM09 SYM10 SYM11 TO SYM13 SYM14 SYM15 SYM31 TO

SYM02 A A SYM07 A A SYM12 A A SYM33 A

A A A A

Figure 22 Degree of Freedom Table in File EXAMPLE.OUT

dinate system as the local system (the usual situation), then the masses are identified as UX, UY, UZ, RX, RY, and RZ.

Joint Mass Output


117

99

SAP2000 Analysis Reference The total unrestrained masses of the structure, Mx, My, and Mz, acting in the global X, Y, and Z directions, are printed in the output file under the heading:
TOTAL UNRESTRAINED MASS AND LOCATION

These three masses may differ even if the three translational masses assigned to each joint are equal, since the restraints for the three translational degrees of freedom at a joint need not be the same. The locations of the centers of mass for Mx, My, and Mz are given with respect to the global origin. These can be used together with the mass values to determine the moments caused by the Acceleration Loads. For more information: See Topic Masses (page 91) in this chapter. See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Displacement and Acceleration Output


Joint displacement output is available for the following types of Analysis Cases: Loads, Modes, Specs, Histories, Moving Loads, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. Joint translations and rotations are printed in the output file under the heading:
JOINT DISPLACEMENTS

The displacements at all joints are printed together for each Analysis Case. Joints are presented in alpha-numeric order. The displacements are given only for the available degrees of freedom. No results are printed for any joint that has only null or unavailable degrees of freedom. The displacements are always referred to the local axes of the joint. They are identified in the output as U1, U2, U3, R1, R2, and R3. However, if all joints use the global coordinate system as the local system (the usual situation), then the displacements are identified as UX, UY, UZ, RX, RY, and RZ. Typical joint displacement output is shown in Figure 23 (page 101).

100

Displacement and Acceleration Output


118

Chapter VI Joints and Degrees of Freedom


J O I N T D I S P L A C E M E N T S

TRANSLATIONS AND ROTATIONS, IN GLOBAL COORDINATES LOAD JOINT 1 2 3 4 5 6 7 8 9 10 11 12 13 14 COMB JOINT 1 2 3 4 5 6 7 8 9 10 11 12 13 14 COMB JOINT 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 -----------------UX .000000 .000000 0.000108 0.000103 .000000 .000000 .000000 3.09E-18 2.26E-18 -6.57E-19 .000000 .000000 -0.000108 -0.000103 UY -0.008161 -0.007954 -0.007493 -0.006662 -0.005493 1.45E-17 2.50E-17 2.77E-17 1.92E-17 1.62E-17 0.008161 0.007954 0.007493 0.006662 UZ 0.076681 0.060396 0.044780 0.030495 0.018238 0.076676 0.060375 0.044741 0.030437 0.018137 0.076681 0.060396 0.044780 0.030495 RX .000000 .000000 .000000 -0.001115 -0.000915 .000000 .000000 .000000 -0.001115 -0.000915 .000000 .000000 .000000 -0.001115 RY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 -2.79E-05 .000000 .000000 .000000 .000000 RZ .000000 .000000 .000000 .000000 4.68E-35 .000000 .000000 .000000 .000000 4.68E-35 .000000 .000000 .000000 .000000

BOTH ------------------ MAX UX .000000 .000000 0.000111 0.000106 .000000 .000000 .000000 5.36E-07 4.92E-07 7.20E-07 .000000 .000000 -0.000105 -0.000100 UY -0.007855 -0.007650 -0.007198 -0.006389 -0.005258 3.16E-05 3.08E-05 2.87E-05 2.46E-05 1.97E-05 0.008467 0.008257 0.007789 0.006936 UZ 0.079786 0.062892 0.046676 0.031820 0.019050 0.079780 0.062871 0.046637 0.031760 0.018946 0.079786 0.062892 0.046676 0.031820 RX .000000 .000000 .000000 -0.001070 -0.000876 .000000 .000000 .000000 -0.001070 -0.000876 .000000 .000000 .000000 -0.001070 RY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 -2.69E-05 .000000 .000000 .000000 .000000 RZ .000000 .000000 .000000 .000000 2.65E-11 .000000 .000000 .000000 .000000 2.65E-11 .000000 .000000 .000000 .000000

BOTH ------------------ MIN UX .000000 .000000 0.000105 0.000100 .000000 .000000 .000000 -5.36E-07 -4.92E-07 -7.20E-07 .000000 .000000 -0.000111 -0.000106 UY -0.008467 -0.008257 -0.007789 -0.006936 -0.005728 -3.16E-05 -3.08E-05 -2.87E-05 -2.46E-05 -1.97E-05 0.007855 0.007650 0.007198 0.006389 UZ 0.073577 0.057900 0.042884 0.029171 0.017425 0.073571 0.057879 0.042846 0.029114 0.017329 0.073577 0.057900 0.042884 0.029171 RX .000000 .000000 .000000 -0.001161 -0.000955 .000000 .000000 .000000 -0.001161 -0.000955 .000000 .000000 .000000 -0.001161 RY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 -2.89E-05 .000000 .000000 .000000 .000000 RZ .000000 .000000 .000000 .000000 -2.65E-11 .000000 .000000 .000000 .000000 -2.65E-11 .000000 .000000 .000000 .000000

...Output Truncated for Brevity...

Figure 23 Joint Displacements in the Output File Displacement and Acceleration Output
119

101

SAP2000 Analysis Reference By default, no displacements are printed. You may specify which Analysis Cases are to be printed. The joint displacements can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. For Response-spectrum cases only, acceleration output will also be provided at any joint for which joint displacements are requested. Joint translational and rotational accelerations are printed in the output file under the heading:
JOINT ACCELERATIONS

The format of the joint acceleration output is identical to that for joint displacements. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Force Output
Joint force output is available for the following types of Analysis Cases: Loads, Modes, Specs, Histories (restricted), Moving Loads (restricted), and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the History case. The forces and moments that act on the joints come from a variety of sources: Applied loads Inertial loads Stiffness forces Spring forces Nllink forces Restraint Forces (Reactions) Constraint forces P-delta forces

102

Force Output
120

Chapter VI Joints and Degrees of Freedom For Histories, Moving Loads, and Combos that contain Histories or Moving Loads, only spring forces and reactions are available. The forces and moments from each different source are computed and printed separately in the output file, as discussed in the subtopics below. The forces and moments are always referred to the local axes of the joint. They are identified in the output as F1, F2, F3, M1, M2, and M3. However, if all joints use the global coordinate system as the local system (the usual situation), then the forces and moments are identified as FX, FY, FZ, MX, MY, and MZ. The values printed are always the forces and moments that act on the joints. Thus a positive value of joint force or moment tends to cause a positive value of joint translation or rotation along the corresponding degree of freedom. The forces and moments for all joints are printed together for each Analysis Case. Joints are presented in alpha-numeric order. Values are given only for the available degrees of freedom. No results are printed for any joint that has only null or unavailable degrees of freedom. By default, no force output is produced. You may specify which Analysis Cases are to be printed. Certain joint forces and moments can be displayed using the SAP2000 graphical interface even if no results have been selected for printing. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. For more information: See Topic Degrees of Freedom (page 83) in this chapter. See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter P-Delta Analysis (page 337). See Chapter The Output Files (page 395).

Applied Loads
The applied loads include all specified loads that act directly on the joints and those that act on elements connected to the joints. These specified loads include Force, Self-Weight, Gravity, Temperature, Concentrated Span, Distributed Span, Prestress, Uniform, Surface Pressure, Pore Pressure, and Rotate Loads. The effects

Force Output
121

103

SAP2000 Analysis Reference of the Restraint Displacement and Spring Displacement Loads are not included here. Joint forces and moments due to applied loads are printed in the output file under the heading:
APPLIED LOADS

Results are given only for joints that have at least one non-zero force or moment value. The applied loads are always zero for Mode and Spec Analysis Cases. Typical applied-load output is shown in Figure 24 (page 105).

Inertial Loads
The inertial loads are the effective, or DAlembert, forces and moments which resist acceleration. For a Vibration Mode of frequency , the acceleration for translational degree of freedom i at a joint is given by 2 u i , where ui is the translation. Hence the inertial force Fi for that degree of freedom is given by:
Fi

= 2

mu

ui

where mui is the mass for degree of freedom i at the joint. Similarly, the inertial moment Mi for rotational degree of freedom i at a joint is given by:
M

= 2

mr

ri

where ri is the rotation and mri is the mass moment of inertia for degree of freedom i at the joint. The displacement at a restrained degree of freedom is always zero in a Vibration Mode, hence the inertial loads are always zero for these degrees of freedom. The inertial loads for Spec and Combo Analysis Cases are just the appropriate combinations of the inertial loads for the contributing Modes. The inertial loads for the Load Cases are always zero. Joint forces and moments due to inertial loads are printed in the output file under the heading:
INERTIAL LOADS

104

Force Output
122

Chapter VI Joints and Degrees of Freedom


A P P L I E D L O A D S

FORCES AND MOMENTS ACTING ON JOINTS, IN GLOBAL COORDINATES LOAD JOINT 1 6 11 16 21 26 COMB 1 -----------------FX .000000 .000000 .000000 .000000 .000000 .000000 FY .000000 .000000 .000000 .000000 .000000 .000000 FZ 167.000000 666.000000 167.000000 167.000000 666.000000 167.000000 MX .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000

STATIC ------------------ MAX FX .000000 .000000 .000000 .000000 .000000 .000000 FY .000000 .000000 .000000 .000000 .000000 .000000 FZ 167.000000 666.000000 167.000000 167.000000 666.000000 167.000000 MX .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000

JOINT 1 6 11 16 21 26 COMB

STATIC ------------------ MIN FX .000000 .000000 .000000 .000000 .000000 .000000 FY .000000 .000000 .000000 .000000 .000000 .000000 FZ 167.000000 666.000000 167.000000 167.000000 666.000000 167.000000 MX .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000

JOINT 1 6 11 16 21 26 COMB JOINT 1 6 11 16 21 26 COMB JOINT 1 6 11 16 21 26

BOTH ------------------ MAX FX .000000 .000000 .000000 .000000 .000000 .000000 FY .000000 .000000 .000000 .000000 .000000 .000000 FZ 167.000000 666.000000 167.000000 167.000000 666.000000 167.000000 MX .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000

BOTH ------------------ MIN FX .000000 .000000 .000000 .000000 .000000 .000000 FY .000000 .000000 .000000 .000000 .000000 .000000 FZ 167.000000 666.000000 167.000000 167.000000 666.000000 167.000000 MX .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000

Figure 24 Applied Loads in the Output File

Force Output
123

105

SAP2000 Analysis Reference Results are given only for joints that have non-zero, unrestrained mass or mass moments of inertia.

Stiffness Forces
The stiffness forces are the elastic forces contributed by any Frame, Shell, Plane, Asolid, and Solid elements that are connected to a joint. These forces are obtained as the product of the element stiffness matrices and the displacements at the connected joints. The stiffness forces are not printed in the output file, but are used to compute the constraint forces and reactions.

Spring Forces
The spring forces are the product of the spring stiffness coefficients and the joint displacements as given by Equation (1). If spring displacements are present, these produce additional forces as given by Equation (2). Spring forces are printed in the output file under the heading:
SPRING FORCES

Results are given only for joints that have non-zero spring stiffness coefficients.

Nllink Forces
The Nllink forces are the elastic forces contributed by any Nllink elements that are connected to a joint. These forces are obtained as the product of the Nllink element stiffness matrices and the displacements at the connected joints. The Nllink stiffness matrices are based upon the effective-stiffness parameters. For two-joint elements, the forces and moments should be self-equilibrating. For single-joint elements, the forces and moments are supporting the structure. Nllink forces are printed in the output file under the heading:
NLLINK FORCES

Results are given only for joints that are connected to Nllink elements.

Restraint Forces (Reactions)


The restraint forces are the forces and moments that are required to enforce the restraints. These reactions are computed as the sum of the applied loads, stiffness forces, spring forces, and Nllink forces acting on the restrained degrees of freedom.

106

Force Output
124

Chapter VI Joints and Degrees of Freedom Restraint forces are printed in the output file under the heading:
RESTRAINT FORCES (REACTIONS)

Results are given only for joints that possess restrained degrees of freedom. Typical reaction output is shown in Figure 25 (page 108).

Constraint Forces
The constraint forces are the forces and moments that are required to enforce the constraints. These are computed as the sum of the applied loads, inertial loads, stiffness forces, spring forces, and Nllink forces acting on the constrained degrees of freedom. These forces and moments should be self-equilibrating over the entire structure if the constraints are properly defined. See the next Subtopic, Global Force Balance Output, for more information. Constraint forces are printed in the output file under the heading:
CONSTRAINT FORCES

Results are given only for joints that possess constrained degrees of freedom.

P-Delta Moments
The P-delta moments are present only when a P-Delta analysis has been performed. For each Frame element, the P-Delta moments are computed as the vector cross product of the acting P-Delta force in the element and the difference between the displacements at the two ends of the element. These individual P-Delta moments are not assigned to specific joints or printed in the output file, but are summed and printed as part of the global force balance.

Global Force Balance Output


For each Analysis Case, the sum of all joint forces and moments acting on the structure should be in equilibrium. The program computes and prints a global force balance for the following types of Analysis Cases: Loads, Modes, Specs, and Combos; no global force balance is computed for Histories, Moving Loads, or Combos that contain Histories or Moving Loads. You should always review the global force balance as a check on the validity of the structural model and the analysis.

Global Force Balance Output


125

107

SAP2000 Analysis Reference


R E S T R A I N T F O R C E S ( R E A C T I O N S )

FORCES AND MOMENTS ACTING ON JOINTS, IN GLOBAL COORDINATES LOAD JOINT 35 36 SPEC JOINT 35 36 COMB 1 -----------------FX 7.35E-13 -7.35E-13 FY -9.73E-11 -9.73E-11 FZ -1000.000 -1000.000 MX 84000.000 84000.000 MY 2026.122 -2026.122 MZ 7.47E-12 -7.47E-12

SPEC -----------------FX 0.972230 0.972230 FY 126.377896 126.377896 FZ 68.323489 68.323489 MX 3978.697 3978.697 MY 89.135262 89.135262 MZ 9.788509 9.788509

STATIC ------------------ MAX FX 7.35E-13 -7.35E-13 FY -9.73E-11 -9.73E-11 FZ -1000.000 -1000.000 MX 84000.000 84000.000 MY 2026.122 -2026.122 MZ 7.47E-12 -7.47E-12

JOINT 35 36 COMB

STATIC ------------------ MIN FX 7.35E-13 -7.35E-13 FY -9.73E-11 -9.73E-11 FZ -1000.000 -1000.000 MX 84000.000 84000.000 MY 2026.122 -2026.122 MZ 7.47E-12 -7.47E-12

JOINT 35 36

COMB DYNAMIC ------------------ MAX JOINT 35 36 FX 0.972230 0.972230 FY 126.377896 126.377896 FZ 68.323489 68.323489 MX 3978.697 3978.697 MY 89.135262 89.135262 MZ 9.788509 9.788509

COMB DYNAMIC ------------------ MIN JOINT 35 36 COMB JOINT 35 36 COMB JOINT 35 36 FX FY -0.972230 -126.377896 -0.972230 -126.377896 FZ -68.323489 -68.323489 MX -3978.697 -3978.697 MY -89.135262 -89.135262 MZ -9.788509 -9.788509

BOTH ------------------ MAX FX 0.972230 0.972230 FY FZ 126.377896 -931.676511 126.377896 -931.676511 MX 87978.697 87978.697 MY 2115.257 -1936.987 MZ 9.788509 9.788509

BOTH ------------------ MIN FX FY -0.972230 -126.377896 -0.972230 -126.377896 FZ -1068.323 -1068.323 MX 80021.303 80021.303 MY 1936.987 -2115.257 MZ -9.788509 -9.788509

...Output has been Edited for Brevity

Figure 25 Restraint Forces (Reactions) in the Output File

108

Global Force Balance Output


126

Chapter VI Joints and Degrees of Freedom For each Analysis Case, the program computes the resultants at the global origin for all joint forces and moments acting on the structure. Separate resultants are computed for each type of joint force: Applied loads Inertial loads Spring forces Nllink forces Restraint Forces (Reactions) Constraint forces P-Delta forces Each force resultant is computed as the sum of the forces acting on all joints in the structure. Each moment resultant is computed as the sum of the moments acting on all joints in the structure, plus the moments about the origin of the forces acting on all joints in the structure. This results in three force and three moment components, all referred to the global coordinate system. The resultants are also computed for the total of all these different forces and moments acting on the structure. These total resultants should be zero if the structure is in exact equilibrium. Due to the approximate nature of computer arithmetic, the totals may not be exactly zero. However, their values relative to the magnitude of the contributing forces gives a measure of the accuracy and stability of the solution. The resultant forces and moments are always printed in the output file for all Analysis Cases under the heading:
GLOBAL FORCE BALANCE

Typical output is shown in Figure 26 (page 110). In addition, relative equilibrium errors are printed in the log file under the heading:
GLOBAL FORCE BALANCE RELATIVE ERRORS

Here the total force and moment components are expressed as a percentage of the maximum possible equilibrium error. The maximum possible error is computed as follows: For each component (FX, FY, FZ, MX, MY, and MZ), the absolute values of the resultants for applied loads, inertial loads, spring forces, Nllink forces, reactions, constraint forces, and P-Delta forces are summed. The maximum of the absolute sums for FX, FY, and FZ is determined. Global Force Balance Output
127

109

SAP2000 Analysis Reference


G L O B A L F O R C E B A L A N C E

TOTAL FORCE AND MOMENT AT THE ORIGIN, IN GLOBAL COORDINATES LOAD 1 -----------------FX .000000 .000000 .000000 3.04E-11 3.04E-11 FY .000000 .000000 -1.95E-10 -2.46E-11 -2.19E-10 FZ 2000.000 .000000 -2000.000 -9.85E-10 2.32E-10 MX .000000 .000000 1.13E-08 1.86E-08 3.00E-08 MY -12000.000 .000000 12000.000 7.33E-09 3.22E-11 MZ .000000 .000000 -1.17E-09 -1.21E-09 -2.38E-09

APPLIED INERTIA REACTNS CONSTRS TOTAL MODE

1 -----------------FX .000000 -4.36E-07 .000000 0.000439 0.000439 FY .000000 -1.18E-06 0.001497 -0.001496 -3.92E-09 FZ MX MY .000000 .000000 .000000 -49672.588 -1.1135E+06 298035.531 49672.588 1.1135E+06 -298035.530 7.42E-05 0.014819 0.002190 4.53E-09 -4.38E-07 0.002633 MZ .000000 1.25E-05 0.008985 -0.030060 -0.021063

APPLIED INERTIA REACTNS CONSTRS TOTAL SPEC

SPEC -----------------FX .000000 0.000784 .000000 0.006039 0.005267 FY .000000 252.768070 252.755793 0.016173 1.03E-11 FZ .000000 136.646868 136.646979 0.004416 1.17E-11 MX .000000 4505.715 4505.613 0.444113 1.14E-09 MY .000000 819.881682 819.881873 0.030539 0.031604 MZ .000000 1516.622 1516.535 0.373549 0.252835

APPLIED INERTIA REACTNS CONSTRS TOTAL COMB

BOTH ------------------ MAX FX .000000 0.000784 .000000 0.006039 0.005267 FY .000000 252.768070 252.755793 0.016173 -2.09E-10 FZ 2000.000 136.646868 -1863.353 0.004416 2.43E-10 MX .000000 4505.715 4505.613 0.444113 3.11E-08 MY -12000.000 819.881682 12819.882 0.030539 0.031604 MZ .000000 1516.622 1516.535 0.373549 0.252835

APPLIED INERTIA REACTNS CONSTRS TOTAL COMB

BOTH ------------------ MIN FX FY FZ .000000 .000000 2000.000 -0.000784 -252.768070 -136.646868 .000000 -252.755793 -2136.647 -0.006039 -0.016173 -0.004416 -0.005267 -2.29E-10 2.20E-10 MX MY .000000 -12000.000 -4505.715 -819.881682 -4505.613 11180.118 -0.444113 -0.030539 2.88E-08 -0.031604 MZ .000000 -1516.622 -1516.535 -0.373549 -0.252835

APPLIED INERTIA REACTNS CONSTRS TOTAL

...Output has been Edited for Brevity

Figure 26 Global Force Balance in the Output File

110

Global Force Balance Output


128

Chapter VI Joints and Degrees of Freedom


G L O B A L F O R C E B A L A N C E R E L A T I V E E R R O R S

PERCENT FORCE AND MOMENT ERROR AT THE ORIGIN, IN GLOBAL COORDINATES LOAD 1 MODE 1 2 3 4 SPEC SPEC FX 7.60E-13 FX 4.42E-07 7.22E-05 0.000232 0.007859 FX 0.001042 FY 5.48E-12 FY 3.94E-12 2.08E-13 3.36E-13 3.36E-14 FY 2.03E-12 FY 5.48E-12 5.48E-12 2.03E-12 2.03E-12 5.22E-12 5.37E-12 FZ 5.79E-12 FZ 4.56E-12 3.59E-14 1.62E-14 9.51E-15 FZ 2.32E-12 FZ 5.79E-12 5.79E-12 2.32E-12 2.32E-12 6.09E-12 5.15E-12 MX 2.20E-11 MX 1.30E-11 1.42E-12 1.23E-13 9.00E-14 MX 1.22E-11 MX 2.20E-11 2.20E-11 1.22E-11 1.22E-11 2.29E-11 1.98E-11 MY 2.36E-14 MY 7.79E-08 1.28E-05 0.000139 0.001394 MY 0.000340 MY 2.36E-14 2.36E-14 0.000340 0.000340 2.32E-05 2.17E-05 MZ 1.75E-12 MZ 6.23E-07 0.000102 0.001113 0.011150 MZ 0.002719 MZ 1.75E-12 1.75E-12 0.002719 0.002719 0.000186 0.000174

COMB MAX/MIN FX STATIC 7.60E-13 STATIC 7.60E-13 DYNAMIC 0.001042 DYNAMIC 0.001042 BOTH 0.000132 BOTH 0.000123

Figure 27 Global Force Balance Relative Errors in the Log File

The maximum of the absolute sums for MX, MY, and MZ is determined. The maximum possible error for the force components is the maximum of the absolute force sums, or the maximum of the absolute moment sums divided by the average moment arm for the structure, whichever is larger. The maximum possible error for the moment components is the maximum of the absolute moment sums, or the maximum of the absolute force sums multiplied by the average moment arm for the structure, whichever is larger. This definition, while complicated, assures that only numerically meaningful equilibrium errors are indicated as such. Typical output is shown in Figure 27 (page 111). It is strongly recommended that you always review the global force balance in the output file and the relative equilibrium errors in the log file. These results can alert you to potential problems with the structural model or the analysis. In particular, you should look for the following: The accuracy of the computer calculations is on the order of 10-15, or 10-13 percent. If the relative force or moment error for the Load Cases is significantly

Global Force Balance Output


129

111

SAP2000 Analysis Reference larger than this, it may indicate that the structure is unstable or that the stiffness matrix is ill-conditioned. You can expect somewhat larger equilibrium errors for Vibration Modes than for Load Cases. For Eigen Modes, the errors generally reflect the convergence tolerance and can be reduced by using a smaller tolerance. For Ritz Modes, the errors indicate the fact that the Ritz modes are not the true Eigen Modes of the structure. These errors can generally be reduced by requesting more modes. Some Modes with large errors may be orthogonal to the Ritz starting vectors and have no effect upon response-spectrum and time-history analyses. The constraint forces should be self-equilibrating, i.e., the resultant constraint forces and moments should be essentially zero. If the values are significantly different from zero (compared to the other resultants), the constraints may be poorly defined. An example of this is a Diaphragm constraint whose joints are not all in the same plane: in-plane forces in the Diaphragm can cause moments that are not captured by the structural model. Such errors in the constraints affect the validity of the model even though the overall equilibrium of the structure may be satisfied. See Topic Force Output (page 102) in this chapter for more information.

112

Global Force Balance Output


130

C h a p t e r VII

Constraints and Welds


Constraints are used to enforce certain types of rigid-body behavior, to connect together different parts of the model, and to impose certain types of symmetry conditions. Welds are used to generate a set of constraints that connect together different parts of the model. Basic Topics for All Users Overview Body Constraint Plane Definition Diaphragm Constraint Plate Constraint Axis Definition Rod Constraint Beam Constraint Equal Constraint Welds

113
131

SAP2000 Analysis Reference Advanced Topics Local Constraint Automatic Master Joints Constraint Output

Overview
A constraint consists of a set of two or more constrained joints. The displacements of each pair of joints in the constraint are related by constraint equations. The types of behavior that can be enforced by constraints are: Rigid-body behavior, in which the constrained joints translate and rotate together as if connected by rigid links. The types of rigid behavior that can be modeled are: Rigid Body: fully rigid for all displacements Rigid Diaphragm: rigid for membrane behavior in a plane Rigid Plate: rigid for plate bending in a plane Rigid Rod: rigid for extension along an axis Rigid Beam: rigid for beam bending on an axis Equal-displacement behavior, in which the translations and rotations are equal at the constrained joints Symmetry and anti-symmetry conditions The use of constraints reduces the number of equations in the system to be solved and will usually result in increased computational efficiency. Most constraint types must be defined with respect to some fixed coordinate system. The coordinate system may be the global coordinate system or an alternate coordinate system, or it may be automatically determined from the locations of the constrained joints. The Local Constraint does not use a fixed coordinate system, but references each joint using its own joint local coordinate system. Welds are used to connect together different parts of the model that were defined separately. Each Weld consists of a set of joints that may be joined. The program searches for joints in each Weld that share the same location in space and constrains them to act as a single joint.

114

Overview
132

Chapter VII Constraints and Welds

Body Constraint
A Body Constraint causes all of its constrained joints to move together as a threedimensional rigid body. Effectively, all constrained joints are connected to each other by rigid links and cannot displace relative to each other. This Constraint can be used to: Model rigid connections, such as where several beams and/or columns frame together Connect together different parts of the structural model that were defined using separate meshes Connect Frame elements that are acting as eccentric stiffeners to Shell elements Welds can be used to automatically generate Body Constraints for the purpose of connecting coincident joints. See Topic Welds (page 131) in this chapter for more information.

Joint Connectivity
Each Body Constraint connects a set of two or more joints together. The joints may have any arbitrary location in space.

Local Coordinate System


Each Body Constraint has its own local coordinate system, the axes of which are denoted 1, 2, and 3. These correspond to the X, Y, and Z axes of the global coordinate system, respectively. The actual orientation of the local axes is not important since the constraint equations are independent of the coordinate system.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Body Constraint. These equations are expressed in terms of the translations (u1, u2, and u3), the rotations (r1, r2, and r3), and the coordinates (x1, x2, and x3), all taken in the Constraint local coordinate system: u1j = u1i + r2i x3 r3i x2 u2j = u2i + r3i x1 r1i x3

Body Constraint
133

115

SAP2000 Analysis Reference u3j = u3i + r1i x2 r2i x1 r1i = r1j r2i = r2j r3i = r3j where x1 = x1j x1i, x2 = x2j x2i, and x3 = x3j x3i.

Plane Definition
The constraint equations for each Diaphragm or Plate Constraint are written with respect to a particular plane. The location of the plane is not important, only its orientation. By default, the plane is determined automatically by the program from the spatial distribution of the constrained joints as follows: The centroid of the constrained joints is determined The second moments of the locations of all of the constrained joints about the centroid are determined The principal values and directions of these second moments are found The direction of the smallest principal second moment is taken as the normal to the constraint plane; if all constrained joints lie in a unique plane, this smallest principal moment will be zero If no unique direction can be found, a horizontal (X-Y) plane is assumed in coordinate system csys; this situation can occur if the joints are coincident or collinear, or if the spatial distribution is more nearly three-dimensional than planar. You may override automatic plane selection by specifying the following: csys: A fixed coordinate system (the default is zero, indicating the global coordinate system) axis: The axis (X, Y, or Z) normal to the plane of the constraint, taken in coordinate system csys. This may be useful, for example, to specify a horizontal plane for a floor with a small step in it.

116

Plane Definition
134

Chapter VII Constraints and Welds

Diaphragm Constraint
A Diaphragm Constraint causes all of its constrained joints to move together as a planar diaphragm that is rigid against membrane deformation. Effectively, all constrained joints are connected to each other by links that are rigid in the plane, but do not affect out-of-plane (plate) deformation. This Constraint can be used to: Model concrete floors (or concrete-filled decks) in building structures, which typically have very high in-plane stiffness Model diaphragms in bridge superstructures The use of the Diaphragm Constraint for building structures eliminates the numerical-accuracy problems created when the large in-plane stiffness of a floor diaphragm is modeled with membrane elements. It is also very useful in the lateral (horizontal) dynamic analysis of buildings, as it results in a significant reduction in the size of the eigenvalue problem to be solved. See Figure 28 (page 118) for an illustration of a floor diaphragm.

Joint Connectivity
Each Diaphragm Constraint connects a set of two or more joints together. The joints may have any arbitrary location in space, but for best results all joints should lie in the plane of the constraint. Otherwise, bending moments may be generated that are restrained by the Constraint, which unrealistically stiffens the structure. If this happens, the constraint forces printed in the output file may not be in equilibrium. For more information, see: Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom. Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

Local Coordinate System


Each Diaphragm Constraint has its own local coordinate system, the axes of which are denoted 1, 2, and 3. Local axis 3 is always normal to the plane of the constraint. The program arbitrarily chooses the orientation of axes 1 and 2 in the plane. The actual orientation of the planar axes is not important since only the normal direction affects the constraint equations. Diaphragm Constraint
135

117

SAP2000 Analysis Reference

Figure 28 Use of the Diaphragm Constraint to Model a Rigid Floor Slab

For more information, see Topic Plane Definition (page 116) in this chapter.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Diaphragm Constraint. These equations are expressed in terms of in-plane translations (u1 and u2), the rotation (r3) about the normal, and the in-plane coordinates (x1 and x2), all taken in the Constraint local coordinate system: u1j = u1i r3i x2 u2j = u2i + r3i x1

118

Diaphragm Constraint
136

Chapter VII Constraints and Welds r3i = r3j where x1 = x1j x1i and x2 = x2j x2i.

Plate Constraint
A Plate Constraint causes all of its constrained joints to move together as a flat plate that is rigid against bending deformation. Effectively, all constrained joints are connected to each other by links that are rigid for out-of-plane bending, but do not affect in-plane (membrane) deformation. This Constraint can be used to: Connect structural-type elements (Frame and Shell) to solid-type elements (Plane and Solid); the rotation in the structural element can be converted to a pair of equal and opposite translations in the solid element by the Constraint Enforce the assumption that plane sections remain plane in detailed models of beam bending

Joint Connectivity
Each Plate Constraint connects a set of two or more joints together. The joints may have any arbitrary location in space. Unlike the Diaphragm Constraint, equilibrium is not affected by whether or not all joints lie in the plane of the Plate Constraint.

Local Coordinate System


Each Plate Constraint has its own local coordinate system, the axes of which are denoted 1, 2, and 3. Local axis 3 is always normal to the plane of the constraint. The program arbitrarily chooses the orientation of axes 1 and 2 in the plane. The actual orientation of the planar axes is not important since only the normal direction affects the constraint equations. For more information, see Topic Plane Definition (page 116) in this chapter.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Plate Constraint. These equations are expressed in terms of the out-of-plane translation (u3), the bending rotations (r1 and r2), and the in-plane coordinates (x1 and x2), all taken in the Constraint local coordinate system: Plate Constraint
137

119

SAP2000 Analysis Reference u3j = u3i + r1i x2 r2i x1 r1i = r1j r2i = r2j where x1 = x1j x1i and x2 = x2j x2i.

Axis Definition
The constraint equations for each Rod or Beam Constraint are written with respect to a particular axis. The location of the axis is not important, only its orientation. By default, the axis is determined automatically by the program from the spatial distribution of the constrained joints as follows: The centroid of the constrained joints is determined The second moments of the locations of all of the constrained joints about the centroid are determined The principal values and directions of these second moments are found The direction of the largest principal second moment is taken as the axis of the constraint; if all constrained joints lie on a unique axis, the two smallest principal moments will be zero If no unique direction can be found, a vertical (Z) axis is assumed in coordinate system csys; this situation can occur if the joints are coincident, or if the spatial distribution is more nearly planar or three-dimensional than linear. You may override automatic axis selection by specifying the following: csys: A fixed coordinate system (the default is zero, indicating the global coordinate system) axis: The axis (X, Y, or Z) of the constraint, taken in coordinate system csys. This may be useful, for example, to specify a vertical axis for a column with a small offset in it.

Rod Constraint
A Rod Constraint causes all of its constrained joints to move together as a straight rod that is rigid against axial deformation. Effectively, all constrained joints main-

120

Axis Definition
138

Chapter VII Constraints and Welds

Figure 29 Use of the Rod Constraint to Model Axially Rigid Beams

tain a fixed distance from each other in the direction parallel to the axis of the rod, but translations normal to the axis and all rotations are unaffected. This Constraint can be used to: Prevent axial deformation in Frame elements Model rigid truss-like links An example of the use of the Rod Constraint is in the analysis of the twodimensional frame shown in Figure 29 (page 121). If the axial deformations in the beams are negligible, a single Rod Constraint could be defined containing the five joints. Instead of five equations, the program would use a single equation to define the X-displacement of the whole floor. However, it should be noted that this will result in the axial forces of the beams being output as zero, as the Constraint will cause the ends of the beams to translate together in the X-direction. Interpretations of such results associated with the use of Constraints should be clearly understood.

Joint Connectivity
Each Rod Constraint connects a set of two or more joints together. The joints may have any arbitrary location in space, but for best results all joints should lie on the axis of the constraint. Otherwise, bending moments may be generated that are re-

Rod Constraint
139

121

SAP2000 Analysis Reference strained by the Constraint, which unrealistically stiffens the structure. If this happens, the constraint forces printed in the output file may not be in equilibrium. For more information, see: Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom. Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

Local Coordinate System


Each Rod Constraint has its own local coordinate system, the axes of which are denoted 1, 2, and 3. Local axis 1 is always the axis of the constraint. The program arbitrarily chooses the orientation of the transverse axes 2 and 3. The actual orientation of the transverse axes is not important since only the axial direction affects the constraint equations. For more information, see Topic Axis Definition (page 120) in this chapter.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Rod Constraint. These equations are expressed only in terms of the axial translation (u1): u1j = u1i

Beam Constraint
A Beam Constraint causes all of its constrained joints to move together as a straight beam that is rigid against bending deformation. Effectively, all constrained joints are connected to each other by links that are rigid for off-axis bending, but do not affect translation along or rotation about the axis. This Constraint can be used to: Connect structural-type elements (Frame and Shell) to solid-type elements (Plane and Solid); the rotation in the structural element can be converted to a pair of equal and opposite translations in the solid element by the Constraint Prevent bending deformation in Frame elements

122

Beam Constraint
140

Chapter VII Constraints and Welds

Joint Connectivity
Each Beam Constraint connects a set of two or more joints together. The joints may have any arbitrary location in space, but for best results all joints should lie on the axis of the constraint. Otherwise, torsional moments may be generated that are restrained by the Constraint, which unrealistically stiffens the structure. If this happens, the constraint forces printed in the output file may not be in equilibrium. For more information, see: Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom

Local Coordinate System


Each Beam Constraint has its own local coordinate system, the axes of which are denoted 1, 2, and 3. Local axis 1 is always the axis of the constraint. The program arbitrarily chooses the orientation of the transverse axes 2 and 3. The actual orientation of the transverse axes is not important since only the axial direction affects the constraint equations. For more information, see Topic Axis Definition (page 120) in this chapter.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Beam Constraint. These equations are expressed in terms of the transverse translations (u2 and u3), the transverse rotations (r2 and r3), and the axial coordinate (x1), all taken in the Constraint local coordinate system: u2j = u2i + r3i x1 u3j = u3i r2i x1 r2i = r2j r3i = r3j where x1 = x1j x1i.

Beam Constraint
141

123

SAP2000 Analysis Reference

Equal Constraint
An Equal Constraint causes all of its constrained joints to move together with the same (or opposite) displacements for each selected degree of freedom, taken in the constraint local coordinate system. The other degrees of freedom are unaffected. The Equal Constraint differs from the rigid-body types of Constraints in that there is no coupling between the rotations and the translations. This Constraint can be used to: Model symmetry and anti-symmetry conditions with respect to a plane Partially connect together different parts of the structural model, such as at expansion joints and hinges For fully connecting meshes, it is better to use the Body Constraint when the constrained joints are not in exactly the same location.

Joint Connectivity
Each Equal Constraint connects a set of two or more joints together. For a given Constraint, if any of the selected degrees of freedom are negative (i.e., opposite), only two constrained joints are allowed for that Constraint. Otherwise any number of constrained joints are permitted. The joints may have any arbitrary location in space, but for best results all joints should share the same location in space if used for connecting meshes. Otherwise, moments may be generated that are restrained by the Constraint, which unrealistically stiffens the structure. If this happens, the constraint forces printed in the output file may not be in equilibrium. Such restrained moments may also be generated when Equal Constraints are used for symmetry purposes. They are necessary to enforce the desired symmetry or anti-symmetry of the displacements when the applied loads are not correspondingly symmetric or anti-symmetric. For more information, see: Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom. Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

124

Equal Constraint
142

Chapter VII Constraints and Welds

Local Coordinate System


Each Equal Constraint uses a fixed coordinate system, csys, that you specify. The default for csys is zero, indicating the global coordinate system. The axes of the fixed coordinate system are denoted X, Y, and Z.

Selected Degrees of Freedom


For each Equal Constraint you may specify a list, cdofs, of up to six degrees of freedom in coordinate system csys that are to be constrained. The degrees of freedom are indicated as UX, UY, UZ, RX, RY, and RZ. A negative sign indicates a degree of freedom that is constrained to be opposite, e.g., UX.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in an Equal Constraint. These equations are expressed in terms of the translations (ux, uy, and uz) and the rotations (rx, ry, and rz), all taken in fixed coordinate system csys. The equations used depend upon the selected degrees of freedom and their signs. Some important cases are described next. Symmetry About a Plane For a structure that is symmetric about a plane, symmetric loading causes symmetric displacements as follows: Forces and displacements parallel to the plane of symmetry are equal Forces and displacements normal to the plane of symmetry are opposite Moments and rotations parallel to the plane of symmetry are opposite Moments and rotations normal to the plane of symmetry are equal As an example, consider a structure that is symmetric with respect to a plane normal to the X axis and subjected to symmetric loading. A separate Equal Constraint must be defined for each pair of joints that is symmetrically located with respect to the plane. The degrees of freedom that would be specified for these Constraints are: UX, UY, UZ, RX, RY, and RZ. The corresponding constraint equations are: uxj = uxi uyj = uyi

Equal Constraint
143

125

SAP2000 Analysis Reference uzj = uzi rxi = rxj ryi = ryj rzi = rzj Any joints on the plane of symmetry should not be constrained, but instead have their UX, RY, and RZ degrees of freedom restrained. Anti-symmetry About a Plane For a structure that is symmetric about a plane, anti-symmetric loading causes anti-symmetric displacements. All degrees of freedom that are equal when symmetric are opposite when anti-symmetric, and all degrees of freedom that are opposite when symmetric are equal when anti-symmetric. Thus the specification of the anti-symmetric degrees of freedom simply uses the opposite signs from the symmetric case. Consider the example above of a structure that is symmetric with respect to a plane normal to the X axis, but now subjected to anti-symmetric loading. A separate Equal Constraint must be defined for each pair of joints that is symmetrically located with respect to the plane. The degrees of freedom that would be specified for these Constraints are: UX, UY, UZ, RX, RY, and RZ. The signs of the constraint equations are corresponding changed from the symmetric case. Partial Connection When joints are being connected, all specified degrees of freedom are positive. For example, consider an idealized hinge connection of eight space-truss members. Only displacements are continuous across the hinge, not rotations. Each truss member is connected to a separate joint (node) at the connection. One Equal Constraint is defined for the eight constrained joints. The degrees of freedom that would be specified for this Constraint are: UX, UY, and UZ. The corresponding constraint equations are: uxj = uxi uyj = uyi uzj = uzi

126

Equal Constraint
144

Chapter VII Constraints and Welds The eight joints should be coincident or the axes of the truss members should all intersect at the same point. Otherwise, moments may be generated that are unrealistically restrained by the Constraint.

Local Constraint
A Local Constraint causes all of its constrained joints to move together with the same (or opposite) displacements for each selected degree of freedom, taken in the separate joint local coordinate systems. The other degrees of freedom are unaffected. The Local Constraint differs from the rigid-body types of Constraints in that there is no coupling between the rotations and the translations. The Local Constraint is the same as the Equal Constraint if all constrained joints have the same local coordinate system. This Constraint can be used to: Model symmetry conditions with respect to a line or a point Model displacements constrained by mechanisms The behavior of this Constraint is dependent upon the choice of the local coordinate systems of the constrained joints.

Joint Connectivity
Each Local Constraint connects a set of two or more joints together. If any of the selected degrees of freedom for a given Constraint are negative (i.e., opposite) only two constrained joints are allowed for that Constraint. Otherwise any number of constrained joints are permitted. The joints may have any arbitrary location in space. If the joints do not share the same location in space, moments may be generated that are restrained by the Constraint. If this happens, the constraint forces printed in the output file may not be in equilibrium. These moments are necessary to enforce the desired symmetry of the displacements when the applied loads are not symmetric, or may represent the constraining action of a mechanism. For more information, see: Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom.

Local Constraint
145

127

SAP2000 Analysis Reference Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

No Local Coordinate System


A Local Constraint does not have its own local coordinate system. The constraint equations are written in terms of constrained joint local coordinate systems, which may differ. The axes of these coordinate systems are denoted 1, 2, and 3.

Selected Degrees of Freedom


For each Local Constraint you may specify a list, ldofs, of up to six degrees of freedom in the joint local coordinate systems that are to be constrained. The degrees of freedom are indicated as U1, U2, U3, R1, R2, and R3. A negative sign indicates a degree of freedom that is constrained to be opposite, e.g., U1.

Constraint Equations
The constraint equations relate the displacements at any two constrained joints (subscripts i and j) in a Local Constraint. These equations are expressed in terms of the translations (u1, u2, and u3) and the rotations (r1, r2, and r3), all taken in joint local coordinate systems. The equations used depend upon the selected degrees of freedom and their signs. Some important cases are described next. Axisymmetry Axisymmetry is a type of symmetry about a line. It is best described in terms of a cylindrical coordinate system having its Z axis on the line of symmetry. The structure, loading, and displacements are each said to be axisymmetric about a line if they do not vary with angular position around the line, i.e., they are independent of the angular coordinate CA. To enforce axisymmetry using the Local Constraint: Model any cylindrical sector of the structure using any axisymmetric mesh of joints and elements Assign each joint a local coordinate system such that local axes 1, 2, and 3 correspond to the coordinate directions +CR, +CA, and +CZ, respectively For each axisymmetric set of joints (i.e., having the same coordinates CR and CZ, but different CA), define a Local Constraint using all six degrees of freedom: U1, U2, U3, R1, R2, and R3

128

Local Constraint
146

Chapter VII Constraints and Welds Restrain joints that lie on the line of symmetry so that, at most, only axial translations (U3) and rotations (R3) are permitted The corresponding constraint equations are: u1j = u1i u2j = u2i u3j = u3i r1i = r1j r2i = r2j r3i = r3j The numeric subscripts refer to the corresponding joint local coordinate systems. Cyclic symmetry Cyclic symmetry is another type of symmetry about a line. It is best described in terms of a cylindrical coordinate system having its Z axis on the line of symmetry. The structure, loading, and displacements are each said to be cyclically symmetric about a line if they vary with angular position in a repeated (periodic) fashion. To enforce cyclic symmetry using the Local Constraint: Model any number of adjacent, representative, cylindrical sectors of the structure; denote the size of a single sector by the angle Assign each joint a local coordinate system such that local axes 1, 2, and 3 correspond to the coordinate directions +CR, +CA, and +CZ, respectively For each cyclically symmetric set of joints (i.e., having the same coordinates CR and CZ, but with coordinate CA differing by multiples of ), define a Local Constraint using all six degrees of freedom: U1, U2, U3, R1, R2, and R3. Restrain joints that lie on the line of symmetry so that, at most, only axial translations (U3) and rotations (R3) are permitted The corresponding constraint equations are: u1j = u1i u2j = u2i u3j = u3i Local Constraint
147

129

SAP2000 Analysis Reference r1i = r1j r2i = r2j r3i = r3j The numeric subscripts refer to the corresponding joint local coordinate systems. For example, suppose a structure is composed of six identical 60 sectors, identically loaded. If two adjacent sectors were modeled, each Local Constraint would apply to a set of two joints, except that three joints would be constrained on the symmetry planes at 0, 60, and 120. If a single sector is modeled, only joints on the symmetry planes need to be constrained. Symmetry About a Point Symmetry about a point is best described in terms of a spherical coordinate system having its Z axis on the line of symmetry. The structure, loading, and displacements are each said to be symmetric about a point if they do not vary with angular position about the point, i.e., they are independent of the angular coordinates SB and SA. Radial translation is the only displacement component that is permissible. To enforce symmetry about a point using the Local Constraint: Model any spherical sector of the structure using any symmetric mesh of joints and elements Assign each joint a local coordinate system such that local axes 1, 2, and 3 correspond to the coordinate directions +SB, +SA, and +SR, respectively For each symmetric set of joints (i.e., having the same coordinate SR, but different coordinates SB and SA), define a Local Constraint using only degree of freedom U3 For all joints, restrain the degrees of freedom U1, U2, R1, R2, and R3 Fully restrain any joints that lie at the point of symmetry The corresponding constraint equations are: u3j = u3i The numeric subscripts refer to the corresponding joint local coordinate systems. It is also possible to define a case for symmetry about a point that is similar to cyclic symmetry around a line, e.g., where each octant of the structure is identical.

130

Local Constraint
148

Chapter VII Constraints and Welds

Figure 30 Use of a Weld to Connect Separate Meshes at Coincident Joints

Welds
A Weld can be used to connect together different parts of the structural model that were defined using separate meshes. A Weld is not a single Constraint, but rather is a set of joints from which the program will automatically generate multiple Body Constraints to connect together coincident joints. Joints are considered to be coincident if the distance between them is less than or equal to a tolerance, tol, that you specify. Setting the tolerance to zero is permissible but is not recommended. One or more Welds may be defined, each with its own tolerance. Only the joints within each Weld will be checked for coincidence with each other. In the most common case, a single Weld is defined that contains all joints in the model; all coincident groups of joints will be welded. However, in situations where structural discontinuity is desired, it may be necessary to prevent the welding of some coincident joints. This may be facilitated by the use of multiple Welds. Figure 30 (page 131) shows a model developed as two separate meshes, A and B. Joints 121 through 125 are associated with mesh A, and Joints 221 through 225 are associated with mesh B. Joints 121 through 125 share the same location in space as Joints 221 through 225, respectively. These are the interfacing joints between the two meshes. To connect these two meshes, a single Weld can be defined containing all joints, or just joints 121 through 125 and 221 through 225. The program would

Welds
149

131

SAP2000 Analysis Reference generate five Body Constraints, each containing two joints, resulting in an integrated model. It is permissible to include the same joint in more than one Weld. This could result in the joints in different Welds being constrained together if they are coincident with the common joint. For example, suppose that Weld 1 contained joints 1,2, and 3, Weld 2 contained joints 3, 4, and 5. If joints 1, 3, and 5 were coincident, joints 1 and 3 would be constrained by Weld 1, and joints 3 and 5 would be constrained by Weld 2. The program would create a single Body Constraint containing joints 1, 3, and 5. One the other hand, if Weld 2 did not contain joint 3, the program would only generate a Body Constraint containing joint 1 and 3 from Weld 1; joint 5 would not be constrained. For more information, see Topic Body Constraint (page 115) in this chapter.

Automatic Master Joints


The program automatically creates an internal master joint for each explicit Constraint, and a master joint for each internal Body Constraint that is generated by a Weld. Each master joint governs the behavior of the corresponding constrained joints. The displacement at a constrained degree of freedom is computed as a linear combination of the displacements of the master joint. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom for more information.

Stiffness, Mass, and Loads


Joint local coordinate systems, springs, masses, and loads may all be applied to constrained joints. Elements may also be connected to constrained joints. The joint and element stiffnesses, masses and loads from the constrained degrees of freedom are be automatically transferred to the master joint in a consistent fashion. The translational stiffness at the master joint is the sum of the translational stiffnesses at the constrained joints. The same is true for translational masses and loads. The rotational stiffness at a master joint is the sum of the rotational stiffnesses at the constrained degrees of freedom, plus the second moment of the translational stiffnesses at the constrained joints for the Body, Diaphragm, Plate, and Beam Constraints. The same is true for rotational masses and loads, except that only the first moment of the translational loads is used. The moments of the translational stiff-

132

Automatic Master Joints


150

Chapter VII Constraints and Welds nesses, masses, and loads are taken about the center of mass of the constrained joints. If the joints have no mass, the centroid is used.

Local Coordinate Systems


Each master joint has two local coordinate systems: one for the translational degrees of freedom, and one for the rotational degrees of freedom. The axes of each local system are denoted 1, 2, and 3. For the Local Constraint, these axes correspond to the local axes of the constrained joints. For other types of Constraints, these axes are chosen to be the principal directions of the translational and rotational masses of the master joint. Using the principal directions eliminates coupling between the mass components in the master-joint local coordinate system. For a Diaphragm or Plate Constraint, the local 3 axes of the master joint are always normal to the plane of the Constraint. For a Beam or Rod Constraint, the local 1 axes of the master joint are always parallel to the axis of the Constraint.

Constraint Output
For each Body, Diaphragm, Plate, Rod, and Beam Constraint having more than two constrained joints, the following information about the Constraint and its master joint is printed in the output file: The translational and rotational local coordinate systems for the master joint The total mass and mass moments of inertia for the Constraint that have been applied to the master joint The center of mass for each of the three translational masses The degrees of freedom are indicated as U1, U2, U3, R1, R2, and R3. These are referred to the two local coordinate systems of the master joint. Typical constraint output is shown in Figure 31 (page 134).

Constraint Output
151

133

SAP2000 Analysis Reference


C O N S T R A I N T CONS C O O R D I N A T E S A N D M A S S E S

1 ================== TYPE = DIAPH, NORMAL DIRECTION = U3 LOCAL COORDINATE SYSTEM FOR CONSTRAINT MASTER U2 U3 R1 R2 .000000 .000000 1.000000 .000000 1.000000 .000000 .000000 1.000000 .000000 1.000000 .000000 .000000 TRANSLATIONAL MASS AND MASS MOMENTS OF INERTIA U2 U3 R1 R2 3.880000 .000000 .000000 .000000 CENTER OF MASS U2 25.000000 12.500000 12.500000

GLOBAL X Y Z

U1 1.000000 .000000 .000000

R3 .000000 .000000 1.000000

U1 3.880000

R3 1011.000

GLOBAL X Y Z CONS

U1 25.000000 12.500000 12.500000

U3 .000000 .000000 .000000

2 ================== TYPE = DIAPH, NORMAL DIRECTION = U3 LOCAL COORDINATE SYSTEM FOR CONSTRAINT MASTER U2 U3 R1 R2 .000000 .000000 1.000000 .000000 1.000000 .000000 .000000 1.000000 .000000 1.000000 .000000 .000000 TRANSLATIONAL MASS AND MASS MOMENTS OF INERTIA U2 U3 R1 R2 3.880000 .000000 .000000 .000000 CENTER OF MASS U2 25.000000 12.500000 25.000000

GLOBAL X Y Z

U1 1.000000 .000000 .000000

R3 .000000 .000000 1.000000

U1 3.880000

R3 1011.000

GLOBAL X Y Z

U1 25.000000 12.500000 25.000000

U3 .000000 .000000 .000000

...Output Edited for Brevity...

Figure 31 Constraint Coordinates and Masses in the Output File

134

Constraint Output
152

C h a p t e r VIII

Material Properties
The Materials are used to define the mechanical, thermal, and density properties used by the Frame, Shell, Plane, Asolid, and Solid elements. Basic Topics for All Users Overview Local Coordinate System Stresses and Strains Isotropic Materials Mass Density Weight Density Design-Type Indicator Advanced Topics Orthotropic Materials Anisotropic Materials Temperature-Dependent Materials Element Material Temperature

135
153

SAP2000 Analysis Reference

Overview
The Material properties are always linear elastic. They may be defined as being isotropic, orthotropic or anisotropic. How the properties are actually utilized depends on the element type. Each Material that you define may be used by more than one element or element type. For the Frame element, the Materials are referenced indirectly through the Section properties. All material properties, except mass and weight density, may be temperature dependent. Properties are given at a series of specified temperatures. Properties at other temperatures are obtained by linear interpolation. For a given execution of the program, the properties used by an element are assumed to be constant regardless of any temperature changes experienced by the structure. Each element may be assigned a material temperature that determines the material properties used for the analysis.

Local Coordinate System


Each Material has its own Material local coordinate system used to define the elastic and thermal properties. This system is significant only for orthotropic and anisotropic materials. Isotropic materials are independent of any particular coordinate system. The axes of the Material local coordinate system are denoted 1, 2, and 3. By default, the Material coordinate system is aligned with the local coordinate system for each element. However, you may specify a set of one or more material angles that rotate the Material coordinate system with respect to the element system for those elements that permit orthotropic or anisotropic properties. For more information: See Topic Material Angle (page 212) in Chapter The Plane Element. See Topic Material Angle (page 226) in Chapter The Asolid Element. See Topic Material Angles (page 237) in Chapter The Solid Element.

136

Overview
154

Chapter VIII Material Properties

Figure 32 Definition of Stress Components in the Material Local Coordinate System

Stresses and Strains


The elastic mechanical properties relate the behavior of the stresses and strains within the Material. The stresses are defined as forces per unit area acting on an elemental cube aligned with the material axes as shown in Figure 32 (page 137). The stresses 11 , 22 , and 33 are called the direct stresses and tend to cause length change, while 12 , 13 , and 23 are called the shear stresses and tend to cause angle change. Not all stress components exist in every element type. For example, the stresses 22 , 33 , and 23 are assumed to be zero in the Frame element, and stress 33 is taken to be zero in the Shell element. The direct strains 11 , 22 , and 33 measure the change in length along the Material local 1, 2, and 3 axes, respectively, and are defined as: 11 = 22 =
du1 dx 1 du 2 dx 2

Stresses and Strains


155

137

SAP2000 Analysis Reference 33 =


du 3 dx 3

where u1, u2, and u3 are the displacements and x1, x2, and x3 are the coordinates in the Material 1, 2, and 3 directions, respectively. The engineering shear strains 12 , 13 , and 23 , measure the change in angle in the Material local 1-2, 1-3, and 2-3 planes, respectively, and are defined as: 12 = 13 = 23 =
du1 dx 2 du1 dx 3 du 2 dx 3

+ + +

du 2 dx 1 du 3 dx 1 du 3 dx 2

Note that the engineering shear strains are equal to twice the tensorial shear strains 12 , 13 , and 23 , respectively. Strains can also be caused by a temperature change, T , from a zero-stress reference temperature. No stresses are caused by a temperature change unless the induced thermal strains are restrained. See Cook, Malkus, and Plesha (1989), or any textbook on elementary mechanics.

Isotropic Materials
The behavior of an isotropic material is independent of the direction of loading or the orientation of the material. In addition, shearing behavior is uncoupled from extensional behavior and is not affected by temperature change. Isotropic behavior is usually assumed for steel and concrete, although this is not always the case. The isotropic mechanical and thermal properties relate strain to stress and temperature change as follows:

138

Isotropic Materials
156

Chapter VIII Material Properties 1 e1 11 22 33 = 12 13 23


-u12 -u12

e1
1

e1
-u12

0 0 0 1

0 0 0 0 1

e1

e1
1

e1

g12
sym.

g12

0 11 a1 22 a1 0 33 a1 + T 0 12 0 13 0 0 23 0 1 g12
0

(Eqn. 1)

where e1 is Youngs modulus of elasticity, u12 is Poissons ratio, g12 is the shear modulus, and a1 is the coefficient of thermal expansion. This relationship holds regardless of the orientation of the Material local 1, 2, and 3 axes. The shear modulus is not directly specified, but instead is defined in terms of Youngs modulus and Poissons ratio as: g12 = e1 2 (1 + u12 )

Note that Youngs modulus must be positive, and Poissons ratio must satisfy the condition: 1< u12 <
1 2

Orthotropic Materials
The behavior of an orthotropic material can be different in each of the three local coordinate directions. However, like an isotropic material, shearing behavior is uncoupled from extensional behavior and is not affected by temperature change. The orthotropic mechanical and thermal properties relate strain to stress and temperature change as follows:

Orthotropic Materials
157

139

SAP2000 Analysis Reference 1 e1 11 22 33 = 12 13 23


-u12 -u13

e2
1

e3
-u23

0 0 0 1

0 0 0 0 1

e2

e3
1

e3

g12
sym.

g13

0 11 a1 22 a2 0 33 a3 + T 0 12 0 13 0 0 23 0 1 g23
0

(Eqn. 2)

where e1, e2, and e3 are the moduli of elasticity; u12, u13, and u23 are the Poissons ratios; g12, g13, and g23 are the shear moduli; and a1, a2, and a3 are the coefficients of thermal expansion. Note that the elastic moduli and the shear moduli must be positive. The Poissons ratios may take on any values provided that the upper-left 3x3 portion of the stressstrain matrix is positive-definite (i.e., has a positive determinant.)

Anisotropic Materials
The behavior of an anisotropic material can be different in each of the three local coordinate directions. In addition, shearing behavior can be fully coupled with extensional behavior and can be affected by temperature change. The anisotropic mechanical and thermal properties relate strain to stress and temperature change as follows:

140

Anisotropic Materials
158

Chapter VIII Material Properties 1 e1 11 22 33 = 12 13 23


-u12 -u13 -u14 -u15

e2
1

e3
-u23

g12
-u24

g13
-u25

e2

e3
1

g12
-u34

g13
-u35

e3

g12
1

g13
-u45

g12
sym.

g13
1

g13

(Eqn. 3) g23 -u26 a1 g23 11 a2 -u36 22 g23 33 + a3 T -u46 a12 12 g23 13 a13 -u56 23 a23 g23 1 g23

-u16

where e1, e2, and e3 are the moduli of elasticity; u12, u13, and u23 are the standard Poissons ratios; u14, u24..., u56 are the shear and coupling Poissons ratios; g12, g13, and g23 are the shear moduli; a1, a2, and a3 are the coefficients of thermal expansion; and a12, a13, and a23 are the coefficients of thermal shear. Note that the elastic moduli and the shear moduli must be positive. The Poissons ratios must be chosen so that the 6x6 stress-strain matrix is positive definite. This means that the determinant of the matrix must be positive. These material properties can be evaluated directly from laboratory experiments. Each column of the elasticity matrix represents the six measured strains due to the application of the appropriate unit stress. The six thermal coefficients are the measured strains due to a unit temperature change.

Temperature-Dependent Properties
All of the mechanical and thermal properties given in Equations 1 to 3 may depend upon temperature. These properties are given at a series of specified material temperatures t. Properties at other temperatures are obtained by linear interpolation between the two nearest specified temperatures. Properties at temperatures outside the specified range use the properties at the nearest specified temperature. See Figure 33 (page 142) for examples. If the Material properties are independent of temperature, you need only specify them at a single, arbitrary temperature.

Temperature-Dependent Properties
159

141

SAP2000 Analysis Reference

Figure 33 Determination of Property Ematt at Temperature Tmatt from Function E(T)

Element Material Temperature


You can assign each element an element material temperature. This is the temperature at which temperature-dependent material properties are evaluated for the element. The properties at this fixed temperature are used for all analyses regardless of any temperature changes experienced by the element during loading. Thus the material properties are independent of the reference temperature and the load temperatures. The element material temperature may be uniform over an element or interpolated from values given at the joints. In the latter case, a uniform material temperature is used that is the average of the joint values. The default material temperature for any element is zero. The properties for a temperature-independent material are constant regardless of the element material temperatures specified.

Mass Density
For each Material you may specify a mass density, m, that is used for calculating the mass of the element. The total mass of the element is the product of the mass density (mass per unit volume) and the volume of the element. This mass is apportioned to each joint of the element. The same mass is applied along of the three

142

Element Material Temperature


160

Chapter VIII Material Properties translational degrees of freedom. No rotational mass moments of inertia are computed. Consistent mass units must be used. Typically the mass density is the same as the weight density divided by the acceleration due to gravity, but this is not required. The mass density property is independent of temperature. For more information: See Topic Mass (page 167) in Chapter The Frame Element. See Topic Mass (page 194) in Chapter The Shell Element. See Topic Mass (page 213) in Chapter The Plane Element. See Topic Mass (page 227) in Chapter The Asolid Element. See Topic Mass (page 238) in Chapter The Solid Element.

Weight Density
For each Material you may specify a weight density, w, that is used for calculating the self-weight of the element. The total weight of the element is the product of the weight density (weight per unit volume) and the volume of the element. This weight is apportioned to each joint of the element. Self-weight is activated using Self-weight Load and Gravity Load. The weight density property is independent of temperature. For more information: See Topic Self-Weight Load (page 283) in Chapter Load Cases. See Topic Gravity Load (page 284) in Chapter Load Cases.

Design-Type Indicator
You may specify a design-type indicator for each Material that indicates how it is to be treated for design by the SAP2000 graphical interface. The three options for this indicator, ides, are: Steel: Frame elements made of this material will be designed according to steel design codes

Weight Density
161

143

SAP2000 Analysis Reference Concrete: Frame elements made of this material will be designed according to concrete design codes Neither: Frame elements made of this material will not be designed

144

Design-Type Indicator
162

C h a p t e r IX

The Frame Element


The Frame element is used to model beam-column and truss behavior in planar and three-dimensional structures. Basic Topics for All Users Overview Joint Connectivity Degrees of Freedom Local Coordinate System Section Properties End Offsets End Releases Mass Self-Weight Load Concentrated Span Load Distributed Span Load Internal Force Output

145
163

SAP2000 Analysis Reference Advanced Topics Advanced Local Coordinate System Non-prismatic Elements Gravity Load Temperature Load Prestress Load Joint Force Output

Overview
The Frame element uses a general, three-dimensional, beam-column formulation which includes the effects of biaxial bending, torsion, axial deformation, and biaxial shear deformations. See Bathe and Wilson (1976). Structures that can be modeled with this element include: Three-dimensional frames Three-dimensional trusses Planar frames Planar grillages Planar trusses A Frame element is modeled as a straight line connecting two joints. Each element has its own local coordinate system for defining section properties and loads, and for interpreting output. The element may be prismatic or non-prismatic. The non-prismatic formulation allows the element length to be divided into any number of segments over which properties may vary. The variation of the bending stiffness may be linear, parabolic, or cubic over each segment of length. The axial, shear, torsional, mass, and weight properties all vary linearly over each segment. End offsets are available to account for the finite size of beam and column intersections. The end offsets may be made partially or fully rigid to model the stiffening effect that can occur when the ends of an element are embedded in beam and column intersections. End releases are also available to model different fixity conditions at the ends of the element.

146

Overview
164

Chapter IX The Frame Element Each Frame element may be loaded by gravity (in any direction), multiple concentrated loads, multiple distributed loads, loads due to prestressing cables, and loads due to temperature change. Element internal forces are produced at the ends of each element and at a userspecified number of equally-spaced output stations along the length of the element.

Joint Connectivity
A Frame element is represented by a straight line connecting two joints, i and j. The two joints must not share the same location in space. The two ends of the element are denoted end I and end J, respectively.

Degrees of Freedom
The Frame element normally activates all six degrees of freedom at both of its connected joints. However, the three rotational degrees of freedom will not be activated at a joint if the element does not provide any moment stiffness or moment load to that joint. This can occur under either of the following conditions: The end offset at that end is zero, and the geometric Section properties j, i33, and i22 are all zero (a is non-zero; as2 and as3 are arbitrary), or The end offset at that end is zero; both bending rotations, R2 and R3, are released at that end; and the torsional rotation, R1, is released at either end When these conditions apply to both ends, the element behaves as a truss member. For more information: See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom. See Topic Section Properties (page 153) in this chapter. See Topic End Offsets (page 162) in this chapter. See Topic End Releases (page 165) in this chapter.

Joint Connectivity
165

147

SAP2000 Analysis Reference

Local Coordinate System


Each Frame element has its own element local coordinate system used to define section properties, loads and output. The axes of this local system are denoted 1, 2 and 3. The first axis is directed along the length of the element; the remaining two axes lie in the plane perpendicular to the element with an orientation that you specify. It is important that you clearly understand the definition of the element local 1-2-3 coordinate system and its relationship to the global X-Y-Z coordinate system. Both systems are right-handed coordinate systems. It is up to you to define local systems which simplify data input and interpretation of results. In most structures the definition of the element local coordinate system is extremely simple. The methods provided, however, provide sufficient power and flexibility to describe the orientation of Frame elements in the most complicated situations. The simplest method, using the default orientation and the Frame element coordinate angle, is described in this topic. Additional methods for defining the Frame element local coordinate system are described in the next topic. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Advanced Local Coordinate System (page 149) in this chapter.

Longitudinal Axis 1
Local axis 1 is always the longitudinal axis of the element, the positive direction being directed from end I to end J.

Default Orientation
The default orientation of the local 2 and 3 axes is determined by the relationship between the local 1 axis and the global Z axis: The local 1-2 plane is taken to be vertical, i.e., parallel to the Z axis The local 2 axis is taken to have an upward (+Z) sense unless the element is vertical, in which case the local 2 axis is taken to be horizontal along the global +X direction

148

Local Coordinate System


166

Chapter IX The Frame Element The local 3 axis is horizontal, i.e., it lies in the X-Y plane An element is considered to be vertical if the sine of the angle between the local 1 axis and the Z axis is less than 10-3. The local 2 axis makes the same angle with the vertical axis as the local 1 axis makes with the horizontal plane. This means that the local 2 axis points vertically upward for horizontal elements.

Coordinate Angle
The Frame element coordinate angle, ang, is used to define element orientations that are different from the default orientation. It is the angle through which the local 2 and 3 axes are rotated about the positive local 1 axis from the default orientation. The rotation for a positive value of ang appears counterclockwise when the local +1 axis is pointing toward you. For vertical elements, ang is the angle between the local 2 axis and the horizontal +X axis. Otherwise, ang is the angle between the local 2 axis and the vertical plane containing the local 1 axis. See Figure 34 (page 150) for examples.

Advanced Local Coordinate System


By default, the element local coordinate system is defined using the element coordinate angle measured with respect to the global +Z and +X directions, as described in the previous topic. In certain modeling situations it may be useful to have more control over the specification of the local coordinate system. This topic describes how to define the orientation of the transverse local 2 and 3 axes with respect to an arbitrary reference vector when the element coordinate angle, ang, is zero. If ang is different from zero, it is the angle through which the local 2 and 3 axes are rotated about the positive local 1 axis from the orientation determined by the reference vector. The local 1 axis is always directed from end I to end J of the element. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Local Coordinate System (page 148) in this chapter.

Advanced Local Coordinate System


167

149

SAP2000 Analysis Reference

Figure 34 The Frame Element Coordinate Angle with Respect to the Default Orientation

Reference Vector
To define the transverse local axes 2 and 3, you specify a reference vector that is parallel to the desired 1-2 or 1-3 plane. The reference vector must have a positive projection upon the corresponding transverse local axis (2 or 3, respectively). This

150

Advanced Local Coordinate System


168

Chapter IX The Frame Element means that the positive direction of the reference vector must make an angle of less than 90 with the positive direction of the desired transverse axis. To define the reference vector, you must first specify or use the default values for: A primary coordinate direction pldirp (the default is +Z) A secondary coordinate direction pldirs (the default is +X). Directions pldirs and pldirp should not be parallel to each other unless you are sure that they are not parallel to local axis 1 A fixed coordinate system csys (the default is zero, indicating the global coordinate system) The local plane, local, to be determined by the reference vector (the default is 12, indicating plane 1-2) You may optionally specify: A pair of joints, plveca and plvecb (the default for each is zero, indicating the center of the element). If both are zero, this option is not used For each element, the reference vector is determined as follows: 1. A vector is found from joint plveca to joint plvecb. If this vector is of finite length and is not parallel to local axis 1, it is used as the reference vector V p 2. Otherwise, the primary coordinate direction pldirp is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 1, it is used as the reference vector V p 3. Otherwise, the secondary coordinate direction pldirs is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 1, it is used as the reference vector V p 4. Otherwise, the method fails and the analysis terminates. This will never happen if pldirp is not parallel to pldirs A vector is considered to be parallel to local axis 1 if the sine of the angle between -3 them is less than 10 . The use of the Frame element coordinate angle in conjunction with coordinate directions that define the reference vector is illustrated in Figure 35 (page 152). The use of joints to define the reference vector is shown in Figure 36 (page 152).

Advanced Local Coordinate System


169

151

SAP2000 Analysis Reference

Figure 35 The Frame Element Coordinate Angle with Respect to Coordinate Directions

Figure 36 Using Joints to Define the Frame Element Local Coordinate System

152

Advanced Local Coordinate System


170

Chapter IX The Frame Element

Determining Transverse Axes 2 and 3


The program uses vector cross products to determine the transverse axes 2 and 3 once the reference vector has been specified. The three axes are represented by the three unit vectors V1 , V2 and V3 , respectively. The vectors satisfy the cross-product relationship: V1 = V2 V3 The transverse axes 2 and 3 are defined as follows: If the reference vector is parallel to the 1-2 plane, then: V3 = V1 V p and V2 = V3 V1 If the reference vector is parallel to the 1-3 plane, then: V2 = V p V1 and V3 = V1 V2 In the common case where the reference vector is perpendicular to axis V1 , the transverse axis in the selected plane will be equal to V p .

Section Properties
A Frame Section is a set of material and geometric properties that describe the cross-section of one or more Frame elements. Sections are defined independently of the Frame elements, and are assigned to the elements. Section properties are of two basic types: Prismatic all properties are constant along the full element length Non-prismatic the properties may vary along the element length Non-prismatic Sections are defined by referring to two or more previously defined prismatic Sections. All of the following subtopics, except the last, describe the definition of prismatic Sections. The last subtopic, Non-prismatic Sections, describes how prismatic Sections are used to define non-prismatic Sections.

Section Properties
171

153

SAP2000 Analysis Reference

Local Coordinate System


Section properties are defined with respect to the local coordinate system of a Frame element as follows: The 1 direction is along the axis of the element. It is normal to the Section and goes through the intersection of the two neutral axes of the Section. The 2 and 3 directions are parallel to the neutral axes of the Section. Usually the 2 direction is taken along the major dimension (depth) of the Section, and the 3 direction along its minor dimension (width), but this is not required. See Topic Local Coordinate System (page 148) in this chapter for more information.

Material Properties
The material properties for the Section are specified by reference to a previouslydefined Material. Isotropic material properties are used, even if the Material selected was defined as orthotropic or anisotropic. The material properties used by the Section are: The modulus of elasticity, e1, for axial stiffness and bending stiffness The shear modulus, g12, for torsional stiffness and transverse shear stiffness The coefficient of thermal expansion, a1, for axial expansion and thermal bending strain The mass density, m, for computing element mass The weight density, w, for computing Self-Weight and Gravity Loads The material properties e1, g12, and a1 are all obtained at the material temperature of each individual Frame element, and hence may not be unique for a given Section. See Chapter Material Properties (page 135) for more information.

Geometric Properties and Section Stiffnesses


Six basic geometric properties are used, together with the material properties, to generate the stiffnesses of the Section. These are: The cross-sectional area, a. The axial stiffness of the Section is given by a e1;

154

Section Properties
172

Chapter IX The Frame Element The moment of inertia, i33, about the 3 axis for bending in the 1-2 plane, and the moment of inertia, i22, about the 2 axis for bending in the 1-3 plane. The corresponding bending stiffnesses of the Section are given by i33 e1 and i22 e1; The torsional constant, j. The torsional stiffness of the Section is given by j g12. Note that the torsional constant is not the same as the polar moment of inertia, except for circular shapes. See Roark and Young (1975) or Cook and Young (1985) for more information. The shear areas, as2 and as3, for transverse shear in the 1-2 and 1-3 planes, respectively. The corresponding transverse shear stiffnesses of the Section are given by as2 g12 and as3 g12. Formulae for calculating the shear areas of typical sections are given in Figure 37 (page 156). Setting a, j, i33, or i22 to zero causes the corresponding section stiffness to be zero. For example, a truss member can be modeled by setting j = i33 = i22 = 0, and a planar frame member in the 1-2 plane can be modeled by setting j = i22 = 0. Setting as2 or as3 to zero causes the corresponding transverse shear deformation to be zero. In effect, a zero shear area is interpreted as being infinite. The transverse shear stiffness is ignored if the corresponding bending stiffness is zero.

Shape Type
For each Section, the six geometric properties (a, j, i33, i22, as2 and as3) may be specified directly, computed from specified Section dimensions, or read from a specified property database file. This is determined by the shape type, sh, specified by the user: If sh=G (general section), the six geometric properties must be explicitly specified If sh=R, P, B, I, C, T, L or 2L, the six geometric properties are automatically calculated from specified Section dimensions as described in Automatic Section Property Calculation below. If sh is any other value (e.g., W27X94 or 2L4X3X1/4), the six geometric properties are obtained from a specified property database file. See Section Property Database Files below.

Section Properties
173

155

SAP2000 Analysis Reference

Figure 37 Shear Area Formulae

156

Section Properties
174

Chapter IX The Frame Element

Automatic Section Property Calculation


The six geometric Section properties can be automatically calculated from specified dimensions for the simple shapes shown in Figure 38 (page 158). The required dimensions for each shape are shown in the figure. Note that the dimension t3 is the depth of the Section in the 2 direction and contributes primarily to i33. Automatic Section property calculation is available for the following shape types: Sh=R: Rectangular Section Sh=P: Pipe Section, or Solid Circular Section if tw=0 (or not specified) Sh=B: Box Section Sh=I: I Section Sh=C: Channel Section Sh=T: T Section Sh=L: Angle Section Sh=2L: Double-angle Section

Section Property Database Files


Geometric Section properties may be obtained from one or more Section property database files. Three database files are supplied with SAP2000: AISC.PRO: American Institute of Steel Construction shapes CISC.PRO: Canadian Institute of Steel Construction shapes SECTIONS.PRO: This is just a copy of AISC.PRO. Additional property database files may be created using the program PROPER, which is available upon request from Computers and Structures, Inc. The geometric properties are stored in the length units specified when the database file was created. These are automatically converted by SAP2000 to the units used in the input data file. Each shape type stored in a database file may be referenced by one or two different labels. For example, the W 36x300 shape type in file AISC.PRO may be referenced either by label W36X300 or by label W920X446. Shape types stored in CISC.PRO may only be referenced by a single label. Section Properties
175

157

SAP2000 Analysis Reference

Figure 38 Automatic Section Property Calculation

158

Section Properties
176

Chapter IX The Frame Element The shape type labels available for a given database file are stored in an associated label file with extension .LBL. For examples, the labels for database file AISC.PRO are stored in file AISC.LBL. The label file is a text file that may be printed or viewed with a text editor. Each line in the label file shows the one or two labels corresponding to a single shape type stored in the database file. You may select one database file to be used when defining a given Frame Section. If shape type sh cannot be found in the database file, an error results. The database file in use can be changed at any time when defining Sections. If no database filename is specified, the default file SECTIONS.PRO is used. You may copy any property database file to SECTIONS.PRO. All Section property database files, including file SECTIONS.PRO, must be located either in the directory that contains the input data file, or in the directory that contains the SAP2000 executable files. If a specified database file is present in both directories, the program will use the file in the input-data-file directory.

Additional Mass and Weight


You may specify mass and/or weight for a Section that acts in addition to the mass and weight of the material. The additional mass and weight are specified per unit of length using the parameters mpl and wpl, respectively. They could be used, for example, to represent the effects of nonstructural material that is attached to a Frame element. The additional mass and weight act regardless of the cross-sectional area of the Section. The default values for mpl and wpl are zero for all shape types.

Non-prismatic Sections
Non-prismatic Sections may be defined for which the properties vary along the element length. You may specify that the element length be divided into any number of segments; these do not need to be of equal length. Most common situations can be modeled using from one to five segments. The variation of the bending stiffnesses may be linear, parabolic, or cubic over each segment of length. The axial, shear, torsional, mass, and weight properties all vary linearly over each segment. Section properties may change discontinuously from one segment to the next. See Figure 39 (page 160) for examples of non-prismatic Sections.

Section Properties
177

159

SAP2000 Analysis Reference

Figure 39 Examples of Non-prismatic Sections

160

Section Properties
178

Chapter IX The Frame Element Segment Lengths The length of a non-prismatic segment may be specified as either a variable length, vl, or an absolute length, l. The default is vl = 1. When a non-prismatic Section is assigned to an element, the actual lengths of each segment for that element are determined as follows: The clear length of the element, Lc , is first calculated as the total length minus the end offsets:
Lc

= L ( ioff + joff

See Topic End Offsets (page 162) in this chapter for more information. If the sum of the absolute lengths of the segments exceeds the clear length, they are scaled down proportionately so that the sum equals the clear length. Otherwise the absolute lengths are used as specified. The remaining length (the clear length minus the sum of the absolute lengths) is divided among the segments having variable lengths in the same proportion as the specified lengths. For example, for two segments with vl = 1 and vl = 2, one third of the remaining length would go to the first segment, and two thirds to the second segment. Starting and Ending Sections The properties for a segment are defined by specifying: The label, seci, of a previously defined prismatic Section that defines the properties at the start of the segment, i.e., at the end closest to joint i. The label, secj, of a previously defined prismatic Section that defines the properties at the end of the segment, i.e., at the end closest to joint j. The starting and ending Sections may be the same if the properties are constant over the length of the segment. The Material would normally be the same for both the starting and ending Sections and only the geometric properties would differ, but this is not required. Variation of Properties Non-prismatic Section properties are interpolated along the length of each segment from the values at the two ends. The variation of the bending stiffnesses, i33e1 and i22e1, are defined by specifying the parameters eivar33 and eivar22, respectively. Assign values of 1, 2, or 3 to Section Properties
179

161

SAP2000 Analysis Reference these parameters to indicate variation along the length that is linear, parabolic, or cubic, respectively. Specifically, the eivar33-th root of the bending stiffness in the 1-2 plane:
eivar33

i33 e1

varies linearly along the length. This usually corresponds to a linear variation in one of the Section dimensions. For example, referring to Figure 38 (page 158): a linear variation in t2 for the rectangular shape would require eivar33=1, a linear variation in t3 for the rectangular shape would require eivar33=3, and a linear variation in t3 for the I-shape would require eivar33=2. The interpolation of the bending stiffness in the 1-2 plane, i22 e1, is defined in the same manner by the parameter eivar22. The remaining properties are assumed to vary linearly between the ends of each segment: Stiffnesses: a e1, j g12, as2 g12, and as3 g12 Mass: am + mpl Weight: aw + wpl If a shear area is zero at either end, it is taken to be zero along the full segment, thus eliminating all shear deformation in the corresponding bending plane for that segment. Effect upon End Offsets Properties vary only along the clear length of the element. Section properties within end offset ioff are constant using the starting Section of the first segment. Section properties within end offset joff are constant using the ending Section of the last segment. See Topic End Offsets (page 162) in this chapter for more information.

End Offsets
Frame elements are modeled as line elements connected at points (joints). However, actual structural members have finite cross-sectional dimensions. When two elements, such as a beam and column, are connected at a joint there is some overlap of the cross sections. In many structures the dimensions of the members are large

162

End Offsets
180

Chapter IX The Frame Element

Figure 40 Frame Element End Offsets

and the length of the overlap can be a significant fraction of the total length of a connecting element. You may specify two end offsets for each element using parameters ioff and joff corresponding to ends I and J, respectively. End offset ioff is the length of overlap for a given element with other connecting elements at joint i. It is the distance from the joint to the face of the connection for the given element. A similar definition applies to end offset joff at joint j. See Figure 40 (page 163). End offsets are automatically calculated by the SAP2000 graphical interface for each element based on the maximum Section dimensions of all other elements that connect to that element at a common joint.

Clear Length
The clear length, denoted Lc , is defined to be the length between the end offsets (support faces) as:
Lc

= L ( ioff + joff

End Offsets
181

163

SAP2000 Analysis Reference where L is the total element length. See Figure 40 (page 163). If end offsets are specified such that the clear length is less than 1% of the total element length, the program will issue a warning and reduce the end offsets proportionately so that the clear length is equal to 1% of the total length. Normally the end offsets should be a much smaller proportion of the total length.

Rigid-end Factor
An analysis based upon the centerline-to-centerline (joint-to-joint) geometry of Frame elements may overestimate deflections in some structures. This is due to the stiffening effect caused by overlapping cross sections at a connection. It is more likely to be significant in concrete than in steel structures. You may specify a rigid-end factor for each element using parameter rigid, which gives the fraction of each end offset that is assumed to be rigid for bending and shear deformation. The length rigidioff, starting from joint i, is assumed to be rigid. Similarly, the length rigidjoff is rigid at joint j. The flexible length L f of the element is given by:
Lf

= L rigid ( ioff + joff

The rigid-zone offsets never affect axial and torsional deformation. The full element length is assumed to be flexible for these deformations. The default value for rigid is zero. The maximum value of unity would indicate that the end offsets are fully rigid. You must use engineering judgment to select the appropriate value for this parameter. It will depend upon the geometry of the connection, and may be different for the different elements that frame into the connection. Typically the value for rigid would not exceed about 0.5.

Effect upon Non-prismatic Elements


At each end of a non-prismatic element, the Section properties are assumed to be constant within the length of the end offset. Section properties vary only along the clear length of the element between support faces. This is not affected by the value of the rigid-end factor, rigid. See Subtopic Non-prismatic Sections (page 159) in this chapter for more information.

164

End Offsets
182

Chapter IX The Frame Element

Effect upon Internal Force Output


All internal forces and moments are output at the faces of the supports and at other equally-spaced points within the clear length of the element. No output is produced within the end offset, which includes the joint. This is not affected by the value of the rigid-end factor, rigid. See Topic Internal Force Output (page 176) in this chapter for more information.

Effect upon End Releases


End releases are always assumed to be at the support faces, i.e., at the ends of the clear length of the element. If a moment or shear release is specified in either bending plane at either end of the element, the end offset is assumed to be rigid for bending and shear in that plane at that end (i.e., it acts as if rigid = 1). This does not affect the values of the rigid-end factor at the other end or in the other bending plane. See Topic End Releases (page 165) in this chapter for more information.

End Releases
Normally, the three translational and three rotational degrees of freedom at each end of the Frame element are continuous with those of the joint, and hence with those of all other elements connected to that joint. However, it is possible to release (disconnect) one or more of the element degrees of freedom from the joint when it is known that the corresponding element force or moment is zero. The releases are always specified in the element local coordinate system, and do not affect any other element connected to the joint. In the example shown in Figure 41 (page 166), the diagonal element has a moment connection at End I and a pin connection at End J. The other two elements connecting to the joint at End J are continuous. Therefore, in order to model the pin condition the rotation R3 at End J of the diagonal element should be released. This assures that the moment is zero at the pin in the diagonal element.

Unstable End Releases


Any combination of end releases may be specified for a Frame element provided that the element remains stable; this assures that all load applied to the element is transferred to the rest of the structure. The following sets of releases are unstable, either alone or in combination, and are not permitted. End Releases
183

165

SAP2000 Analysis Reference

Figure 41 Frame Element End Releases

Releasing U1 at both ends; Releasing U2 at both ends; Releasing U3 at both ends; Releasing R1 at both ends; Releasing R2 at both ends and U3 at either end; Releasing R3 at both ends and U2 at either end.

Effect of End Offsets


End releases are always applied at the support faces, i.e., at the ends of the element clear length. The presence of a moment or shear release will cause the end offset to be rigid in the corresponding bending plane at the corresponding end of the element. See Topic End Offsets (page 162) in this chapter for more information.

166

End Releases
184

Chapter IX The Frame Element

Effect upon Prestress Load


Certain end releases (e.g., axial) are not compatible with the presence of prestressing cables, even though the program will permit both to be present. End releases affect only the portion of the prestress load that is uniformly distributed over the element, not the loads that are applied to the joints. See Topic Prestress Load (page 174) in this chapter for more information.

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Frame element is lumped at the joints i and j. No inertial effects are considered within the element itself. The total mass of the element is equal to the integral along the length of the mass density, m, multiplied by the cross-sectional area, a, plus the additional mass per unit length, mpl. For non-prismatic elements, the mass varies linearly over each non-prismatic segment of the element, and is constant within the end offsets. The total mass is apportioned to the two joints in the same way a similarlydistributed transverse load would cause reactions at the ends of a simply-supported beam. The effects of end releases are ignored when apportioning mass. The total mass is applied to each of the three translational degrees of freedom: UX, UY, and UZ. No mass moments of inertia are computed for the rotational degrees of freedom. For more information: See Topic Mass Density (page 142) in Chapter Material Properties. See Topic Section Properties (page 153) in this chapter for the definition of a and mpl. See Subtopic Non-prismatic Sections (page 159) in this chapter. See Topic End Offsets (page 162) in this chapter. See Chapter Static and Dynamic Analysis (page 291).

Mass
185

167

SAP2000 Analysis Reference

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For a Frame element, the self-weight is a force that is distributed along the length of the element. The magnitude of the self-weight is equal to the weight density, w, multiplied by the cross-sectional area, a, plus the additional weight per unit length, wpl. For non-prismatic elements, the self-weight varies linearly over each non-prismatic segment of the element, and is constant within the end offsets. Self-Weight Load always acts downward, in the global Z direction. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. For more information: See Topic Weight Density (page 143) in Chapter Material Properties for the definition of w. See Topic Section Properties (page 153) in this chapter for the definition of a and wpl.. See Subtopic Non-prismatic Sections (page 159) in this chapter. See Topic End Offsets (page 162) in this chapter. See Topic Self-Weight Load (page 283) in Chapter Load Cases.

Gravity Load
Gravity Load can be applied to each Frame element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information: See Topic Self-Weight Load (page 168) in this chapter for the definition of self-weight for the Frame element. See Topic Gravity Load (page 284) in Chapter Load Cases.

168

Self-Weight Load
186

Chapter IX The Frame Element

Concentrated Span Load


The Concentrated Span Load is used to apply concentrated forces and moments at arbitrary locations on Frame elements. The direction of loading may be specified in a fixed coordinate system (global or alternate coordinates) or in the element local coordinate system. The location of the load may be specified in one of the following ways: Specifying a relative distance, rd, measured from joint i. This must satisfy . 0 rd 1 The relative distance is the fraction of element length; Specifying an absolute distance, d, measured from joint i. This must satisfy 0 d L, where L is the element length. Any number of concentrated loads may be applied to each element. Loads given in fixed coordinates are transformed to the element local coordinate system. See Figure 42 (page 170). Multiple loads that are applied at the same location are added together. See Chapter Load Cases (page 279) for more information.

Distributed Span Load


The Distributed Span Load is used to apply distributed forces and moments on Frame elements. The load intensity may be uniform or trapezoidal. The direction of loading may be specified in a fixed coordinate system (global or alternate coordinates) or in the element local coordinate system. See Chapter Load Cases (page 279) for more information.

Loaded Length
Loads may apply to full or partial element lengths. Multiple loads may be applied to a single element. The loaded lengths may overlap, in which case the applied loads are additive. A loaded length may be specified in one of the following ways: Specifying two relative distances, rda and rdb, measured from joint i. They must satisfy 0 rda < rdb 1. The relative distance is the fraction of element length;

Concentrated Span Load


187

169

SAP2000 Analysis Reference

Figure 42 Examples of the Definition of Concentrated Span Loads

Specifying two absolute distances, da and db, measured from joint i. They must satisfy 0 da < db L, where L is the element length; Specifying no distances, which indicates the full length of the element.

Load Intensity
The load intensity is a force or moment per unit of length. Except for the case of projected loads described below, the intensity is measured per unit of element length. For each force or moment component to be applied, a single load value may be given if the load is uniformly distributed. Two load values are needed if the load intensity varies linearly over its range of application (a trapezoidal load). See Figure 43 (page 171) and Figure 44 (page 172).

170

Distributed Span Load


188

Chapter IX The Frame Element

Figure 43 Examples of the Definition of Distributed Span Loads

Distributed Span Load


189

171

SAP2000 Analysis Reference

Figure 44 Examples of Distributed Span Loads

172

Distributed Span Load


190

Chapter IX The Frame Element

Projected Loads
A distributed snow or wind load produces a load intensity (force per unit of element length) that is proportional to the sine of the angle between the element and the direction of loading. This is equivalent to using a fixed load intensity that is measured per unit of projected element length. The fixed intensity would be based upon the depth of snow or the wind speed; the projected element length is measured in a plane perpendicular to the direction of loading. Distributed Span Loads may be specified as acting upon the projected length. The program handles this by reducing the load intensity according to the angle, , between the element local 1 axis and the direction of loading. Projected force loads are scaled by sin , and projected moment loads are scaled by cos . The reduced load intensities are then applied per unit of element length. The scaling of the moment loads is based upon the assumption that the moment is caused by a force acting upon the projected element length. The resulting moment is always perpendicular to the force, thus accounting for the use of the cosine instead of the sine of the angle. The specified intensity of the moment should be computed as the product of the force intensity and the perpendicular distance from the element to the force. The appropriate sign of the moment must be given.

Temperature Load
The Temperature Load creates thermal strain in the Frame element. This strain is given by the product of the Material coefficient of thermal expansion and the temperature change of the element. The temperature change is measured from the element Reference Temperature to the element Load Temperature. Three independent Load Temperature fields may be specified: Temperature, t, which is constant over the cross section and produces axial strains Temperature gradient, t2, which is linear in the local 2 direction and produces bending strains in the 1-2 plane Temperature gradient, t3, which is linear in the local 3 direction and produces bending strains in the 1-3 plane Temperature gradients are specified as the change in temperature per unit length. The temperature gradients are positive if the temperature increases (linearly) in the

Temperature Load
191

173

SAP2000 Analysis Reference positive direction of the element local axis. The gradient temperatures are zero at the neutral axes, hence no axial strain is induced. Each of the three Load Temperature fields may be constant along the element length or interpolated from values given at the joints. The Reference Temperature gradients are always taken to be zero, hence the temperature changes that produce the bending strain are equal to the Load Temperature gradients. See Chapter Load Cases (page 279) for more information.

Prestress Load
Any of the Frame elements in the model can be subjected to loading produced by one or more Prestressing cables. This load always acts in the local 1-2 plane of the element. See Chapter Load Cases (page 279) for more information.

Prestressing Cables
Any number of Prestressing cables may act on a single element. Each cable is subject to the following assumptions and specifications: The tension, t, is assumed to be constant along the length and does not change with element deformation. The drape configuration is assumed to be parabolic. It is specified by giving three dimensions measured from the local 1 axis , as shown in Figure 45 (page 175): The drapes at the two ends, di and dj, measured in the positive 2 direction The drape at the center, dc, measured in the negative 2 direction The cable is assumed to act in a narrow duct within the element, so that the transverse deflection of the Frame element and the Prestressing cable are the same. This affects the P-delta force in the element.

Prestress Load
Each Prestressing cable produces a set of self-equilibrating forces and moments that are proportional to the cable tension, t:

174

Prestress Load
192

Chapter IX The Frame Element

Figure 45 Prestressing Cable Profile

Tensile forces acting on joints i and j; Moments acting on joints i and j that are proportional to the drapes di and dj, respectively; Shear forces acting on joints i and j that are proportional to the slopes of the cable at ends i and j, respectively; A uniform, distributed load acting on the element that is proportional to the curvature of the cable. The sum of these forces and moments for all Prestressing cables acting on a Frame element forms the unscaled prestress load for that element. This load has no effect upon the structure until scaled and applied using the Prestress Load specification. All cables acting on a given element are applied simultaneously and scaled by the same factor. Only the uniform load acting on the element is affected by end releases and other properties of the element. The forces and moments acting on joints i and j are independent of all element properties. See Topic End Releases (page 165) in this chapter for more information.

Effect upon P-Delta Analysis


It is assumed that the transverse deflections of a Frame element and its Prestressing cables are the same. This means that the net P-Delta axial force acting on a Frame

Prestress Load
193

175

SAP2000 Analysis Reference element is the sum of the axial forces in the element and all of its cables, taken in the P-Delta load combination. The tension in a Prestressing cable tends to stiffen a Frame element against transverse deflection when a P-Delta analysis is performed. If the element is allowed to deform axially the cable will also produce compression in the element, reducing the net axial force in the element. Thus the P-Delta axial force due to a Prestressing cable alone will vary from zero to the scaled tension force in the cable, depending upon the degree of axial restraint. The net effect will always be some degree of stiffening (zero or more); it will never cause buckling. If the actual Prestressing cables are not constrained to deflect transversely with the Frame element, special modeling procedures are needed when P-Delta effects are important. Each such Frame element is replaced by two elements connected to the same joints: The first element represents the beam in the absence of prestressing. It is given the real Material and Section properties and is assigned all Loads except the Prestress Load. The second element represents the Prestressing cables. It is given arbitrary Material properties and all geometric Section properties are set to zero. The Prestressing cables and the Prestress Load are assigned to this element. This model can create a negative P-Delta axial force in the first element due to prestressing alone, and may cause buckling of that element. See Chapter P-Delta Analysis (page 337) for the definition of P-Delta axial force and the P-delta load combination for more information.

Internal Force Output


The Frame element internal forces are the forces and moments that result from integrating the stresses over an element cross section. These internal forces are: P, the axial force V2, the shear force in the 1-2 plane V3, the shear force in the 1-3 plane T, the axial torque M2, the bending moment in the 1-3 plane (about the 2 axis) M3, the bending moment in the 1-2 plane (about the 3 axis)

176

Internal Force Output


194

Chapter IX The Frame Element These internal forces and moments are present at every cross section along the length of the element. The sign convention is illustrated in Figure 46 (page 178). Positive internal forces and axial torque acting on a positive 1 face are oriented in the positive direction of the element local coordinate axes. Positive internal forces and axial torque acting on a negative face are oriented in the negative direction of the element local coordinate axes. A positive 1 face is one whose outward normal (pointing away from element) is in the positive local 1 direction. Positive bending moments cause compression at the positive 2 and 3 faces and tension at the negative 2 and 3 faces. The positive 2 and 3 faces are those faces in the positive local 2 and 3 directions, respectively, from the neutral axis. The internal forces and moments are computed at equally-spaced output points along the length of the element. The nseg parameter specifies the number of equal segments (or spaces) along the length of the element between the output points. For the default value of 2, output is produced at the two ends and at the midpoint of the element. See Effect of End Offsets below. The Frame element internal forces are computed for all Analysis Cases: Loads, Modes, Specs, Histories, Moving Loads, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. These can be printed in the output file and displayed using the SAP2000 graphical interface. See Subtopic Internal Forces in the Output File below. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. For more information: See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter Bridge Analysis (page 357). See Chapter The Output Files (page 395).

Effect of End Offsets


When end offsets are present, internal forces and moments are output at the faces of the supports and at nseg 1 equally-spaced points within the clear length of the element. No output is produced within the length of the end offset, which includes the

Internal Force Output


195

177

SAP2000 Analysis Reference

Figure 46 Frame Element Internal Forces and Moments

178

Internal Force Output


196

Chapter IX The Frame Element joint. Output will only be produced at joints i or j when the corresponding end offset is zero. See Topic End Offsets (page 162) in this chapter for more information.

Internal Forces in the Output File


Frame element internal forces and moments are printed in the output file under the heading:
FRAME ELEMENT INTERNAL FORCES

All internal forces for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case internal forces are printed at all output points. The locations of the output points are identified by the relative distance of the points from joint i. The relative distance is a fraction of the element length (between zero and one.) Typical Frame element output is shown in Figure 47 (page 180). By default, no internal forces are printed. You may specify which Analysis Cases are to be printed. The internal forces can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. See Chapter The Output Files (page 395) for more information.

Joint Force Output


The Frame element joint forces are concentrated forces and moments acting at the joints of the element that represent the effect of the rest of the structure upon the element. Element joint forces must not be confused with internal forces and moments which, like stresses, act throughout the element, not just at the joints. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

Joint Force Output


197

179

SAP2000 Analysis Reference


F R A M E ELEM LOAD REL DIST 0.00000 0.50000 1.00000 SPEC E L E M E N T I N T E R N A L F O R C E S 12.000000

1 ================== LENGTH = 1 -----------------P V2 -9.73E-11 -1.00000E+3 -9.73E-11 -1.00000E+3 -9.73E-11 -1.00000E+3 SPEC -----------------V2 68.323489 68.323489 68.323489 V3 0.972230 0.972230 0.972230 MAX V3 0.972230 0.972230 0.972230 MIN V3 -0.972230 -0.972230 -0.972230 V3 7.35E-13 7.35E-13 7.35E-13

T 2.02612E+3 2.02612E+3 2.02612E+3

M2 -1.35E-12 3.06E-12 7.47E-12

M3 7.20000E+4 7.80000E+4 8.40000E+4

REL DIST P 0.00000 126.377896 0.50000 126.377896 1.00000 126.377896 COMB

T 89.135262 89.135262 89.135262

M2 1.878252 3.955129 9.788509

M3 3.19034E+3 3.58275E+3 3.97870E+3

BOTH ------------------

REL DIST P V2 0.00000 126.377896 -931.676511 0.50000 126.377896 -931.676511 1.00000 126.377896 -931.676511 COMB BOTH ------------------

T 2.11526E+3 2.11526E+3 2.11526E+3

M2 1.878252 3.955129 9.788509

M3 7.51903E+4 8.15828E+4 8.79787E+4

REL DIST P V2 0.00000 -126.377896 -1.06832E+3 0.50000 -126.377896 -1.06832E+3 1.00000 -126.377896 -1.06832E+3 ELEM LOAD REL DIST 0.00000 0.50000 1.00000 SPEC

T 1.93699E+3 1.93699E+3 1.93699E+3 12.000000

M2 -1.878252 -3.955129 -9.788509

M3 6.88097E+4 7.44172E+4 8.00213E+4

2 ================== LENGTH = 1 -----------------P V2 -9.73E-11 -1.00000E+3 -9.73E-11 -1.00000E+3 -9.73E-11 -1.00000E+3 SPEC -----------------V2 68.323489 68.323489 68.323489 V3 0.972230 0.972230 0.972230

V3 T -7.35E-13 -2.02612E+3 -7.35E-13 -2.02612E+3 -7.35E-13 -2.02612E+3

M2 1.35E-12 -3.06E-12 -7.47E-12

M3 7.20000E+4 7.80000E+4 8.40000E+4

REL DIST P 0.00000 126.377896 0.50000 126.377896 1.00000 126.377896 ...Output Edited for Brevity...

T 89.135262 89.135262 89.135262

M2 1.878252 3.955129 9.788509

M3 3.19034E+3 3.58275E+3 3.97870E+3

Figure 47 Frame Element Internal Forces in the Output File

180

Joint Force Output


198

Chapter X

The Shell Element


The Shell element is used to model shell, membrane, and plate behavior in planar and three-dimensional structures. Basic Topics for All Users Overview Joint Connectivity Degrees of Freedom Local Coordinate System Section Properties Mass Self-Weight Load Uniform Load Internal Force and Stress Output Advanced Topics Advanced Local Coordinate System Gravity Load

181
199

SAP2000 Analysis Reference Surface Pressure Load Temperature Load Joint Force Output

Overview
The Shell element is a three- or four-node formulation that combines separate membrane and plate-bending behavior. The four-joint element does not have to be planar. The membrane behavior uses an isoparametric formulation that includes translational in-plane stiffness components and a rotational stiffness component in the direction normal to the plane of the element. See Taylor and Simo (1985) and Ibrahimbegovic and Wilson (1991). The plate bending behavior includes two-way, out-of-plane, plate rotational stiffness components and a translational stiffness component in the direction normal to the plane of the element. By default, a thick-plate (Mindlin/Reissner) formulation is used which includes the effects of transverse shearing deformation. Optionally, you may choose a thin-plate (Kirchhoff) formulation that neglects transverse shearing deformation. Structures that can be modeled with this element include: Three-dimensional shells, such as tanks and domes Plate structures, such as floor slabs Membrane structures, such as shear walls For each Shell element in the structure, you can choose to model pure membrane, pure plate, or full shell behavior. It is generally recommended that you use the full shell behavior unless the entire structure is planar and is adequately restrained. Each Shell element has its own local coordinate system for defining Material properties and loads, and for interpreting output. Temperature-dependent, orthotropic material properties are allowed. Each element may be loaded by gravity and uniform loads in any direction; surface pressure on the top, bottom, and side faces; and loads due to temperature change. An variable, four-to-eight-point numerical integration formulation is used for the Shell stiffness. Stresses and internal forces and moments, in the element local coordinate system, are evaluated at the 2-by-2 Gauss integration points and extrapolated

182

Overview
200

Chapter X The Shell Element to the joints of the element. An approximate error in the element stresses or internal forces can be estimated from the difference in values calculated from different elements attached to a common joint. This will give an indication of the accuracy of a given finite-element approximation and can then be used as the basis for the selection of a new and more accurate finite element mesh.

Joint Connectivity
Each Shell element may have either of the following shapes, as shown in Figure 48 (page 184): Quadrilateral, defined by the four joints j1, j2, j3, and j4. Triangular, defined by the three joints j1, j2, and j3. The quadrilateral formulation is the more accurate of the two. The triangular element is recommended for transitions only. The stiffness formulation of the threenode element is reasonable; however, its stress recovery is poor. The use of the quadrilateral element for meshing various geometries and transitions is illustrated in Figure 49 (page 185). The locations of the joints should be chosen to meet the following geometric conditions: The inside angle at each corner must be less than 180. Best results for the quadrilateral will be obtained when these angles are near 90, or at least in the range of 45 to 135. The aspect ratio of an element should not be too large. For the triangle, this is the ratio of the longest side to the shortest side. For the quadrilateral, this is the ratio of the longer distance between the midpoints of opposite sides to the shorter such distance. Best results are obtained for aspect ratios near unity, or at least less than four. The aspect ratio should not exceed ten. For the quadrilateral, the four joints need not be coplanar. A small amount of twist in the element is accounted for by the program. The angle between the normals at the corners gives a measure of the degree of twist. The normal at a corner is perpendicular to the two sides that meet at the corner. Best results are obtained if the largest angle between any pair of corners is less than 30. This angle should not exceed 45. These conditions can usually be met with adequate mesh refinement. The accuracy of the thick-plate formulation is more sensitive to large aspect ratios and mesh distortion than is the thin-plate formulation. Joint Connectivity
201

183

SAP2000 Analysis Reference

Figure 48 Shell Element Joint Connectivity and Face Definitions

184

Joint Connectivity
202

Chapter X The Shell Element

Figure 49 Mesh Examples Using the Quadrilateral Shell Element

Degrees of Freedom
The Shell element always activates all six degrees of freedom at each of its connected joints. When the element is used as a pure membrane, you must ensure that restraints or other supports are provided to the degrees of freedom for normal translation and bending rotations. When the element is used as a pure plate, you must ensure that restraints or other supports are provided to the degrees of freedom for inplane translations and the rotation about the normal.

Degrees of Freedom
203

185

SAP2000 Analysis Reference The use of the full shell behavior (membrane plus plate) is recommended for all three-dimensional structures. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom for more information.

Local Coordinate System


Each Shell element has its own element local coordinate system used to define Material properties, loads and output. The axes of this local system are denoted 1, 2 and 3. The first two axes lie in the plane of the element with an orientation that you specify; the third axis is normal. It is important that you clearly understand the definition of the element local 1-2-3 coordinate system and its relationship to the global X-Y-Z coordinate system. Both systems are right-handed coordinate systems. It is up to you to define local systems which simplify data input and interpretation of results. In most structures the definition of the element local coordinate system is extremely simple. The methods provided, however, provide sufficient power and flexibility to describe the orientation of Shell elements in the most complicated situations. The simplest method, using the default orientation and the Shell element coordinate angle, is described in this topic. Additional methods for defining the Shell element local coordinate system are described in the next topic. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Advanced Local Coordinate System (page 187) in this chapter.

Normal Axis 3
Local axis 3 is always normal to the plane of the Shell element. This axis is directed toward you when the path j1-j2-j3 appears counterclockwise. For quadrilateral elements, the element plane is defined by the vectors that connect the midpoints of the two pairs of opposite sides.

186

Local Coordinate System


204

Chapter X The Shell Element

Default Orientation
The default orientation of the local 1 and 2 axes is determined by the relationship between the local 3 axis and the global Z axis: The local 3-2 plane is taken to be vertical, i.e., parallel to the Z axis The local 2 axis is taken to have an upward (+Z) sense unless the element is horizontal, in which case the local 2 axis is taken along the global +Y direction The local 1 axis is horizontal, i.e., it lies in the X-Y plane The element is considered to be horizontal if the sine of the angle between the local -3 3 axis and the Z axis is less than 10 . The local 2 axis makes the same angle with the vertical axis as the local 3 axis makes with the horizontal plane. This means that the local 2 axis points vertically upward for vertical elements.

Element Coordinate Angle


The Shell element coordinate angle, ang, is used to define element orientations that are different from the default orientation. It is the angle through which the local 1 and 2 axes are rotated about the positive local 3 axis from the default orientation. The rotation for a positive value of ang appears counterclockwise when the local +3 axis is pointing toward you. For horizontal elements, ang is the angle between the local 2 axis and the horizontal +Y axis. Otherwise, ang is the angle between the local 2 axis and the vertical plane containing the local 3 axis. See Figure 50 (page 188) for examples.

Advanced Local Coordinate System


By default, the element local coordinate system is defined using the element coordinate angle measured with respect to the global +Z and +Y directions, as described in the previous topic. In certain modeling situations it may be useful to have more control over the specification of the local coordinate system. This topic describes how to define the orientation of the tangential local 1 and 2 axes, with respect to an arbitrary reference vector when the element coordinate angle, ang, is zero. If ang is different from zero, it is the angle through which the local 1 and 2 axes are rotated about the positive local 3 axis from the orientation deter-

Advanced Local Coordinate System


205

187

SAP2000 Analysis Reference

Figure 50 The Shell Element Coordinate Angle with Respect to the Default Orientation

mined by the reference vector. The local 3 axis is always normal to the plane of the element. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Local Coordinate System (page 186) in this chapter.

188

Advanced Local Coordinate System


206

Chapter X The Shell Element

Reference Vector
To define the tangential local axes, you specify a reference vector that is parallel to the desired 3-1 or 3-2 plane. The reference vector must have a positive projection upon the corresponding tangential local axis (1 or 2, respectively). This means that the positive direction of the reference vector must make an angle of less than 90 with the positive direction of the desired tangential axis. To define the reference vector, you must first specify or use the default values for: A primary coordinate direction pldirp (the default is +Z) A secondary coordinate direction pldirs (the default is +Y). Directions pldirs and pldirp should not be parallel to each other unless you are sure that they are not parallel to local axis 3 A fixed coordinate system csys (the default is zero, indicating the global coordinate system) The local plane, local, to be determined by the reference vector (the default is 32, indicating plane 3-2) You may optionally specify: A pair of joints, plveca and plvecb (the default for each is zero, indicating the center of the element). If both are zero, this option is not used For each element, the reference vector is determined as follows: 1. A vector is found from joint plveca to joint plvecb. If this vector is of finite length and is not parallel to local axis 3, it is used as the reference vector V p 2. Otherwise, the primary coordinate direction pldirp is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 3, it is used as the reference vector V p 3. Otherwise, the secondary coordinate direction pldirs is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 3, it is used as the reference vector V p 4. Otherwise, the method fails and the analysis terminates. This will never happen if pldirp is not parallel to pldirs A vector is considered to be parallel to local axis 3 if the sine of the angle between -3 them is less than 10 .

Advanced Local Coordinate System


207

189

SAP2000 Analysis Reference

Figure 51 Shell Element Local Coordinate System Using Coordinate Directions

The use of the coordinate direction method is illustrated in Figure 51 (page 190) for the case where local = 32. A special option is available for backward compatibility with previous versions of the program. If pldirp is set to zero, the reference vector V p is directed from the midpoint of side j1-j3 to the midpoint of side j2-j4 (or side j2-j3 for the triangle). This is illustrated in Figure 48 (page 184), where the reference vector would be identical to local axis 1. With this option, the orientation of the tangential local axes is very dependent upon the mesh used.

Determining Tangential Axes 1 and 2


The program uses vector cross products to determine the tangential axes 1 and 2 once the reference vector has been specified. The three axes are represented by the three unit vectors V1 , V2 and V3 , respectively. The vectors satisfy the cross-product relationship: V1 = V2 V3

190

Advanced Local Coordinate System


208

Chapter X The Shell Element The tangential axes 1 and 2 are defined as follows: If the reference vector is parallel to the 3-1 plane, then: V2 = V3 V p and V1 = V2 V3 If the reference vector is parallel to the 3-2 plane, then: V1 = V p V3 and V2 = V3 V1 In the common case where the reference vector is parallel to the plane of the element, the tangential axis in the selected local plane will be equal to V p .

Section Properties
A Shell Section is a set of material and geometric properties that describe the cross-section of one or more Shell elements. Sections are defined independently of the Shell elements, and are referenced during the definition of the elements.

Section Type
The Section type, specified by the parameter type, determines the type of behavior modeled by the corresponding Shell elements: type = MEMBR: Pure membrane behavior; only the in-plane forces and the normal (drilling) moment can be supported type = PLATE: Pure plate behavior; only the bending moments and the transverse force can be supported type = SHELL: Full shell behavior, a combination of membrane and plate behavior; all forces and moments can be supported It is generally recommended that you use the full shell behavior unless the entire structure is planar and is adequately restrained.

Thickness Formulation
The thickness formulation, specified by the parameter thicktype, determines whether or not transverse shearing deformations are included in the plate-bending behavior of a plate or shell element:

Section Properties
209

191

SAP2000 Analysis Reference thicktype = THICK: A thick-plate (Mindlin/Reissner) formulation is used which includes the effects of transverse shear deformation thicktype = THIN: A thin-plate (Kirchhoff) formulation is used that neglects transverse shearing deformation Shearing deformations tend to be important when the thickness is greater than about one-tenth to one-fifth of the span. They can also be quite significant in the vicinity of bending-stress concentrations, such as near sudden changes in thickness or support conditions, and near holes or re-entrant corners. Even for thin-plate bending problems where shearing deformations are truly neglibible, the thick-plate formulation tends to be more accurate, although somewhat stiffer, than the thin-plate formulation. However, the accuracy of the thick-plate formulation is more sensitive to large aspect ratios and mesh distortion than is the thin-plate formulation. It is generally recommended that you use the default thick-plate formulation unless you are using a distorted mesh and you know that shearing deformations will be small, or unless you are trying to match a theoretical thin-plate solution. The thickness formulation has no effect upon membrane behavior, only upon plate-bending behavior.

Material Properties
The material properties for each Section are specified by reference to a previouslydefined Material. Orthotropic properties are used, even if the Material selected was defined as anisotropic. The material properties used by the Shell Section are: The moduli of elasticity, e1, e2, and e3 The shear modulus, g12, g13, and g23 The Poissons ratios, u12, u13, and u23 The coefficients of thermal expansion, a1 and a2 The mass density, m, for computing element mass The weight density, w, for computing Self-Weight and Gravity Loads The properties e3, u13, and u23 are condensed out of the matrial matrix by assuming a state of plane stress in the element. The resulting, modified values of e1, e2, g12, and u12 are used to compute the membrane and plate-bending stiffnesses.

192

Section Properties
210

Chapter X The Shell Element

Figure 52 Shell Element Material Angle

The shear moduli, g13 and g23, are used to compute the transverse shearing stiffness if the thick-plate formulation is used. The coefficients of thermal expansion, a1 and a2, are used for membrane expansion and thermal bending strain. All material properties (except the densities) are obtained at the material temperature of each individual element. See Chapter Material Properties (page 135) for more information.

Material Angle
The material local coordinate system and the element (Shell Section) local coordinate system need not be the same. The local 3 directions always coincide for the two systems, but the material 1 axis and the element 1 axis may differ by the angle a as shown in Figure 52 (page 193). This angle has no effect for isotropic material properties since they are independent of orientation. See Topic Local Coordinate System (page 136) in Chapter Material Properties for more information.

Section Properties
211

193

SAP2000 Analysis Reference

Thickness
Each Section has a constant membrane thickness and a constant bending thickness. The membrane thickness, th, is used for calculating: The membrane stiffness for full-shell and pure-membrane Sections The element volume for the element self-weight and mass calculations The bending thickness, thb, is use for calculating: The plate-bending and transverse-shearing stiffnesses for full-shell and pureplate Sections Normally these two thicknesses are the same and you only need to specify th. However, for some applications, such as modeling corrugated surfaces, the membrane and plate-bending behavior cannot be adequately represented by a homogeneous material of a single thickness. For this purpose, you may specify a value of thb that is different from th.

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Shell element is lumped at the element joints. No inertial effects are considered within the element itself. The total mass of the element is equal to the integral over the plane of the element of the mass density, m, multiplied by the thickness, th. The total mass is apportioned to the joints in a manner that is proportional to the diagonal terms of the consistent mass matrix. See Cook, Malkus, and Plesha (1989) for more information. The total mass is applied to each of the three translational degrees of freedom: UX, UY, and UZ. No mass moments of inertia are computed for the rotational degrees of freedom. For more information: See Topic Mass Density (page 142) in Chapter Material Properties. See Subtopic Thickness (page 194) in this chapter for the definition of th. See Chapter Static and Dynamic Analysis (page 291).

194

Mass
212

Chapter X The Shell Element

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For a Shell element, the self-weight is a force that is uniformly distributed over the plane of the element. The magnitude of the self-weight is equal to the weight density, w, multiplied by the thickness, th. Self-Weight Load always acts downward, in the global Z direction. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. For more information: See Topic Weight Density (page 143) in Chapter Material Properties for the definition of w. See Subtopic Thickness (page 194) in this chapter for the definition of th. See Topic Self-Weight Load (page 283) in Chapter Load Cases.

Gravity Load
Gravity Load can be applied to each Shell element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information: See Topic Self-Weight Load (page 195) in this chapter for the definition of self-weight for the Shell element. See Topic Gravity Load (page 284) in Chapter Load Cases.

Uniform Load
Uniform Load is used to apply uniformly distributed forces to the midsurfaces of the Shell elements. The direction of the loading may be specified in a fixed coordinate system (global or Alternate Coordinates) or in the element local coordinate system.

Self-Weight Load
213

195

SAP2000 Analysis Reference

Figure 53 Example of Uniform Load Acting on the Projected Area of the Midsurface

Load intensities are given as forces per unit area. Load intensities specified in different coordinate systems are converted to the element local coordinate system and added together. The total force acting on the element in each local direction is given by the total load intensity in that direction multiplied by the area of the midsurface. This force is apportioned to the joints of the element. Forces given in fixed coordinates can optionally be specified to act on the projected area of the midsurface, i.e., the area that can be seen along the direction of loading. The specified load intensity is automatically multiplied by the cosine of the angle between the direction of loading and the normal to the element (the local 3 direction). This can be used, for example, to apply distributed snow or wind loads. See Figure 53 (page 196). See Chapter Load Cases (page 279) for more information.

Surface Pressure Load


The Surface Pressure Load is used to apply external pressure loads upon any of the six faces of the Shell element. The definition of these faces is shown in Figure 48 (page 184). Surface pressure always acts normal to the face. Positive pressures are directed toward the interior of the element.

196

Surface Pressure Load


214

Chapter X The Shell Element The pressure may be constant over a face or interpolated from values given at the joints. The values given at the joints are obtained from Joint Patterns, and need not be the same for the different faces. Joint Patterns can be used to easily apply hydrostatic pressures. The bottom and top faces are denoted Faces 5 and 6, respectively. The top face is the one visible when the +3 axis is directed toward you and the path j1-j2-j3 appears counterclockwise. The pressure acting on the bottom or top face is integrated over the plane of the element and apportioned to the corner joints.. The sides of the element are denoted Faces 1 to 4 (1 to 3 for the triangle), counting counterclockwise from side j1-j2 when viewed from the top. The pressure acting on a side is multiplied by the thickness, th, integrated along the length of the side, and apportioned to the two joints on that side. For more information: See Topic Thickness (page 194) in this chapter for the definition of th. See Chapter Load Cases (page 279).

Temperature Load
The Temperature Load creates thermal strain in the Shell element. This strain is given by the product of the Material coefficient of thermal expansion and the temperature change of the element. The temperature change is measured from the element Reference Temperature to the element Load Temperature. Two independent Load Temperature fields may be specified: Temperature, t, which is constant through the thickness and produces membrane strains Temperature gradient, t3, which is linear in the thickness direction and produces bending strains The temperature gradient is specified as the change in temperature per unit length. The temperature gradient is positive if the temperature increases (linearly) in the positive direction of the element local 3 axis. The gradient temperature is zero at the midsurface, hence no membrane strain is induced. Each of the two Load Temperature fields may be constant over the plane of the element or interpolated from values given at the joints.

Temperature Load
215

197

SAP2000 Analysis Reference The Reference Temperature gradient is always taken to be zero, hence the temperature change that produces the bending strain is equal to the Load Temperature gradient. See Chapter Load Cases (page 279) for more information.

Internal Force and Stress Output


The Shell element internal forces (also called stress resultants) are the forces and moments that result from integrating the stresses over the element thickness. These internal forces are: Membrane direct forces:
F11

+ th / 2 + th / 2

th / 2 th / 2 + th / 2

11 22

dx 3

(Eqns. 1)

F22

= =

dx 3

Membrane shear force:


F12

th / 2

12

dx 3

Plate bending moments:


M 11

+ thb/ 2 + thb/ 2

thb/ 2 thb/ 2 + thb/ 2

11 22

dx 3

M 22

= =

dx 3

Plate twisting moment:


M 12

thb/ 2

12

dx 3

Plate transverse shear forces:


V13

+ thb/ 2 + thb/ 2

thb/ 2 thb/ 2

13 23

dx 3

V 23

dx 3

where x3 represents the thickness coordinate measured from the midsurface of the element.

198

Internal Force and Stress Output


216

Chapter X The Shell Element It is very important to note that these stress resultants are forces and moments per unit of in-plane length. They are present at every point on the midsurface of the element. The transverse shear forces are computed from the moments using the equilibrium equations:
V13

= =

dM 11 dx 1 dM 12 dx 1

dM 12 dx 2 dM 22 dx 2

V 23

where x1 and x2 are in-plane coordinates parallel to the local 1 and 2 axes. The sign conventions for the stresses and internal forces are illustrated in Figure 54 (page 200). Stresses acting on a positive face are oriented in the positive direction of the element local coordinate axes. Stresses acting on a negative face are oriented in the negative direction of the element local coordinate axes. A positive face is one whose outward normal (pointing away from element) is in the positive local 1 or 2 direction. Positive internal forces correspond to a state of positive stress that is constant through the thickness. Positive internal moments correspond to a state of stress that varies linearly through the thickness and is positive at the bottom. Thus: 11 = 22 = 12 = 13 = 23 =
F11

th
F22

12

M 11
3

thb
12

x3

(Eqns. 2)

M 22
3

th
F12

thb
12

x3

M 12
3

th
V13

thb

x3

thb
V 23

thb

33 = 0 The transverse shear stresses given here are average values. The actual shear stress distribution is parabolic, being zero at the top and bottom surfaces and taking a maximum or minimum value at the midsurface of the element. Internal Force and Stress Output
217

199

SAP2000 Analysis Reference

Figure 54 Shell Element Stresses and Internal Forces

200

Internal Force and Stress Output


218

Chapter X The Shell Element The stresses and internal forces are evaluated at the standard 2-by-2 Gauss integration points of the element and extrapolated to the joints. Although they are reported at the joints, the stresses and internal forces exist throughout the element. See Cook, Malkus, and Plesha (1989) for more information. The Shell element stresses and internal forces are computed for the following types of Analysis Cases: Loads, Modes, Specs, Histories, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. No results are available for Moving Loads or Combos that contain Moving Loads. Principal values and the associated principal directions are also computed for the Loads and Modes. The angle given is measured counterclockwise (when viewed from the top) from the local 1 axis to the direction of the maximum principal value. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. Shell element stresses and internal forces are printed in the output file under the heading:
SHELL ELEMENT INTERNAL FORCES & STRESSES

All results for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case the internal forces are printed at the element joints, followed by the stresses at the joints. The internal forces and stresses can be interpolated over the whole element from the values printed at the joints. The internal forces are labeled F11, F22, F12, M11, M22, M12, V13, and V23, corresponding to the terms defined in Equations 1 above. The stresses are labeled S11, S22, S12, S13, and S23, corresponding to the terms defined in Equations 2 above. Stress values at the top and bottom surface are indicated. Typical Shell element output is shown in Figure 55 (page 202). By default, no internal forces or stresses are printed. You may specify which Analysis Cases are to be printed. The internal forces and stresses can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. For more information: See Topic Stresses and Strains (page 137) in Chapter Material Properties.

Internal Force and Stress Output


219

201

SAP2000 Analysis Reference


S H E L L ELEM LOAD JOINT 10 31 25 32 JOINT 10 31 25 32 JOINT 10 31 25 32 JOINT 10 31 25 32 JOINT 10 31 25 32 JOINT 10 31 25 32 ELEM LOAD JOINT 4 5 9 10 JOINT 4 5 9 10 E L E M E N T I N T E R N A L F O R C E S & S T R E S S E S

1 ================== TYPE = SHELL 1 -----------------F11 -1.63E-11 -1.62E-11 -1.63E-11 -1.62E-11 M11 -8000.000 -10000.000 -8000.010 -9999.990 V13 -147.129704 -147.129704 -147.128035 -147.128035 S11-TOP 333.333333 416.666667 333.333751 416.666249 S11-BOT -333.333333 -416.666667 -333.333751 -416.666249 S13-AVG -12.260809 -12.260809 -12.260670 -12.260670 F22 -1.22E-13 1.53E-13 -1.22E-13 1.53E-13 F12 1.38E-13 1.38E-13 -1.38E-13 -1.38E-13 F-MAX -1.20E-13 1.54E-13 -1.21E-13 1.54E-13 M-MAX 12.085230 97.842783 12.085625 97.842326 V-MAX 147.129704 147.129704 147.128035 147.128035 S12-TOP 4.884265 4.884265 -4.884216 -4.884216 S12-BOT -4.884265 -4.884265 4.884216 4.884216 S-TOP-MAX 333.404794 416.723366 333.405209 416.722948 S-BOT-MAX 0.503551 4.076783 0.503568 4.076764 S-AVG-MAX 12.260809 12.260809 12.260670 12.260670 S-TOP-MIN -0.503551 -4.076783 -0.503568 -4.076764 S-BOT-MIN -333.404794 -416.723366 -333.405209 -416.722948 F-MIN -1.63E-11 -1.62E-11 -1.63E-11 -1.62E-11 M-MIN -8001.715 -10001.361 -8001.725 -10001.351 ANGLE 89.511243 89.517398 -89.511244 -89.517399 ANGLE -89.161783 -89.334903 89.161792 89.334909 ANGLE 179.999986 -179.999986 179.999986 -179.999986 ANGLE 0.838217 0.665097 -0.838208 -0.665091 ANGLE -89.161783 -89.334903 89.161792 89.334909 ANGLE 179.999986 -179.999986 179.999986 -179.999986

M22 M12 10.370186 -117.222361 96.481989 -117.222361 10.370617 117.221187 96.481558 117.221187 V23 3.59E-05 -3.59E-05 3.59E-05 -3.59E-05 S22-TOP -0.432091 -4.020083 -0.432109 -4.020065 S22-BOT 0.432091 4.020083 0.432109 4.020065 S23-AVG 2.99E-06 -2.99E-06 2.99E-06 -2.99E-06

3 ================== TYPE = MEMBRANE 1 -----------------F11 F22 F12 1824.191 279.849372 -79.513613 1790.814 146.341527 -43.335783 -46.975988 -187.937212 -104.469157 -80.351616 -321.443470 -68.289377 S11-TOP 304.031827 298.469000 -7.829331 -13.391936 S22-TOP 46.641562 24.390255 -31.322869 -53.573912 S12-TOP -13.252269 -7.222631 -17.411526 -11.381563 F-MAX F-MIN 1828.274 275.766246 1791.955 145.200317 8.564515 -243.477715 -62.352389 -339.442697 S-TOP-MAX 304.712348 298.659202 1.427419 -10.392065 S-TOP-MIN 45.961041 24.200053 -40.579619 -56.573783 ANGLE -2.939630 -1.508485 -27.997121 -14.765796 ANGLE -2.939630 -1.508485 -27.997121 -14.765796

...Output Edited for Brevity...

Figure 55 Shell Element Internal Forces and Stresses in the Output File

202

Internal Force and Stress Output


220

Chapter X The Shell Element See Subtopic Thickness (page 194) in this chapter for the definition of th and thb. See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Joint Force Output


The Shell element joint forces are concentrated forces and moments acting at the joints of the element that represent the effect of the rest of the structure upon the element. Element joint forces must not be confused with internal forces and moments which, like stresses, act throughout the element, not just at the joints. Furthermore, the internal forces are defined on a per-unit-of-length basis. Element joints forces are concentrated values of force and moment. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

Joint Force Output


221

203

SAP2000 Analysis Reference

204

Joint Force Output


222

C h a p t e r XI

The Plane Element


The Plane element is used to model plane-stress and plane-strain behavior in twodimensional solids. Advanced Topics Overview Joint Connectivity Degrees of Freedom Local Coordinate System Stresses and Strains Material Properties Material Angle Thickness Mass Self-Weight Load Gravity Load Surface Pressure Load Pore Pressure Load

205
223

SAP2000 Analysis Reference Temperature Load Stress Output Joint Force Output

Overview
The Plane element is a three- to nine-node element for modeling two-dimensional solids of uniform thickness. It is based upon an isoparametric formulation (Hollings and Wilson, 1977). The element must be planar and must always exist parallel to one of the global principal planes (i.e., parallel to the X-Y, Y-Z, or Z-X planes). Structures that can be modeled with this element include: Thin, planar structures in a state of plane stress Long, prismatic structures in a state of plane strain The stresses and strains are assumed not to vary in the thickness direction. Each Plane element has its own local coordinate system that is aligned with the global system. The local coordinate system is used for defining Material properties and loads, and for interpreting output. Temperature-dependent, orthotropic material properties are allowed. Each element may be loaded by gravity (in any direction); surface pressure on the side faces; pore pressure within the element; and loads due to temperature change. An eight-point numerical integration scheme is used for the Plane. Stresses in the element local coordinate system are evaluated at the integration points and extrapolated to the joints of the element. An approximate error in the stresses can be estimated from the difference in values calculated from different elements attached to a common joint. This will give an indication of the accuracy of the finite element approximation and can then be used as the basis for the selection of a new and more accurate finite element mesh. The nine-node option is the recommended formulation.

Joint Connectivity
Each Plane element may have either of the following shapes, as shown in Figure 56 (page 207):

206

Overview
224

Chapter XI The Plane Element

Figure 56 Plane Element Joint Connectivity and Face Definitions

Quadrilateral, defined by the four corner joints j1, j3, j7, and j9, plus the optional midside joints j2, j4, j6, and j8, and the optional center joint j5. Triangular, defined by the three joints j1, j3, and j7 only.

Joint Connectivity
225

207

SAP2000 Analysis Reference All joints for a given element must lie in a single plane that is parallel to one of the principal global planes: X-Y, Y-Z, or Z-X. The path j1-j3-j7 must appear counterclockwise when viewed from the positive global coordinate direction that is perpendicular to the plane of the element, i.e., when viewed from +Z, +X, or +Y, respectively. For the best accuracy, the use of the nine-node quadrilateral is recommended. However, the variable node option of the element is very useful for generating transition interfaces between coarse and fine finite element meshes. See Figure 57 (page 209) for examples. The locations of the joints should be chosen to meet the following geometric conditions: The inside angle at each corner must be less than 180. Best results for the quadrilateral will be obtained when these angles are near 90, or at least in the range of 45 to 135. The aspect ratio of an element should not be too large. For the triangle, this is the ratio of the longest side to the shortest side. For the quadrilateral, this is the ratio of the longer distance between the midpoints of opposite sides to the shorter such distance. Best results are obtained for aspect ratios near unity, or at least less than four. The aspect ratio should not exceed ten. If a midside joint is present, it should be near the center of the side, or at least between the quarter points. Best results are obtained if the side is straight. If the side is curved, the distance from the midside joint to the straight line between the corner joints should be no larger than half the distance between the corner joints. The position of the center joint with respect to each pair of opposite midside joints is subject to the same rules as is the position of a midside joint with respect to the corner joints.

Degrees of Freedom
The Plane element activates the three translational degrees of freedom at each of its connected joints. Rotational degrees of freedom are not activated. This element contributes stiffness only to the degrees of freedom in the plane of the element. It is necessary to provide restraints or other supports for the translational degrees of freedom that are normal to this plane; otherwise, the structure will be unstable.

208

Degrees of Freedom
226

Chapter XI The Plane Element

Figure 57 Examples of Plane Element Mesh Transitions

See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom for more information.

Degrees of Freedom
227

209

SAP2000 Analysis Reference

Figure 58 Plane Element Local Coordinate System and Thickness Definition

Local Coordinate System


Each Plane element has its own element local coordinate system used to define Material properties, loads and output. The axes of this local system are denoted 1, 2 and 3. These axes are always parallel to the axes of the global coordinate system, with the 3 direction being perpendicular to the plane of the element. The local axes all have the same positive sense as do the corresponding global axes. Thus: For an element parallel to the global X-Y plane, the local 1-2-3 system is the same as the global X-Y-Z system For an element parallel to the global Y-Z plane, the local 1-2-3 system is the same as the global Y-Z-X system For an element parallel to the global Z-X plane, the local 1-2-3 system is the same as the global Z-X-Y system The local coordinate system for elements parallel to the Y-Z plane is shown in Figure 58 (page 210).

210

Local Coordinate System


228

Chapter XI The Plane Element

Stresses and Strains


The Plane element models the mid-plane of a structure having uniform thickness, and whose stresses and strains do not vary in the thickness direction. Transverse shear stresses ( 12 , 13 ) and shear strains ( 12 , 13 ) are assumed to be zero. Displacements in the thickness (local 3) direction have no effect on the element. Plane-stress is appropriate for structures that are thin compared to their planar dimensions. The thickness normal stress ( 33 ) is assumed to be zero. The thickness normal strain ( 33 ) may not be zero due to Poisson effects. Plane-strain is appropriate for structures that are thick compared to their planar dimensions. The thickness normal strain ( 33 ) is assumed to be zero. The thickness normal stress ( 33 ) may not be zero due to Poisson effects. See Topic Stresses and Strains (page 137) in Chapter Material Properties for more information.

Material Properties
The material properties for each Plane element are specified by reference to a previously-defined Material. Orthotropic properties are used, even if the Material selected was defined as anisotropic. The material properties used by the Plane element are: The moduli of elasticity, e1, e2, and e3 The shear modulus, g12 The Poissons ratios, u12, u13 and u23 The coefficients of thermal expansion, a1, a2, and a3 The mass density, m, for computing element mass The weight density, w, for computing Self-Weight and Gravity Loads The properties e3, u13, u23, and a3 are not used for plane stress. They are used to compute the thickness-normal stress ( 33 ) in plane strain. All material properties (except the densities) are obtained at the material temperature of each individual element. See Chapter Material Properties (page 135) for more information. Stresses and Strains
229

211

SAP2000 Analysis Reference

Figure 59 Plane Element Material Angle

Material Angle
The material local coordinate system and the element local coordinate system need not be the same. The local 3 directions always coincide for the two systems, but the material 1 axis and the element 1 axis may differ by the angle a as shown in Figure 59 (page 212). This angle has no effect for isotropic material properties since they are independent of orientation. See Topic Local Coordinate System (page 136) in Chapter Material Properties for more information.

Thickness
Each element has a uniform thickness, th. This may be the actual thickness, particularly for plane-stress elements; or it may be a representative portion, such as a unit thickness of an infinitely-thick plane-strain element. See Figure 58 (page 210). The element thickness is used for calculating the element stiffness, mass, and loads. Hence, joint forces computed from the element are proportional to this thickness.

212

Material Angle
230

Chapter XI The Plane Element

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Plane element is lumped at the element joints. No inertial effects are considered within the element itself. The total mass of the element is equal to the integral over the plane of the element of the mass density, m, multiplied by the thickness, th. The total mass is apportioned to the joints in a manner that is proportional to the diagonal terms of the consistent mass matrix. See Cook, Malkus, and Plesha (1989) for more information. The total mass is applied to each of the three translational degrees of freedom (UX, UY, and UZ) even though the element contributes stiffness to only two of these degrees of freedom. For more information: See Topic Mass Density (page 142) in Chapter Material Properties. See Chapter Static and Dynamic Analysis (page 291).

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For a Plane element, the self-weight is a force that is uniformly distributed over the plane of the element. The magnitude of the self-weight is equal to the weight density, w, multiplied by the thickness, th. Self-Weight Load always acts downward, in the global Z direction. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. For more information: See Topic Weight Density (page 143) in Chapter Material Properties for the definition of w. See Topic Thickness (page 212) in this chapter for the definition of th. See Topic Self-Weight Load (page 283) in Chapter Load Cases.

Mass
231

213

SAP2000 Analysis Reference

Gravity Load
Gravity Load can be applied to each Plane element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information: See Topic Self-Weight Load (page 213) in this chapter for the definition of self-weight for the Plane element. See Topic Gravity Load (page 284) in Chapter Load Cases.

Surface Pressure Load


The Surface Pressure Load is used to apply external pressure loads upon any of the three or four side faces of the Plane element. The definition of these faces is shown in Figure 56 (page 207). Surface pressure always acts normal to the face. Positive pressures are directed toward the interior of the element. The pressure may be constant over a face or interpolated from values given at the joints. The values given at the joints are obtained from Joint Patterns, and need not be the same for the different faces. Joint Patterns can be used to easily apply hydrostatic pressures. The pressure acting on a side is multiplied by the thickness, th, integrated along the length of the side, and apportioned to the two or three joints on that side. See Chapter Load Cases (page 279) for more information.

Pore Pressure Load


The Pore Pressure Load is used to model the drag and buoyancy effects of a fluid within a solid medium, such as the effect of water upon the solid skeleton of a soil. Scalar fluid-pressure values are given at the element joints by Joint Patterns, and interpolated over the element. The total force acting on the element is the integral of the gradient of this pressure field over the plane of the element, multiplied by the thickness, th. This force is apportioned to each of the joints of the element. The

214

Gravity Load
232

Chapter XI The Plane Element forces are typically directed from regions of high pressure toward regions of low pressure. See Chapter Load Cases (page 279) for more information.

Temperature Load
The Temperature Load creates thermal strain in the Plane element. This strain is given by the product of the Material coefficient of thermal expansion and the temperature change of the element. The temperature change is measured from the element Reference Temperature to the element Load Temperature. Temperature changes are assumed to be constant through the element thickness. See Chapter Load Cases (page 279) for more information.

Stress Output
The Plane element stresses are evaluated at the standard 2-by-2 Gauss integration points of the element and extrapolated to the joints. See Cook, Malkus, and Plesha (1989) for more information. The stresses are computed for the following types of Analysis Cases: Loads, Modes, Specs, Histories, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. No results are available for Moving Loads or Combos that contain Moving Loads. Principal values and their associated principal directions in the element local 1-2 plane are also computed for the Loads and Modes. The angle given is measured counterclockwise (when viewed from the +3 direction) from the local 1 axis to the direction of the maximum principal value. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. Plane element stresses are printed in the output file under the heading:
PLANE ELEMENT STRESSES

All results for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case the stresses are printed at all element joints. Temperature Load
233

215

SAP2000 Analysis Reference The stresses are labeled S11, S22, S33, and S12 corresponding to 11 , 22 , 33 , and 12 , respectively, in the element local coordinate system. Stresses 12 and 13 are always zero and are not printed. Typical Plane element output is shown in Figure 60 (page 217). By default, no stresses are printed. You may specify which Analysis Cases are to be printed. The stresses can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. For more information: See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Joint Force Output


The Plane element joint forces are concentrated forces acting at the joints of the element that represent the effect of the rest of the structure upon the element. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

216

Joint Force Output


234

Chapter XI The Plane Element


P L A N E ELEM LOAD E L E M E N T S T R E S S E S

1 ================== TYPE = PLANE STRESS 1 -----------------S22 0.465601 -0.409666 -1.284933 -4.70E-13 2.63E-12 5.72E-12 -0.465601 0.409666 1.284933 S33 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 S12 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 ANGLE -27.266570 -9.175411 -5.325693 -45.000000 -45.000000 -45.000000 -62.733430 -80.824589 -84.674307

JOINT S11 1 20.255067 2 83.333333 3 146.411600 6 4.09E-12 7 2.35E-12 8 6.06E-13 11 -20.255067 12 -83.333333 13 -146.411600 JOINT 1 2 3 6 7 8 11 12 13 ELEM LOAD

S-MAX S-MIN 27.413397 -6.692729 85.576726 -2.653058 147.706315 -2.579648 13.888889 -13.888889 13.888889 -13.888889 13.888889 -13.888889 6.692729 -27.413397 2.653058 -85.576726 2.579648 -147.706315

2 ================== TYPE = PLANE STRESS 1 -----------------S22 0.465601 -0.409666 -1.284933 -4.70E-13 2.63E-12 5.72E-12 -0.465601 0.409666 1.284933 S33 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 S12 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 ANGLE -27.266570 -9.175411 -5.325693 -45.000000 -45.000000 -45.000000 -62.733430 -80.824589 -84.674307

JOINT S11 16 20.255067 17 83.333333 18 146.411600 21 4.09E-12 22 2.35E-12 23 6.06E-13 26 -20.255067 27 -83.333333 28 -146.411600 JOINT 16 17 18 21 22 23 26 27 28

S-MAX S-MIN 27.413397 -6.692729 85.576726 -2.653058 147.706315 -2.579648 13.888889 -13.888889 13.888889 -13.888889 13.888889 -13.888889 6.692729 -27.413397 2.653058 -85.576726 2.579648 -147.706315

...Output Edited for Brevity...

Figure 60 Plane Element Stresses in the Output File

Joint Force Output


235

217

SAP2000 Analysis Reference

218

Joint Force Output


236

C h a p t e r XII

The Asolid Element


The Asolid element is used to model axisymmetric solids under axisymmetric loading. Advanced Topics Overview Joint Connectivity Degrees of Freedom Local Coordinate System Stresses and Strains Material Properties Material Angle Arc and Thickness Mass Self-Weight Load Gravity Load Surface Pressure Load Pore Pressure Load

219
237

SAP2000 Analysis Reference Temperature Load Rotate Load Stress Output Joint Force Output

Overview
The Asolid element is a three- to nine-node element for modeling axisymmetric structures under axisymmetric loading. It is based upon an isoparametric formulation (Hollings and Wilson, 1977). The element models a representative two-dimensional cross section of the threedimensional axisymmetric solid. The axis of symmetry must be one of the global axes, and the element must exist in one of the global principal planes (in the X-Y, Y-Z, or Z-X planes). The geometry, loading, displacements, stresses, and strains are assumed not to vary in the circumferential direction. Any displacements that occur in the circumferential direction do not affect the element.. Each Asolid element has its own local coordinate system that is aligned with the global system. The local coordinate system is used for defining Material properties and loads, and for interpreting output. Temperature-dependent, orthotropic material properties are allowed. Each element may be loaded by gravity (in any direction); centrifugal force; surface pressure on the side faces; pore pressure within the element; and loads due to temperature change. An eight-point numerical integration scheme is used for the Asolid. Stresses in the element local coordinate system are evaluated at the integration points and extrapolated to the joints of the element. An approximate error in the stresses can be estimated from the difference in values calculated from different elements attached to a common joint. This will give an indication of the accuracy of the finite element approximation and can then be used as the basis for the selection of a new and more accurate finite element mesh. The nine-node option is the recommended formulation.

220

Overview
238

Chapter XII The Asolid Element

Joint Connectivity
All joints for a given element must lie in one of the principal global planes. The axis of symmetry and radial direction are as follows: Element Plane X-Y Y-Z Z-X Radial Direction Local 1 X Y Z Axis of Symmetry Local 2 Y Z X Circumferential Direction Local 3 Z X Y

The radial coordinate must not be negative for any of the elements joints. Each Asolid element may have either of the following shapes, as shown in Figure 61 (page 222): Quadrilateral, defined by the four corner joints j1, j3, j7, and j9, plus the optional midside joints j2, j4, j6, and j8, and the optional center joint j5. Triangular, defined by the three joints j1, j3, and j7 only. The path j1-j3-j7 must appear counterclockwise when viewed from the positive global coordinate direction that is perpendicular to the plane of the element, i.e., when viewed from +Z, +X, or +Y, respectively, corresponding to the table above. For the best accuracy, the use of the nine-node quadrilateral is recommended. However, the variable node option of the element is very useful for generating transition interfaces between coarse and fine finite element meshes. See Figure 62 (page 223) for examples. The locations of the joints should be chosen to meet the following geometric conditions: The inside angle at each corner must be less than 180. Best results for the quadrilateral will be obtained when these angles are near 90, or at least in the range of 45 to 135. The aspect ratio of an element should not be too large. For the triangle, this is the ratio of the longest side to the shortest side. For the quadrilateral, this is the ratio of the longer distance between the midpoints of opposite sides to the

Joint Connectivity
239

221

SAP2000 Analysis Reference

Figure 61 Asolid Element Joint Connectivity and Face Definitions

shorter such distance. Best results are obtained for aspect ratios near unity, or at least less than four. The aspect ratio should not exceed ten. If a midside joint is present, it should be near the center of the side, or at least between the quarter points. Best results are obtained if the side is straight. If the side is curved, the distance from the midside joint to the straight line between

222

Joint Connectivity
240

Chapter XII The Asolid Element

Figure 62 Examples of Asolid Element Mesh Transitions

the corner joints should be no larger than half the distance between the corner joints. The position of the center joint with respect to each pair of opposite midside joints is subject to the same rules as is the position of a midside joint with respect to the corner joints. Joint Connectivity
241

223

SAP2000 Analysis Reference Asolid elements should not be connected to joints that are connected to other types of elements unless special provisions are made to enforce axisymmetric conditions upon these joints.

Degrees of Freedom
The Asolid element activates the three translational degrees of freedom at each of its connected joints. Rotational degrees of freedom are not activated. This element contributes stiffness only to the degrees of freedom in the plane of the element. It is necessary to provide restraints or other supports for the translational degrees of freedom that are normal to this plane; otherwise, the structure will be unstable. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom for more information.

Local Coordinate System


Each Asolid element has its own element local coordinate system used to define Material properties, loads and output. The axes of this local system are denoted 1, 2 and 3. These axes are always parallel to the axes of the global coordinate system with the same positive sense. Local axis 1 is parallel to the radial direction, axis 2 is parallel to the axis of symmetry, and axis 3 is tangent to the circumferential direction of the axisymmetric solid. See the table in Topic Joint Connectivity (page 221). The local coordinate system for elements parallel to the Y-Z plane is shown in Figure 63 (page 225).

Stresses and Strains


The Asolid element models the mid-plane of a representative sector of an axisymmetric structure whose stresses and strains do not vary in the circumferential direction. Circumferential shear stresses ( 12 , 13 ) and shear strains ( 12 , 13 ) are assumed to be zero. Displacements in the circumferential (local 3) direction have no effect on the element. Displacements in the radial (local 1) direction cause circumferential normal strains:

224

Degrees of Freedom
242

Chapter XII The Asolid Element

Figure 63 Asolid Element Local Coordinate System and Arc Definition

33 =

u1 r

where u1 is the radial displacement, and r is the radius at the point in question. The circumferential normal stress ( 33 ) is computed as usual from the three normal strains. See Topic Stresses and Strains (page 137) in Chapter Material Properties for more information.

Material Properties
The material properties for each Asolid element are specified by reference to a previously-defined Material. Orthotropic properties are used, even if the Material selected was defined as anisotropic. The material properties used by the Asolid element are: The moduli of elasticity, e1, e2, and e3 The shear modulus, g12 Material Properties
243

225

SAP2000 Analysis Reference

Figure 64 Asolid Element Material Angle

The Poissons ratios, u12, u13 and u23 The coefficients of thermal expansion, a1, a2, and a3 The mass density, m, for computing element mass The weight density, w, for computing Self-Weight and Gravity Loads All material properties (except the densities) are obtained at the material temperature of each individual element. See Chapter Material Properties (page 135) for more information.

Material Angle
The material local coordinate system and the element local coordinate system need not be the same. The local 3 directions always coincide for the two systems, but the material 1 axis and the element 1 axis may differ by the angle a as shown in Figure 64 (page 226). This angle has no effect for isotropic material properties since they are independent of orientation.

226

Material Angle
244

Chapter XII The Asolid Element See Topic Local Coordinate System (page 136) in Chapter Material Properties for more information.

Arc and Thickness


The Asolid element represents a solid that is created by rotating the elements planar shape through 360 about the axis of symmetry. However, the analysis considers only a representative sector of the solid. You can specify the size of the sector, in degrees, using the parameter arc. For example, arc=360 models the full structure, and arc=90 models one quarter of it. See Figure 63 (page 225). Setting arc=0, the default, models a one-radian sector. One radian is the same as 180/, or approximately 57.3. The element thickness (circumferential extent), h, increases with the radial distance, r, from the axis of symmetry:
h

arc
180

Clearly the thickness varies over the plane of the element. The element thickness is used for calculating the element stiffness, mass, and loads. Hence, joint forces computed from the element are proportional to arc.

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Asolid element is lumped at the element joints. No inertial effects are considered within the element itself. The total mass of the element is equal to the integral over the plane of the element of the product of the mass density, m, multiplied by the thickness, h. The total mass is apportioned to the joints in a manner that is proportional to the diagonal terms of the consistent mass matrix. See Cook, Malkus, and Plesha (1989) for more information. The total mass is applied to each of the three translational degrees of freedom (UX, UY, and UZ) even though the element contributes stiffness to only two of these degrees of freedom. For more information: See Topic Mass Density (page 142) in Chapter Material Properties. Arc and Thickness
245

227

SAP2000 Analysis Reference See Chapter Static and Dynamic Analysis (page 291).

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For an Asolid element, the self-weight is a force that is distributed over the plane of the element. The magnitude of the self-weight is equal to the weight density, w, multiplied by the thickness, h. Self-Weight Load always acts downward, in the global Z direction. If the downward direction corresponds to the radial or circumferential direction of an Asolid element, the Self-Weight Load for that element will be zero, since self-weight acting in these directions is not axisymmetric. Non-zero Self-Weight Load will only exist for elements whose axial (local 2) direction is vertical. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. For more information: See Topic Weight Density (page 143) in Chapter Material Properties for the definition of w. See Topic Arc and Thickness (page 227) in this chapter for the definition of h. See Topic Self-Weight Load (page 283) in Chapter Load Cases.

Gravity Load
Gravity Load can be applied to each Asolid element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. However, only the components of Gravity load acting in the axial direction of an Asolid element will be non-zero. Components in the radial or circumferential direction will be set to zero, since gravity acting in these directions is not axisymmetric. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information:

228

Self-Weight Load
246

Chapter XII The Asolid Element See Topic Self-Weight Load (page 228) in this chapter for the definition of self-weight for the Asolid element. See Topic Gravity Load (page 284) in Chapter Load Cases.

Surface Pressure Load


The Surface Pressure Load is used to apply external pressure loads upon any of the three or four side faces of the Asolid element. The definition of these faces is shown in Figure 61 (page 222). Surface pressure always acts normal to the face. Positive pressures are directed toward the interior of the element. The pressure may be constant over a face or interpolated from values given at the joints. The values given at the joints are obtained from Joint Patterns, and need not be the same for the different faces. Joint Patterns can be used to easily apply hydrostatic pressures. The pressure acting on a side is multiplied by the thickness, h, integrated along the length of the side, and apportioned to the two or three joints on that side. See Chapter Load Cases (page 279) for more information.

Pore Pressure Load


The Pore Pressure Load is used to model the drag and buoyancy effects of a fluid within a solid medium, such as the effect of water upon the solid skeleton of a soil. Scalar fluid-pressure values are given at the element joints by Joint Patterns, and interpolated over the element. The total force acting on the element is the integral of the gradient of this pressure field, multiplied by the thickness h, over the plane of the element. This force is apportioned to each of the joints of the element. The forces are typically directed from regions of high pressure toward regions of low pressure. See Chapter Load Cases (page 279) for more information.

Temperature Load
The Temperature Load creates thermal strain in the Asolid element. This strain is given by the product of the Material coefficient of thermal expansion and the temperature change of the element. The temperature change is measured from the eleSurface Pressure Load
247

229

SAP2000 Analysis Reference ment Reference Temperature to the element Load Temperature. Temperature changes are assumed to be constant through the element thickness. See Chapter Load Cases (page 279) for more information.

Rotate Load
Rotate Load is used to apply centrifugal force to Asolid elements. Each element is assumed to rotate about its own axis of symmetry at a constant angular velocity. The angular velocity creates a load on the element that is proportional to its mass, its distance from the axis of rotation, and the square of the angular velocity. This load acts in the positive radial direction, and is apportioned to each joint of the element. No Rotate Load will be produced by an element with zero mass density. Since Rotate Loads assume a constant rate of rotation, it does not make sense to use a Load Case that contains Rotate Load in a time-history analysis unless that Load Case is applied quasi-statically (i.e., with a very slow time variation). For more information: See Topic Mass Density (page 142) in Chapter Material Properties. See Chapter Load Cases (page 279).

Stress Output
The Asolid element stresses are evaluated at the standard 2-by-2 Gauss integration points of the element and extrapolated to the joints. See Cook, Malkus, and Plesha (1989) for more information. The stresses are computed for the following types of Analysis Cases: Loads, Modes, Specs, Histories, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. No results are available for Moving Loads or Combos that contain Moving Loads. Principal values and their associated principal directions in the element local 1-2 plane are also computed for the Loads and Modes. The angle given is measured counterclockwise (when viewed from the +3 direction) from the local 1 axis to the direction of the maximum principal value.

230

Rotate Load
248

Chapter XII The Asolid Element It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. Asolid element stresses are printed in the output file under the heading:
ASOLID ELEMENT STRESSES

All results for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case the stresses are printed at all element joints. The stresses are labeled S11, S22, S33, and S12 corresponding to 11 , 22 , 33 , and 12 , respectively, in the element local coordinate system. Stresses 12 and 13 are always zero and are not printed. Typical Asolid element output is shown in Figure 65 (page 232). By default, no stresses are printed. You may specify which Analysis Cases are to be printed. The stresses can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. For more information: See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Joint Force Output


The Asolid element joint forces are concentrated forces acting at the joints of the element that represent the effect of the rest of the structure upon the element. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

Joint Force Output


249

231

SAP2000 Analysis Reference


A S O L I D ELEM E L E M E N T S T R E S S E S

1 ==================

LOAD INTPRES -----------------JOINT 1 2 3 12 13 14 23 24 25 JOINT 1 2 3 12 13 14 23 24 25 S11 -0.939522 -0.836460 -0.733398 -0.939522 -0.836460 -0.733398 -0.939522 -0.836460 -0.733398 S-MAX 0.157630 0.114524 0.071418 0.157629 0.114523 0.071418 0.157630 0.114524 0.071418 S22 0.157629 0.114523 0.071418 0.157629 0.114523 0.071418 0.157629 0.114523 0.071418 S-MIN -0.939523 -0.836461 -0.733399 -0.939522 -0.836460 -0.733398 -0.939523 -0.836461 -0.733399 S33 1.255412 1.065966 0.876520 1.255412 1.065966 0.876520 1.255412 1.065966 0.876520 S12 0.000920 0.000829 0.000739 -2.89E-13 -2.32E-13 -1.76E-13 -0.000920 -0.000829 -0.000739 ANGLE 89.951947 89.950033 89.947423 -90.000000 -90.000000 -90.000000 -89.951947 -89.950033 -89.947423

LOAD LOGTEMP -----------------JOINT 1 2 3 12 13 14 23 24 25 ELEM S11 -9.421631 -9.169759 -8.917887 -9.421631 -9.169759 -8.917887 -9.421631 -9.169759 -8.917887 S22 -172.393810 -150.661663 -128.929517 -172.393810 -150.661663 -128.929517 -172.393810 -150.661663 -128.929517 S33 -136.017533 -106.777127 -77.536722 -136.017533 -106.777127 -77.536722 -136.017533 -106.777127 -77.536722 S12 -0.187845 -0.172932 -0.158019 -1.28E-11 -9.82E-12 -6.79E-12 0.187845 0.172932 0.158019

2 ==================

LOAD INTPRES -----------------JOINT 3 4 5 14 15 16 25 26 27 S11 -0.634081 -0.561526 -0.488971 -0.634081 -0.561526 -0.488971 -0.634081 -0.561526 -0.488971 S22 0.168448 0.117673 0.066898 0.168448 0.117673 0.066898 0.168448 0.117673 0.066898 S33 0.971652 0.797344 0.623036 0.971652 0.797344 0.623036 0.971652 0.797344 0.623036 S12 0.000681 0.000613 0.000545 -1.46E-13 -2.06E-14 1.05E-13 -0.000681 -0.000613 -0.000545

...Output Edited for Brevity...

Figure 65 Asolid Element Stresses in the Output File

232

Joint Force Output


250

C h a p t e r XIII

The Solid Element


The Solid element is used to model three-dimensional solid structures. Advanced Topics Overview Joint Connectivity Degrees of Freedom Local Coordinate System Stresses and Strains Material Properties Material Angles Incompatible Bending Modes Mass Self-Weight Load Gravity Load Surface Pressure Load Pore Pressure Load Temperature Load

233
251

SAP2000 Analysis Reference Stress Output Joint Force Output

Overview
The Solid element is an eight-node element for modeling three-dimensional structures and solids. It is based upon an isoparametric formulation that includes nine optional incompatible bending modes. The incompatible bending modes significantly improve the bending behavior of the element if the element geometry is of a rectangular form. Improved behavior is exhibited even with non-rectangular geometry. The local coordinate system for each Solid element is identical to the global system. The local coordinate system is used for defining Material properties and loads, and for interpreting output. Temperature-dependent, anisotropic material properties are allowed. Each element may be loaded by gravity (in any direction); surface pressure on the faces; pore pressure within the element; and loads due to temperature change. An 2 x 2 x 2 numerical integration scheme is used for the Solid. Stresses in the element local coordinate system are evaluated at the integration points and extrapolated to the joints of the element. An approximate error in the stresses can be estimated from the difference in values calculated from different elements attached to a common joint. This will give an indication of the accuracy of the finite element approximation and can then be used as the basis for the selection of a new and more accurate finite element mesh.

Joint Connectivity
Each Solid element has six quadrilateral faces, with a joint located at each of the eight corners as shown in Figure 66 (page 235). It is important to note the relative position of the eight joints: the paths j1-j2-j3 and j5-j6-j7 should appear counterclockwise when viewed along the direction from j5 to j1. Mathematically stated, the three vectors: V12 , from joints j1 to j2, V13 , from joints j1 to j3,

234

Overview
252

Chapter XIII The Solid Element

Figure 66 Solid Element Joint Connectivity and Face Definitions

V15 , from joints j1 to j5, must form a positive triple product, that is:
( V12

V13 ) V15 > 0

The locations of the joints should be chosen to meet the following geometric conditions: The inside angle at each corner of the faces must be less than 180. Best results will be obtained when these angles are near 90, or at least in the range of 45 to 135. The aspect ratio of an element should not be too large. This is the ratio of the longest dimension of the element to its shortest dimension. Best results are obtained for aspect ratios near unity, or at least less than four. The aspect ratio should not exceed ten. These conditions can usually be met with adequate mesh refinement.

Joint Connectivity
253

235

SAP2000 Analysis Reference

Degrees of Freedom
The Solid element activates the three translational degrees of freedom at each of its connected joints. Rotational degrees of freedom are not activated. This element contributes stiffness to all of these translational degrees of freedom. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom for more information.

Local Coordinate System


Each Solid element has its own element local coordinate system used to define Material properties, loads and output. The axes of this local system are denoted 1, 2 and 3. These axes always correspond with the global coordinate axes X, Y and Z, respectively, regardless of the orientation of the element.

Stresses and Strains


The Solid element models a general state of stress and strain in a three-dimensional solid. All six stress and strain components are active for this element. See Topic Stresses and Strains (page 137) in Chapter Material Properties for more information.

Material Properties
The material properties for each Solid element are specified by reference to a previously-defined Material. Fully anisotropic material properties are used. The material properties used by the Solid element are: The moduli of elasticity, e1, e2, and e3 The shear moduli, g12, g13, and g23 All of the Poissons ratios, u12, u13, u23, ..., u56 The coefficients of thermal expansion, a1, a2, a3, a12, a13, and a23 The mass density, m, used for computing element mass The weight density, w, used for computing Self-Weight and Gravity Loads

236

Degrees of Freedom
254

Chapter XIII The Solid Element All material properties (except the densities) are obtained at the material temperature of each individual element. See Chapter Material Properties (page 135) for more information.

Material Angles
The material local coordinate system and the element local coordinate system need not be the same. The material coordinate system is oriented with respect to the element coordinate system using the three angles a, b, and c according to the following procedure: The material system is first aligned with the element system; The material system is then rotated about its +3 axis by angle a; The material system is next rotated about the resulting +2 axis by angle b; The material system is lastly rotated about the resulting +1 axis by angle c; This is shown in Figure 67 (page 238). These angles have no effect for isotropic material properties since they are independent of orientation. See Topic Local Coordinate System (page 136) in Chapter Material Properties for more information.

Incompatible Bending Modes


By default each Solid element includes nine incompatible bending modes in its stiffness formulation. These incompatible bending modes significantly improve the bending behavior of the element if the element geometry is of a rectangular form. Improved behavior is exhibited even with non-rectangular geometry. If an element is severely distorted, the inclusion of the incompatible modes should be suppressed. The element then uses the standard isoparametric formulation. Incompatible bending modes may also be suppressed in cases where bending is not important, such as in typical geotechnical problems.

Material Angles
255

237

SAP2000 Analysis Reference

Figure 67 Solid Element Material Angles

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Solid element is lumped at the element joints. No inertial effects are considered within the element itself. The total mass of the element is equal to the integral of the mass density, m, over the volume of the element. The total mass is apportioned to the joints in a manner that is proportional to the diagonal terms of the consistent mass matrix. See Cook, Malkus, and Plesha (1989) for more information. The total mass is applied to each of the three translational degrees of freedom (UX, UY, and UZ). For more information: See Topic Mass Density (page 142) in Chapter Material Properties. See Chapter Static and Dynamic Analysis (page 291).

238

Mass
256

Chapter XIII The Solid Element

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For a Solid element, the self-weight is a force that is uniformly distributed over the volume of the element. The magnitude of the self-weight is equal to the weight density, w. Self-Weight Load always acts downward, in the global Z direction. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. For more information: See Topic Weight Density (page 143) in Chapter Material Properties for the definition of w. See Topic Self-Weight Load (page 283) in Chapter Load Cases.

Gravity Load
Gravity Load can be applied to each Solid element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information: See Topic Self-Weight Load (page 239) in this chapter for the definition of self-weight for the Solid element. See Topic Gravity Load (page 284) in Chapter Load Cases.

Surface Pressure Load


The Surface Pressure Load is used to apply external pressure loads upon any of the six faces of the Solid element. The definition of these faces is shown in Figure 66 (page 235). Surface pressure always acts normal to the face. Positive pressures are directed toward the interior of the element. The pressure may be constant over a face or interpolated from values given at the joints. The values given at the joints are obtained from Joint Patterns, and need not

Self-Weight Load
257

239

SAP2000 Analysis Reference be the same for the different faces. Joint Patterns can be used to easily apply hydrostatic pressures. The pressure acting on a given face is integrated over the area of that face, and the resulting force is apportioned to the four corner joints of the face. See Chapter Load Cases (page 279) for more information.

Pore Pressure Load


The Pore Pressure Load is used to model the drag and buoyancy effects of a fluid within a solid medium, such as the effect of water upon the solid skeleton of a soil. Scalar fluid-pressure values are given at the element joints by Joint Patterns, and interpolated over the element. The total force acting on the element is the integral of the gradient of this pressure field over the volume of the element. This force is apportioned to each of the joints of the element. The forces are typically directed from regions of high pressure toward regions of low pressure. See Chapter Load Cases (page 279) for more information.

Temperature Load
The Temperature Load creates thermal strain in the Solid element. This strain is given by the product of the Material coefficient of thermal expansion and the temperature change of the element. The temperature change is measured from the element Reference Temperature to the element Load Temperature. Temperature changes are assumed to be constant through the element thickness. See Chapter Load Cases (page 279) for more information.

Stress Output
The Solid element stresses are evaluated at the standard 2 x 2 x 2 Gauss integration points of the element and extrapolated to the joints. See Cook, Malkus, and Plesha (1989) for more information. The stresses are computed for the following types of Analysis Cases: Loads, Modes, Specs, Histories, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically re-

240

Pore Pressure Load


258

Chapter XIII The Solid Element quested in the definition of the individual History cases. No results are available for Moving Loads or Combos that contain Moving Loads. Principal values and their associated principal directions in the element local coordinate system are also computed for the Loads and Modes. Three direction cosines each are given for the directions of the maximum and minimum principal stresses. The direction of the middle principal stress is perpendicular to the maximum and minimum principal directions. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. Solid element stresses are printed in the output file under the heading:
SOLID ELEMENT STRESSES

All results for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case the stresses are printed at all element joints. The stresses are labeled S11, S22, S33, S12, S13, and S23 corresponding to 11 , 22 , 33 , 12 , 13 , and 23 , respectively, in the element local coordinate system. Typical Solid element output is shown in Figure 68 (page 242). By default, no stresses are printed. You may specify which Analysis Cases are to be printed. The stresses can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. For more information: See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

Joint Force Output


The Solid element joint forces are concentrated forces acting at the joints of the element that represent the effect of the rest of the structure upon the element. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

Joint Force Output


259

241

SAP2000 Analysis Reference


S O L I D ELEM LOAD JOINT 3 18 4 19 8 23 9 24 JOINT 3 18 4 19 8 23 9 24 ELEM LOAD JOINT 8 23 9 24 13 28 14 29 JOINT 8 23 9 24 13 28 14 29 E L E M E N T S T R E S S E S

1 ================== 1 -----------------S11 -0.364752 -0.364752 -0.199823 -0.199823 1.966476 1.966476 -0.837325 -0.837325 S-MAX 208.757887 208.757887 209.207102 209.207102 18.186219 18.186219 12.357765 12.357765 S22 207.803651 207.803651 208.298440 208.298440 0.811970 0.811970 0.317181 0.317181 S-MID 5.648477 5.648477 -0.142270 -0.142270 1.949535 1.949535 -0.819458 -0.819458 S33 6.559145 6.559145 -3.171692 -3.171692 7.053933 7.053933 -3.666480 -3.666480 S-MIN -0.408320 -0.408320 -4.137907 -4.137907 -10.303376 -10.303376 -15.724930 -15.724930 S12 -0.247394 0.247394 -0.247394 0.247394 0.247394 -0.247394 0.247394 -0.247394 S13 -0.494788 0.494788 0.494788 -0.494788 -0.494788 0.494788 0.494788 -0.494788 S23 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889

MAX-1 MAX-2 MAX-3 MIN-1 MIN-2 MIN-3 -.001 .998 -.069 .996 .007 .084 .001 .998 -.069 .996 -.007 -.084 -.001 .998 -.065 -.120 .065 .991 .001 .998 -.065 .120 .065 .991 -.033 -.624 .780 .009 .781 .625 .033 -.624 .780 -.009 .781 .625 -.010 .755 -.655 -.036 .654 .755 .010 .755 -.655 .036 .654 .755

2 ================== 1 -----------------S11 -1.966476 -1.966476 0.837325 0.837325 0.364752 0.364752 0.199823 0.199823 S-MAX 10.303376 10.303376 15.724930 15.724930 0.408320 0.408320 4.137907 4.137907 S22 -0.811970 -0.811970 -0.317181 -0.317181 -207.803651 -207.803651 -208.298440 -208.298440 S-MID -1.949535 -1.949535 0.819458 0.819458 -5.648477 -5.648477 0.142270 0.142270 S33 -7.053933 -7.053933 3.666480 3.666480 -6.559145 -6.559145 3.171692 3.171692 S-MIN -18.186219 -18.186219 -12.357765 -12.357765 -208.757887 -208.757887 -209.207102 -209.207102 S12 -0.247394 0.247394 -0.247394 0.247394 0.247394 -0.247394 0.247394 -0.247394 MAX-1 .009 -.009 .036 -.036 .996 .996 .120 -.120 MAX-2 .781 .781 -.654 -.654 .007 -.007 -.065 -.065 S13 -0.494788 0.494788 0.494788 -0.494788 -0.494788 0.494788 0.494788 -0.494788 MAX-3 -.625 -.625 .755 .755 -.084 .084 .991 .991 S23 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889 -13.888889

MIN-1 MIN-2 MIN-3 .033 .624 .780 -.033 .624 .780 -.010 .755 .655 .010 .755 .655 -.001 .998 .069 .001 .998 .069 -.001 .998 .065 .001 .998 .065

...Output Edited for Brevity...

Figure 68 Solid Element Stresses in the Output File

242

Joint Force Output


260

C h a p t e r XIV

The Nllink Element


The Nllink element is used to model local structural nonlinearities such as gaps, dampers, isolators, and the like. Nonlinear behavior is exhibited only during nonlinear time-history analyses. For all other analyses, the Nllink element behaves linearly. Advanced Topics Overview Joint Connectivity Zero-Length Elements Degrees of Freedom Local Coordinate System Advanced Local Coordinate System Internal Deformations Nlprop Properties Nonlinear Deformation Loads Mass Self-Weight Load

243
261

SAP2000 Analysis Reference Gravity Load Internal Force and Deformation Output Joint Force Output

Overview
The Nllink element is used to model local structural nonlinearities. Nonlinear behavior is only exhibited during nonlinear time-history analyses. For all other analyses, the Nllink element behaves linearly. Each Nllink element may be either a one-joint grounded spring or a two-joint link. Properties for either type of element are defined in the same way. Each element is assumed to be composed of six separate springs, one for each of six deformational degrees-of freedom (axial, shear, torsion, and pure bending). Each of these springs possesses a dual set of properties: Linear effective-stiffness and effective-damping properties used for all linear analyses An optional nonlinear force-deformation relationship used only for nonlinear time-history analyses If the optional nonlinear properties are not specified for a given degree of freedom, the linear stiffness (but not damping) properties are used for nonlinear time-history analyses. The linear effective damping property is only used for response-spectrum analyses and linear time-history analyses. The nonlinear force-deformation relationships of these springs may be coupled or uncoupled, depending on the type of behavior modeled. A set of properties for all six degrees of freedom is called an Nlprop. Each Nlprop consists of mass, weight, and up to six linear and nonlinear force-deformation relationships that may be used by one or more Nllink elements. The types of nonlinear behavior that can be modeled with this element include: Viscoelastic damping Gap (compression only) and hook (tension only) Uniaxial plasticity

244

Overview
262

Chapter XIV The Nllink Element Biaxial-plasticity base isolator Friction-pendulum base isolator Each element has its own local coordinate system for defining the forcedeformation properties and for interpreting output. Each Nllink element may be loaded by gravity (in any direction). Available output includes the deformation across the element, and the internal forces at the joints of the element.

Joint Connectivity
Each Nllink element may take one of the following two configurations: A link connecting two joints, i and j; it is permissible for the two joints to share the same location in space A spring connecting a single joint, j, to ground

Zero-Length Elements
The following types of Nllink elements are considered to be of zero length: Single-joint elements Two-joint elements with the distance from joint i to joint j being less than or equal to the zero-length tolerance, zero, that you specify.
-3 The default value for zero is 10 . The purpose of this tolerance is to account for numerical round-off in the specification and generation of joint coordinates. For example, if joint coordinates are specified to the nearest millimeter, then a possible value for zero might be 2 or 3 mm.

Two-joint elements having a length greater than the tolerance zero are considered to be of finite length. Whether an element is of zero length or finite length affects the definition of the element local coordinate system.

Joint Connectivity
263

245

SAP2000 Analysis Reference

Degrees of Freedom
The Nllink element always activates all six degrees of freedom at its one or two connected joints. To which joint degrees of freedom the element contributes stiffness depends upon the properties you assign to the element. You must ensure that restraints or other supports are provided to those joint degrees of freedom that receive no stiffness. For more information: See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom. See Topic Nlprop Properties (page 254) in this chapter.

Local Coordinate System


Each Nllink element has its own element local coordinate system used to define force-deformation properties and output. The axes of this local system are denoted 1, 2 and 3. The first axis is directed along the length of the element and corresponds to extensional deformation. The remaining two axes lie in the plane perpendicular to the element and have an orientation that you specify; these directions correspond to shear deformation. It is important that you clearly understand the definition of the element local 1-2-3 coordinate system and its relationship to the global X-Y-Z coordinate system. Both systems are right-handed coordinate systems. It is up to you to define local systems which simplify data input and interpretation of results. In most structures the definition of the element local coordinate system is extremely simple. The methods provided, however, provide sufficient power and flexibility to describe the orientation of Nllink elements in the most complicated situations. The simplest method, using the default orientation and the Nllink element coordinate angle, is described in this topic. Additional methods for defining the Nllink element local coordinate system are described in the next topic. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Advanced Local Coordinate System (page 249) in this chapter.

246

Degrees of Freedom
264

Chapter XIV The Nllink Element

Longitudinal Axis 1
Local axis 1 is the longitudinal axis of the element, corresponding to extensional deformation. This axis is determined as follows: For elements of finite length this axis is automatically defined as the direction from joint i to joint j For zero-length elements the local 1 axis defaults to the +Z global coordinate direction (upward) For the definition of zero-length elements, see Topic Zero-Length Elements (page 245) in this chapter.

Default Orientation
The default orientation of the local 2 and 3 axes is determined by the relationship between the local 1 axis and the global Z axis. The procedure used here is identical to that for the Frame element: The local 1-2 plane is taken to be vertical, i.e., parallel to the Z axis The local 2 axis is taken to have an upward (+Z) sense unless the element is vertical, in which case the local 2 axis is taken to be horizontal along the global +X direction The local 3 axis is always horizontal, i.e., it lies in the X-Y plane An element is considered to be vertical if the sine of the angle between the local 1 -3 axis and the Z axis is less than 10 . The local 2 axis makes the same angle with the vertical axis as the local 1 axis makes with the horizontal plane. This means that the local 2 axis points vertically upward for horizontal elements.

Coordinate Angle
The Nllink element coordinate angle, ang, is used to define element orientations that are different from the default orientation. It is the angle through which the local 2 and 3 axes are rotated about the positive local 1 axis from the default orientation. The rotation for a positive value of ang appears counterclockwise when the local +1 axis is pointing toward you. The procedure used here is identical to that for the Frame element.

Local Coordinate System


265

247

SAP2000 Analysis Reference

Figure 69 The Nllink Element Coordinate Angle with Respect to the Default Orientation

For vertical elements, ang is the angle between the local 2 axis and the horizontal +X axis. Otherwise, ang is the angle between the local 2 axis and the vertical plane containing the local 1 axis. See Figure 69 (page 248) for examples.

248

Local Coordinate System


266

Chapter XIV The Nllink Element

Advanced Local Coordinate System


By default, the element local coordinate system is defined using the element coordinate angle measured with respect to the global +Z and +X directions, as described in the previous topic. In certain modeling situations it may be useful to have more control over the specification of the local coordinate system. This topic describes how to define the orientation of the transverse local 2 and 3 axes with respect to an arbitrary reference vector when the element coordinate angle, ang, is zero. If ang is different from zero, it is the angle through which the local 2 and 3 axes are rotated about the positive local 1 axis from the orientation determined by the reference vector. This topic also describes how to change the orientation of the local 1 axis from the default global +Z direction for zero-length elements. The local 1 axis is always directed from joint i to joint j for elements of finite length. For more information: See Chapter Coordinate Systems (page 21) for a description of the concepts and terminology used in this topic. See Topic Local Coordinate System (page 245) in this chapter.

Axis Reference Vector


To define the local 1 axis for zero-length elements, you specify an axis reference vector that is parallel to and has the same positive sense as the desired local 1 axis. The axis reference vector has no effect upon finite-length elements. To define the axis reference vector, you must first specify or use the default values for: A coordinate direction axdir (the default is +Z) A fixed coordinate system csys (the default is zero, indicating the global coordinate system). This will be the same coordinate system that is used to define the plane reference vector, as described below You may optionally specify: A pair of joints, axveca and axvecb (the default for each is zero, indicating the center of the element). If both are zero, this option is not used For each element, the axis reference vector is determined as follows: Advanced Local Coordinate System
267

249

SAP2000 Analysis Reference 1. A vector is found from joint axveca to joint axvecb. If this vector is of finite length, it is used as the reference vector Va 2. Otherwise, the coordinate direction axdir is evaluated at the center of the element in fixed coordinate system csys, and is used as the reference vector Va The center of a zero-length element is taken to be at joint j. The local 1 axis is given by the vector Va after it has been normalized to unit length.

Plane Reference Vector


To define the transverse local axes 2 and 3, you specify a plane reference vector that is parallel to the desired 1-2 or 1-3 plane. The procedure used here is identical to that for the Frame element. The reference vector must have a positive projection upon the corresponding transverse local axis (2 or 3, respectively). This means that the positive direction of the reference vector must make an angle of less than 90 with the positive direction of the desired transverse axis. To define the reference vector, you must first specify or use the default values for: A primary coordinate direction pldirp (the default is +Z) A secondary coordinate direction pldirs (the default is +X). Directions pldirs and pldirp should not be parallel to each other unless you are sure that they are not parallel to local axis 1 A fixed coordinate system csys (the default is zero, indicating the global coordinate system). This will be the same coordinate system that was used to define the axis reference vector, as described above The local plane, local, to be determined by the reference vector (the default is 12, indicating plane 1-2) You may optionally specify: A pair of joints, plveca and plvecb (the default for each is zero, indicating the center of the element). If both are zero, this option is not used For each element, the reference vector is determined as follows: 1. A vector is found from joint plveca to joint plvecb. If this vector is of finite length and is not parallel to local axis 1, it is used as the reference vector Vp.

250

Advanced Local Coordinate System


268

Chapter XIV The Nllink Element

Figure 70 The Nllink Element Coordinate Angle with Respect to Coordinate Directions

2. Otherwise, the primary coordinate direction pldirp is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 1, it is used as the reference vector Vp. 3. Otherwise, the secondary coordinate direction pldirs is evaluated at the center of the element in fixed coordinate system csys. If this direction is not parallel to local axis 1, it is used as the reference vector Vp. 4. Otherwise, the method fails and the analysis terminates. This will never happen if pldirp is not parallel to pldirs A vector is considered to be parallel to local axis 1 if the sine of the angle between -3 them is less than 10 . The use of the Nllink element coordinate angle in conjunction with coordinate directions that define the reference vector is illustrated in Figure 70 (page 251). The use of joints to define the reference vector is shown in Figure 71 (page 252).

Determining Transverse Axes 2 and 3


The program uses vector cross products to determine the transverse axes 2 and 3 once the reference vector has been specified. The three axes are represented by the Advanced Local Coordinate System
269

251

SAP2000 Analysis Reference

Figure 71 Using Joints to Define the Nllink Element Local Coordinate System

three unit vectors V1 , V2 and V3 , respectively. The vectors satisfy the cross-product relationship: V1 = V2 V3 The transverse axes 2 and 3 are defined as follows: If the reference vector is parallel to the 1-2 plane, then: V3 = V1 V p and V2 = V3 V1 If the reference vector is parallel to the 1-3 plane, then: V2 = V p V1 and V3 = V1 V2 In the common case where the reference vector is perpendicular to axis V1 , the transverse axis in the selected plane will be equal to Vp.

252

Advanced Local Coordinate System


270

Chapter XIV The Nllink Element

Internal Deformations
Six independent internal deformations are defined for the Nllink element. These are calculated from the relative displacements of joint j with respect to: Joint i for a two-joint element The ground for a single-joint element For two-joint link elements the internal deformations are defined as: Axial: Shear in the 1-2 plane: Shear in the 1-3 plane: Torsion: Pure bending in the 1-3 plane: Pure bending in the 1-2 plane: where: u1i, u2i, u3i, r1i, r2i, and r3i are the translations and rotations at joint i u1j, u2j, u3j, r1j, r2j, and r3j are the translations and rotations at joint j dj2 is the distance you specify from joint j to the location where the shear deformation du2 is measured (the default is zero) dj3 is the distance you specify from joint j to the location where the shear deformation du3 is measured (the default is zero)
L

du1 = u1j u1i du2 = u2j u2i dj2 r3j (L dj2) r3i du3 = u3j u3i + dj3 r2j + (L dj3) r2i dr1 = r1j r1i dr2 = r2i r2j dr3 = r3j r3i

is the length of the element

All translations, rotations, and deformations are expressed in terms of the element local coordinate system. Note that shear deformation can be caused by rotations as well as translations. These definitions ensure that all deformations will be zero under rigid-body motions of the element. It is important to note that the negatives of the rotations r2i and r2j have been used for the definition of shear and bending deformations in the 1-3 plane. This provides consistent definitions for shear and moment in both the Nllink and Frame elements. Three of these internal deformations are illustrated in Figure 72 (page 254).

Internal Deformations
271

253

SAP2000 Analysis Reference

Figure 72 Internal Deformations for a Two-Joint Nllink Element

For one-joint grounded-spring elements the internal deformations are the same as above, except that the translations and rotations at joint i are taken to be zero: Axial: Shear in the 1-2 plane: Shear in the 1-3 plane: Torsion: Pure bending in the 1-3 plane: Pure bending in the 1-2 plane: du1 = u1j du2 = u2j dj2 r3j du3 = u3j + dj3 r2j dr1 = r1j dr2 = r2j dr3 = r3j

Nlprop Properties
An Nlprop is a set of structural properties that can be used to define the behavior of one or more Nllink elements. Nlprops are defined independently of the Nllink elements and are referenced during the definition of the elements.

254

Nlprop Properties
272

Chapter XIV The Nllink Element Each Nlprop specifies the optional nonlinear force-deformation relationships for the six internal deformations. These nonlinear properties are used only during a nonlinear time-history analysis. Effective-stiffness and effective-damping properties may also be specified. These properties are used for all linear analyses, which include: static analysis, P-delta analysis, modal analysis, moving-load analysis, response-spectrum analysis, harmonic steady-state analysis, and linear or periodic time-history analysis. The effective stiffness is also used during nonlinear time-history analyses for all degrees of freedom for which nonlinear properties are not specified. The effective damping is never used for nonlinear time-history analysis. Mass and weight properties may also be specified.

Local Coordinate System


Nlprop properties are defined with respect to the local coordinate system of the Nllink element. The local 1 axis is the longitudinal direction of the element and corresponds to extensional and torsional deformations. The local 2 and 3 directions correspond to shear and bending deformations. See Topic Local Coordinate System (page 245) in this chapter.

Internal Nonlinear Springs


Each Nlprop is assumed to be composed of six internal nonlinear springs, one for each of six internal deformations. Each spring may actually consist of several components, including springs and dashpots. The force-deformation relationships of these springs may be coupled or independent of each other. Figure 73 (page 256) shows the springs for three of the deformations: axial, shear in the 1-2 plane, and pure-bending in the 1-2 plane. It is important to note that the shear spring is located a distance dj2 from joint j. All shear deformation is assumed to occur in this spring; the links connecting this spring to the joints (or ground) are rigid in shear. Deformation of the shear spring can be caused by rotations as well as translations at the joints. The force in this spring will produce a linearly-varying moment along the length. This moment is taken to be zero at the shear spring, which acts as a moment hinge. The moment due to shear is independent of, and additive to, the constant moment in the element due to the pure-bending spring.

Nlprop Properties
273

255

SAP2000 Analysis Reference

Figure 73 Three of the Six Independent Nonlinear Springs in an Nllink Element

The other three springs that are not shown are for torsion, shear in the 1-3 plane, and pure-bending in the 1-3 plane. The shear spring is located a distance dj3 from joint j. The values of dj2 and dj3 may be different, although normally they would be the same for the friction-pendulum (Isolator2) element.

Spring Force-Deformation Relationships


There are six force-deformation relationships that govern the behavior of the element, one for each of the internal springs: Axial: Shear: Torsional: fu1 vs. du1 fu2 vs. du2 , fu3 vs. du3 fr1 vs. dr1

Pure bending: fr2 vs. dr2 , fr3 vs. dr3 where fu1, fu2, and fu3 are the internal-spring forces; and fr1, fr2, and fr3 are the internalspring moments.

256

Nlprop Properties
274

Chapter XIV The Nllink Element

Figure 74 Nllink Element Internal Forces and Moments, Shown Acting at the Joints

Each of these relationships may be zero, linear only, or linear/nonlinear for a given Nlprop. These relationships may be independent or coupled. The forces and moments may be related to the deformation rates (velocities) as well as to the deformations.

Element Internal Forces


The Nllink element internal forces, P, V 2, V 3, and the internal moments, T , M 2, M 3, have the same meaning as for the Frame element. These are illustrated in Figure 74 (page 257). These can be defined in terms of the spring forces and moments as: Axial: Shear in the 1-2 plane: Shear in the 1-3 plane: P = fu1 V2 = fu2 , V3 = fu3 , M3s = (d dj2) fu2 M2s = (d dj3) fu3

Nlprop Properties
275

257

SAP2000 Analysis Reference Torsion: Pure bending in the 1-3 plane: Pure bending in the 1-2 plane: T = fr1 M2b = fr2 M3b = fr3
M2

where d is the distance from joint j. The total bending-moment resultants M 3 composed of shear and pure-bending parts:
M2 M3

and

= M 2s + M 2b = M 3s + M 3b

These internal forces and moments are present at every cross section along the length of the element. See Topic Internal Force Output (page 176) in Chapter The Frame Element.

Linear Force-Deformation Relationships


If each of the internal springs behaves linearly, the spring force-deformation relationships can be expressed in matrix form as: f u1 k u1 f u2 f u3 = f r1 f r2 f r3
0
k u2

0 0
k u3

0 0 0
k r1

0 0 0 0
k r2

sym.

d u1 d 0 u2 0 d u3 0 d r1 0 d r2 k r3 d r3
0

(Eqn. 1)

where ku1, ku2, ku3, kr1, kr2, and kr3 are the linear stiffness coefficients of the internal springs. This can be recast in terms of the element internal forces and displacements at joint j for a one-joint element as:

258

Nlprop Properties
276

Chapter XIV The Nllink Element (Eqn. 2) k u1 P V2 V3 = T M 2 M 3 j


0
k u2

0 0
k u3

0 0 0
k r1

0 0

dj3 k u 3
0
k r2

sym.

+ dj3

k u3

u1 dj2 k u 2 u 2 0 u3 r 0 1 0 r2 2 k r 3 + dj2 k u2 r3 j
0

This relationship also holds for a two-joint element if all displacements at joint i are zero. Similar relationships hold for linear damping behavior, except that the stiffness terms are replaced with damping coefficients, and the displacements are replaced with the corresponding velocities. Consider an example where the equivalent shear and bending springs are to be computed for a prismatic beam with a section bending stiffness of EI in the 1-2 plane. The stiffness matrix at joint j for the 1-2 bending plane is: 6L u 2 V2 EI 12 = 3 2 M 3 j L 6L 4L r3 j From this it can be determined that the equivalent shear spring has a stiffness of
k u2 EI
3 L

= 12

located at dj2 = , and the equivalent pure-bending spring has a stiff2


EI L

ness of k r 3 =

For an element that possesses a true moment hinge in the 1-2 bending plane, the pure-bending stiffness is zero, and dj2 is the distance to the hinge. See Figure 75 (page 260).

Linear Effective Stiffness


For each Nlprop you may specify six linear effective-stiffness coefficients, ke, one for each of the internal springs. The linear effective stiffness represents the total elastic stiffness for the Nllink element that is used for all linear analyses: static analysis, P-delta analysis, modal analysis, moving-load analysis, response-spectrum analysis, harmonic steady-state Nlprop Properties
277

259

SAP2000 Analysis Reference

Figure 75 Location of Shear Spring at a Moment Hinge or Point of Inflection

analysis, and linear or periodic time-history analysis. The actual nonlinear properties are ignored for these types of analysis. The linear effective stiffness is also used for all linear degrees of freedom during a nonlinear time-history analysis. The effective force-deformation relationships for the Nlprops are given by Equation 1 above with the appropriate values of ke substituted for ku1, ku2, ku3, kr1, kr2, and kr3. The effective stiffness properties are not directly used for nonlinear degrees of freedom during nonlinear time-history analysis. However, these analyses do make use of the vibration modes that are computed based on the effective stiffness. During time integration the behavior of these modes is modified so that the structural response reflects the actual stiffness and other nonlinear parameters specified. The rate of convergence of the nonlinear iteration may be improved by changing the effective stiffness. Following are some guidelines for selecting the linear effective stiffness. You should deviate from these as necessary to achieve your modeling and analysis goals. In particular, you should consider whether you are more interested in the results to be obtained from linear analyses, or in obtaining modes that are used as the basis for nonlinear time-history analyses. When carrying out analyses based on the UBC 94 code, the effective stiffness should usually be the code-defined maximum effective stiffness

260

Nlprop Properties
278

Chapter XIV The Nllink Element For Gap and Hook elements the effective stiffness should usually be zero or k, depending on whether the element is likely to be open or closed, respectively, in normal service For Damper elements, the effective stiffness should usually be zero For other elements, the stiffness should be between zero and k If you have chosen an artificially large value for k, be sure to use a much smaller value for ke to help avoid numerical problems in nonlinear time-history analyses In the above, k is the nonlinear stiffness property for a given degree of freedom. See subtopic Nonlinear Properties below. For more information, see Topic Nonlinear Time-History Analysis (page 328) in Chapter Static and Dynamic Analysis.

Linear Effective Damping


For each Nlprop you may specify six linear effective-damping coefficients, ce, one for each of the internal springs. By default, each coefficient ce is equal to zero. The linear effective damping represents the total viscous damping for the Nllink element that is used for response-spectrum analyses, and for linear and periodic time-history analyses. The actual nonlinear properties are ignored for these types of analysis. Effective damping can be used to represent energy dissipation due to nonlinear damping, plasticity, or friction. The effective force/deformation-rate relationships for the Nlprops are given by Equation 1 above with the appropriate values of ce substituted for ku1, ku2, ku3, kr1, kr2, and kr3, and deformation rates substituted for the corresponding deformations. The effective damping values are converted to modal damping ratios assuming proportional damping, i.e., the modal cross-coupling damping terms are ignored. These effective modal-damping values are added to any other modal damping that you specify directly. The program will not permit the total damping ratio for any mode to exceed 99.995%. Important Note: Modal cross-coupling damping terms can be very significant for some structures. A linear analysis based on effective-damping properties may grossly overestimate the amount of damping present in the structure. Nonlinear time-history analysis is strongly recommended to determine the effect of added energy dissipation devices.

Nlprop Properties
279

261

SAP2000 Analysis Reference Nonlinear time-history analysis does not use the effective damping values since it accounts for energy dissipation in the elements directly, and correctly accounts for the effects of modal cross-coupling.

Nonlinear Properties
The nonlinear properties for each Nlprop must be of one of the six types described below. The type determines which degrees of freedom may be nonlinear and the kinds of nonlinear force-deformation relationships available for those degrees of freedom. Every degree of freedom may have linear effective-stiffness and effective-damping properties specified, as described above in Subtopics Linear Effective Stiffness and Linear Effective Damping. During nonlinear time-history analysis, the nonlinear force-deformation relationships are used at all degrees of freedom for which nonlinear properties were specified. For all other degrees of freedom, the linear effective stiffnesses are used during a nonlinear time-history analysis. Nonlinear properties are not used for any other type of analysis. Linear effective stiffnesses are used for all degrees of freedom for all linear analyses. Each nonlinear force-deformation relationship includes a stiffness coefficient, k. This represents the linear stiffness when the nonlinear effect is negligible, e.g., for rapid loading of the Damper; for a closed Gap or Hook; or in the absence of yielding or slipping for the Plastic1, Isolator1, or Isolator2 properties. If k is zero, no nonlinear force can be generated for that degree of freedom, with the exception of the pendulum force in the Isolator2 property. IMPORTANT! You may sometimes be tempted to specify very large values for k, particularly for Damper, Gap, and Hook properties. Resist this temptation! If you want to limit elastic deformations in a particular internal spring, it is usually suffi2 4 cient to use a value of k that is from 10 to 10 times as large as the corresponding stiffness in any connected elements. Larger values of k may cause numerical difficulties during solution. See the additional discussion for the Damper property below. Damper Property For each deformational degree of freedom you may specify independent damping properties. The damping properties are based on the Maxwell model of viscoelas-

262

Nlprop Properties
280

Chapter XIV The Nllink Element

Figure 76 Damper, Gap, and Hook Property Types, Shown for Axial Deformations

ticity (Malvern, 1969) having a nonlinear damper in series with a spring. See Figure 76 (page 263). If you do not specify nonlinear properties for a degree of freedom, that degree of freedom is linear using the effective stiffness, which may be zero. The nonlinear force-deformation relationship is given by:
f

=k

dk

& = c d c cexp

where k is the spring constant, c is the damping coefficient, cexp is the damping ex& ponent, d k is the deformation across the spring, and d c is the deformation rate across the damper. The damping exponent must be positive; the practical range is between 0.2 and 2.0. The spring and damping deformations sum to the total internal deformation:
d

= dk + dc

If pure damping behavior is desired, the effect of the spring can be made negligible by making it sufficiently stiff. The spring stiffness should be large enough so that the characteristic time of the spring-dashpot system, given by = c / k (when cexp = 1), is an order of magnitude smaller than the size of the load steps. The load steps are the time intervals over which the load is changing. The stiffness should not be made excessively large or else numerical sensitivity may result.

Nlprop Properties
281

263

SAP2000 Analysis Reference Gap Property For each deformational degree of freedom you may specify independent gap (compression-only) properties. See Figure 76 (page 263). All internal deformations are independent. The opening or closing of a gap for one deformation does not affect the behavior of the other deformations. If you do not specify nonlinear properties for a degree of freedom, that degree of freedom is linear using the effective stiffness, which may be zero. The nonlinear force-deformation relationship is given by:
f

k ( d + open ) = 0

if

+ open < 0

otherwise

where k is the spring constant, and open is the initial gap opening, which must be zero or positive. Hook Property For each deformational degree of freedom you may specify independent hook (tension-only) properties. See Figure 76 (page 263). All internal deformations are independent. The opening or closing of a hook for one deformation does not affect the behavior of the other deformations. If you do not specify nonlinear properties for a degree of freedom, that degree of freedom is linear using the effective stiffness, which may be zero. The nonlinear force-deformation relationship is given by:
f

k ( d open ) = 0

if

open > 0

otherwise

where k is the spring constant, and open is the initial hook opening, which must be zero or positive. Plastic1 Property For each deformational degree of freedom you may specify independent uniaxialplasticity properties. The plasticity model is based on the hysteretic behavior proposed by Wen (1976). See Figure 77 (page 265).

264

Nlprop Properties
282

Chapter XIV The Nllink Element

Figure 77 Plastic1 Property Type for Uniaxial Deformation

All internal deformations are independent. The yielding at one degree of freedom does not affect the behavior of the other deformations. If you do not specify nonlinear properties for a degree of freedom, that degree of freedom is linear using the effective stiffness, which may be zero. The nonlinear force-deformation relationship is given by:
f

= ratio k

+ (1 ratio ) yield

where k is the elastic spring constant, yield is the yield force, ratio is the specified ratio of post-yield stiffness to elastic stiffness (k), and z is an internal hysteretic variable. This variable has a range of | z | 1, with the yield surface represented by . | z | = 1 The initial value of z is zero, and it evolves according to the differential equation:
z

&=

& k d (1 | z | exp & yield d

if

d z

& >0

otherwise

where exp is an exponent greater than or equal to unity. Larger values of this exponent increases the sharpness of yielding as shown in Figure 78 (page 266). The & practical limit for exp is about 20. The equation for z is equivalent to Wens model with A = 1 and = = 05. . Nlprop Properties
283

265

SAP2000 Analysis Reference

Figure 78 Definition of Parameters for Plastic1 Property

Isolator1 Property This is a biaxial hysteretic isolator that has coupled plasticity properties for the two shear deformations, and linear effective-stiffness properties for the remaining four deformations. The plasticity model is based on the hysteretic behavior proposed by Wen (1976), and Park, Wen and Ang (1986), and recommended for base-isolation analysis by Nagarajaiah, Reinhorn and Constantinou (1991). See Figure 79 (page 267). For each shear deformation degree of freedom you may independently specify either linear or nonlinear behavior: If both shear degrees of freedom are nonlinear, the coupled force-deformation relationship is given by:
f f

u2 u3

= ratio2 k2 d u 2 + (1 ratio2 ) yield2 = ratio3 k3 d u 3 + (1 ratio3 ) yield3

z2 z3

where k2 and k3 are the elastic spring constants, yield2 and yield3 are the yield forces, ratio2 and ratio3 are the ratios of post-yield stiffnesses to elastic stiffnesses (k2 and k3), and z 2 and z 3 are internal hysteretic variables. These vari-

266

Nlprop Properties
284

Chapter XIV The Nllink Element

Figure 79 Isolator1 Property for Biaxial Shear Deformation

ables have a range of


z2
2

z2

+ z 3 2 1, with the yield surface represented by


z2

+ z 3 2 = 1. The initial values of

and

z3

are zero, and they evolve ac-

cording to the differential equations: k2 d & & 1 a 2 z 2 2 a 3 z 2 z 3 yield2 u 2 = & 3 a 2 z 2 z 3 1 a 3 z 3 2 & z k3 d u 3 yield3
z2

Where:
a2

1 = 0

if

d u2 z 2

&

>0

otherwise

Nlprop Properties
285

267

SAP2000 Analysis Reference 1 = 0


if
d u3 z 3

&

a3

>0

otherwise

These equations are equivalent to those of Park, Wen and Ang (1986) with A = 1 and = = 05. . If only one shear degree of freedom is nonlinear, the above equations reduce to the uniaxial plasticity behavior of the Plastic1 property with exp = 2 for that degree of freedom. A linear spring relationship applies to the axial deformation, the three moment deformations, and to any shear deformation without nonlinear properties. All linear degrees of freedom use the corresponding effective stiffness, which may be zero. Isolator2 Property This is a biaxial friction-pendulum isolator that has coupled friction properties for the two shear deformations, post-slip stiffness in the shear directions due the pendulum radii of the slipping surfaces, gap behavior in the axial direction, and linear effective-stiffness properties for the three moment deformations. See Figure 80 (page 269). The friction model is based on the hysteretic behavior proposed by Wen (1976), and Park, Wen and Ang (1986), and recommended for base-isolation analysis by Nagarajaiah, Reinhorn and Constantinou (1991). The pendulum behavior is as recommended by Zayas and Low (1990). The friction forces and pendulum forces are directly proportional to the compressive axial force in the element. The element cannot carry axial tension. The axial force, P, is always nonlinear, and is given by:
P

u1

k1 d u1 = 0

if

d u1

<0

otherwise

Stiffness k1 must be positive in order to generate nonlinear shear force in the element. For each shear deformation degree of freedom you may independently specify either linear or nonlinear behavior: If both shear degrees of freedom are nonlinear, the friction and pendulum effects for each shear deformation act in parallel:

268

Nlprop Properties
286

Chapter XIV The Nllink Element

Figure 80 Isolator2 Property for Friction-Pendulum Shear Behavior

f f

u2 u3

= =

f f

u2 f u3 f

+ +

f f

u2 p u3 p

The frictional force-deformation relationships are given by:


f f

u2 f u3 f

= P 2 = P 3

z2 z3

where 2 and 3 are friction coefficients, and z 2 and z 3 are internal hysteretic variables. The friction coefficients are velocity-dependent according to:
r v 2 = fast2 ( fast2 slow2 ) e

3 = fast3 ( fast3 slow3 ) e r v

Nlprop Properties
287

269

SAP2000 Analysis Reference where slow2 and slow3 are the friction coefficients at zero velocity, fast2 and fast3 are the friction coefficients at fast velocities, v is the resultant velocity of sliding:
v

d u2

&

& + d u3 2

r is an effective inverse velocity given by:


r

& & rate2 d u 2 2 + rate3 d u 3 2


v
2

and rate2 and rate3 are the inverses of characteristic sliding velocities. For a Teflon-steel interface the coefficient of friction normally increases with sliding velocity (Nagarajaiah, Reinhorn, and Constantinou, 1991). The internal hysteretic variables have a range of surface represented by
z2
2

z2

+ z 3 2 1, with the yield

+ z 3 2 = 1. The initial values of z 2 and z 3 are zero, k2 d & u2 P 2 2 z3 & k3 d u 3 P 3

and they evolve according to the differential equations: & z 2 1 a 2 z 2 = & z 3 a 2 z 2 z 3


2

a 3
1

z2 z3

a3

where k2 and k3 are the elastic shear stiffnesses of the slider in the absence of sliding, and
a2

a3

1 = 0 1 = 0

if

d u2 z 2

& &

>0 >0

otherwise if
d u3 z 3

otherwise

These equations are equivalent to those of Park, Wen and Ang (1986) with A = 1 and = = 05. . This friction model permits some sliding at all non-zero levels of shear force; the amount of sliding becomes much larger as the shear force approaches the yield value of P . Sliding at lower values of shear force can be minimized by using larger values of the elastic shear stiffnesses.

270

Nlprop Properties
288

Chapter XIV The Nllink Element The pendulum force-deformation relationships are given by:
f

u2 p u3 p

=P =P

d u2

radius2
d u3

radius3

A zero radius indicates a flat surface, and the corresponding shear force is zero. Normally the radii in the two shear directions will be equal (spherical surface), or one radius will be zero (cylindrical surface). However, it is permitted to specify unequal non-zero radii. If only one shear degree of freedom is nonlinear, the above frictional equations reduce to:
f

=P

z
 rate d

= fast ( fast slow ) e


z

&=

& k d (1 z 2 ) & P d

if

d z

& >0

otherwise

The above pendulum equation is unchanged for the nonlinear degree of freedom. A linear spring relationship applies to the three moment deformations, and to any shear deformation without nonlinear properties. All linear degrees of freedom use the corresponding effective stiffness, which may be zero. The axial degree of freedom is always nonlinear for nonlinear time-history analyses.

Nonlinear Deformation Loads


A nonlinear deformation load is a set of forces and/or moments on the structure that activates a nonlinear internal deformation of an Nllink element. A nonlinear internal deformation is an Nllink internal deformation for which nonlinear properties have been specified. Nonlinear deformation loads are used as starting load vectors for Ritz-vector analysis. Their purpose is to generate Modes that can adequately represent nonlinear behavior when performing nonlinear time-history analyses. A separate nonlinear deformation load should be used for each nonlinear internal deformation of each Nllink element. Nonlinear Deformation Loads
289

271

SAP2000 Analysis Reference When requesting a Ritz-vector analysis, you may specify that the program use built-in nonlinear deformation loads, or you may define your own Load Cases for this purpose. In the latter case you may need up to six of these Load Cases per Nllink element in the model. The built-in nonlinear deformation loads for a single two-joint Nllink element are shown in Figure 81 (page 273). Each set of forces and/or moments is selfequilibrating. This tends to localize the effect of the load, usually resulting in a better set of Ritz-vectors. For a single-joint element, only the forces and/or moments acting on joint j are needed. It is strongly recommended that mass or mass moment of inertia be present at each degree of freedom that is acted upon by a force or moment from a nonlinear deformation load. This is needed to generate the appropriate Ritz vectors. For more information: See Topic Internal Deformations (page 253) in this chapter. See Topic Nlprop Properties (page 254) in this chapter. See Topic Mass (page 272) in this chapter. See Topic Ritz-Vector Analysis (page 299) in Chapter Static and Dynamic Analysis. See Topic Nonlinear Time-History Analysis (page 328) in Chapter Static and Dynamic Analysis.

Mass
In a dynamic analysis, the mass of the structure is used to compute inertial forces. The mass contributed by the Nllink element is lumped at the joints i and j. No inertial effects are considered within the element itself. For each Nlprop, you may specify a total translational mass, m. Half of the mass is assigned to the three translational degrees of freedom at each of the elements one or two joints. For single-joint elements, half of the mass is assumed to be grounded. You may additionally specify total rotational mass moments of inertia, mr1, mr2, and mr3, about the three local axes of each element. Half of each mass moment of inertia is assigned to each of the elements one or two joints. For single-joint elements, half of each mass moment of inertia is assumed to be grounded.

272

Mass
290

Chapter XIV The Nllink Element

Figure 81 Built-in Nonlinear Deformation Loads for a Two-joint Nllink Element Mass
291

273

SAP2000 Analysis Reference The rotational inertias are defined in the element local coordinate system, but will be transformed by the program to the local coordinate systems for joint i and j. If the three inertias are not equal and element local axes are not parallel to the joint local axes, then cross-coupling inertia terms will be generated during this transformation. These will be discarded by the program, resulting in some error. It is strongly recommended that there be mass corresponding to each nonlinear deformation load in order to generate appropriate Ritz vectors for nonlinear timehistory analysis. Note that rotational inertia is needed as well as translational mass for nonlinear shear deformations if either the element length or dj is non-zero. For more information: See Chapter Static and Dynamic Analysis (page 291). See Topic Nonlinear Deformation Loads (page 271) in this chapter.

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. For each Nlprop, a total self-weight, w, may be defined. Half of this weight is assigned to each joint of each Nllink element using that Nlprop. For single-joint elements, half of the weight is assumed to be grounded. Self-Weight Load always acts downward, in the global Z direction. You may scale the self-weight by a single scale factor that applies equally to all elements in the structure. See Topic Self-Weight Load (page 283) in Chapter Load Cases for more information.

Gravity Load
Gravity Load can be applied to each Nllink element to activate the self-weight of the element. Using Gravity Load, the self-weight can be scaled and applied in any direction. Different scale factors and directions can be applied to each element. If all elements are to be loaded equally and in the downward direction, it is more convenient to use Self-Weight Load. For more information:

274

Self-Weight Load
292

Chapter XIV The Nllink Element See Topic Self-Weight Load (page 274) in this chapter for the definition of self-weight for the Nllink element. See Topic Gravity Load (page 284) in Chapter Load Cases.

Internal Force and Deformation Output


Nllink element internal forces and deformations are computed for the following types of Analysis Cases: Loads, Modes, Specs, Histories, and Combos. History results are the maximum and minimum values attained during the analysis, and are only available if specifically requested in the definition of the individual History cases. No results are available for Moving Loads or Combos that contain Moving Loads. Results for linear analyses are based upon the linear effective-stiffness and effective-damping properties and do not include any nonlinear effects. Only the results for nonlinear Histories include nonlinear behavior. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. The element internal forces were defined in Subtopic Element Internal Forces (page 257) of this chapter. The internal deformations were defined in Topic Internal Deformations (page 253) of this chapter. Nllink element internal forces and deformations are printed in the output file under the heading:
NLLINK ELEMENT INTERNAL FORCES & DEFORMS

All results for a single element are printed together. Elements are presented in alpha-numeric order. For each Analysis Case internal forces are printed at the element joints, followed by the internal deformations. The element internal forces are labeled P, V2, V3, T, M2, and M3 in the output. The internal deformations are labeled U1, U2, U3, R1, R2, and R3 in the output, corresponding to the values of du1, du2, du3, dr1, dr2, and dr3. Typical Nllink element output is shown in Figure 82 (page 276). By default, no internal forces or deformations are printed. You may specify which Analysis Cases are to be printed. The internal forces and deformations can always be displayed using the SAP2000 graphical interface even if no results have been selected for printing. Internal Force and Deformation Output
293

275

SAP2000 Analysis Reference


N L L I N K ELEM MODE JOINT 8 15 JOINT 15 MODE JOINT 8 15 JOINT 15 ELEM MODE JOINT 5 14 JOINT 14 MODE JOINT 5 14 JOINT 14 E L E M E N T I N T E R N A L F O R C E S & D E F O R M S

N1 ================== TYPE = TWO-JOINT LINK 1 -----------------P .000000 .000000 U1 5.70E-16 V2 -7339.063 -7339.063 U2 -0.733906 V3 .000000 .000000 U3 .000000 T .000000 .000000 R1 .000000 M2 .000000 .000000 R2 .000000 M3 .000000 .000000 R3 -0.000139

2 -----------------P .000000 .000000 U1 9.20E-16 V2 -15370.177 -15370.177 U2 -1.537018 V3 .000000 .000000 U3 .000000 T .000000 .000000 R1 .000000 M2 .000000 .000000 R2 .000000 M3 .000000 .000000 R3 0.000489

N2 ================== TYPE = TWO-JOINT LINK 1 -----------------P .000000 .000000 U1 1.36E-16 V2 -5608.591 -5608.591 U2 -0.560859 V3 .000000 .000000 U3 .000000 T .000000 .000000 R1 .000000 M2 .000000 .000000 R2 .000000 M3 .000000 .000000 R3 -0.000184

2 -----------------P .000000 .000000 U1 -2.50E-17 V2 11274.612 11274.612 U2 1.127461 V3 .000000 .000000 U3 .000000 T .000000 .000000 R1 .000000 M2 .000000 .000000 R2 .000000 M3 .000000 .000000 R3 0.000779

...Output Edited for Brevity...

Figure 82 Nllink Element Internal Forces and Deformations in the Output File

For more information: See Chapter Load Cases (page 279). See Chapter Static and Dynamic Analysis (page 291). See Chapter The Output Files (page 395).

276

Internal Force and Deformation Output


294

Chapter XIV The Nllink Element

Joint Force Output


The Nllink element joint forces are concentrated forces and moments acting at the joints of the element that represent the effect of the rest of the structure upon the element. The Nllink element joint forces do not include any nonlinear effects. Element joint forces must not be confused with internal forces and moments which, like stresses, act throughout the element, not just at the joints. See Topic Element Joint Force Output (page 402) in Chapter The Output Files for more information.

Joint Force Output


295

277

SAP2000 Analysis Reference

278

Joint Force Output


296

C h a p t e r XV

Load Cases
A Load Case is a specified spatial distribution of forces, displacements, temperatures, and other effects that act upon the structure. The Load Cases form the basis for most of the loading used by the different types of analyses that can be performed by the program. Basic Topics for All Users Overview Load Cases for Analysis Defining Load Cases Coordinate Systems and Load Components Force Load Restraint Displacement Load Spring Displacement Load Self-Weight Load Concentrated Span Load Distributed Span Load Uniform Load

279
297

SAP2000 Analysis Reference Advanced Topics Gravity Load Prestress Load Surface Pressure Load Pore Pressure Load Temperature Load Reference Temperature Rotate Load

Overview
Each Load Case may consist of an arbitrary combination of the available load types: Concentrated forces and moments acting at the joints Displacements of the grounded ends of restraints at the joints Displacements of the grounded ends of springs at the joints Self-weight and/or gravity acting on all element types Concentrated or distributed forces and moments acting on the Frame elements Distributed forces acting on the Shell elements Surface pressure acting on the Shell, Plane, Asolid, and Solid elements Pore pressure acting on the Plane, Asolid, and Solid elements Thermal expansion acting on the Frame, Shell, Plane, Asolid, and Solid elements Prestress cable forces acting on Frame elements Centrifugal forces acting on Asolid elements For practical purposes, it is usually most convenient to restrict each Load Case to a single type of load, using Combos and Histories to create more complicated combinations. Any number of Load Cases may be specified.

280

Overview
298

Chapter XV Load Cases

Load Cases for Analysis


The Load Cases are the basic loadings used by the different analysis types as follows: Static analysis The program automatically computes the linear response to each Load Case unless steady-state analysis has been requested. Steady-state analysis Each Load Case represents a spatial load distribution applied at a harmonic frequency, cyc. The program automatically computes the linear response to each Load Case. P-delta analysis A specified linear combination of the Load Cases is applied to the structure to determine the P-delta effect. Vibration-mode analysis Selected Load Cases may be used to bias the Ritz vectors obtained. Load Cases do not affect eigenvectors. Time-history analysis The Load Cases may be multiplied by timedependent functions and combined to form a complete load that varies with time and space. For more information: See Chapter Static and Dynamic Analysis (page 291). See Chapter P-Delta Analysis (page 337).

Defining Load Cases


You specify a unique label, name, for each Load Case to be defined. Within each Load Case, any number of joints or elements may be loaded by any number of different load types. Loads may be applied to the joints and elements, individually or in arrays. Multiple applications of the same type of load to a joint or element are additive. See the individual topics that follow for how each type of load adds. Loads of a given type may be removed from the joints and elements, individually or in arrays. Subsequent applications of that same type of load to the joints or elements are additive from zero.

Load Cases for Analysis


299

281

SAP2000 Analysis Reference

Coordinate Systems and Load Components


Certain types of loads, such as temperature and pressure, are scalars that are independent of any coordinate system. Forces and displacements, however, are vectors whose components depend upon the coordinate system in which they are specified. Vector loads may be specified with respect to any fixed coordinate system. The fixed coordinate system to be used is specified as csys. If csys is zero (the default), the global system is used. Otherwise csys refers to an Alternate Coordinate System. The X, Y, and Z components of a force or translation in a fixed coordinate system are specified as ux, uy, and uz, respectively. The X, Y, and Z components of a moment or rotation are specified as rx, ry, and rz, respectively. Most vector loads may also be specified with respect to joint and element local coordinate systems. Unlike fixed coordinate systems, the local coordinate systems may vary from joint to joint and element to element. The 1, 2, and 3 components of a force or translation in a local coordinate system are specified as u1, u2, and u3, respectively. The 1, 2, and 3 components of a moment or rotation are specified as r1, r2, and r3, respectively. You may use a different coordinate system, as convenient, for each application of a given type of load to a particular joint or element. The program will convert all these loads to a single coordinate system and add them together to get the total load. See Chapter Coordinate Systems (page 21) for more information.

Force Load
Force Load applies concentrated forces and moments to the joints. You may specify components ux, uy, uz, rx, ry, and rz in any fixed coordinate system csys, and components u1, u2, u3, r1, r2, and r3 in the joint local coordinate system. Joint Patterns can be used to specify the distribution of component values. Force values are additive after being converted to the joint local coordinate system. See Topic Force Load (page 93) in Chapter Joints and Degrees of Freedom for more information.

282

Coordinate Systems and Load Components


300

Chapter XV Load Cases

Restraint Displacement Load


Restraint Displacement Load applies specified ground displacements (translations and rotations) along the restrained degrees of freedom at the joints. You may specify components ux, uy, uz, rx, ry, and rz in any fixed coordinate system csys, and components u1, u2, u3, r1, r2, and r3 in the joint local coordinate system. Joint Patterns can be used to specify the distribution of component values. Displacement values are additive after being converted to the joint local coordinate system. See Topic Restraint Displacement Load (page 93) in Chapter Joints and Degrees of Freedom for more information.

Spring Displacement Load


Spring Displacement Load applies specified displacements (translations and rotations) at the grounded end of spring supports at the joints. You may specify components ux, uy, uz, rx, ry, and rz in any fixed coordinate system csys, and components u1, u2, u3, r1, r2, and r3 in the joint local coordinate system. Joint Patterns can be used to specify the distribution of component values. Displacement values are additive after being converted to the joint local coordinate system. See Topic Spring Displacement Load (page 95) in Chapter Joints and Degrees of Freedom for more information.

Self-Weight Load
Self-Weight Load activates the self-weight of all elements in the model. Selfweight always acts downward, in the global Z direction. You may scale the selfweight by a single scale factor that applies to the whole structure. No Self-Weight Load can be produced by an element with zero weight. For more information: See Topic Upward and Horizontal Directions (page 23) in Chapter Coordinate Systems. See Topic Self-Weight Load (page 168) in Chapter The Frame Element. See Topic Self-Weight Load (page 195) in Chapter The Shell Element. See Topic Self-Weight Load (page 213) in Chapter The Plane Element. See Topic Self-Weight Load (page 228) in Chapter The Asolid Element. Restraint Displacement Load
301

283

SAP2000 Analysis Reference See Topic Self-Weight Load (page 239) in Chapter The Solid Element. See Topic Self-Weight Load (page 274) in Chapter The Nllink Element.

Gravity Load
Gravity Load activates the self-weight of the Frame, Shell, Plane, Asolid, Solid, and Nllink elements. For each element to be loaded, you may specify the gravitational multipliers ux, uy, and uz in any fixed coordinate system csys. Multiplier values are additive after being converted to the global coordinate system. Each element produces a Gravity Load, having three components in system csys, equal to its self-weight multiplied by the factors ux, uy, and uz. This load is apportioned to each joint of the element. For example, if uz = 2, twice the self-weight is applied to the structure acting in the negative Z direction of system csys. No Gravity Load can be produced by an element with zero weight. The difference between Self-Weight Load and Gravity Load is: Self-Weight Load acts equally on all elements of the structure and always in the global Z direction Gravity Load may have a different magnitude and direction for each element in the structure Both loads are proportional to the self-weight of the individual elements. For more information: See Topic Gravity Load (page 168) in Chapter The Frame Element. See Topic Gravity Load (page 195) in Chapter The Shell Element. See Topic Gravity Load (page 214) in Chapter The Plane Element. See Topic Gravity Load (page 228) in Chapter The Asolid Element. See Topic Gravity Load (page 239) in Chapter The Solid Element. See Topic Gravity Load (page 274) in Chapter The Nllink Element.

Concentrated Span Load


Concentrated Span Load applies concentrated forces and moments at arbitrary locations on Frame elements. You may specify components ux, uy, uz, rx, ry, and rz in any fixed coordinate system csys, and components u1, u2, u3, r1, r2, and r3 in

284

Gravity Load
302

Chapter XV Load Cases the Frame element local coordinate system. Force values are additive after being converted to the Frame element local coordinate system. See Topic Concentrated Span Load (page 169) in Chapter The Frame Element for more information.

Distributed Span Load


Distributed Span Load applies distributed forces and moments at arbitrary locations on Frame elements. You may specify components ux, uy, uz, rx, ry, and rz in any fixed coordinate system csys, and components u1, u2, u3, r1, r2, and r3 in the Frame element local coordinate system. Force values are additive after being converted to the Frame element local coordinate system. See Topic Distributed Span Load (page 169) in Chapter The Frame Element for more information.

Prestress Load
Prestress Load applies the forces and moments caused by prestressing cables in Frame elements. For each element you may specify a scale factor, p, that multiplies the effect of all cables that act on that element. The scale factors are additive. See Topic Prestress Load (page 174) in Chapter The Frame Element for more information.

Uniform Load
Uniform Load applies uniformly distributed forces to the midsurface of Shell elements. You may specify components ux, uy, and uz in any fixed coordinate system csys, and components u1, u2, and u3 in the element local coordinate system. Force values are additive after being converted to the element local coordinate system. See Topic Uniform Load (page 195) in Chapter The Shell Element for more information.

Distributed Span Load


303

285

SAP2000 Analysis Reference

Surface Pressure Load


Surface Pressure Load applies an external pressure to any of the outer faces of the Shell, Plane, Asolid, and Solid elements. The load on each face of an element is specified independently. You may specify pressures, p, that are uniform over an element face or interpolated from pressure values given by Joint Patterns. Hydrostatic pressure fields may easily be specified using Joint Patterns. Pressure values are additive. For more information: See Topic Surface Pressure Load (page 196) in Chapter The Shell Element. See Topic Surface Pressure Load (page 214) in Chapter The Plane Element. See Topic Surface Pressure Load (page 229) in Chapter The Asolid Element. See Topic Surface Pressure Load (page 239) in Chapter The Solid Element.

Pore Pressure Load


Pore Pressure Load models the drag and buoyancy effects of a fluid within a solid medium, such as the effect of water upon the solid skeleton of a soil. Pore Pressure Load may be used with Shell, Asolid, and Solid elements. Scalar fluid-pressure values are given at the element joints by Joint Patterns, and interpolated over the element. These pressure values may typically be obtained by flow-net analysis, such as illustrated in Figure 83 (page 287). Hydrostatic pressure fields may easily be specified using Joint Patterns. Pressure values are additive. The total force acting on the element is the integral of the gradient of this pressure field over the volume of the element. This force is apportioned to each of the joints of the element. The forces are typically directed from regions of high pressure toward regions of low pressure. Note that although pressures are specified, it is the pressure gradient over an element that causes the load. Thus a uniform pressure field over an element will cause no load. Pressure differences between elements also cause no load. For this reason, it is important that the pore-pressure field be continuous over the structure.

286

Surface Pressure Load


304

Chapter XV Load Cases

Figure 83 Flow-net Analysis of an Earth Dam to Obtain Pore Pressures

The displacements, stresses, and reactions due to Pore Pressure Load represent the response of the solid medium, not that of the combined fluid and solid structure. In the case of soil, the stresses obtained are the usual effective stresses of soil mechanics (Terzaghi and Peck, 1967). Note, however, that the total soil weight and mass density should be used for the material properties. For more information: See Topic Pore Pressure Load (page 214) in Chapter The Plane Element. See Topic Pore Pressure Load (page 229) in Chapter The Asolid Element. See Topic Pore Pressure Load (page 240) in Chapter The Solid Element.

Temperature Load
Temperature Load creates thermal strains in the Frame, Shell, Plane, Asolid, and Solid elements. These strains are given by the product of the Material coefficients of thermal expansion and the temperature change of the element. The temperature change is measured from the element Reference Temperature to the element Load Temperature. The Load Temperature may be different for each Load Case. You may specify Load temperatures, t, that are uniform over an element or that are interpolated from Temperature Load
305

287

SAP2000 Analysis Reference values given by Joint Patterns. Temperature values are additive and add from zero, not from the Reference Temperature. In any Load Case where the Load Temperature is not specified for a given element, it is assumed to be equal to the Reference Temperature, and hence causes no load. Note, however, that when Load Temperatures are specified they add from zero, not from the Reference Temperature. Load Temperature gradients may also be specified in the two transverse directions of the Frame element, and in the thickness direction of the Shell element. These gradients induce bending strains in the elements. Temperature gradients are specified as the change in temperature per unit length. The Reference Temperature gradients are always assumed to be zero. The Load Temperature gradients may be different for each Load Case. You may specify temperature gradients, t2 and/or t3, that are uniform over an element or that are interpolated from values given by Joint Patterns. Temperature gradient values are additive. For more information: See Topic Temperature Load (page 173) in Chapter The Frame Element. See Topic Temperature Load (page 197) in Chapter The Shell Element. See Topic Temperature Load (page 215) in Chapter The Plane Element. See Topic Temperature Load (page 229) in Chapter The Asolid Element. See Topic Temperature Load (page 240) in Chapter The Solid Element.

Reference Temperature
Each Frame, Shell, Plane, Asolid, and Solid element has a single Reference Temperature field that is used for all Load Cases. This is the temperature at which the unloaded element is assumed to be stress-free. The Reference Temperature is used as part of the Temperature Load. You may specify Reference Temperatures, t, that are uniform over an element or that are interpolated from values given by Joint Patterns. Temperature values are additive. If no Reference Temperature is given for an element, a value of zero is assumed. The Reference Temperature gradients for the Frame and Shell elements are always assumed to be zero.

288

Reference Temperature
306

Chapter XV Load Cases

Rotate Load
Rotate Load applies centrifugal force to Asolid elements. You may specify an angular velocity, r, for each element. The centrifugal force is proportional to the square of the angular velocity. The angular velocities are additive. The load on the element is computed from the total angular velocity. See Topic Rotate Load (page 230) in Chapter The Asolid Element.

Rotate Load
307

289

SAP2000 Analysis Reference

308

C h a p t e r XVI

Static and Dynamic Analysis


Static and dynamic analyses are used to determine the response of the structure to various types of loading. This chapter describes the basic types of analysis available for SAP2000. See Chapter P-Delta Analysis (page 337) for information on that subject. See Chapter Bridge Analysis (page 357) for information on moving-load analysis. Basic Topics for All Users Overview Analysis Cases Static Analysis Acceleration Loads Eigenvector Analysis Ritz-vector Analysis Modal Analysis Output Functions Response-Spectrum Analysis Response-Spectrum Analysis Output

291
309

SAP2000 Analysis Reference Advanced Topics Harmonic Steady-State Analysis Time-History Analysis Nonlinear Time-History Analysis Combos

Overview
Many different types of analysis are available using program SAP2000. These include: Static analysis P-delta analysis Modal analysis for vibration modes, using eigenvectors or Ritz vectors Response-spectrum analysis for seismic response Time-history dynamic response analysis: linear, nonlinear, and periodic Moving-load analysis for bridge vehicle live loads Harmonic steady-state analysis These different types of analyses can all be performed in the same execution of the program, and the results combined for output, with the following exceptions: Modal analysis is required in order to perform response-spectrum and/or timehistory analyses Only one type of modal analysis may be performed in a single run: eigenvalue analysis or Ritz-vector analysis When harmonic steady-state analysis is performed, no other analysis type is permitted P-delta analysis, when performed, affects the results of all other analyses performed in the same execution of the program.

Analysis Cases
Each different analysis performed or combination of analysis results is called an analysis case. You assign a label to each analysis case as part of its definition. These labels can be used to create additional combinations and to control output.

292

Overview
310

Chapter XVI Static and Dynamic Analysis The different types of analysis cases are: Load Case, or simply Load a basic spatial distribution of loading, and the corresponding result of a static analysis or a harmonic steady-state analysis Mode an eigenvector or Ritz-vector, and the corresponding frequency, resulting from vibration-mode analysis Specs the basic result of a response-spectrum analysis History the basic result of a time-history analysis Moving Load the basic result of a moving-load analysis Combo a combination or envelope of Loads, Modes, Specs, Histories, Moving Loads, and/or other Combos You may define any number of each different type of analysis case to be computed in a single execution of the program. For example, you could request lots of Loads, many Modes, several Specs, multiple Moving Loads, a handful of Histories, and a couple of Combos. P-delta analysis is not considered to be an analysis case. When requested, a single P-delta analysis is performed. It effectively modifies the characteristics of the structure, and thus affects the results of all the analysis cases. For more information: See Chapter Load Cases (page 279). See Topic Harmonic Steady State Analysis (page 294) in this chapter. See Topic Eigenvector Analysis (page 296) in this chapter. See Topic Ritz-vector Analysis (page 299) in this chapter. See Topic Response-Spectrum Analysis (page 310) in this chapter. See Topic Time History Analysis (page 319) in this chapter. See Topic Moving Load Cases (page 383) in Chapter Bridge Analysis. See Topic Combos (page 333) in this chapter. See Chapter P-Delta Analysis (page 337).

Static Analysis
The static analysis of a structure involves the solution of the system of linear equations represented by:

Static Analysis
311

293

SAP2000 Analysis Reference K u =r where K is the stiffness matrix, r is the vector of applied loads, and u is the vector of resulting displacements. See Bathe and Wilson (1976). For each Load Case that you define, the program automatically creates the load vector r and solves for the static displacements u. The Load Cases do not yield static results when steady-state analysis is requested or when a frequency shift has been used for eigenvalue analysis. See Chapter Load Cases (page 279) for more information.

Harmonic Steady-State Analysis


A very common type of loading is of the form r ( t ) = p cos(t ), where is the circular frequency of the excitation, so that r varies with respect to time; however, the spatial distribution of load p does not vary as a function of time. For the case of zero damping, the equilibrium equations for the structural system are of the following form: && K u( t ) + M u( t ) = r ( t ) = p cos(t ) where K is the stiffness matrix and M is the diagonal mass matrix. The steady-state solution of this equation requires that the joint displacements u and accelerations && u are of the following form: u( t ) = a cos(t ) && u( t ) = 2 a cos(t ) Therefore, the response amplitude a is given by the solution of the following set of linear equations:
[

K 2 M ] a = p

It is of interest to note that the solution for static loads is nothing more than a solution of this equation with zero frequency. For each Load Case that you define, the program automatically creates the spatial load vector p and solves for the steady-state displacement amplitudes a. The Load Case may be composed of the same types of loading used for static analysis, except that the constant-velocity Rotate Load for the Asolid element would not be meaningful here.

294

Harmonic Steady-State Analysis


312

Chapter XVI Static and Dynamic Analysis The displacements printed by the program are amplitudes a (the maximum displacements) which vary as cos(t ). All resulting forces and stresses are also maximum values, which vary as cos(t ). The limitations of the harmonic steady-state analysis option are: The structural damping is assumed to be zero. The structure may be analyzed for any number of spatial distributions of steady-state loads. However, the excitation frequency of all the loads in any one execution of the program must be the same. If the excitation frequency corresponds to a natural frequency of the structure, the system will go into resonance, resulting in an infinite response. In such cases the solution of the system will fail as the matrix [ K 2 M ] will be singular. The solver will report that the structure is unstable. Harmonic steady-state analysis is requested by specifying a positive value for the cyclic frequency cyc. The circular frequency = 2 cyc. Setting cyc to zero (the default) enables static analysis. P-delta, response-spectrum, time-history, and moving-load analyses may not be performed when the program is in harmonic steady-state analysis mode. See Chapter Load Cases (page 279) for more information.

Acceleration Loads
The program automatically computes three Acceleration Loads that act on the structure due to unit translational accelerations in each of the three global directions. They are determined by dAlemberts principal, and are denoted mx, my, and mz. These loads are used for applying ground accelerations in response-spectrum and time-history analyses, and can be used as starting load vectors for Ritz-vector analysis. These loads are computed for each joint and element and summed over the whole structure. The Acceleration Loads for the joints are simply equal to the negative of the joint translational masses in the joint local coordinate system. These loads are transformed to the global coordinate system. The Acceleration Loads for all elements except for the Asolid are the same in each direction and are equal to the negative of the element mass. No coordinate transformations are necessary.

Acceleration Loads
313

295

SAP2000 Analysis Reference For the Asolid element, the Acceleration Load in the global direction corresponding to the axial (local 2) direction is equal to the negative of the element mass. The Acceleration Loads in the radial and circumferential directions are zero, since translations in the corresponding global directions are not axisymmetric. The Acceleration Loads can be transformed into any coordinate system. In a fixed coordinate system (global or Alternate), the Acceleration Loads along the positive X, Y, and Z axes are denoted UX, UY, and UZ, respectively. In a local coordinate system defined for a response-spectrum or time-history analysis, the Acceleration Loads along the positive local 1, 2, and 3 axes are denoted U1, U2, and U3, respectively.

Eigenvector Analysis
Eigenvector analysis determines the undamped free-vibration mode shapes and frequencies of the system. These natural Modes provide an excellent insight into the behavior of the structure. They can also be used as the basis for response-spectrum or time-history analyses, although Ritz vectors are recommended for this purpose. Eigenvector analysis involves the solution of the generalized eigenvalue problem:
[

K 2 M ] = 0

where K is the stiffness matrix, M is the diagonal mass matrix, 2 is the diagonal matrix of eigenvalues, and is the matrix of corresponding eigenvectors (mode shapes). Each eigenvalue-eigenvector pair is called a natural Vibration Mode of the structure. The Modes are identified by numbers from 1 to n in the order in which the modes are found by the program. The eigenvalue is the square of the circular frequency, , for that Mode (unless a frequency shift is used, see below). The cyclic frequency, f, and period, T, of the Mode are related to by:
T

1
f

and

You may specify the number of modes to be found, a convergence tolerance, and the frequency range of interest. These parameters are described in the following subtopics.

296

Eigenvector Analysis
314

Chapter XVI Static and Dynamic Analysis

Number of Modes
You may specify the number of Modes, n, to be found. The program will seek the n lowest-frequency (longest-period) Modes. If a non-zero frequency shift has been specified, the program will seek the n Modes with frequencies closest to shift. The number of Modes actually found, n, is limited by: The number of Modes requested, n The number of Modes present in the specified frequency range; see below The number of mass degrees of freedom in the model A mass degree of freedom is any active degree of freedom that possesses translational mass or rotational mass moment of inertia. The mass may have been assigned directly to the joint or may come from connected elements. Only the Modes that are actually found will be available for any subsequent response-spectrum or time-history analysis processing. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom.

Frequency Range
You may specify a restricted frequency range in which to seek the Vibration Modes by using the parameters: shift: The center of the cyclic frequency range, known as the shift frequency cut: The radius of the cyclic frequency range, known as the cutoff frequency The program will only seek Modes with frequencies
|
f f

that satisfy:

shift | cut

The default value of cut = 0 does not restrict the frequency range of the Modes. Modes are found in order of increasing distance of frequency from the shift. This continues until the cutoff is reached, the requested number of Modes is found, or the number of mass degrees of freedom is reached. A stable structure will possess all positive natural frequencies. When performing a seismic analysis and most other dynamic analyses, the lower-frequency modes are usually of most interest. It is then appropriate to use a shift of zero, resulting in the lowest-frequency modes of the structure being calculated. If the shift is not zero, Eigenvector Analysis
315

297

SAP2000 Analysis Reference response-spectrum and time-history analyses may be performed; however, static, moving-load, and p-delta analyses are not allowed. If the dynamic loading is known to be of high frequency, such as that caused by vibrating machinery, it may be most efficient to use a positive shift near the center of the frequency range of the loading. A structure that is unstable when unloaded will have some modes with zero frequency. These modes may correspond to rigid-body motion of an inadequately supported structure, or to mechanisms that may be present within the structure. It is not possible to compute the static response of such a structure. However, by using a small negative shift, the lowest-frequency vibration modes of the structure, including the zero-frequency instability modes, can be found. This does require some mass to be present that is activated by each instability mode. A structure that has buckled under P-Delta load will have some modes with zero or negative frequency. If the program detects a negative-frequency mode it will stop the analysis since the results will be meaningless. During equation solution, the number of frequencies less than the shift is determined and printed in the log file. If any frequencies exist that are less than a zero or negative shift, the analysis is halted. If a positive shift is used, the program must perform an eigen-analysis to find negative frequencies. These may not be found if the cutoff radius is too small, or if not enough modes were requested. In order to avoid this problem whenever a positive shift is used with a P-Delta analysis, it is recommended that a preliminary P-Delta analysis be performed using a zero shift. When using a frequency shift, the stiffness matrix is modified by subtracting from it the mass matrix multiplied by 0 2 , where 0 = 2 shift. If the shift is very near a natural frequency of the structure, the solution becomes unstable and will be halted during equation solution. The circular frequency, , of a Vibration Mode is determined from the shifted eigenvalue, , as: = + 0 2

Convergence Tolerance
SAP2000 solves for the eigenvalue-eigenvectors pairs using an accelerated subspace iteration algorithm. During the solution phase, the program prints the approximate eigenvalues after each iteration. As the eigenvectors converge they are

298

Eigenvector Analysis
316

Chapter XVI Static and Dynamic Analysis removed from the subspace and new approximate vectors are introduced. For details of the algorithm, see Wilson and Tetsuji (1983). You may specify the relative convergence tolerance, tol, to control the solution; the -5 default value is tol = 10 . The iteration for a particular Mode will continue until the relative change in the eigenvalue between successive iterations is less than 2 tol, that is until:
1 i+1

i tol i+1

where is the eigenvalue relative to the frequency shift, and i and i + 1 denote successive iteration numbers. In the usual case where the frequency shift is zero, the test for convergence becomes approximately the same as: Ti+1 Ti tol Ti+1
or

i+1
f

i+1

tol

provided that the difference between the two iterations is small. Note that the error in the eigenvectors will generally be larger than the error in the eigenvalues. The relative error in the global force balance for a given Mode gives a measure of the error in the eigenvector. This error can usually be reduced by using a smaller value of tol, at the expense of more computation time. See Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

Ritz-vector Analysis
Research has indicated that the natural free-vibration mode shapes are not the best basis for a mode-superposition analysis of structures subjected to dynamic loads. It has been demonstrated (Wilson, Yuan, and Dickens, 1982) that dynamic analyses based on a special set of load-dependent Ritz vectors yield more accurate results than the use of the same number of natural mode shapes. The reason the Ritz vectors yield excellent results is that they are generated by taking into account the spatial distribution of the dynamic loading, whereas the direct use of the natural mode shapes neglects this very important information.

Ritz-vector Analysis
317

299

SAP2000 Analysis Reference The spatial distribution of the dynamic load vector serves as a starting load vector to initiate the procedure. The first Ritz vector is the static displacement vector corresponding to the starting load vector. The remaining vectors are generated from a recurrence relationship in which the mass matrix is multiplied by the previously obtained Ritz vector and used as the load vector for the next static solution. Each static solution is called a generation cycle. When the dynamic load is made up of several independent spatial distributions, each of these may serve as a starting load vector to generate a set of Ritz vectors. Each generation cycle creates as many Ritz vectors as there are starting load vectors. If a generated Ritz vector is redundant or does not excite any mass degrees of freedom, it is discarded and the corresponding starting load vector is removed from all subsequent generation cycles. Standard eigensolution techniques are used to orthogonalize the set of generated Ritz vectors, resulting in a final set of Ritz-vector Modes. Each Ritz-vector Mode consists of a mode shape and frequency. The full set of Ritz-vector Modes can be used as a basis to represent the dynamic displacement of the structure. Once the stiffness matrix is triangularized it is only necessary to statically solve for one load vector for each Ritz vector required. This results in an extremely efficient algorithm. Also, the method automatically includes the advantages of the proven numerical techniques of static condensation, Guyan reduction, and static correction due to higher-mode truncation. The algorithm is detailed in Wilson (1985). When a sufficient number of Ritz-vector Modes have been found, some of them may closely approximate natural mode shapes and frequencies. In general, however, Ritz-vector Modes do not represent the intrinsic characteristics of the structure in the same way the natural Modes do. The Ritz-vector modes are biased by the starting load vectors. You may specify the number of Modes to be found, the starting load vectors to be used, and the number of generation cycles to be performed for each starting load vector. These parameters are described in the following subtopics.

Number of Modes
You may specify the total number of Modes, n, to be found. The total number of Modes actually found, n, is limited by: The number of Modes requested, n

300

Ritz-vector Analysis
318

Chapter XVI Static and Dynamic Analysis The number of mass degrees of freedom present in the model The number of natural free-vibration modes that are excited by the starting load vectors (some additional natural modes may creep in due to numerical noise) A mass degree of freedom is any active degree of freedom that possesses translational mass or rotational mass moment of inertia. The mass may have been assigned directly to the joint or may come from connected elements. Only the Modes that are actually found will be available for any subsequent response-spectrum or time-history analysis processing. See Topic Degrees of Freedom (page 83) in Chapter Joints and Degrees of Freedom.

Starting Load Vectors


You may specify any number of starting load vectors. Each starting load vector may be one of the following: An Acceleration Load in the global X, Y, or Z direction A Load Case A built-in nonlinear deformation load, as described below For response-spectrum analysis, only the Acceleration Loads are needed. For time-history analysis, one starting load vector is needed for each Load Case or Acceleration Load that is used in any History. If nonlinear time-history analysis is to be performed, an additional starting load vector is needed for each independent nonlinear deformation. You may specify that the program use the built-in nonlinear deformation loads, or you may define your own Load Cases for this purpose. See Topic Nonlinear Deformation Loads (page 271) in Chapter The Nllink Element for more information. If you define your own starting load vectors, do the following for each nonlinear deformation: Explicitly define a Load Case that consists of a set of self-equilibrating forces that activates the desired nonlinear deformation Specify that Load Case as a starting load vector The number of such Load Cases required is equal to the number of independent nonlinear deformations in the model.

Ritz-vector Analysis
319

301

SAP2000 Analysis Reference If several Nllink elements act together, you may be able to use fewer starting load vectors. For example, suppose the horizontal motion of several base isolators are coupled with a diaphragm. Only three starting load vectors acting on the diaphragm are required: two perpendicular horizontal loads and one moment about the vertical axis. Independent Load Cases may still be required to represent any vertical motions or rotations about the horizontal axes for these isolators. It is strongly recommended that mass (or mass moment of inertia) be present at every degree of freedom that is loaded by a starting load vector. This is automatic for Acceleration Loads, since the load is caused by mass. If a Load Case or nonlinear deformation load acts on a non-mass degree of freedom, the program issues a warning. Such starting load vectors may generate inaccurate Ritz vectors, or even no Ritz vectors at all. Generally, the more starting load vectors used, the more Ritz vectors must be requested to cover the same frequency range. Thus including unnecessary starting load vectors is not recommended. In each generation cycle, Ritz vectors are found in the order in which the starting load vectors are specified. In the last generation cycle, only as many Ritz vectors will be found as required to reach the total number of Modes, n. For this reason, the most important starting load vectors should be specified first, especially if the number of starting load vectors is not much smaller than the total number of Modes. If no starting load vectors are explicitly specified, the three Acceleration Loads are used by default. This is adequate for response-spectrum analysis and linear seismic time-history analysis with no other loads applied. For more information: See Topic Nonlinear Time-History Analysis (page 328) in this chapter. See Chapter Load Cases (page 279).

Number of Generation Cycles


You may specify the maximum number of generation cycles, ncyc, to be performed for each starting load vector. This enables you to obtain more Ritz vectors for some starting load vectors than others. By default, the number of generation cycles performed for each starting load vector is unlimited, i.e., until the total number, n, of requested Ritz vectors have been found. As an example, suppose that two linear time-history analyses are to be performed:

302

Ritz-vector Analysis
320

Chapter XVI Static and Dynamic Analysis (1) Gravity load is applied quasi-statically to the structure using Load Cases DL and LL (2) Seismic load is applied in all three global directions starting from the conditions at the end of the first History The starting load vectors required are the three Acceleration Loads and Load Cases DL and LL. The first generation cycle creates the static solution for each starting load vector. This is all that is required for Load Cases DL and LL in the first History, hence for these starting load vectors ncyc = 1 should be specified. Additional Modes may be required to represent the dynamic response to the seismic loading, hence an unlimited number of cycles should be specified for these starting load vectors. If 12 Modes are requested (n = 12), there will be one each for DL and LL, three each for two of the Acceleration Loads, and four for the Acceleration Load that was specified first as a starting load vector. Starting load vectors corresponding to nonlinear deformation loads may often need only a limited number of generation cycles. Many of these loads affect only a small local region and excite only high-frequency natural modes that may respond quasistatically to typical seismic excitation. If this is the case, you may be able to specify ncyc = 1 or 2 for these starting load vectors. More cycles may be required if you are particularly interested in the dynamic behavior in the local region. You must use your own engineering judgment to determine the number of Ritz vectors to be generated for each starting load vector. No simple rule can apply to all cases.

Modal Analysis Output


Various properties of the Vibration Modes are printed in the output file. This information is the same regardless of whether you use eigenvector or Ritz-vector analysis, and is described in the following subtopics. Typical modal-analysis output is shown in Figure 84 (page 305).

Periods and Frequencies


The following time-properties are printed for each Mode: Period, T, in units of time Cyclic frequency, f, in units of cycles per time; this is the inverse of T

Modal Analysis Output


321

303

SAP2000 Analysis Reference Circular frequency, , in units of radians per time; = 2 f Eigenvalue, 2, in units of radians-per-time squared These can be found in the output file under the heading:
MODAL PERIODS AND FREQUENCIES

Participation Factors
The modal participation factors are the dot products of the three Acceleration Loads with the modes shapes. The participation factors for Mode n corresponding to Acceleration Loads in the global X, Y, and Z directions are given by:
f

xn yn zn

= nT mx = nT m y = nT m z

where n is the mode shape and mx, my, and, mz are the unit Acceleration Loads. These factors are the generalized loads acting on the Mode due to each of the Acceleration Loads. They are referred to the global coordinate system and can be found in the output file under the heading:
MODAL PARTICIPATION FACTORS

These values are called factors because they are related to the mode shape and to a unit acceleration. The modes shapes are each normalized, or scaled, with respect to the mass matrix such that: n T M n =1 The actual magnitudes and signs of the participation factors are not important. What is important is the relative values of the three factors for a given Mode.

Participating Mass Ratios


The participating mass ratio for a Mode provides a measure of how important the Mode is for computing the response to the Acceleration Loads in each of the three global directions. Thus it is useful for determining the accuracy of responsespectrum analyses and seismic time-history analyses. The participating mass ratio

304

Modal Analysis Output


322

Chapter XVI Static and Dynamic Analysis


M O D A L MODE P E R I O D S PERIOD (TIME) 2.425536 2.407723 2.134032 0.233621 0.229738 0.168388 0.131499 0.115861 A N D F R E Q U E N C I E S FREQUENCY (RAD/TIME) 2.590431 2.609596 2.944279 26.894795 27.349324 37.313777 47.781207 54.230275 EIGENVALUE (RAD/TIME)**2 6.710334 6.809991 8.668776 723.329999 747.985515 1392.318 2283.044 2940.923

FREQUENCY (CYC/TIME) 0.412280 0.415330 0.468596 4.280440 4.352780 5.938672 7.604615 8.631016

1 2 3 4 5 6 7 8 M O D A L

P A R T I C I P A T I O N

F A C T O R S

FOR UNIT ACCELERATION LOADS IN GLOBAL COORDINATES MODE 1 2 3 4 5 6 7 8 M O D A L MODE PERIOD 2.425536 2.407723 2.134032 0.233621 0.229738 0.168388 0.131499 0.115861 UX 1.391631 -1.781824 -0.312429 0.027199 8.72E-05 -6.46E-05 -0.000283 0.000830 UY 1.732188 1.426170 -0.418085 -0.000259 0.025523 -0.000624 0.002020 0.000853 UZ -7.71E-05 -0.000268 0.000114 -0.017982 0.056865 2.172956 0.072018 0.373158 R A T I O S CUMULATIVE SUM (PERCENT) UX UY UZ 37.1724 57.5919 98.1122 96.6323 99.9858 99.9874 100.0000 99.9874 100.0000 99.9999 100.0000 99.9999 100.0000 100.0000 100.0000 100.0000 R A T I O S 0.0000 0.0000 0.0000 0.0062 0.0683 90.6986 90.7982 93.4709

P A R T I C I P A T I N G PERIOD

M A S S

INDIVIDUAL MODE (PERCENT) UX UY UZ 37.1724 60.9398 1.8736 0.0142 0.0000 0.0000 0.0000 0.0000 57.5919 39.0404 3.3551 0.0000 0.0125 0.0000 0.0001 0.0000 0.0000 0.0000 0.0000 0.0062 0.0621 90.6304 0.0996 2.6727

1 2 3 4 5 6 7 8 M O D A L

2.425536 2.407723 2.134032 0.233621 0.229738 0.168388 0.131499 0.115861 L O A D

P A R T I C I P A T I O N STATIC (PERCENT) DYNAMIC (PERCENT)

LOAD, ACC, OR NLLINK/DEF (TYPE) (NAME) LOAD ACC ACC ACC NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF NLLINK/DEF GRAVITY UX UY UZ 1/U1 1/U2 1/U3 2/U1 2/U2 2/U3 3/U1 3/U2 3/U3

100.0000 -> 93.4709<- EXCLUDES LOAD ON NON-MASS DOF 100.0000 100.0000 100.0000 100.0000 99.0042 93.4709 100.0000 31.6238 100.0000 67.2915 100.0000 63.9055 100.0000 37.9576 100.0000 67.2915 100.0000 89.9700 5.3432 0.8109 100.0000 67.2915 100.0000 57.5064

Figure 84 Modal Analysis Output in the Output File Modal Analysis Output
323

305

SAP2000 Analysis Reference provides no information about the accuracy of time-history analyses subjected to other loads. The participating mass ratios for Mode n corresponding to Acceleration Loads in the global X, Y, and Z directions are given by:
rxn

xn ) x

r yn

yn ) y

r zn

zn ) z

where fxn, fyn, and fzn are the participation factors defined in the previous subtopic; and Mx, My, and Mz are the total unrestrained masses acting in the X, Y, and Z directions. The participating mass ratios are expressed as percentages and printed in the output file under the heading:
MODAL PARTICIPATING MASS RATIOS

The cumulative sums of the participating mass ratios for all Modes up to Mode n are printed with the individual values for Mode n. This provides a simple measure of how many Modes are required to achieve a given level of accuracy for groundacceleration loading. If all eigen Modes of the structure are present, the participating mass ratio for each of the three Acceleration Loads should generally be 100%. However, this may not be the case in the presence of Asolid elements or certain types of Constraints where symmetry conditions prevent some of the mass from responding to translational accelerations.

Static and Dynamic Load Participation Ratios


The static and dynamic load participation ratios provide a measure of how adequate the calculated modes are for representing the response to time-history analyses. These two measures are printed in the output file for each of the following spatial load vectors: The three unit Acceleration Loads All static Load Cases

306

Modal Analysis Output


324

Chapter XVI Static and Dynamic Analysis All nonlinear deformation loads The first two represent spatial loads that you can explicitly specify in a time-history analysis, whereas the last represents loads that can act implicitly in a nonlinear time-history analysis. The load participation ratios, expressed as percentages, are printed in the output file under the heading:
MODAL LOAD PARTICIPATION RATIOS

For more information: See Topic Nonlinear Deformation Loads (page 271) in Chapter The Nllink Element. See Chapter Load Cases (page 279). See Topic Acceleration Loads (page 295) in this chapter. See Topic Time-History Analysis (page 319) in this chapter. See Topic Nonlinear Time-History Analysis (page 328) in this chapter. Static Load Participation Ratio The static load participation ratio measures how well the calculated modes can represent the response to a given static load. This measure was first presented by Wilson (1997). For a given spatial load vector p, the participation factor for Mode n is given by
f

= nT p

where n is the mode shape (vector) of Mode n. This factor is the generalized load acting on the Mode due to load p. Note that f n is just the usual participation factor when p is one of the unit acceleration loads. The static participation ratio for this mode is given by fn = n uT p
2

rn

where u is the static solution given by Ku = p. This ratio gives the fraction of the total strain energy in the exact static solution that is contained in Mode n. Note that the denominator can also be represented as u T Ku. Modal Analysis Output
325

307

SAP2000 Analysis Reference Finally, the cumulative sum of the static participation ratios for all the calculated modes is printed in the output file: nT p n n=1 = uT p
N
2

S = rn n=1

where N is the number of modes found. This value gives the fraction of the total strain energy in the exact static solution that is captured by the N modes. When solving for static solutions using quasi-static time-history analysis, the value of R S should be close to 100% for any applied static Loads, and also for all nonlinear deformation loads if the analysis is nonlinear. Note that when Ritz-vectors are used, the value of R S will always be 100% for all starting load vectors. This may not be true when eigenvectors are used. In fact, even using all possible eigenvectors will not give 100% static participation if load p acts on any massless degrees-of-freedom. Dynamic Load Participation Ratio The dynamic load participation ratio measures how well the calculated modes can represent the response to a given dynamic load. This measure was developed for SAP2000, and it is an extension of the concept of participating mass ratios. It is assumed that the load acts only on degrees of freedom with mass. Any portion of load vector p that acts on massless degrees of freedom cannot be represented by this measure and is ignored in the following discussion. For a given spatial load vector p, the participation factor for Mode n is given by
f

= nT p

where n is the mode shape for Mode n. Note that f n is just the usual participation factor when p is one of the unit acceleration loads. The dynamic participation ratio for this mode is given by
rn

( n)
f

aT p

308

Modal Analysis Output


326

Chapter XVI Static and Dynamic Analysis where a is the acceleration given by Ma = p. The acceleration a is easy to calculate since M is diagonal. The values of a and p are taken to be zero at all massless degrees of freedom. Note that the denominator can also be represented as a T Ma . Finally, the cumulative sum of the dynamic participation ratios for all the calculated modes is printed in the output file:
N

D = rn = n=1 n=1

( n
N

aT p

where N is the number of modes found. When p is one of the unit acceleration loads, D is the usual mass participation ratio, and R D is the usual cumulative mass participation ratio.

When R D is 100%, the calculated modes should be capable of exactly representing the solution to any time-varying application of spatial load p. If R D is less than 100%, the accuracy of the solution will depend upon the frequency content of the time-function multiplying load p. Normally it is the high frequency response that is not captured when R D is less than 100%. The dynamic load participation ratio only measures how the modes capture the spatial characterisitcs of p, not its temporal characteristics. For this reason, R D serves only as a qualitative guide as to whether enough modes have been computed. You must still examine the response to each different dynamic loading with varying number of modes to see if enough modes have been used.

Functions
A Function is a series of digitized abscissa-ordinate pairs that may represent: Pseudo-spectral acceleration vs. period for response-spectrum analysis, or Load vs. time for time-history analysis You may define any number of Functions, assigning each one a unique label. You may scale the abscissa and/or ordinate values whenever the Function is used. The abscissa of a Function is always time or period (which has time units). The abscissa-ordinate pairs must be specified in order of increasing time value.

Functions
327

309

SAP2000 Analysis Reference If the increment between time values (abscissas) is constant and the Function starts at time zero, you need only specify the time increment, dt, and the successive function values (ordinates) starting at time zero. The function values are specified as: f0, f1, f2, ..., fn corresponding to times: 0, dt, 2 dt, ..., n dt where n + 1 is the number of values given. If the time increment is not constant or the Function does not start at time zero, you must specify the pairs of time and function value as: t0, f0, t1, f1, t2, f2, ..., tn, fn where n + 1 is the number of pairs of values given.

Response-Spectrum Analysis
The dynamic equilibrium equations associated with the response of a structure to ground motion are given by: & && && && && K u( t ) + C u( t ) + M u( t ) = m x u gx ( t ) + m y u gy ( t ) + m z u gz ( t ) where K is the stiffness matrix; C is the proportional damping matrix; M is the di& && agonal mass matrix; u, u, and u are the relative displacements, velocities, and accelerations with respect to the ground; mx, my, and mz are the unit Acceleration Loads; && && && and u gx , u gy , and u gz are the components of uniform ground acceleration. Response-spectrum analysis seeks the likely maximum response to these equations rather than the full time history. The earthquake ground acceleration in each direction is given as a digitized response-spectrum curve of pseudo-spectral acceleration response versus period of the structure. Even though accelerations may be specified in three directions, only a single, positive result is produced for each response quantity. The response quantities include displacements, forces, and stresses. Each computed result represents a statistical measure of the likely maximum magnitude for that response quantity. The actual response can be expected to vary within a range from this positive value to its negative.

310

Response-Spectrum Analysis
328

Chapter XVI Static and Dynamic Analysis No correspondence between two different response quantities is available. No information is available as to when this extreme value occurs during the seismic loading, or as to what the values of other response quantities are at that time. Response-spectrum analysis is performed using mode superposition (Wilson and Button, 1982). Modes may have been computed using eigenvector analysis or Ritz-vector analysis. Ritz vectors are recommended since they give more accurate results for the same number of Modes. Any number of response-spectrum analyses can be performed in a single execution of the program. Each analysis case is called a Spec, to which you assign a unique label. Each Spec can differ in the acceleration spectra applied and in the way that results are combined. Results from each Spec can be output directly or used in Combos. The following subtopics describe in more detail the parameters that you use to define each Spec.

Local Coordinate System


Each Spec has its own response-spectrum local coordinate system used to define the directions of ground acceleration loading. The axes of this local system are denoted 1, 2, and 3. By default these correspond to the global X, Y, and Z directions, respectively. You may change the orientation of the local coordinate system by specifying: A fixed coordinate system csys (the default is zero, indicating the global coordinate system) A coordinate angle, ang (the default is zero) The local 3 axis is always the same as the Z axis of coordinate system csys. The local 1 and 2 axes coincide with the X and Y axes of csys if angle ang is zero. Otherwise, ang is the angle from the X axis to the local 1 axis, measured counterclockwise when the +Z axis is pointing toward you. This is illustrated in Figure 85 (page 312).

Response-Spectrum Curve
The response-spectrum curve for a given direction is defined by digitized points of pseudo-spectral acceleration response versus period of the structure. The shape of

Response-Spectrum Analysis
329

311

SAP2000 Analysis Reference

Figure 85 Definition of Response Spectrum Local Coordinate System

the curve is given by specifying the name of a Function. All values for the abscissa and ordinate of this Function must be zero or positive. If no Function is specified, a constant function of unit acceleration value for all periods is assumed. You may specify a scale factor sf to multiply the ordinate (pseudo spectral acceleration response) of the function. This is often needed to convert values given in terms of the acceleration due to gravity to units consistent with the rest of the model. See Figure 86 (page 313). The response-spectrum curve chosen should reflect the damping that is present in the structure being modeled. Note that the damping is inherent in the responsespectrum curve itself. It is not affected by the damping ratio, damp, used for the CQC or GMC method of modal combination, although normally these two damping values should be the same. If the response-spectrum curve is not defined over a period range large enough to cover the Vibration Modes of the structure, the curve is extended to larger and smaller periods using a constant acceleration equal to the value at the nearest defined period. See Topic Functions (page 309) in this chapter for more information.

312

Response-Spectrum Analysis
330

Chapter XVI Static and Dynamic Analysis

Figure 86 Digitized Response-Spectrum Curve

Modal Combination
For a given direction of acceleration, the maximum displacements, forces, and stresses are computed throughout the structure for each of the Vibration Modes. These modal values for a given response quantity are combined to produce a single, positive result for the given direction of acceleration using one of the following methods. Use the parameter modc to specify which method to use. CQC Method Specify modc = CQC to combine the modal results by the Complete Quadratic Combination technique described by Wilson, Der Kiureghian, and Bayo (1981). This is the default method of modal combination. The CQC method takes into account the statistical coupling between closelyspaced Modes caused by modal damping. Increasing the modal damping increases the coupling between closely-spaced modes. If the damping is zero for all Modes, this method degenerates to the SRSS method. Modal damping comes from two sources:

Response-Spectrum Analysis
331

313

SAP2000 Analysis Reference You may specify a CQC damping ratio, damp, measured as a fraction of critical damping: 0 damp < 1. This damping equally affects all Modes, and should reflect the damping that is present in the structure being modeled. Note that the value of damp does not affect the response-spectrum curve, which is developed independently for an assumed value of structural damping. Normally these two damping values should be the same. You may specify non-zero effective-damping coefficients for any Nllink elements in the model. These effective damping values are converted to modal damping ratios by assuming proportional damping, i.e., by ignoring any modal cross-coupling terms. These effective modal-damping values will generally be different for each Mode, depending upon how much deformation each Mode causes in the Nllink elements. Only effective damping, not actual damping, in the Nllink elements affects the CQC results. The total damping ratio for each Mode is the sum of these two sources. The program will automatically make sure that the total is less than one. GMC Method Specify modc = GMC to combine the modal results by the General Modal Combination technique. This method is the same as the complete modal combination procedure described by Equation 3.31 in Gupta (1990). The GMC method takes into account the statistical coupling between closely-spaced Modes similarly to the CQC method, but also includes the correlation between modes with rigid-response content. Increasing the modal damping increases the coupling between closely-spaced modes. As with the CQC method, modal damping comes from the two sources: You may specify a GMC damping ratio, damp, such that damping equally affects all Modes.
0

damp < 1. This

Additional modal damping is computed from the effective-damping coefficients for any Nllink elements in the model See the CQC method above for more information on modal damping. In addition, the GMC method requires you to specify two frequencies, f1 and f2, which define the rigid-response content of the ground motion. These must satisfy 0 < f1 < f2. The rigid-response parts of all modes are assumed to be perfectly correlated.

314

Response-Spectrum Analysis
332

Chapter XVI Static and Dynamic Analysis The GMC method assumes no rigid response below frequency f1, full rigid response above frequency f2, and an interpolated amount of rigid response for frequencies between f1 and f2. Frequencies f1 and f2 are properties of the seismic input, not of the structure. Gupta defines f1 as: f1 =
S

Amax
S Vmax

where S Amax is the maximum spectral acceleration and S Vmax is the maximum spectral velocity for the ground motion considered. The default value for f1 is unity. Gupta defines f2 as: f2 = f1 +
3 1 2 3
f

where f r is the rigid frequency of the seismic input, i.e., that frequency above which the spectral acceleration is essentially constant and equal to the value at zero period (infinite frequency). Others have defined f2 as: f2 =
f

The default value for f2 is zero, indicating infinite frequency. For the default value of f2, the GMC method gives results similar to the CQC method. SRSS Method Specify modc = SRSS to combine the modal results by taking the square root of the sum of their squares. This method does not take into account any coupling of Modes as do the CQC and GMC methods. Absolute Sum Method Specify modc = ABS to combine the modal results by taking the sum of their absolute values. This method is usually over-conservative.

Directional Combination
For each displacement, force, or stress quantity in the structure, modal combination produces a single, positive result for each direction of acceleration. These directional values for a given response quantity are combined to produce a single, posi-

Response-Spectrum Analysis
333

315

SAP2000 Analysis Reference tive result. Use the directional combination scale factor, dirf, to specify which method to use. SRSS Method Specify dirf = 0 to combine the directional results by taking the square root of the sum of their squares. This method is invariant with respect to coordinate system, i.e., the results do not depend upon your choice of coordinate system when the given response-spectrum curves are the same. This is the recommended method for directional combination, and is the default. Absolute Sum Method Specify dirf = 1 to combine the directional results by taking the sum of their absolute values. This method is usually over-conservative. Scaled Absolute Sum Method Specify 0 < dirf < 1 to combine the directional results by the scaled absolute sum method. Here, the directional results are combined by taking the maximum, over all directions, of the sum of the absolute values of the response in one direction plus dirf times the response in the other directions. For example, if dirf = 0.3, the spectral response, R, for a given displacement, force, or stress would be:
R

= max ( R1 , R 2 , R 3 )

where:
R1 R2 R3

= R1 + 03 ( R 2 + R 3 ) . = R 2 + 03 ( R1 + R 3 ) . = R 3 + 03 ( R1 + R 2 ) .

and R1 , R 2 , and R 3 are the modal-combination values for each direction. The results obtained by this method will vary depending upon the coordinate system you choose. Results obtained using dirf = 0.3 are comparable to the SRSS method (for equal input spectra in each direction), but may be as much as 8% unconservative or 4% over-conservative, depending upon the coordinate system. Larger values of dirf tend to produce more conservative results.

316

Response-Spectrum Analysis
334

Chapter XVI Static and Dynamic Analysis

Response-Spectrum Analysis Output


Certain information is printed in the output file for each response-spectrum analysis. This information is described in the following subtopics. Typical response-spectrum analysis output is shown in Figure 87 (page 318).

Damping and Accelerations


The modal damping and the ground accelerations acting in each direction are printed for every Mode under the heading:
RESPONSE SPECTRUM ACCELERATIONS

The damping value printed for each Mode is the sum of the specified CQC or GMC damping ratio, damp, plus the modal damping contributed by effective damping in the Nllink elements, if any. The accelerations printed for each Mode are the actual values as interpolated at the modal period from the response-spectrum curves after scaling by the specified values of sf and tf. The accelerations are always referred to the local axes of the response-spectrum analysis. They are identified in the output as U1, U2, and U3.

Modal Amplitudes
The response-spectrum modal amplitudes give the multipliers of the mode shapes that contribute to the displaced shape of the structure for each direction of Acceleration. For a given Mode and a given direction of acceleration, this is the product of the modal participation factor and the response-spectrum acceleration, divided by 2 the eigenvalue, , of the Mode. These values are printed in the output file under the heading:
RESPONSE SPECTRUM MODAL AMPLITUDES

The acceleration directions are always referred to the local axes of the responsespectrum analysis. They are identified in the output as U1, U2, and U3. For more information: See the previous Subtopic Damping and Acceleration for the definition of the response-spectrum accelerations. See Topic Modal Analysis Output (page 303) in this chapter for the definition of the modal participation factors and the eigenvalues. Response-Spectrum Analysis Output
335

317

SAP2000 Analysis Reference


R E S P O N S E S P E C T R U M A C C E L E R A T I O N S

IN RESPONSE-SPECTRUM LOCAL COORDINATES SPEC MODE 1 2 3 1 -----------------PERIOD 0.414601 0.375298 0.243572 DAMP-RATIO 0.050000 0.050000 0.050000 U1 29.443875 28.919579 26.554969 M O D A L U2 29.443875 28.919579 26.554969 U3 .000000 .000000 .000000

R E S P O N S E

S P E C T R U M

A M P L I T U D E S

IN RESPONSE-SPECTRUM LOCAL COORDINATES SPEC MODE 1 2 3 1 -----------------PERIOD 0.414601 0.375298 0.243572 U1 -0.168018 -0.231232 0.010526 U2 -0.286686 0.135573 -0.006012 M O D A L U3 .000000 .000000 .000000 C O R R E L A T I O N S

R E S P O N S E

S P E C T R U M

PARTIAL MATRIX SHOWING CORRELATION FACTORS BETWEEN NEARBY MODES SPEC MODE I 1 2 3 1 -----------------PERIOD I I+1 I+2 0.414601 1.000 0.501 0.032 0.375298 1.000 0.049 0.243572 1.000 S P E C T R U M I+3 I+4 I+5 I+6 I+7 I+8 I+9

R E S P O N S E

B A S E

R E A C T I O N S

IN RESPONSE-SPECTRUM LOCAL COORDINATES SPEC 1 ------------------

FOR EACH MODE, DUE TO SPECTRAL ACCELERATION IN DIRECTION U1: MODE 1 2 3 F1 50.572403 145.250875 1.847642 F2 86.290877 -85.161739 -1.055259 F3 .000000 .000000 .000000 M1 -1830.315 1793.046 20.956121 M2 1072.735 3058.354 36.678180 M3 2492.263 -2132.818 -26.348178

FOR EACH MODE, DUE TO SPECTRAL ACCELERATION IN DIRECTION U2: MODE 1 2 3 F1 86.290877 -85.161739 -1.055259 F2 147.236736 49.931002 0.602698 F3 .000000 .000000 .000000 M1 -3123.037 -1051.277 -11.968838 M2 1830.390 -1793.137 -20.948303 M3 4252.508 1250.488 15.048447

COMBINED FOR ALL MODES AND ALL DIRECTIONS OF SPECTRAL ACCELERATION: F1 195.935537 F2 197.196505 F3 .000000 M1 4174.970 M2 4133.414 M3 5515.265

SPEC

Figure 87 Response-Spectrum Analysis Output in the Output File

318

Response-Spectrum Analysis Output


336

Chapter XVI Static and Dynamic Analysis

Modal Correlation Factors


When the type of modal combination is either CQC or GMC, a partial modal correlation matrix is printed out. This matrix shows the coupling assumed between closely-spaced modes. The correlation factors are always between zero and one. The correlation factors coupling each mode with the next nine higher modes are printed in the output file under the heading:
RESPONSE SPECTRUM MODAL CORRELATIONS

The correlation matrix is symmetric.

Base Reactions
The base reactions are the total forces and moments about the global origin required of the supports (Restraints and Springs) to resist the inertia forces due to responsespectrum loading. These are printed in the output file under the heading:
RESPONSE SPECTRUM BASE REACTIONS

These are printed separately for each individual Mode and each direction of loading without any combination. The total response-spectrum reactions are then printed after performing modal combination and directional combination. The reaction forces and moments are always referred to the local axes of the response-spectrum analysis. They are identified in the output as F1, F2, F3, M1, M2, and M3.

Time-History Analysis
Time-history analysis is used to determine the dynamic response of a structure to arbitrary loading. The dynamic equilibrium equations to be solved are given by: & && K u( t ) + C u( t ) + M u( t ) = r ( t ) where K is the stiffness matrix; C is the proportional damping matrix; M is the di&& & agonal mass matrix; u, u, and u are the relative displacements, velocities, and accelerations with respect to the ground; and r is the applied load. Any number of time-history analyses can be performed in a single execution of the program. Each analysis case is called a History, to which you assign a unique label. Each History can differ in the load applied and in the type of analysis to be performed. Time-History Analysis
337

319

SAP2000 Analysis Reference Three types of time-history analysis are available: Linear transient: The structure starts with zero initial conditions or with the conditions at the end of a previous linear transient History that you specify. All elements are assumed to behave linearly for the duration of the analysis. Periodic: The initial conditions are adjusted to be equal to the those at the end of the period of analysis. All elements are assumed to behave linearly for the duration of the analysis. Nonlinear transient: The structure starts with zero initial conditions or with the conditions at the end of a previous nonlinear transient History that you specify. The Nllink elements may exhibit nonlinear behavior during the analysis. All other elements behave linearly. P-Delta analysis, if requested, is performed before any time-history analyses. This topic describes features that are common to all three types of analysis. Additional features applying only to nonlinear transient analysis are described in the Topic Nonlinear Time-History Analysis (page 328) in this chapter.

Loading
The load, r(t), applied in a given History may be an arbitrary function of space and time. It can be written as a finite sum of spatial load vectors, p i , multiplied by time functions, f i ( t ), as: r ( t ) = f i ( t ) pi
i

(Eqn. 1)

The program uses Load Cases and/or Acceleration Loads to represent the spatial load vectors. The time functions can be arbitrary functions of time or periodic functions such as those produced by wind or sea wave loading. If Acceleration Loads are used, the displacements, velocities, and accelerations are all measured relative to the ground. The time functions associated with the Acceleration Loads mx, my, and mz are the corresponding components of uniform ground && && && acceleration, u gx , u gy , and u gz . Defining the Spatial Load Vectors To define the spatial load vector, pi, for a single term of the loading sum of Equation 1, you may specify either: The label of a Load Case using the parameter load, or

320

Time-History Analysis
338

Chapter XVI Static and Dynamic Analysis

Figure 88 Definition of History Acceleration Local Coordinate System

An Acceleration Load using the parameters csys, ang, and acc, where: csys is a fixed coordinate system (the default is zero, indicating the global coordinate system) ang is a coordinate angle (the default is zero) acc is the Acceleration Load (U1, U2, or U3) in the acceleration local coordinate system as defined below Each Acceleration Load in the loading sum may have its own acceleration local coordinate system with local axes denoted 1, 2, and 3. The local 3 axis is always the same as the Z axis of coordinate system csys. The local 1 and 2 axes coincide with the X and Y axes of csys if angle ang is zero. Otherwise, ang is the angle from the X axis to the local 1 axis, measured counterclockwise when the +Z axis is pointing toward you. This is illustrated in Figure 88 (page 321). The response-spectrum local axes are always referred to as 1, 2, and 3. The global Acceleration Loads mx, my, and mz are transformed to the local coordinate system for loading. It is generally recommended, but not required, that the same coordinate system be used for all Acceleration Loads applied in a given History. Load Cases and Acceleration Loads may be mixed in the loading sum. For more information: Time-History Analysis
339

321

SAP2000 Analysis Reference See Chapter Load Cases (page 279). See Topic Acceleration Loads (page 295) in this chapter. Defining the Time Functions To define the time function, fi(t), for a single term of the loading sum of Equation 1, you may specify: The label of a Function, using the parameter func, that defines the shape of the time variation (the default is zero, indicating the built-in ramp function defined below) A scale factor, sf, that multiplies the ordinate values of the Function (the default is unity) A time-scale factor, tf, that multiplies the time (abscissa) values of the Function (the default is unity) An arrival time, at, when the Function begins to act on the structure (the default is zero) The time function, fi(t), is related to the specified Function, func(t), by: fi(t) = sf func(t) The analysis time, t, is related to the time scale, t, of the specified Function by: t = at + tf t If the arrival time is positive, the application of Function func is delayed until after the start of the analysis. If the arrival time is negative, that portion of Function func occurring before t = at / tf is ignored. For a Function func defined from initial time t0 to final time tn, the value of the Function for all time t < t0 is taken as zero, and the value of the Function for all time t > tn is held constant at fn, the value at tn. If no Function is specified, or func = 0, the built-in ramp function is used. This function increases linearly from zero at t = 0 to unity at t = 1 and for all time thereafter. When combined with the scaling parameters, this defines a function that increases linearly from zero at t = at to a value of sf at t = at + tf and for all time thereafter, as illustrated in Figure 88 (page 323). This function is most commonly used to gradually apply static loads, but can also be used to build up triangular pulses and more complicated functions.

322

Time-History Analysis
340

Chapter XVI Static and Dynamic Analysis

Figure 89 Built-in Ramp Function before and after Scaling

See Topic Functions (page 309) in this chapter for more information.

Mode Superposition
The standard mode-superposition method of response analysis is used by the program to solve the dynamic equilibrium equations of motion for the complete structure. The Modes used can be the undamped free-vibration Modes (eigenvectors) or the load-dependent Ritz-vector Modes. If all of the spatial load vectors, p i , are used as starting load vectors for Ritz-vector analysis, then the Ritz vectors will always produce more accurate results than if the same number of eigenvectors is used. Since the Ritz-vector algorithm is faster than the eigenvector algorithm, the former is recommended for time-history analyses. It is up to you to determine if the Modes calculated by the program are adequate to represent the time-history response to the applied load. You should check: That enough Modes have been computed That the Modes cover an adequate frequency range That the participating mass ratios are adequate if Acceleration Loads are applied Time-History Analysis
341

323

SAP2000 Analysis Reference That the modes shapes adequately represent all desired deformations That the global force balance for each Mode is acceptable Special considerations must be made for the Modes used when performing nonlinear analysis. See Topic Nonlinear Time-History Analysis (page 328) in this chapter for more information. For more information: See Topic Eigenvector Analysis (page 296) in this chapter. See Topic Ritz-vector Analysis (page 299) in this chapter. See Subtopic Participating Mass Ratios (page 304) in this chapter See Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom.

Modal Damping
The damping in the structure is modeled using modal damping, also known as proportional or classical damping. The damping in each Mode may be specified using the parameter damp, which is measured as a fraction of critical damping. The damping value for each Mode must satisfy:
0

damp < 1

For a nonlinear transient analysis, this is the only source of modal damping. See Topic Nonlinear Time-History Analysis (page 328) for important considerations about using modal damping with nonlinear transient analysis. For linear transient and periodic analyses, additional modal damping may come from any non-zero effective-damping coefficients that may be specified for Nllink elements in the model. These effective damping values are converted to modal damping ratios assuming proportional damping, i.e., by ignoring any modal crosscoupling terms. These effective modal-damping values will generally be different for each Mode, depending upon how much deformation each Mode causes in the Nllink elements. Only effective damping, not the actual damping in the Nllink elements, affects the linear time-history results. The total damping ratio for each Mode is the sum of these two sources. The program automatically makes sure that the total is less than one.

324

Time-History Analysis
342

Chapter XVI Static and Dynamic Analysis

Time Steps
Time-history analysis is performed at discrete time steps. You may specify the number of time steps with parameter nstep and the size of the time steps with parameter dt. The time span over which the analysis is carried out is given by nstepdt. For periodic analysis, the period of the cyclic loading function is assumed to be equal to this time span. Responses are calculated at the end of each dt time increment, resulting in nstep+1 values for each output response quantity. For linear transient and periodic analysis, closed-form integration of the modal equations is used to compute the response, assuming linear variation of the time functions, f i ( t ), between the input data time points. Therefore, numerical instability problems are never encountered, and the time increment may be any sampling value that is deemed fine enough to capture the maximum response values. Onetenth of the time period of the highest mode is usually recommended; however, a larger value may give an equally accurate sampling if the contribution of the higher modes is small. For nonlinear transient analysis, closed-form integration is again used to compute the response, with the forces from the nonlinear Nllink elements applied to the right-hand side of the dynamic modal equilibrium equations. Since these forces depend upon the dynamic response, the solution is iterated at each time step until the nonlinear forces converge. The accuracy of the results may depend upon the size of the time step, improving with decreasing dt. See Topic Nonlinear Time-History Analysis (page 328) in this chapter for more information.

Initial Conditions
The initial conditions describe the state of the structure at the beginning of a History. These include: Displacements and velocities Internal forces and stresses Internal state variables for the Nllink elements, if any Energy values for the structure The accelerations are not considered initial conditions, but are computed from the equilibrium equation. Time-History Analysis
343

325

SAP2000 Analysis Reference For each linear and nonlinear transient analysis to be performed, you may use the parameter prev to specify the initial conditions to be used. It is not applicable to periodic analyses. If prev = 0 (the default), zero initial conditions are assumed for all displacements, velocities, energies, and the internal state variables. The initial acceleration of the structure is computed by equilibrium using the applied loads and/or ground accelerations at time t = 0. Otherwise prev must be the label of a previous History of the same type (linear or nonlinear transient) as the current History. The initial conditions of the current History are taken to be the conditions that existed at the end of History prev. This includes the displacements, velocities, energies, and the internal state variables. In addition, the applied loads and/or ground accelerations that existed at the end of History prev are held constant and are added to the loads and/or ground accelerations specified for the current History. The initial acceleration of the structure is computed from equilibrium using the total applied loads and/or ground accelerations at time t = 0. For example, suppose three linear transient Histories are defined for a single execution of the program by the following data: (1) name=GRAV, nstep=40, dt=1, prev=0, damp=0.99 for all Modes, and loading comes from the following two Load Cases: load=DL, func=0, tf=10 load=LL, func=0, tf=10 (2) name=LPRIET, nstep=2000, dt=0.02, prev=GRAV, damp=0.05 for all Modes, and loading comes from the following two Acceleration Loads: acc=U1, func=LPRIET1, sf=386.4 acc=U2, func=LPRIET2, sf=386.4 (3) name=ELCENT, nstep=2000, dt=0.02, prev=GRAV, damp=0.05 for all Modes, and loading comes from the following two Acceleration Loads: acc=U1, func=ELCENT1, sf=386.4 acc=U2, func=ELCENT2, sf=386.4 The first History, GRAV, starts with zero initial conditions (it must, since there is no previous History). Dead load and live load are applied slowly using a ramp function over a length of 10 seconds and held for an additional 30 seconds. The length of

326

Time-History Analysis
344

Chapter XVI Static and Dynamic Analysis the ramp function was chosen to be much longer than the longest period of the structure. A large damping value is used to prevent oscillation. This is especially important for nonlinear analysis where oscillation may cause unrealistic hysteric behavior. The second and third Histories, LPRIET and ELCENT each start from History GRAV and apply ground accelerations to the structure with 5% damping. Gravity loads are still acting on the structure throughout Histories LPRIET and ELCENT since they were present at the end of History GRAV. This approach is particularly useful for nonlinear analysis where the behavior of certain Nllink elements, especially the Gap, Hook and Isolat2 types, is strongly dependent on the total force or displacement acting on the elements.

Analysis Results
Two types of results are available from a time-history analysis: traces and envelopes. These are described below. Traces Time-history traces are plots of response quantities as functions of time. Using the SAP2000 graphical user interface, you may plot one or more response quantities against time, or any two response quantities against each other. Time-history traces are always available after an analysis. The following type of response quantities may be plotted: Joint displacements, velocities, and accelerations Joint reactions and spring forces Frame, Shell, and Nllink element internal forces Shell, Plane, Asolid, and Solid element stresses Base reactions for ground-acceleration loading Solution energies: input, kinetic, potential, modal damping, nonlinear damping, and nonlinear hysteretic Envelopes Time-history envelopes are the maximum and minimum values attained during a time-history analysis for each response quantity. Envelope values may be printed in the output file, used in Combos, and displayed and printed in the SAP2000 graphical user interface. Time-History Analysis
345

327

SAP2000 Analysis Reference For each time-history case, you may specify whether or not envelope values are to be computed. By default, envelope values are not computed and thus they cannot be printed, displayed, or used in Combos. If you request envelopes for a given History, values are computed for all of the following quantities: Joint displacements Joint reactions and spring forces Frame, Shell, and Nllink element internal forces Shell, Plane, Asolid, and Solid element stresses

Nonlinear Time-History Analysis


The method of nonlinear time-history analysis used in SAP2000 is an extension of the Fast Nonlinear Analysis (FNA) method developed by Wilson (Ibrahimbegovic and Wilson, 1989; Wilson, 1993). The method is extremely efficient and is designed to be used for structural systems which are primarily linear elastic, but which have a limited number of predefined nonlinear elements. In SAP2000, all nonlinearity is restricted to the Nllink elements. A short description of the method follows. The dynamic equilibrium equations of a linear elastic structure with predefined nonlinear elements subjected to an arbitrary load can be written as: & && K L u( t ) + C u( t ) + M u( t ) + r N ( t ) = r ( t ) where K L is the stiffness matrix for the linear elastic elements (all elements except the Nllinks); C is the proportional damping matrix; M is the diagonal mass matrix; r N is the vector of forces from the nonlinear degrees of freedom in the Nllink ele& && ments; u, u, and u are the relative displacements, velocities, and accelerations with respect to the ground; and r is the vector of applied loads. See Topic Time-History Analysis (page 319) in this chapter for the definition of r.

Nllink Effective Stiffness


For the purposes of analysis, a linear effective stiffness is defined for each degree of freedom of the nonlinear elements. The effective stiffness at nonlinear degrees of freedom is arbitrary, but generally varies between zero and the maximum nonlinear stiffness of that degree of freedom. The equilibrium equation can then be rewritten as:

328

Nonlinear Time-History Analysis


346

Chapter XVI Static and Dynamic Analysis & && K u( t ) + C u( t ) + M u( t ) = r ( t ) [ r N ( t ) K N u( t ) ] where K = K L + K N , with K L being the stiffness of all the linear elements and for the linear degrees of freedom of the Nllink elements, and K N being the linear effective-stiffness matrix for all of the nonlinear degrees of freedom. See Chapter The Nllink Element (page 243) for more information.

Mode Superposition
Modal analysis is performed using the full stiffness matrix, K, and the mass matrix, M. It is strongly recommended that the Ritz-vector method be used to perform the modal analysis. Using standard techniques, the equilibrium equation can be written in modal form as: & && 2 a( t) + a( t) + I a( t) = q( t) q N ( t) where 2 is the diagonal matrix of squared structural frequencies given by: 2 = T K is the modal damping matrix which is assumed to be diagonal: = T C I is the identity matrix which satisfies: I = T M q ( t ) is the vector of modal applied loads: q( t) = T r ( t) q N ( t ) is the vector of modal forces from the nonlinear elements: q N ( t ) = T [ r N ( t ) K N u( t ) ] a ( t ) is the vector of modal displacement amplitudes such that: u( t ) = a ( t ) and is the matrix of mode shapes.

Nonlinear Time-History Analysis


347

329

SAP2000 Analysis Reference It should be noted that, unlike linear dynamic analysis, the above modal equations are coupled. In general the nonlinear forces, q N ( t ), will couple the modes since they are functions of the modal displacements, a ( t ). It is important to recognize that the solution to these modal equations is dependent on being able to adequately represent the nonlinear forces by the modal forces, q N ( t ). This is not automatic, but requires the following special considerations: Mass and/or mass moments of inertia should be present at all nonlinear degrees of freedom. The Ritz-vector method should be used to determine the Modes, unless all possible structural Modes are found using eigenvector analysis The Ritz starting load vectors should include a nonlinear deformation load for each independent nonlinear degree of freedom A sufficient number of Ritz-vectors should be sought to capture the deformation in the nonlinear elements completely For more information: See Topic Ritz-vector Analysis (page 299) in this chapter. See Chapter The Nllink Element (page 243).

Modal Damping
It is also important to note that the assumption of proportional modal damping is being made with respect to the total stiffness matrix, K, which includes the effective stiffness from the nonlinear elements. If non-zero modal damping is to be used, then the effective stiffness specified for these elements is important. The effective stiffness should be selected such that the modes for which these damping values are specified are realistic. In general it is recommended that either the initial stiffness of the element be used as the effective stiffness or the secant stiffness obtained from tests at the expected value of the maximum displacement be used. Initially open gap and hook elements and all damper elements should generally be specified with zero effective stiffness.

Iterative Solution
The nonlinear modal equations are solved iteratively in each time step. The program assumes that the right-hand sides of the equations vary linearly during a time step, and uses exact, closed-form integration to solve these equations in each itera-

330

Nonlinear Time-History Analysis


348

Chapter XVI Static and Dynamic Analysis tion. The iterations are carried out until the solution converges. If convergence cannot be achieved, the program divides the time step into smaller substeps and tries again. Several parameters are available for you to control the iteration process. In general, the use of the default values is recommended since this will solve most problems. If convergence cannot be achieved, inaccurate results are obtained, or the solution takes too long, changing these control parameters may help. However, you should first check that reasonable loads and properties have been specified, and that appropriate Modes have been obtained, preferably using the Ritz vector method. The parameters that are available to control iteration and substepping are: The relative force convergence tolerance, ftol The relative energy convergence tolerance, etol The maximum allowed substep size, dtmax The minimum allowed substep size, dtmin The maximum number of force iterations permitted for small substeps, itmax The maximum number of force iterations permitted for large substeps, itmin The convergence factor, cf These parameters are used in the iteration and substepping algorithm as described in the following. Force Convergence Check Each time step of length dt is divided into substeps as needed to achieve convergence. In each substep, the solution is iterated until the change in the right-hand side of the modal equations, expressed as a fraction of the right-hand side, becomes less than the force tolerance, ftol. If this does not occur within the permitted number of iterations, the substep size is halved and the iteration is tried again.
-5 The default value for ftol is 10 . It must satisfy ftol > 0.

Energy Convergence Check If force convergence occurs within the permitted number of iterations, the work done by the nonlinear forces is compared with the work done by all the other force terms in the modal equilibrium equations. If the difference, expressed as a fraction of the total work done, is greater than the energy tolerance, etol, the substep size is halved and the iteration is tried again. Nonlinear Time-History Analysis
349

331

SAP2000 Analysis Reference This energy check essentially measures how close to linear is the variation of the nonlinear force over the time step. It is particularly useful for detecting sudden changes in nonlinear behavior, such as the opening and closing of gaps or the onset of yielding and slipping. Setting etol greater than unity turns off this energy check.
-5 The default value for etol is 10 . It must satisfy etol > 0.

Maximum and Minimum Substep Sizes If the substep meets both the force and energy convergence criteria, the results of the substep are accepted, and the next substep is attempted using twice the previous substep length. The substep size is never increased beyond dtmax. When the substep size is halved because of failure to meet either the force or energy convergence criteria, the resulting substep size will never be set less than dtmin. If the failed substep size is already dtmin, the results for the remaining time steps in the current History are set to zero and a warning message is issued.
-9 The default value for dtmax is dt. The default value for dtmin is dtmax10 . They must satisfy 0 < dtmin dtmax dt.

Maximum Number of Iterations The maximum number of iterations permitted for force iteration varies between itmin and itmax. The actual number permitted for a given substep is chosen automatically by the program to achieve a balance between iteration and substepping. The number of iterations permitted tends to be larger for smaller substeps. The default values for itmin and itmax are 2 and 100, respectively. They must satisfy 2 itmin itmax. Convergence Factor Under-relaxation of the force iteration may be used by setting the convergence factor, cf, to a value less than unity. Smaller values increase the stability of the iteration, but require more iterations to achieve convergence. This is generally only needed when Damper-type elements are present with nonlinear damping exponents. Specifying cf to be greater than unity may reduce the number of iterations required for certain types of problems, but may cause instability in the iteration and is not recommended. The default value for cf is 1. It must satisfy cf > 0.

332

Nonlinear Time-History Analysis


350

Chapter XVI Static and Dynamic Analysis

Static Period
Normally all modes are treated as being dynamic. In each time step, the response of a dynamic mode has two parts: Forced response, which is directly proportional to the modal load Transient response, which is oscillatory, and which depends on the displacements and velocities of the structure at the beginning of the time step You may optionally specify that high-frequency (short period) modes be treated as static, so that they follow the load without any transient response. This is done by specifying a static period, tstat, such that all modes with periods less than tstat are considered to be static modes. The default for tstat is zero, meaning that all modes are considered to be dynamic. Although tstat can be used for any nonlinear time-history analysis, it is of most use for quasistatic analyses. If the default iteration parameters do not work for such an analysis, you may try using the following parameters as a starting point: tstat greater than the longest period of the structure itmax = itmin 1000 dtmax = dtmin = dt ftol 10-6 cf = 0.1 This causes all modes to be treated as static, and uses iteration rather than substepping to find a solution. The choice of parameters to achieve convergence is very problem dependent, and you should experiment to find the best values to use for each different model.

Combos
A Combo is a combination of the results from Loads, Modes, Specs, Histories, Moving Loads, and/or previously-defined Combos. Combo results include all displacements and forces at the joints and internal forces or stresses in the elements. You may specify any number of Combos. To each one of these you assign a unique label.

Combos
351

333

SAP2000 Analysis Reference Each Combo produces a pair of values for each response quantity: a maximum and a minimum. These two values may be equal for certain type of Combos, as discussed below. Each contributing analysis case supplies one or two values to the Combo for each response quantity: Loads, Modes, and additive Combos of Loads and Modes provide a single value. For the purposes of defining the Combos below, this single value can also be considered to be two equal values All other analysis cases (including Combos) provide two values: a maximum and minimum. For some types of Combos, both values are used. For other types of Combos, only the value with the larger absolute value is used. For contributing response-spectrum analyses, the maximum value used is the positive Spec value, and the minimum value is just the negative of the maximum. For contributing time-history analyses, the values used are the maximum and minimum values attained at any time during the analysis. These values are only available for a given History if you specified that envelopes be computed for that History. Otherwise, zero values will be used for that History! See Subtopic Analysis Results (page 327) under Topic Time-History Analysis for more information. For contributing moving-load analyses, the values used are the maximum and minimum values obtained for any vehicle loading of the lanes permitted by the parameters of the analysis. Each contributing analysis case is multiplied by a scale factor, sf, before being included in a Combo. Four types of Combos are available. For each individual response quantity (force, stress, or displacement component) the two Combo values are calculated as follows: Additive type: The Combo maximum is an algebraic linear combination of the maximum values for each of the contributing cases. Similarly, Combo minimum is an algebraic linear combination of the minimum values for each of the contributing cases. Absolute type: The Combo maximum is the sum of the larger absolute values for each of the contributing cases. The Combo minimum is the negative of the Combo maximum.

334

Combos
352

Chapter XVI Static and Dynamic Analysis SRSS type: The Combo maximum is the square root of the sum of the squares of the larger absolute values for each of the contributing cases. The Combo minimum is the negative of the Combo maximum. Envelope type: The Combo maximum is the maximum of all of the maximum values for each of the contributing cases. Similarly, Combo minimum is the minimum of all of the minimum values for each of the contributing cases. Only additive Combos of single-valued analysis cases produce a single-valued result, i.e., the maximum and minimum values are equal. Thus only Combos of Loads, Modes, and other single-valued Combos are single-valued. All other Combos will generally have different maximum and minimum values. For example, suppose that the values, after scaling, for the displacement at a particular joint are 3.5 for Load Case LL and are 2.0 for Spec QUAKE. Suppose that these two cases have been included in an additive-type Combo called COMB1 and an envelope-type Combo called COMB2. The results for the displacement at the joint are computed as follows: COMB1: The maximum is 3.5 + 2.0 = 5.5, and the minimum is 3.5 2.0 = 1.5 COMB2: The maximum is max (3.5, 2.0) = 3.5, and the minimum is min (3.5, 2.0) = 2.0 As another example, suppose that Load Cases GRAV, WINDX and WINDY are gravity load and two perpendicular, transverse wind loads, respectively; and that a response-spectrum analysis named EQ has been performed. The following four Combos could be defined: WIND: An SRSS-type Combo of the two wind loads, WINDX and WINDY. The maximum and minimum results produced for each response quantity are equal and opposite GRAVEQ: An additive-type Combo of the gravity load, GRAV, and the response-spectrum results, EQ. The Combo automatically accounts for the positive and negative senses of the earthquake load GRAVWIN: An additive-type Combo of the gravity load, GRAV, and the wind load given by Combo WIND, which already accounts for the positive and negative senses of the load SEVERE: An envelope-type Combo that produces the worst case of the two additive Combos, GRAVEQ and GRAVWIN Suppose that the values of axial force in a frame element, after scaling, are 10, 5, 3, and 7 for cases GRAV, WINDX, WINDY, and EQ, respectively. The following results for axial force are obtained for the Combos above: Combos
353

335

SAP2000 Analysis Reference WIND:


maximum

+ 3 2 = 58 , .

minimum

= 58 . = 10 58 = 42 . . = min(3,42) = 3 .

GRAVEQ: maximum = 10 + 7 = 17

, minimum

= 10 7 = 3

GRAVWIN: maximum = 10 + 58 = 158 , . . SEVERE: maximum = max(17 ,158) = 17 .

minimum

, minimum

As you can see, using Combos of Combos gives you considerable power and flexibility in how you can combine the results of the various analysis cases. Moving Load Cases should not normally be added together, in order to avoid multiple loading of the lanes. Additive combinations of Moving Loads should only be defined within the Moving Load Case itself. Therefore, it is recommended that only a single Moving Load be included in any additive-, absolute-, or SRSS-type Combo, whether referenced directly as a Moving Load or indirectly through another Combo. Multiple Moving Loads may be included in any envelope-type Combo, since they are not added. When Combos are used for design, they may be treated somewhat differently than has been described here for output purposes. For example, every time step in a History may be considered under certain circumstances. Similarly, corresponding response quantities at the same location in a Moving Load case may be used for design purposes. See the SAP2000 Steel Design Manual and the SAP2000 Concrete Design Manual for more information.

336

Combos
354

C h a p t e r XVII

P-Delta Analysis
The P-Delta analysis option causes the transverse bending stiffness of all Frame elements to be modified to account for the P-Delta effect under large axial force. This affects the results of all static, dynamic, and bridge analyses performed in the same execution of the program. See Chapters Static and Dynamic Analysis (page 291) and Bridge Analysis (page 357) for information on these other types of analysis. Advanced Topics Overview Geometric Nonlinearity The P-Delta Effect Equilibrium Equations P-Delta Axial Forces Iterative Analysis Frame Element Effect upon Other Analyses Buckling Practical Application

337
355

SAP2000 Analysis Reference

Overview
The P-Delta analysis option accounts for the effect of a large axial load upon the transverse bending behavior of Frame elements. Axial compression reduces the flexural stiffness of Frame elements, and axial tension stiffens them. This is a type of geometric nonlinearity known as the P-Delta effect. P-delta analysis does not include large-strain or large-rotation effects. This option is particularly useful for considering the effect of gravity loads upon the lateral stiffness of building structures, as required by certain design codes (ACI 1995; AISC 1994). It can also be used for the analysis of some cable structures, such as suspension bridges, cable-stayed bridges, and guyed towers. Other applications are possible. P-delta analysis, when requested, is performed before all static, dynamic, and bridge analyses in the same execution of the program. The axial forces in the Frame elements that cause the P-delta effect may be: directly specified, and/or computed from a specified set of applied static loads The P-Delta effect essentially modifies the characteristics of the structure, affecting the results of all other analyses performed, including: static Loads, Modes, Specs, bridge influence lines, Moving Loads, Combos, and Histories. The P-Delta analysis option provides considerable power and flexibility for analyzing the P-Delta effect in buildings, bridges and other structures. To make the most effective and efficient use of this capability, you should fully understand the assumptions and restrictions that underlie the implementation of this feature in SAP2000. Some of the more important assumptions include: The P-Delta effect is analyzed only in Frame elements. Other types of elements may still be present in the model. Only the large-stress effect of an axial force upon transverse bending and shear deformation is considered. All deflections, strains and rotations are assumed to be small. The transverse deflected shape of a Frame element is assumed to be cubic in bending and linear in shear between the reduced rigid zone offsets. The P-Delta axial forces are assumed to be constant along the element length. P-Delta analysis under applied loads is iterative in nature, and may considerably increase computation time. Including a P-Delta analysis may make interpretation of

338

Overview
356

Chapter XVII P-Delta Analysis the results more difficult. It is strongly recommended that you perform a preliminary linear analysis to check your model for correctness before using the P-Delta analysis option.

Geometric Nonlinearity
When the load acting on a structure and the resulting deflections are small enough, the load-deflection relationship for the structure is linear. For the most part, SAP2000 analyses assume such linear behavior. This permits the program to form the equilibrium equations using the original (undeformed) geometry of the structure. Strictly speaking, the equilibrium equations should actually refer to the geometry of the structure after deformation. The linear equilibrium equations are independent of the applied load and the resulting deflection. Thus the results of different static and/or dynamic loads can be superposed (scaled and added), resulting in great computational efficiency. If the load on the structure and/or the resulting deflections are large, then the loaddeflection behavior may become nonlinear. Several causes of this nonlinear behavior can be identified: Large-stress effect: when large stresses (or forces and moments) are present within a structure, equilibrium equations written for the original and the deformed geometries may differ significantly, even if the deformations are very small. The P-Delta effect is an example of this large stress effect. Large-displacement effect: when a structure undergoes large deformation (in particular, large strains and rotations), the usual engineering stress and strain measures no longer apply, and the equilibrium equations must be written for the deformed geometry. This is true even if the stresses are small. Material nonlinearity: when a material is strained beyond its proportional limit, the stress-strain relationship is no longer linear. Plastic materials strained beyond the yield point may exhibit history-dependent behavior. Material nonlinearity may affect the load-deflection behavior of a structure even when the equilibrium equations for the original geometry are still valid. The large-stress and large-displacement effects are both termed geometric (or kinematic) nonlinearity, as distinguished from material nonlinearity. Kinematic nonlinearity may also be referred to as second-order geometric effects. Other sources of nonlinearity are also possible, including nonlinear loads, boundary conditions and constraints. Geometric Nonlinearity
357

339

SAP2000 Analysis Reference

Figure 90 Geometry for Cantilever Beam Example

The P-Delta analysis option can be used to analyze the P-Delta effect, one type of large-stress effect. Other types of large-stress effects are not usually important for structures, and are not considered by the program.

The P-Delta Effect


The P-Delta effect refers specifically to the nonlinear geometric effect of a large tensile or compressive direct stress upon transverse bending and shear behavior. A compressive stress tends to make a structural member more flexible in transverse bending and shear, whereas a tensile stress tends to stiffen the member against transverse deformation. The basic concepts behind the P-Delta effect are illustrated in the following example. Consider a cantilever beam subject to an axial load P and a transverse tip load F as shown in Figure 90 (page 340). The internal axial force throughout the member is also equal to P.

340

The P-Delta Effect


358

Chapter XVII P-Delta Analysis If equilibrium is examined in the original configuration (using the undeformed geometry), the moment at the base is M = FL, and decreases linearly to zero at the loaded end. If, instead, equilibrium is considered in the deformed configuration, there is an additional moment caused by the axial force P acting on the transverse tip displacement, . The moment no longer varies linearly along the length; the variation depends instead upon the deflected shape. The moment at the base is now M = FL P. The moment diagrams for various cases are shown in Figure 91 (page 342). Note that only the transverse deflection is considered in the deformed configuration. Any change in moment due to a change in length of the member is neglected here. If the beam is in tension, the moment at the base and throughout the member is reduced, hence the transverse bending deflection, , is also reduced. Thus the member is effectively stiffer against the transverse load F. Conversely, if the beam is in compression, the moment throughout the member, and hence the transverse bending deflection, , are now increased. The member is effectively more flexible against the load F. If the compressive force is large enough, the transverse stiffness goes to zero and hence the deflection tends to infinity; the structure is said to have buckled. The theoretical value of P at which this occurs is called the Euler buckling load for the beam; it is denoted by Pcr and is given by the formula
Pcr

EI
2

4L

where EI is the bending stiffness of the beam section. The exact P-Delta effect of the axial load upon the transverse deflection and stiffness is a rather complicated function of the ratio of the force P to the buckling load Pcr . The true deflected shape of the beam, and hence the effect upon the moment diagram, is described by cubic functions under zero axial load, hyperbolic functions under tension, and trigonometric functions under compression. The P-Delta effect can be present in any other beam configuration, such as simplysupported, fixed-fixed, etc. The P-Delta effect may apply locally to individual members, or globally to the structural system as a whole. The key feature is that a large axial force, acting upon a small transverse deflection, produces a significant moment that affects the behavior of the member or structure.

The P-Delta Effect


359

341

SAP2000 Analysis Reference

Figure 91 Moment Diagrams for Cantilever Beam Examples

If the deflection is small, then the moment produced is proportional to the deflection.

342

The P-Delta Effect


360

Chapter XVII P-Delta Analysis

Equilibrium Equations
The equilibrium equations are manifested in SAP2000 in two ways: The element stiffness matrices, which are assembled (combined) to form the overall structure stiffness matrix. The element stress-displacement relations, which are used to determine the internal element stresses or forces for output. Only Frame element stiffness and stress-transformation matrices are modified to include the P-Delta effect. The P-Delta option may be used with models containing any or all types of elements. For Shell, Plane, Asolid, Solid, and Nllink elements the linear elastic stiffnesses will always be used, and the element stresses will not reflect the P-Delta effect within those elements. Frame elements should be used to model those portions of any structure that may carry the significant P-Delta loads.

P-Delta Axial Forces


SAP2000 modifies the equilibrium equations to include the P-Delta effect due to a single set of axial forces acting in the Frame elements of the model. These modified equations are then used for all subsequent analyses that are applied to the structure. The results for these different analyses may then be superposed (except for nonlinear time-history analysis). These axial forces that create the P-Delta effect may come from two sources: You may directly specify the axial forces in the Frame elements You may specify a combination of applied static loads from which the axial forces are computed by an iterative analysis The P-Delta axial forces from these two sources are added together. The total P-Delta axial force in each Frame element is only used to modify the equilibrium equations. It is printed in the output file for each element. It is not included in the axial force results that are reported for the various analysis cases in the output file or by the SAP2000 graphical interface.

Equilibrium Equations
361

343

SAP2000 Analysis Reference

Directly Specified Axial Forces


Direct specification of the P-delta forces can be used for any structure, but is most appropriate for cable structures where the tensions are large and well-known, and where large deformations of the structure are required to generate these tensions. For other situations, it is generally better to allow the program to compute the axial forces from applied loads. No iterative analysis is required to include the effect of directly specified P-Delta axial forces. You may directly specify the P-Delta axial force acting in any Frame element using the following parameters: The P-Delta axial force, p A fixed coordinate system, csys (the default is zero, indicating the global coordinate system) The projection, px, of the P-Delta axial force upon the X axis of csys The projection, py, of the P-Delta axial force upon the Y axis of csys The projection, pz, of the P-Delta axial force upon the Z axis of csys Normally only one of the parameters p, px, py, or pz should be given for each Frame element. If you do choose to specify more than one value, they are additive:
P0

=p+

px
cx

py
c

pz
cz

where P0 is the P-Delta axial force, and cx, cy, and cz are the cosines of the angles between the local 1 axis of the Frame element and the X, Y, and Z axes of coordinate system csys, respectively. To avoid division by zero, you may not specify the projection upon any axis of csys that is perpendicular to the local 1 axis of the element. The use of the P-delta axial force projections is convenient, for example, when specifying the tension in the main cable of a suspension bridge, since the horizontal component of the tension is usually the same for all elements. It is important when directly specifying P-Delta axial forces that you include all significant forces in the structure. The program does not check for equilibrium of the specified P-Delta axial forces. In a suspension bridge, for example, the cable tension is supported at the anchorages, and it is usually sufficient to consider the PDelta effect only in the main cable (and possibly the towers). On the other hand, the

344

P-Delta Axial Forces


362

Chapter XVII P-Delta Analysis cable tension in a cable-stayed bridge is taken up by the deck and tower, and it is usually necessary to consider the P-Delta effect in all three components.

P-Delta Load Combination


In a complicated structure, especially one that is redundant, the P-Delta axial forces in the Frame elements may not be known a priori. This is typically the case in building structures where the P-Delta effect is due primarily to gravity load. For these situations, the program can compute the P-Delta axial forces from a single static load combination that you define. This P-Delta load combination is an arbitrary combination of the usual static Load Cases. Each Load Case may be scaled before it is added into this load combination. These Load Cases may include any type of loading. The P-Delta load combination is independent of any Combos that may have been defined. Forces from response-spectrum, time-history and bridge moving-load analyses cannot be included in this load combination. An iterative analysis is required to determine the P-Delta axial forces in the Frame elements due to the specified P-Delta Load Combinations. It is important that the strains and rotations remain small during this analysis; otherwise, meaningful results may not be obtained. For more information: See Topic Iterative Analysis (page 345) in this chapter. See Chapter Load Cases (page 279).

Iterative Analysis
When a P-Delta load combination is specified, an iterative analysis is required to determine the P-Delta axial forces in the Frame elements. A preliminary analysis is performed to estimate the axial forces throughout the structure. The equilibrium equations are then re-formed and re-solved taking these axial forces into account. This second analysis may produce different axial forces in the members if the modified stiffness causes a force redistribution. Additional iterations, each re-forming and re-solving the equilibrium equations, may be required, until the axial forces and the structural deflections converge, that is, until they do not significantly change from one iteration to the next.

Iterative Analysis
363

345

SAP2000 Analysis Reference Such an iterative type of solution is typical for nonlinear problems. The type of iteration described above is called direct iteration (Bathe 1982; Cook, Malkus, and Plesha 1989; Zienkiewicz and Taylor 1991). Because each iteration requires forming and solving the equilibrium equations, the total solution time may be considerably longer than that required for a linear problem. Directly specified P-Delta axial forces, if present, are included in every iteration. They are added to the computed P-Delta axial forces. Only the directly specified axial forces are present for the initial iteration. The final equilibrium equations obtained after convergence of the iteration process will be valid for any loading situation that does not change the axial forces in the members. The results for different loads can then be superposed as for a linear structure. In general, however, the axial forces throughout a structure may depend upon the applied loads. In such a case, a separate iterative analysis may be required for each set of applied loads, and the results of these analyses should not be superposed.

Convergence Criterion
You may specify a relative displacement tolerance, told, which measures convergence. The default value is 10-3. If the relative change in displacement from one iteration to the next is less than the tolerance, then no further iterations are performed. The relative change in displacement is defined as the ratio of the maximum change in displacement to the largest displacement in either iteration. Note that rotations and translations are treated equally.

Maximum Number of Iterations


You may specify the maximum number of iterations, itmax, that the program is allowed to perform. This is used to prevent excessive computational time, since each iteration requires about as much computational effort as a linear static analysis. The initial iteration is termed the zero-th iteration. It is a standard linear analysis that is always performed whether or not a P-Delta load combination is specified. Parameter itmax limits the number of additional iterations performed that correct for the effect of the P-Delta load combination. Setting itmax to zero is the same as not specifying the P-Delta load combination. Setting told to zero will force all itmax additional iterations to be performed.

346

Iterative Analysis
364

Chapter XVII P-Delta Analysis

Convergence Failure
If convergence has not been obtained after the maximum number of iterations has been performed, then the results of the analysis may be meaningless, and they should be viewed with great skepticism. Failure to converge may be due to several causes: Too few iterations were permitted. A reasonable number is usually 2 to 5, although more may be required, depending on the particular problem at hand. A convergence tolerance that is too small is used. A reasonable value depends on the particular problem. Beware, however, that using a value that is too large may result in convergence to meaningless results. The structure is near buckling. The structure should be stiffened against buckling, or the magnitude of the P-Delta load combination reduced.

Frame Element
The implementation of the P-Delta effect in the Frame element is described in the following subtopics.

Small Deflections
All deflections are assumed to be small. In particular, the magnitude of all strains and rotations are assumed to be very much less than unity. The P-Delta effect does not capture any change in axial stiffness of the member, such as might occur after buckling. This would require consideration of largedisplacement effects.

Cubic Deflected Shape


The P-Delta effect is integrated along the length of each Frame element, taking into account the deflection within the element. For this purpose the transverse deflected shape is assumed to be cubic for bending and linear for shear between the rigid ends of the element. The length of the rigid ends is the product of the rigid-end factor and the end offsets, and is usually zero. See Topic End Offsets (page 162) in Chapter The Frame Element for more information. The true deflected shape may differ somewhat from this assumed cubic/linear deflection in the following situations: Frame Element
365

347

SAP2000 Analysis Reference The element has non-prismatic Section properties. In this case the P-Delta deflected shape is computed as if the element were prismatic using the average of the properties over the length of the element Loads are acting along the length of the element. In this case the P-Delta deflected shape is computed using the equivalent fixed-end forces applied to the ends of the element. A large P-force is acting on the element. The true deflected shape is actually described by trigonometric functions under large compression, and by hyperbolic functions under large tension. The assumed cubic shape is usually a good approximation to these shapes except under a compressive P-force near the buckling load with certain end restraints. Excellent results, however, can be obtained by dividing any structural member into two or more Frame elements. See the SAP2000 Verification Manual for more detail.

Computed P-Delta Axial Forces


The following discussion applies only to P-Delta axial forces computed from the P-Delta load combination. Directly specified P-Delta axial forces are not affected by any of the following considerations. The P-Delta axial force in each Frame element is determined from the axial displacements computed in the previous iteration. For meaningful results, it is important to use realistic values for the axial stiffness of these elements. The axial stiffness is determined from the Section properties that define the cross-sectional area and the modulus of elasticity. Using values that are too small may underestimate the P-Delta effect. Using values that are too large may make the P-Delta force in the element very sensitive to the iteration process. Elements that have an axial force release, or that are constrained against axial deformation by a Constraint, will have a zero P-Delta axial force and hence no PDelta effect. The P-Delta axial force also includes loads that act within the element itself. These may include Self-Weight and Gravity Loads, Concentrated and Distributed Span Loads, Prestress Load, and Temperature Load. The P-Delta axial force is assumed to be constant over the length of each Frame element. If the P-Delta load combination includes loads that cause the axial force to vary, then the average axial force is used for computing the P-Delta effect. If the difference in axial force between the two ends of an element is small compared to

348

Frame Element
366

Chapter XVII P-Delta Analysis the average axial force, then this approximation is usually reasonable. This would normally be the case for the columns in a building structure. If the difference is large, then the element should be divided into many smaller Frame elements wherever the P-Delta effect is important. An example of the latter case could be a flagpole under self-weight. For more information: See Topic Section Properties (page 153) in Chapter The Frame Element. See Topic End Releases (page 165) in Chapter The Frame Element. See Chapter Constraints and Welds (page 113).

Prestress
When Prestress Load is included in the P-Delta load combination, the combined tension in the prestressing cables tends to stiffen the Frame elements against transverse deflections. This is true regardless of any axial end releases. Axial compression of the Frame element due to Prestress Load may reduce this stiffening effect, perhaps to zero. See Topic Prestress Load (page 174) in Chapter The Frame Element for more information.

Effect upon Other Analyses


The P-Delta effect essentially modifies the characteristics of the structure, affecting the results of all other analyses performed, including: static Loads, Modes, Specs, Histories, bridge influence lines, Moving Loads, and Combos. Because the same constant stiffness matrix is used, these analyses are all linear, resulting in great computational efficiency, and permitting superposition of the results (except for nonlinear time-history analysis).

Dynamic Analyses
Eigenvectors and Ritz-vectors are obtained using the stiffness matrix as modified for the P-Delta effect. Compressive P-Delta axial forces tend to soften the structure, lengthening the periods of vibration. Tensile axial forces tend to shorten the periods. In structures where the P-Delta axial forces are both compressive and tensile, the effect on the periods is not easily predicted.

Effect upon Other Analyses


367

349

SAP2000 Analysis Reference Response-spectrum analyses and time-history analyses are performed using these eigenvectors or Ritz vectors as a basis. Thus the static P-Delta effect is automatically included. Any additional P-Delta effect due to dynamic axial forces is not accounted for, since this would require many separate, nonlinear P-Delta analyses taking great computational effort. See Chapter Static and Dynamic Analysis (page 291) for more information.

Harmonic Steady-State Analysis


Direct specification of the P-Delta axial forces may be used with harmonic steadystate analysis. These forces are assumed to be constant during the period of loading. You should not specify a P-Delta load combination since this load would be harmonically varying, causing a varying P-Delta effect that is not accounted for by the program. See Topic Harmonic Steady-State Analysis (page 294) in Chapter Static and Dynamic Analysis for more information

Bridge Moving-Load Analysis


Bridge influence lines and moving-load response are computed using the stiffness matrix as modified for the P-Delta effect. The additional P-Delta effect due to vehicle moving loads is not accounted for, since this would require many separate nonlinear P-Delta analyses taking great computational effort. If the axial forces due to vehicle moving loads are significant in comparison to the dead-load axial forces, it may be necessary to specify P-Delta axial forces that include some average of the moving-load forces. This may require a prior SAP2000 Moving-Load analysis to determine these values. See Chapter Bridge Analysis (page 357) for more information.

Buckling
If compressive P-Delta axial forces are present and are large enough, the structure may buckle. Local buckling of individual members or global buckling of the whole structure are possible. The program makes no distinction between local and global buckling.

350

Buckling
368

Chapter XVII P-Delta Analysis If the program detects that buckling has occurred, the analysis is terminated and no results are produced. This is because the analysis of a structure that has buckled requires consideration of large-displacement effects that are not modeled by SAP2000.

Detection of Buckling
Buckling may be detected during any iteration at one of two possible stages: During the formation of the stiffness matrix: if a sufficiently large, compressive P-Delta axial force is present in a Frame element, a negative value may be created on the diagonal of the stiffness matrix. During the solution of the equilibrium equations: a zero or negative value may be produced on the diagonal of the stiffness matrix. In either case, the analysis will be terminated immediately.

Estimating the Buckling Load


SAP2000 does not provide a direct method of determining the buckling load of a structure. It may be estimated, however, by performing a series of runs, each time increasing the magnitude of the P-Delta load combination, until buckling is detected. It is important to keep unchanged the relative contributions from each Load Case to the P-load, increasing all Load Case scale factors by the same amount between runs. It is important to understand that there is no single buckling load for a structure. Rather, there is a different buckling load corresponding to each spatial distribution of loads. If buckling of the structure is a concern under various loading situations, the buckling load should be estimated separately for each situation, as described above, by starting with different P-Delta load combinations.

Local Buckling
The buckling of individual members within a structure can be detected provided that they are adequately modeled. For some support conditions, a single Frame element may adequately represent a structural member. However, for best results for all support conditions, two Frame elements should be used to represent any structural member for which buckling may be a problem. The improvement obtained by using more than two elements does not usually warrant the additional effort.

Buckling
369

351

SAP2000 Analysis Reference

Practical Application
This section provides some additional guidelines regarding practical use of the PDelta analysis option. See also the SAP2000 Verification Manual for example problems.

Preliminary Linear Analysis


After the structural model is developed, including the basic Load Cases, it is strongly recommended that a linear, static analysis be performed before running any P-Delta analyses. The results of this preliminary static analysis should be used to check the model for correctness before incorporating the complexity of the nonlinear P-Delta effect. Examination of the deflected shape and the element forces and stresses will help detect errors in geometry, element connectivity, material properties, restraints and constraints. It is useful to define a Combo that is equivalent to the P-Delta load combination. The Frame element axial force distribution due to this Combo should be examined to make sure that the axial stiffnesses are reasonable, and that the axial force does not vary too much within any single Frame element.

Building Structures
For most building structures, especially tall buildings, the P-Delta effect of most concern occurs in the columns due to gravity load, including dead and live load. The column axial forces are compressive, making the structure more flexible against lateral loads. Building codes (ACI 1995; AISC 1994) normally recognize two types of P-Delta effects: the first due to the overall sway of the structure and the second due to the deformation of the member between its ends. The former effect is often significant; it can be accounted for fairly accurately by considering the total vertical load at a story level, which is due to gravity loads and is unaffected by any lateral loads. The latter effect is significant only in very slender columns or columns bent in single curvature (not the usual case); this requires consideration of axial forces in the members due to both gravity and lateral loads. SAP2000 can analyze both of these P-Delta effects. However, it is recommended that the former effect be accounted for in the SAP2000 analysis, and the latter effect be accounted for in design by using the applicable building-code moment-

352

Practical Application
370

Chapter XVII P-Delta Analysis magnification factors (White and Hajjar 1991). This is how the SAP2000 design processors for steel frames and concrete frames are set up. The P-Delta effect due to the sway of the structure can be accounted for accurately and efficiently, even if each column is modeled by a single Frame element, by using the factored dead and live loads as the P-Delta load combination. The iterative PDelta analysis should converge rapidly, usually requiring only a single iteration. As an example, suppose that the building code requires the following load combinations to be considered for design: (1) 1.4 dead load (2) 1.2 dead load + 1.6 live load (3) 1.2 dead load + 0.5 live load + 1.3 wind load (4) 1.2 dead load + 0.5 live load 1.3 wind load (5) 0.9 dead load + 1.3 wind load (6) 0.9 dead load + 1.3 wind load For this case, the P-Delta effect due to overall sway of the structure can usually be accounted for, conservatively, by specifying the P-Delta load combination to be 1.2 times the dead load plus 0.5 times the live load. This will accurately account for this effect in load combinations 3 and 4 above, and will conservatively account for this effect in load combinations 5 and 6. This P-Delta effect is not generally important in load combinations 1 and 2 since there is no lateral load. The P-Delta effect due to the deformation of the member between its ends can be accurately analyzed only when separate SAP2000 runs are made for each load combination above, with the P-Delta load combination specified equal to the combination being analyzed. Six runs would be needed for the example above. Also, at least two Frame elements per column should be used. Again, it is recommended that this effect be accounted for instead by using the SAP2000 design processors.

Cable Structures
The P-Delta effect can be a very important contributor to the stiffness of suspension bridges, cable-stayed bridges, and other cable structures. The lateral stiffness of cables is due almost entirely to tension, since they are very flexible in bending when unstressed.

Practical Application
371

353

SAP2000 Analysis Reference The easiest way to model this is by directly specifying the P-Delta axial forces if they are known. When this is not possible, a P-Delta load combination can be used provided that the cable geometry does not change too much upon loading. In many cable structures, the tension in the cables is due primarily to gravity load, and it is relatively unaffected by other loads. If this is the case, it is appropriate to define the P-Delta load combination to be a realistic combination of the dead load and live load. It is important to use realistic values for the P-Delta load combination, since the lateral stiffness of the cables is approximately proportional to the PDelta axial forces. Frame elements are used to model cables. A single element is sufficient between points of concentrated load. Additional elements may be needed if significant distributed loads, including self weight, act upon the cable. Concentrated loads should only be applied at joints, not as Concentrated Span loads, since cables kink at such loads. Each Frame cable element should be given a small, realistic bending stiffness. Otherwise the structure may be unstable in the zero-th iteration before the tensile PDelta axial forces can provide lateral stiffness. For the same reason, moment endreleases should generally not be used for cable elements. The geometry of a loaded cable is strongly dependent upon the type of loading applied. Because SAP2000 only considers small deflections, it is important to define the cable geometry (joint coordinates) to be close to what is expected after the structure is loaded. It may be necessary to correct the geometry after one or more preliminary runs that determine the shape of the cable under the P-Delta load combination. If the stretching or rotation of the cable is large (say more than a few hundredths of a percent) it may not be possible to obtain meaningful results with a PDelta load combination. The P-Delta effect only affects transverse stiffness, not axial stiffness. Therefore, Frame elements representing a cable can carry compression as well as tension; this type of behavior is generally unrealistic. You should review the analysis results to make sure that this does not occur. Because convergence tends to be slower for stiffening than softening structures, the nonlinear P-Delta analysis may require many iterations. Five to ten iterations would not be unusual.

354

Practical Application
372

Chapter XVII P-Delta Analysis

Guyed Towers
In guyed towers and similar structures, the cables are under a large tension produced by mechanical methods that shorten the length of the cables. These structures can be analyzed by the same methods discussed above for cabled bridges. A Temperature load causing a decrease in the temperature of the cables can be used to produce the requisite shortening. The P-Delta load combination should include this load, and may also include other loads that cause significant axial force in the cables, such as gravity and wind loads. Several analyses may be required to determine the magnitude of the temperature change needed to produce the desired amount of cable tension. Alternatively, the P-Delta axial force can be directly specified in the cables and in the tower.

Practical Application
373

355

SAP2000 Analysis Reference

374

C h a p t e r XVIII

Bridge Analysis
Bridge Analysis can be used to compute influence lines for traffic lanes on bridge structures and to analyze these structures for the response due to vehicle live loads. Advanced Topics Overview Modeling the Bridge Structure Roadways and Lanes Spatial Resolution Influence Lines Vehicles Vehicle Classes Moving Load Cases Influence Line Tolerance Exact and Quick Response Calculation Moving Load Response Control Correspondence Computational Considerations

357
375

SAP2000 Analysis Reference

Overview
Bridge Analysis can be used to determine the response of bridge structures due to the weight of Vehicle live loads. Considerable power and flexibility is provided for determining the maximum and minimum displacements and forces due to multiple-lane loads on complex structures, such as highway interchanges. The effects of Vehicle live loads can be combined with static and dynamic loads, and envelopes of the response can be computed. The bridge to be analyzed is modeled with Frame elements representing the superstructure, substructure and other components of interest. Displacements, reactions, spring forces, and Frame-element internal forces can be determined due to the influence of Vehicle live loads. Other element types (Shell, Plane, Asolid, Solid, and Nllink) may be used; they contribute to the stiffness of the structure, but they are not analyzed for the effect of Vehicle load. Lanes are defined on the superstructure that represent where the live loads can act. These Lanes need not be parallel nor of the same length, so that complex traffic patterns may be considered. The program computes conventional influence lines for all response quantities due to the loading of each Lane. These influence lines may be displayed using the SAP2000 graphical interface. You may select Vehicle live loads from a set of standard highway and railway Vehicles, or you may create your own Vehicle live loads. Vehicles move in both directions along each Lane of the bridge. Vehicles are automatically located at such positions along the length of the Lanes to produce the maximum and minimum response quantities throughout the structure. Each Vehicle live load may be allowed to act on every lane or be restricted to certain lanes. The program can automatically find the maximum and minimum response quantities throughout the structure due to placement of different Vehicles in different Lanes. For each maximum or minimum extreme response quantity, the corresponding values for the other components of response can also be computed. In summary, the procedure to perform a Bridge Analysis is to: Model the structural behavior of the bridge with Frame elements Define traffic Lanes describing where the Vehicle live loads act Define the different Vehicle live loads that may act on the bridge Define Vehicle Classes (groups) containing one or more Vehicles that must be considered interchangeably

358

Overview
376

Chapter XVIII Bridge Analysis Define Moving Load cases that assign Vehicle Classes to act on the traffic Lanes in various combinations Specify for which joints and Frame elements the Moving Load response is to be calculated The most extreme (maximum and minimum) displacements, reactions, spring forces, and Frame element internal forces are automatically computed for each Moving Load case defined.

Modeling the Bridge Structure


You should model the bridge structure primarily with Frame elements as described in the following.

Frame Elements
In simple cases you may define a two-dimensional model with longitudinal elements representing the superstructure and roadway, and vertical elements representing the piers and supports. For curved bridge structures these Frame elements need not exist in a single plane. Elements directed in the third, transverse direction may also be used for modeling the bents and other features. Figure 92 (page 360) shows an example of a bridge model. Specify appropriate Section properties to represent the total effective-stiffness properties of the superstructure and substructure members. These elements should be placed along the neutral axis of the members they represent. The results of the Bridge Analysis will report the Frame element internal forces and moments which can then be used to design the actual sections. Moving-load response will only be calculated for those elements you specifically request. For more information: See Topic Local Coordinate System (page 148) in Chapter The Frame Element. See Topic Section Properties (page 153) in Chapter The Frame Element. See Topic Vehicles (page 372) in this chapter. See Topic Moving Load Response Control (page 392) in this chapter.

Modeling the Bridge Structure


377

359

SAP2000 Analysis Reference

Figure 92 Frame Element Model of a Bridge Structure

Supports
Supports can be modeled using either springs or restraints. Moving-load response will only be calculated for those springs or restraints you specifically request.

360

Modeling the Bridge Structure


378

Chapter XVIII Bridge Analysis See Topic Moving Load Response Control (page 392) in this chapter for more information.

Bearings and Expansion Joints


Effective modeling of support conditions at bearings and expansion joints requires careful consideration of the continuity of each translational and rotational component of displacement. Continuous components require that the corresponding degrees-of-freedom remain connected across the bearing or expansion joint; degrees-of-freedom representing discontinuous components must be disconnected. You can achieve this by two principal methods: (1) Attaching elements to separate joints at the same location (which automatically disconnects all degrees-of-freedom between the elements) and constraining together the connected degrees-of-freedom using an Equal or Local Constraint, or (2) Attaching several elements to a common joint (which automatically connects the degrees-of-freedom between the elements) and using Frame element end releases to free the unconnected degrees-of-freedom. Both methods are acceptable for static analysis. For dynamic analysis, method (1) is recommended since method (2) does not properly distribute the mass on either side of the joint. Typically the vertical and transverse translations and the torsional rotation would be connected, while the longitudinal translations and the bending and in-plane rotations would be disconnected. However, the appropriate use of constrained or released degrees-of-freedom depends on the details of each individual bearing or joint. See Figure 93 (page 362) for examples. For more information: See Topic End Releases (page 165) in Chapter The Frame Element. See Topics Equal Constraint (page 123) and Local Constraint (page 127) in Chapter Constraints and Welds.

Other Element Types


Shell, Plane, Asolid, Solid, and Nllink elements should not generally be used in models subjected to Vehicle loads. If you do use these types of elements, you should do so with caution and with complete understanding of the following implications: Modeling the Bridge Structure
379

361

SAP2000 Analysis Reference

Figure 93 Modeling of Bearings and Expansion Joints

362

Modeling the Bridge Structure


380

Chapter XVIII Bridge Analysis Vehicle live loads can only be applied to Frame elements. Thus live loads cannot be represented as acting directly on bridge decks modeled with Shell or other element types. All elements present in the structure contribute to the stiffness and may carry part of the load. However, element internal forces (stresses) due to Vehicle live loads are computed only for Frame elements. Therefore, the presence of other element types may result in an underestimate of the internal forces in Frame elements if these are intended to represent the complete behavior of the substructure or superstructure. The corresponding response in the other element types will not be reported. This approach may be unconservative for all element types.

Roadways and Lanes


The Vehicle live loads are considered to act in traffic Lanes transversely spaced across the bridge roadway. These Lanes are supported by Frame elements representing the bridge deck. The number of Lanes and their transverse spacing can be chosen to satisfy the appropriate design-code requirements. For simple bridges with a single roadway, the Lanes will usually be parallel and evenly spaced, and will run the full length of the bridge structure. For complex structures, such as interchanges, multiple roadways may be considered; these roadways can merge and split. Lanes need not be parallel nor be of the same length. The number of Lanes across the roadway may vary along the length to accommodate merges.

Roadways
Typically each roadway is modeled with a single string (or chain) of Frame elements running along the length of the roadway. These elements should possess Section properties representing the full width and depth of the bridge deck. They are modeled as a normal part of the overall structure and are not explicitly identified as being roadway elements.

Lanes
A traffic Lane on a roadway has its length represented by a consecutive set of some or all of the roadway elements. The transverse position of the Lane center line is specified by its eccentricity relative to the roadway elements. Each Lane across the roadway width will usually refer to the same set of roadway elements, but will typiRoadways and Lanes
381

363

SAP2000 Analysis Reference cally have a different eccentricity. The eccentricity for a given Lane may also vary along the length. A Lane is thus defined by listing, in sequence, the labels of a chain of Frame elements that already exist as part of the structure. Each Lane is said to run in a particular direction, namely from the first element in the listed sequence to the second element, and so on, to the last element. This direction may be the same or different for different Lanes using the same roadway elements, depending on the order in which each Lane is defined. It is independent of the direction that traffic travels.

Eccentricities
The sign of a Lane eccentricity is defined as follows: in an elevation view of the bridge where the Lane runs from left to right, Lanes located behind the roadway elements have positive eccentricity. Alternatively, to a driver traveling on the roadway in the direction that the Lane runs, a Lane to the left of the roadway elements has a positive eccentricity. The use of eccentricities is primarily important for the determination of axial torsion in the bridge deck and transverse bending in the substructure; secondary effects may also be found in more complex structures. Although the modeling of lane eccentricities is generally realistic and advantageous, some savings in computation time, memory requirements, and disk storage space can be realized by using zero eccentricities for all elements in all Lanes.

Modeling Guidelines
Although roadway elements are not explicitly defined as such, they can be identified as those Frame elements in the structure that are referred to by one or more Lane definitions. Since the Vehicle live loads will be applied to the roadway Frame elements through the use of the Lanes, the modeling of roadway elements should adhere to the following guidelines: They should be located at the neutral axis of the bridge deck They should be parallel to the direction of traffic, or approximately so They should form one or more (nearly) contiguous chains of elements. To be contiguous, the end of one element should be located at the beginning of the next element in the chain. The two elements may be attached to the same joint, or to two different joints at the same location. If they are not contiguous, the gap between adjacent elements should be small, especially in the longitudinal direction; gaps in the transverse and vertical direction are not usually significant

364

Roadways and Lanes


382

Chapter XVIII Bridge Analysis

Figure 94 Definition of Lanes for a Simple Bridge

They must not be vertical Each Lane should be a consecutive set of some or all of the roadway elements, chosen to form a (nearly) contiguous chain or path

Examples
Figure 93 (page 365) shows a simple 24 ft wide bridge carrying two opposing 12 ft traffic Lanes. The roadway, and also each traffic Lane, are modeled by four Frame elements (1, 2, 3, 4) running along the center line of the bridge from east to west. Roadways and Lanes
383

365

SAP2000 Analysis Reference

Figure 95 Definition of Lanes for a Two-Roadway Merge

The eccentricities are constant at +6 ft and 6 ft for the eastbound and westbound Lanes, respectively. A second example showing a simple portion of an interchange is presented in Figure 94 (page 366). Here two 12 ft wide roadways (A and B) merge into a single 24 ft wide roadway (C), which then tapers down to a 12 ft width. Each roadway is

366

Roadways and Lanes


384

Chapter XVIII Bridge Analysis modeled with a single chain of elements. Elements representing the substructure and other structural members are not shown. Two 12 ft wide traffic Lanes are defined: the first runs the full length of roadways A and C; the second runs the full length of roadway B and the 24 ft wide portion of roadway C. The chain of elements defining the first Lane is: 8, 7, 6, 5, 4, 3, 2, 1; the chain defining the second Lane is: 11, 10, 9, 5, 4, 3. Both Lanes run eastward. The eccentricities at the centers of the elements are zero except for: +3 ft for element 2 and +6 ft for elements 3, 4 and 5 in the first Lane; and 6 ft for elements 3, 4 and 5 in the second Lane. A significant transverse gap exists between element 5 and elements 6 and 9 in Lanes 1 and 2, respectively. Significant transverse gaps also exist on either side of element 2, which is taken to be parallel to the direction of traffic in order to obtain the most meaningful definitions for moments and torque. Since no longitudinal gaps exists, the Vehicle loads will be modeled adequately. However, appropriate stiffness connections must still be made to tie the roadways together at the gaps. This can be done using elements or rigid Body Constraints. See Topic Body Constraints (page 115) in Chapter Constraints and Welds for more information.

Spatial Resolution
The accuracy of the Bridge Analysis is determined by the spatial resolution (the number of load and response points) of the Lanes.

Load and Output Points


The program applies the Vehicle loads at a finite set of fixed load points along the traffic Lanes. Likewise, Frame element internal forces are computed and output at fixed output points along all Frame elements. The accuracy of a Vehicle live-load analysis is dependent upon the resolution, i.e., the number of load points and output points used. Increasing the resolution increases the likelihood of obtaining accurate values and locations for the maximum and minimum displacements and forces in the structure; it also significantly increases computation time, memory, and disk storage requirements. The number of load and output points for each Frame element is determined by the number, nseg, of output segments specified for that element:

Spatial Resolution
385

367

SAP2000 Analysis Reference

Figure 96 Load and Output Points for a Single Frame Element with nseg = 4

The output points for each element are the nseg+1 equally spaced points along the clear length of the element. The clear length is that length between the rigid zone offsets at either end of the element, if any. The load points for a Lane element are the same as the output points, plus an additional point at each joint of the element wherever there is a non-zero rigid zone offset. Thus the number of load points will be from nseg+1 to nseg+3. See Figure 95 (page 368) for an example. Only Lane elements possess load points. Displacements, reactions, and spring-forces are only available at the joints. The accuracy of these results may still be dependent upon the number of load points.

368

Spatial Resolution
386

Chapter XVIII Bridge Analysis See Topic Internal Force Output (page 176) in Chapter The Frame Element for more information.

Resolution
The resolution of a live-load analysis may be increased by increasing the number of Frame elements, the number of output segments for each Frame element, or both. Increasing the number of output segments, nseg, is the simplest way to increase the resolution; it is also the most computationally efficient. See Topic Computational Considerations (page 393) in this chapter for more information. Increasing the number of Frame elements as a way to increase resolution is not generally recommended. However, other factors may govern the number of Frame elements that need to be used in a given region, such as: Curved roadways: more than one element may be needed between supports, diaphragms, or cross-braces. You should experiment to determine the number of elements required to adequately represent the stiffness and loading of the curved roadway. Dynamic analyses: element masses are lumped at the joints, hence a sufficient number of Frame elements must be defined to represent the governing vibration modes of the structure Non-constant Lane eccentricities For example, a simple span represented by a single element with nseg=8 has the same effective number of load and response points as four elements with nseg=2. This is true because load and response points at the interior joints are duplicated in the latter case. Both meshes have the same resolution for live-load analysis and will produce the same results. The former mesh is easier to define and will be somewhat more efficient. However, the latter mesh is superior when dynamics are to be considered.

Modeling Guidelines
A suggested approach for achieving adequate resolution for live-load analysis is as follows: Devise a SAP2000 bridge model that uses an appropriate number of Frame elements to capture significant structural behavior and inertia properties, and to properly represent the traffic lanes and spans

Spatial Resolution
387

369

SAP2000 Analysis Reference Perform a preliminary analysis using a moderate number of output segments (say nseg=2 for all elements) to assess the correctness of the model and to check for adequate local resolution Correct the model as required, adding more Frame elements if necessary Perform another analysis using the corrected model with an increased number of output segments, nseg, where needed Repeat the previous steps as necessary Checking the model for adequate resolution should ideally be done using the SAP2000 graphical interface to examine the influence lines and the Moving Load response. The influence lines will give a more critical view of the resolution of the model, although the Moving Load results are of more practical interest.

Influence Lines
SAP2000 automatically computes influence lines for the following response quantities: Frame element internal forces at the output points Joint displacements Reactions Spring forces For each response quantity in the structure, there is one influence line for each traffic Lane. An influence line can be viewed as a curve of influence values plotted at the load points along a traffic Lane. For a given response quantity at a given location in the structure, the influence value at a load point is the value of that response quantity due to a unit concentrated downward force acting at that load point. The influence line thus shows the influence upon the given response quantity of a unit force moving along the traffic lane. Figure 96 (page 371) shows some simple examples of influence lines. Influence lines may exhibit discontinuities (jumps) at the output point when it is located at a load point on the traffic lane. Discontinuities may also occur where the structure itself is not continuous (e.g., expansion joints).

370

Influence Lines
388

Chapter XVIII Bridge Analysis

Figure 97 Examples of Influence Lines for One-Span and Two-Span Beams

SAP2000 uses influence lines to compute the response to vehicle live loads. Influence lines are also of interest in their own right for understanding the sensitivity of various response quantities to traffic loads.

Influence Lines
389

371

SAP2000 Analysis Reference Influence lines can be displayed using the SAP2000 graphical user interface. They are plotted along the Lane elements with the influence values plotted in the vertical direction. A positive influence value due to gravity load is plotted upward. Influence values are linearly interpolated between the known values at the load points. Influence values may also be written to a text file from the graphical interface. Influence lines are available after any analysis for which traffic Lanes were defined. It is not necessary to define Vehicles, Classes, Moving Load cases, or response control in order to get influence lines.

Vehicles
Any number of Vehicle live loads, or simply Vehicles, may be defined to act on the traffic Lanes. You may use standard types of Vehicles known to the program, or design your own using the general Vehicle specification.

Direction of Loads
All vehicle live loads represent weight and are assumed to act downward, in the Z global coordinate direction. See Upward and Horizontal Directions (page 23) in Chapter Coordinate Systems.

Application of Loads
Each Vehicle consists of one or more concentrated (point) and/or uniform line loads. These act on the center line of the lane, i.e., along lines parallel to the Lane elements, horizontally offset from the Lane elements by the lane eccentricity. Each concentrated or uniform load is considered to represent a range of values from zero up to a specified maximum. When computing a response quantity (force or displacement) the maximum value of load is used where it increases the severity of the response, and zero is used where the load would have a relieving effect. Thus the specified load values for a given Vehicle may not always be applied proportionally. This is a conservative approach that accounts for Vehicles that are not fully loaded. The maximum and minimum values of a response quantity are computed using the corresponding influence line. Concentrated loads are multiplied by the influence value at the point of application to obtain the corresponding response; uniform

372

Vehicles
390

Chapter XVIII Bridge Analysis loads are multiplied by the influence values and integrated over the length of application. Loads acting in regions of positive influence value only add to the maximum response; they never affect the minimum response. Similarly, loads acting in regions of negative influence value only subtract from the minimum response. Thus the maximum response is always positive (or zero); the minimum response is always negative (or zero). By way of example, consider the influence line for the moment at the center of the left span shown in Figure 97(b) (page 371). Any concentrated load or portion of a uniform load that acts on the left span would contribute only to the positive maximum value of the moment response. Loads acting on the right span would not decrease this maximum, but would contribute to the negative minimum value of this moment response.

General Vehicle
The general Vehicle may represent an actual vehicle or a notional vehicle used by a design code. Most trucks and trains can be modeled by the SAP2000 general Vehicle. The general Vehicle consists of n axles with specified distances between them. Concentrated loads may exist at the axles. Uniform loads may exist between pairs of axles, in front of the first axle, and behind the last axle. The distance between any one pair of axles may vary over a specified range; the other distances are fixed. The leading and trailing uniform loads are of infinite extent. Additional floating concentrated loads may be specified that are independent of the position of the axles. Because the applied loads never decrease the severity of the computed response, the effect of a shorter Vehicle is captured by a longer Vehicle that includes the same loads and spacings as the longer Vehicle. Only the longer Vehicle need be considered in such cases. Specification To define a Vehicle, you may specify: n1 positive distances, d, between the pairs of axles; one inter-axle distance may be specified as a range from dmin to dmax, where 0 < dmin dmax, and dmax = 0 is used to represent a maximum distance of infinity n concentrated loads, p, at the axles

Vehicles
391

373

SAP2000 Analysis Reference

Figure 98 General Vehicle Definition

n+1 uniform loads, w: the leading load, the inter-axle loads, and the trailing load Floating concentrated loads; either: A single floating load, px, for all response quantities, or A pair of floating loads: Load pm for span moments in the Lane elements. This load receives special treatment for span moments over the supports, as described below, and

374

Vehicles
392

Chapter XVIII Bridge Analysis Load pxm for all response quantities except span moments in the Lane elements Whether or not this Vehicle is to be used for: Negative span moments over the supports in the Lane elements Vertical forces in interior piers and/or interior supports Response quantities other than the two types above The number of axles, n, may be zero, in which case only a single uniform load and the floating concentrated loads can be specified. These parameters are illustrated in Figure 98 (page 374). Specific examples are given in the next subtopic using the standard Vehicles. Additional detail is provided in the following. Moving the Vehicle When a Vehicle is applied to a traffic Lane, the axles are moved along the length of the lane to where the maximum and minimum values are produced for every response quantity in every element. Usually this location will be different for each response quantity. For asymmetric (front to back) Vehicles, both directions of travel are considered. Vehicle Response Categories In order to satisfy certain requirements of the AASHTO HL design vehicular live load (AASHTO, 1996 b), the available response quantities are divided into the following categories: (1) Negative span moments over the supports in Lane elements only. A negative span moment is defined as that moment which causes tension in the upward-most face of a Lane element: Negative M3 if the local +2 axis is most upward (the default) Positive M3 if the local 2 axis is most upward Negative M2 if the local +3 axis is most upward Positive M2 if the local 3 axis is most upward SAP2000 considers all negative span moments in the Lane elements to be in this category without regard for the location of the supports (piers). (2) Reactions at interior supports (piers). This takes two forms:

Vehicles
393

375

SAP2000 Analysis Reference A compressive axial force in vertical Frame elements, where vertical is defined here as being within 15 of the Z axis The most upward local component of reactions and spring forces. For example: Positive F3 if the joint local +3 axis is most upward (the default) Negative F2 if the joint local 2 axis is most upward The program automatically determines if these force components correspond to an interior support from the shape of the influence line. If the absolute minimum influence value does not occur at either end of the influence line, the support is considered to be interior. (3) All other response quantities not included in (1) and (2) You may specify for each Vehicle whether or not to calculate each of these three categories of response using the parameters supmom, intsup, and other, respectively. Each of these parameters may take either value Y (yes) or N (no). The default is Y for all three. This enables you to define different vehicles for the different categories of response. You must be sure that each category of response is captured by at least one Vehicle in each Lane when defining the Moving Load cases. Floating Concentrated Loads The floating concentrated loads (px, pm, and pxm) are placed at the point of maximum positive influence value (if any) and the point of minimum negative influence value (if any). You may specify either px, or the pair pm and pxm. Floating load px is used equally for all response quantities. Floating load pm is only used for span moments in the Lane elements. A span moments is defined as the moment in the most vertical local plane of a Lane element: M3 if the local 2 axis is most vertical (the default) M2 if the local 3 axis is most vertical For negative span moments, as defined in category (1) under Vehicle Response Categories above, two floating concentrated loads of magnitude pm are located in two different spans. This is used to meet the requirements of the AASHTO HS Lane Load (AASHTO, 1996 a) for negative moments in continuous spans. The program automatically determines the spans from the shape of the influence line by using the

376

Vehicles
394

Chapter XVIII Bridge Analysis two most negative influence values (if any) that are separated by at least one local maximum. A single value of pm is used for positive span moments. Floating load pxm is used for all response quantities except the span moments in the Lane elements.

Standard Vehicles
The following standard vehicle types are available in SAP2000 to represent vehicular live loads specified in various design codes. The type of vehicle is specified using the parameter type. Hn-44 and HSn-44 Vehicles specified with type = Hn-44 and type = HSn-44 represent the AASHTO standard H and HS Truck Loads, respectively. The n in the type is an integer scale factor that specifies the nominal weight of the Vehicle in tons. Thus H15-44 is a nominal 15 ton H Truck Load, and HS20-44 is a nominal 20 ton HS Truck Load. These Vehicles are illustrated in Figure 98 (page 378). The effect of an H Vehicle is included in an HS Vehicle of the same nominal weight. If you are designing for both H and HS Vehicles, only the HS Vehicle is needed. Hn-44L and HSn-44L Vehicles specified with type = Hn-44L and type = HSn-44L represent the AASHTO standard H and HS Lane Loads, respectively. The n in the type is an integer scale factor that specifies the nominal weight of the Vehicle in tons. Thus H1544 is a nominal 15 ton H Lane Load, and HS20-44 is a nominal 20 ton HS Lane Load. These Vehicles are illustrated in Figure 98 (page 378). The Hn-44L and HSn-44L Vehicles are identical. AML Vehicles specified with type = AML represent the AASHTO standard Alternate Military Load. This Vehicle consists of two 24 kip axles spaced 4 ft apart. HL-93K, HL-93M and HL-93S Vehicles specified with type = HL-93K represent the AASHTO standard HL-93 Load consisting of the code-specified design truck and the design lane load.

Vehicles
395

377

SAP2000 Analysis Reference

Figure 99 AASHTO Standard H and HS Vehicles

378

Vehicles
396

Chapter XVIII Bridge Analysis

Figure 100 AASHTO Standard HL Vehicles Vehicles


397

379

SAP2000 Analysis Reference Vehicles specified with type = HL-93M represent the AASHTO standard HL-93 Load consisting of the code-specified design tandem and the design lane load. Vehicles specified with type = HL-93S represent the AASHTO standard HL-93 Load consisting of two code-specified design trucks and the design lane load, all scaled by 90%. The axle spacing for each truck is fixed at 14 ft. The spacing between the rear axle of the lead truck and the lead axle of the rear truck varies from 50 ft to the length of the Lane. This vehicle is only used for negative moment over supports and reactions at interior piers, so supmom = Y, intsup = Y, and other = N. A dynamic load allowance may be specified for each Vehicle using the parameter im. This is the additive percentage by which the concentrated truck or tandem axle loads will be increased. The uniform lane load is not affected. Thus if im = 33, all concentrated axle loads for the vehicle will be multiplied by the factor 1.33. These Vehicles are illustrated in Figure 100 (page 379) for im = 0. P5, P7, P9, P11, and P13 Vehicles specified with type = P5, type = P7, type = P9, type = P11, and type = P13 represent the Caltrans standard Permit Loads. These Vehicles are illustrated in Figure 100 (page 381). The effect of a shorter Caltrans Permit Load is included in any of the longer Permit Loads. If you are designing for all of these permit loads, only the P13 Vehicle is needed. Cooper E 80 Vehicles specified with type = COOPERE80 represent the AREA standard Cooper E 80 train load. This Vehicle is illustrated in Figure 101 (page 382). UICn Vehicles specified with type = UICn represent the European UIC (or British RU) train load. The n in the type is an integer scale factor that specifies magnitude of the uniform load in kN/m. Thus UIC80 is the full UIC load with an 80 kN/m uniform load, and UIC60 is the UIC load with an 60 kN/m uniform load. The concentrated loads are not affected by n. This Vehicle is illustrated in Figure 101 (page 382).

380

Vehicles
398

Chapter XVIII Bridge Analysis

Figure 101 Caltrans Standard Permit Vehicles Vehicles


399

381

SAP2000 Analysis Reference

Figure 102 Standard Train Vehicles

382

Vehicles
400

Chapter XVIII Bridge Analysis RL Vehicles specified with type = RL represent the British RL train load. This Vehicle is illustrated in Figure 102 (page 382).

Vehicle Classes
The designer is often interested in the maximum and minimum response of the bridge to the most extreme of several types of Vehicles rather than the effect of the individual Vehicles. For this purpose, Vehicle Classes are defined that may include any number of individual Vehicles. The maximum and minimum force and displacement response quantities for a Vehicle Class will be the maximum and minimum values obtained for any individual Vehicle in that Class. Only one Vehicle ever acts at a time. All Vehicle loads are applied to the traffic Lanes through the use of Vehicle Classes. If it is desired to apply an individual Vehicle load, you must define a Vehicle Class that contains only that single Vehicle. For example, the you may need to consider the most severe of a Truck Load and the corresponding Lane Load, say the HS20-44 and HS20-44L loads. A Vehicle Class can be defined to contain these two Vehicles. Additional Vehicles, such as the Alternate Military Load type AML, could be included in the Class as appropriate. Different members of the Class may cause the most severe response at different locations in the structure. For HL-93 loading, you would first define three Vehicles, one each of the standard types HL-93K, HL-93M, and HL-93S. You then could define a single Vehicle Class containing all three Vehicles.

Moving Load Cases


The final step in the definition of the vehicle live loading is the application of the Vehicle Classes to the traffic Lanes. This is done by creating independent Moving Load cases. Each Moving Load case consists of a set of assignments that specify how the Classes are assigned to the Lanes. Each assignment in a Moving Load case requires the following data: A Vehicle Class, class A scale factor, sf, multiplying the effect of class (the default is unity) Vehicle Classes
401

383

SAP2000 Analysis Reference A list, lanes, of one or more Lanes in which class may act (the default is all Lanes) The minimum number, lmin, of Lanes lanes in which class must act (the default is zero) The maximum number, lmax, of Lanes lanes in which class may act (the default is all of lanes) The program looks at all of the assignments in a Moving Load case, and tries every possible permutation of loading the traffic Lanes with Vehicle Classes that is permitted by the assignments. No Lane is ever loaded by more than one Class at a time. You may specify multiple-lane scale factors, rf1, rf2, rf3, ..., for each Moving Load case that multiply the effect of each permutation depending upon the number of loaded Lanes. For example, the effect of a permutation that loads two Lanes is multiplied by rf2. The maximum and minimum response quantities for a Moving Load case will be the maximum and minimum values obtained for any permutation permitted by the assignments. Usually the permutation producing the most severe response will be different for different response quantities. The results for a Moving Load case include the response due to loads ranging from zero up to the specified Vehicle live loads. As a result, the envelope always includes zero response: the maximum will never be negative and the minimum will never be positive. Thus there is no need to consider smaller vehicles that differ only by a scale factor. For example, the range of response for an HS20-44 vehicle includes the response for an HS15-44 vehicle. The concepts of assignment can be clarified with the help of the following examples.

Example 1 AASHTO HS Loading


Consider a four-lane bridge designed to carry AASHTO HS20-44 Truck and Lane Loads, and the Alternate Military Load (AASHTO, 1996 a). Suppose that it is required that the number of Lanes loaded be that which produces the most severe response in every member. Only one of the three Vehicle loads is allowed per lane. Load intensities may be reduced by 10% and 25% when three or four Lanes are loaded, respectively. Generally, loading all of the Lanes will produce the most severe moments and shears along the span and axial forces in the piers. However, the most severe torsion

384

Moving Load Cases


402

Chapter XVIII Bridge Analysis of the bridge deck and transverse bending of the piers will usually be produced by loading only those Lanes possessing eccentricities of the same sign. Assume that the bridge structure and traffic Lanes have been defined. Three Vehicles are defined: name = HSK, type = HS20-44 name = HSL, type = HS20-44L name = AML, type = AML where name is an arbitrary label assigned to each Vehicle. The three Vehicles are assigned to a single Vehicle Class, with an arbitrary label of name = HS, so that the most severe of these three Vehicle loads will be used for every situation. A single Moving Load case is then defined that seeks the maximum and minimum responses throughout the structure for the most severe of loading all four Lanes, any three Lanes, any two Lanes or any single Lane. This can be accomplished using a single assignment. The parameters for the assignment are: class = HS sf = 1 lanes = 1, 2, 3, 4 lmin = 1 lmax = 4 The scale factors for the loading of multiple Lanes in the set of assignments are rf1 = 1, rf2 = 1, rf3 = 0.9, and rf4 = 0.75. There are fifteen possible permutations assigning the single Vehicle Class HS to any one, two, three, or four Lanes. These are presented in the following table: Permutation 1 2 3 4 5 HS HS Lane 1 HS HS HS HS Lane 2 Lane 3 Lane 4 Scale Factor 1.00 1.00 1.00 1.00 1.00

Moving Load Cases


403

385

SAP2000 Analysis Reference

Permutation 6 7 8 9 10 11 12 13 14 15

Lane 1

Lane 2 HS

Lane 3 HS HS

Lane 4

Scale Factor 1.00

HS HS

1.00 1.00 1.00

HS HS HS HS HS HS HS HS HS HS HS HS HS HS HS HS

HS

1.00 0.90

HS HS HS HS

0.90 0.90 0.90 0.75

An HS in a Lane column of this table indicates application of Class HS; a blank indicates that the Lane is unloaded. The scale factor for each permutation is determined by the number of Lanes loaded.

Example 2 AASHTO HL Loading


Consider a four-lane bridge designed to carry AASHTO HL-93 loading (AASHTO, 1996 b). The approach is the same as used for AASHTO HS loading in the previous example. Only the multiple-lane scale factors and the Vehicles differ. Three Vehicles are defined: name = HLK, type = HL-93K name = HLM, type = HL-93M name = HLS, type = HL-93S where name is an arbitrary label assigned to each Vehicle. The three Vehicles are assigned to a single Vehicle Class, with an arbitrary label of name = HL, so that the most severe of these three Vehicle loads will be used for every situation. By definition of the standard Vehicle type HL-93S, Vehicle HLS will only be used when computing negative moments over supports or the reaction

386

Moving Load Cases


404

Chapter XVIII Bridge Analysis at interior piers. The other two Vehicles will be considered for all response quantities. A single Moving Load case is then defined that is identical to that of the previous example, except that class = HL, and the scale factors for multiple Lanes are rf1 = 1.2, rf2 = 1, rf3 = 0.85, and rf4 = 0.65. There are again fifteen possible permutations assigning the single Vehicle Class HL to any one, two, three, or four Lanes. These are similar to the permutations of the previous example, with the scale factors changed as appropriate.

Example 3 Caltrans Permit Loading


Consider the four-lane bridge of the previous examples now subject to Caltrans Combination Group I PW (Caltrans, 1995). Here the permit load(s) are to be used alone in a single traffic Lane, or in combination with one HS or Alternate Military Load in a separate traffic lane, depending upon which is more severe. Four Vehicles are defined: name = HSK, type = HS20-44 name = HSL, type = HS20-44L name = AML, type = AML name = P13, type = P13 where name is an arbitrary label assigned to each Vehicle. The first three Vehicles are assigned to a Vehicle Class that is given the label name = HS, as in Example 1. The last Vehicle is assigned as the only member of a Vehicle Class that is given the label name = P13. Note that the effects of SAP2000 Vehicle types P5, P7, P9, and P11 are captured by Vehicle type P13. Combination Group I PW is then represented as a single Moving Load case consisting of the assignment of Class P13 to any single Lane with or without Class HS being assigned to any other single Lane. This can be accomplished using two assignments. A scale factor of unity is used regardless of the number of loaded Lanes. The first assignment assigns Class P13 to any single Lane: class = P13 sf = 1 lanes = 1, 2, 3, 4 Moving Load Cases
405

387

SAP2000 Analysis Reference lmin = 1 lmax = 1 The second assignment assigns Class HS to any single Lane, or to no Lane at all: class = HS sf = 1 lanes = 1, 2, 3, 4 lmin = 0 lmax = 1 There are sixteen possible permutations for these two assignments such that no Lane is loaded by more than one Class at a time. These are presented in the following table:

Permutation 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Lane 1 P P P P HS

Lane 2

Lane 3

Lane 4

Scale Factor 1.00

HS HS HS P P P P HS HS P HS P P P HS P HS P

1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

HS

HS

388

Moving Load Cases


406

Chapter XVIII Bridge Analysis

Permutation 15 16

Lane 1

Lane 2

Lane 3 HS

Lane 4 P P

Scale Factor 1.00 1.00

Example 4 Restricted Caltrans Permit Loading


Consider the four-Lane bridge and the Caltrans permit loading of Example 3, but subject to the following restrictions: The permit Vehicle is only allowed in Lane 1 or Lane 4 The Lane adjacent to the Lane occupied by the permit Vehicle must be empty Two Moving Load cases are required, each containing two assignments. A scale factor of unity is used regardless of the number of loaded Lanes. The first Moving Load case considers the case where the permit Vehicle occupies Lane 1. The first assignment assigns Class P13 to Lane 1 class = P13 sf = 1 lanes = 1 lmin = 1 lmax = 1 The second assignment assigns Class HS to either Lane 3 or 4, or to no Lane at all: class = HS sf = 1 lanes = 3, 4 lmin = 0 lmax = 1 These assignments permits the following three permutations:

Moving Load Cases


407

389

SAP2000 Analysis Reference

Permutation 1 2 3

Lane 1 P P P

Lane 2

Lane 3

Lane 4

Scale Factor 1.00

HS HS

1.00 1.00

Similarly, the second Moving Load case considers the case where the permit Vehicle occupies Lane 4. The first assignment assigns Class P13 to Lane 4 class = P13 sf = 1 lanes = 4 lmin = 1 lmax = 1 The second assignment assigns Class HS to either Lane 1 or 2, or to no Lane at all: class = HS sf = 1 lanes = 1, 2 lmin = 0 lmax = 1 These assignments permits the following three permutations: Permutation 1 2 3 HS HS Lane 1 Lane 2 Lane 3 Lane 4 P P P Scale Factor 1.00 1.00 1.00

An envelope-type Combo that includes only these two Moving Load cases would produce the most severe response for the six permutations above. See Topic Combos (page 333) in Chapter Static and Dynamic Analysis for more information.

390

Moving Load Cases


408

Chapter XVIII Bridge Analysis

Influence Line Tolerance


SAP2000 simplifies the influence lines used for response calculation in order to increase efficiency. A relative tolerance is used to reduce the number of load points by removing those that are approximately duplicated or that can be approximately linearly-interpolated. The default value of this tolerance permits response errors on the order of 0.01%. Setting the tolerance to zero will provide exact results to within the resolution of the analysis.

Exact and Quick Response Calculation


For the purpose of moving a Vehicle along a lane, each axle is placed on every load point in turn. When another axle falls between two load points, the effect of that axle is determined by linear interpolation of the influence values. The effect of uniform loads is computed by integrating the linearly-interpolated segments of the influence line. This method is exact to within the resolution of the analysis, but is computationally intensive if there are many load points. A Quick method is available which may be much faster than the usual Exact method, but it may also be less accurate. The Quick method approximates the influence line by using a limited number of load points in each span. For purposes of this discussion, a span is considered to be a region where the influence line is all positive or all negative. The degree of approximation to be used is specified by the parameter quick, which may be any non-negative integer. The default value is quick = 0, which indicates to use the full influence line, i.e., the Exact method. Positive values indicate increasing degrees of refinement for the Quick method. For quick = 1, the influence line is simplified by using only the maximum or minimum value in each span, plus the zero points at each end of the span. For quick = 2, an additional load point is used on either side of the maximum/minimum. Higher degrees of refinement use additional load points. The number of points used in a span can be quick as many as 2 +1, but not more than the number of load points available in the span for the Exact method. It is strongly recommended that quick = 0 be used for all final analyses. For preliminary analyses, quick = 1, 2, or 3 is usually adequate, with quick = 2 often providing a good balance between speed and accuracy. The effect of parameter quick upon speed and accuracy is problem-dependent, and you should experiment to determine the best value to use for each different model. Influence Line Tolerance
409

391

SAP2000 Analysis Reference

Moving Load Response Control


By default, no Moving Load response is calculated for any joint or element, since this calculation is computationally intensive. You must explicitly request the Moving Load response that you want calculated. For each joint, you may explicitly request the following types of results to be calculated: Displacements Reactions, and/or Spring forces For each Frame element, you may explicitly request that the internal forces be calculated. Each of the selected joint or Frame-element response quantities is calculated for all Moving Load cases. If the displacements, reactions, spring forces, or internal forces are not calculated for a given joint or Frame element, no Moving Load response can be printed or plotted for that joint or element. Likewise, no response can be printed or plotted for any Combo that contains a Moving Load case.

Correspondence
For each maximum or minimum Frame-element response quantity computed, the corresponding values for the other five internal force and moment components may be determined. For example, the shear, moment, and torque that occur at the same time as the maximum axial force in a Frame element may be computed. These corresponding response quantities are only used for steel and concrete design in the SAP2000 graphical user interface. They cannot be printed or displayed. When Moving Load cases are printed or displayed, the extreme values of each response quantity are given without correspondence. By default, no corresponding quantities are computed for the Frame elements, since this significantly increases the computation time for moving-load response. You may specify that correspondence is to be calculated, in which case it will be done for all Frame elements for which moving-load response is requested, and for all Moving Load cases.

392

Moving Load Response Control


410

Chapter XVIII Bridge Analysis

Computational Considerations
The computation of influence lines requires a moderate amount of computer time and a large amount of disk storage compared with the execution of other typical SAP2000 analyses. The amount of computer time is approximately proportional to 2 N L, where N is the number of structure degrees-of-freedom, and L is the number of load points. The amount of disk storage required is approximately proportional to NL. Increasing the resolution of the analysis by increasing the number of Frame elements (holding nseg constant) causes approximately proportional increases in N 3 and L, and hence increases computation time by about L and storage space by about 2 L . Holding the number of elements constant and increasing nseg instead increases computation time and storage space each by approximately L. Clearly the latter method is more efficient. If all traffic Lanes have zero eccentricities everywhere, computation time and storage space for the influence lines are cut in half. The computation of Moving Load response may require a large amount of computer time compared with the execution of other typical SAP2000 analyses. The amount of disk storage needed (beyond the influence lines) is small. The computation time for Moving Load response is proportional to the number of response points. To obtain the same effective increase in resolution, increasing the values of nseg produces fewer additional response points than does increasing the number of elements, since the latter approach introduces duplicate response points at the joints. Hence changing nseg is more efficient. Considerable savings in computation time can also be realized by restricting Moving Load calculations to only those joints and Frame elements of significant interest. The computation time for Moving Load response is also directly proportional to the number of Lanes. It is not, however, sensitive to whether or not Lane eccentricities are present. For each Vehicle load, the computation time is approximately proportional to the square of the number of axles. It is also proportional to L, the effective number of load points. Larger values of the truck influence tolerance tend to produce smaller values of L compared to L. The value of L will be different for each response quantity; it tends to be smaller for structures with simple spans than with continuous spans. The value of L is not sensitive to whether increases in resolution are obtained by increasing the number of elements or the number of output segments.

Computational Considerations
411

393

SAP2000 Analysis Reference

394

Computational Considerations
412

C h a p t e r XIX

The Output Files


Three output text files are produced during a SAP2000 analysis: the input echo file, the analysis log file, and the results output file. Additional output may be obtained using the SAP2000 graphical user interface. Basic Topics for All Users Overview The Input Echo (.EKO) File The Analysis Log (.LOG) File The Results Output (.OUT) File Advanced Topics Element Joint Force Output Joint and Element Output Control Pagination Control

395
413

SAP2000 Analysis Reference

Overview
An analysis performed by SAP2000 produces three output text files. These are: The input echo file, having file extension .EKO; this is a tabulation of the input data text file as read and checked by the analysis The analysis log file, having file extension .LOG; this contains analysis statistics, warnings, and error messages The results output file, having file extension .OUT; this contains displacements, forces, stresses and other results of the analysis When importing a SAP90 or SAP2000 input data text file, only the input echo file is produced. For an input data file named EXAMPLE.SDB or EXAMPLE.S2K, these files would be called EXAMPLE.EKO, EXAMPLE.LOG, and EXAMPLE.OUT, respectively. These are ASCII text files with an 80-character wide output format. Smaller files may be viewed with a text editor such as Windows Notepad, or any other editor or word-processing program in text mode. For larger problems, you may need to use a large-capacity editor or word-processing program. You may use any standard Windows printer to print these files.

The Input Echo (.EKO) File


The input echo file, more simply called the echo file, is created during the datachecking phase of the program. It contains a tabulation of the input data file as read and interpreted by SAP2000, as well as the details of any warnings or errors detected during this phase. The contents of this file are self-explanatory. It is recommended that you carefully review the contents of the echo file for errors, warnings, and the correctness of the model. If errors are present, the analysis phase of the program will not be executed. The model must be corrected and the program executed again. If only warnings are present, the analysis phase may proceed. You should check these warnings carefully before reviewing the results of the analysis. All data that is used by the analysis phase is presented in the input echo file. As such, this file forms a record of the analysis being performed.

396

Overview
414

Chapter XIX The Output Files

The Analysis Log (.LOG) File


The analysis log file, more simply called the log file, is created during the analysis phase of the program. It contains analysis statistics and results summaries, as well as the details of any warnings or errors detected during this phase. It is strongly recommended that you carefully review the contents of the log file for errors, warnings, and the results summaries, including the global force balance relative errors. If errors are present, the analysis phase of the program will have terminated before completion. If only warnings are present, the analysis phase should have completed. In either case, you should check these errors and warnings carefully before examining the results of the analysis. The analysis phase is divided into many sub-phases. The time and date at the beginning and end of each of these sub-phases is reported. The date is given in year/month/day format, and the time is given in hour:minute:second format. The following additional information is provided in the analysis log file: The amount of memory allocated for data storage, in bytes The number of elements formed The number of Welds and Constraints formed The number of equations formed, the size of the stiffness matrix, and the number of loads solved The number of natural frequencies of the model below frequency cyc if harmonic steady-state analysis is performed; see Topic Harmonic Steady-State Analysis (page 294) in Chapter Static and Dynamic Analysis for more information The number of natural frequencies of the model below frequency shift if eigenvector analysis is performed with a frequency shift; see Topic Eigenvector Analysis (page 296) in Chapter Static and Dynamic Analysis for more information For modal analysis, the control parameters and the number of Modes found For eigenvector analysis, the number of iterations performed For Ritz-vector analysis, the number of generation cycles performed and the reasons for discarding the starting load vectors (if any) For response-spectrum analysis, the number of analyses performed

The Analysis Log (.LOG) File


415

397

SAP2000 Analysis Reference


------------------------------------------------------- 1996/02/01 16:36:51.06 MEMORY AVAILABLE FOR DATA (BYTES) = 2000000

------------------------------------------------------- 1996/02/01 16:36:51.17 J O I N T E L E M E N T F O R M A T I O N = = P H A S E 14 0

NUMBER OF JOINT ELEMENTS FORMED NUMBER OF SPRING ELEMENTS FORMED

------------------------------------------------------- 1996/02/01 16:36:51.66 F R A M E E L E M E N T F O R M A T I O N = P H A S E 35

NUMBER OF FRAME

ELEMENTS FORMED

------------------------------------------------------- 1996/02/01 16:36:53.75 E Q U A T I O N F O R M A T I O N A N D = = = = = = = S O L U T I O N 49 8 1 418 3556 49 2 P H A S E

TOTAL NUMBER OF EQUILIBRIUM EQUATIONS APPROXIMATE EFFECTIVE BAND WIDTH NUMBER OF EQUATION STORAGE BLOCKS MAXIMUM BLOCK SIZE (NUMBER OF TERMS) SIZE OF STIFFNESS FILE (BYTES) NUMBER OF EQUATIONS TO REDUCE NUMBER OF LOAD CONDITIONS TO SOLVE

------------------------------------------------------- 1996/02/01 16:36:58.09 J O I N T G L O B A L O U T P U T F O R C E P H A S E B A L A N C E R E L A T I V E E R R O R S

PERCENT FORCE AND MOMENT ERROR AT THE ORIGIN, IN GLOBAL COORDINATES LOAD 1 2 COMB 1 FX .000000 .000000 FX .000000 FY 7.38E-17 9.97E-16 FY 7.38E-17 FZ 4.72E-14 7.98E-15 FZ 4.84E-14 MX 6.62E-15 2.50E-14 MX 1.67E-14 MY .000000 .000000 MY .000000 MZ .000000 .000000 MZ .000000

------------------------------------------------------- 1996/02/01 16:37:02.54 F R A M E E L E M E N T O U T P U T P H A S E = = 35 35

NUMBER OF FRAME NUMBER OF FRAME

ELEMENTS SAVED ELEMENTS PRINTED

------------------------------------------------------- 1996/02/01 16:37:04.62 ...Output Edited for Brevity...

Figure 103 Typical Contents of the Analysis Log (.LOG) File

398

The Analysis Log (.LOG) File


416

Chapter XIX The Output Files For each time-history analysis, the control parameters and the number of time steps completed For each nonlinear time-history analysis, the number of substeps and iterations completed and other statistics For moving-load analysis, the control parameters and solution statistics The global force balance relative errors; see Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom for more information The number of elements processed for output Typical output is shown in Figure 103 (page 398).

The Results Output (.OUT) File


The results output file, more simply called the output file, is created during the analysis phase of the program. The results given for the joints and elements are subject to your control. See Topic Joint and Element Output Control (page 405) in this chapter for more information. The remaining results are always given by the program when appropriate. It is recommended that you carefully review the input echo (.EKO) file and the analysis log (.LOG) file for errors, warnings, and correctness of the model before looking at the results output file. The output file does not contain any warnings or error messages. The results in the output file are divided into various sections, each with a heading that describes the type of results that follow. These sections are described below in the same order that they appear in the output file: The following output section is present if there are any Body, Diaphragm, Plate, Rod, and Beam Constraints that constrain more than two joints: CONSTRAINT COORDINATES AND MASSES See Topic Constraint Output (page 133) in Chapter Constraints and Welds. The following output section is always present: DISPLACEMENT DEGREES OF FREEDOM

The Results Output (.OUT) File


417

399

SAP2000 Analysis Reference See Topic Degree of Freedom Output (page 97) in Chapter Joints and Degrees of Freedom. The following output sections are present if the structure has any mass: ASSEMBLED JOINT MASSES TOTAL UNRESTRAINED MASS AND LOCATION See Topic Joint Mass Output (page 98) in Chapter Joints and Degrees of Freedom. The following output sections are present if an eigenvector or Ritz-vector modal analysis has been performed: MODAL PERIODS AND FREQUENCIES MODAL PARTICIPATION FACTORS MODAL PARTICIPATING MASS RATIOS MODAL LOAD PARTICIPATION RATIOS See Topic Modal Analysis Output (page 303) in Chapter Static and Dynamic Analysis. The following output sections are present if response-spectrum analyses have been performed: RESPONSE SPECTRUM ACCELERATIONS RESPONSE SPECTRUM MODAL AMPLITUDES RESPONSE SPECTRUM MODAL CORRELATIONS RESPONSE SPECTRUM BASE REACTIONS See Topic Response-Spectrum Analysis Output (page 317) in Chapter Static and Dynamic Analysis. The following output section is present whenever joint displacements are requested: JOINT DISPLACEMENTS See Topic Joint Displacement Output (page 100) in Chapter Joints and Degrees of Freedom.

400

The Results Output (.OUT) File


418

Chapter XIX The Output Files The following output sections are present whenever joint applied loads are requested and the corresponding loads exist: APPLIED LOADS INERTIAL LOADS See Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom. The following output sections are present whenever joint reactions are requested and the corresponding types of forces exist: SPRING FORCES NLLINK FORCES RESTRAINT FORCES (REACTIONS) CONSTRAINT FORCES See Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom. The following output section is always present: GLOBAL FORCE BALANCE See Topic Global Force Balance Output (page 107) in Chapter Joints and Degrees of Freedom. The following output sections are present whenever joint forces are requested for the corresponding type of element: FRAME SHELL PLANE SOLID ELEMENT JOINT FORCES ELEMENT JOINT FORCES ELEMENT JOINT FORCES ELEMENT JOINT FORCES

ASOLID ELEMENT JOINT FORCES NLLINK ELEMENT JOINT FORCES See Topic Joint Force Output (page 402) in this chapter. See Topic Joint Force Output (page 179) in Chapter The Frame Element. See Topic Joint Force Output (page 203) in Chapter The Shell Element. See Topic Joint Force Output (page 216) in Chapter The Plane Element.

The Results Output (.OUT) File


419

401

SAP2000 Analysis Reference See Topic Joint Force Output (page 231) in Chapter The Asolid Element. See Topic Joint Force Output (page 241) in Chapter The Solid Element. See Topic Joint Force Output (page 277) in Chapter The Nllink Element. The following output sections are present whenever internal forces or stresses are requested for the corresponding type of element: FRAME ELEMENT INTERNAL FORCES SHELL ELEMENT INTERNAL FORCES & STRESSES PLANE ELEMENT STRESSES ASOLID ELEMENT STRESSES SOLID ELEMENT STRESSES NLLINK ELEMENT INTERNAL FORCES & DEFORMS See Topic Internal Force Output (page 176) in Chapter The Frame Element. See Topic Internal Force and Stress Output (page 198) in Chapter The Shell Element. See Topic Stress Output (page 215) in Chapter The Plane Element. See Topic Stress Output (page 230) in Chapter The Asolid Element. See Topic Stress Output (page 240) in Chapter The Solid Element. See Topic Internal Force and Deformation Output (page 275) in Chapter The Nllink Element. Typical output is shown in the respective topics referenced above.

Element Joint Force Output


The element joint forces are concentrated forces and moments acting at the joints of the element that represent the effect of the rest of the structure upon the element and that cause the deformation of the element. The moments will always be zero for the solid-type elements: Plane, Asolid, and Solid. A positive value of force or moment tends to cause a positive value of translation or rotation of the element along the corresponding joint degree of freedom. See Figure 104 (page 403).

402

Element Joint Force Output


420

Chapter XIX The Output Files

Figure 104 Element Joint Forces when All Joints Have the Global Coordinate System for Their Local Coordinate Systems

Element Joint Force Output


421

403

SAP2000 Analysis Reference Element joint forces must not be confused with internal forces and moments which, like stresses, act within the volume of the element. For a given element, the vector of element joint forces, f, is computed as: f =K u r where K is the element stiffness matrix, u is the vector of element joint displacements, and r is the vector of element applied loads as apportioned to the joints. The applied loads may include Force, Self-Weight, Gravity, Concentrated Span, Distributed Span, Uniform, Surface Pressure, Pore Pressure, Temperature, Prestress, and Rotate Loads, as appropriate, that act directly on the given element. The element joint forces are always referred to the local axes of the individual joints. They are identified in the output as F1, F2, F3, M1, M2, and M3. However, if all joints use the global coordinate system as the local system (the usual situation), then the forces and moments are identified as FX, FY, FZ, MX, MY, and MZ. Element joint forces and moments for the different types of elements are printed in the output file under the headings:
FRAME SHELL PLANE ELEMENT JOINT FORCES ELEMENT JOINT FORCES ELEMENT JOINT FORCES

ASOLID ELEMENT JOINT FORCES SOLID ELEMENT JOINT FORCES

NLLINK ELEMENT JOINT FORCES

All joint forces for a single element are printed together. Elements of a given type are presented in alpha-numeric order. For each Analysis Case the joint forces are printed at all element joints. Typical element joint force output is shown for the Frame element in Figure 105 (page 406), the Shell element in Figure 106 (page 407), the Plane element in Figure 107 (page 408), the Asolid element in Figure 108 (page 409), the Solid element in Figure 109 (page 410), and the Nllink element in Figure 110 (page 411). By default, no element joint forces are printed. You may specify which Analysis Cases are to be printed for each element. Available Analysis Cases include Loads,

404

Element Joint Force Output


422

Chapter XIX The Output Files Modes, Specs, and Combos, but not Histories, Moving Loads, or Combos that contain Histories or Moving Loads. The element joint forces are available through the SAP2000 graphical user interface using the Group Joint Force Sum. This sum is the resultant of the forces and moments acting on one or more joints of one or more elements. It is important to note that the Response Spectrum results are always positive, and that the correspondence between different values has been lost. For more information: See Topic Joint and Element Output Control (page 405) in this chapter Chapter The Frame Element (page 145) Chapter The Shell Element (page 181) Chapter The Plane Element (page 205) Chapter The Asolid Element (page 219) Chapter The Solid Element (page 233) Chapter The Nllink Element (page 243)

Joint and Element Output Control


By default, no joint and element results are written to the results output (.OUT) file. You may select the types of analysis results to be printed for each of the following analysis cases: Load Cases Modes Specs Histories Moving Loads Combos The printed History results are just the maximum and minimum values of each response quantity attained during the History. For the time-variation of the response, use the SAP2000 graphical user interface.

Joint and Element Output Control


423

405

SAP2000 Analysis Reference


F R A M E E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM LOAD JOINT 33 35 SPEC JOINT 33 35 COMB JOINT 33 35 COMB JOINT 33 35 ELEM LOAD JOINT 34 36 SPEC JOINT 34 36 COMB JOINT 34 36 COMB JOINT 34 36 1 ================== 1 -----------------FX -7.35E-13 7.35E-13 FY 9.73E-11 -9.73E-11 FZ 1000.000 -1000.000 MX -72000.000 84000.000 MY -2026.122 2026.122 MZ 1.35E-12 7.47E-12

SPEC -----------------FX 0.972230 0.972230 FY 126.377896 126.377896 FZ 68.323489 68.323489 MX 3190.341 3978.697 MY 89.135262 89.135262 MZ 1.878252 9.788509

BOTH ------------------ MAX FX 0.972230 0.972230 FY FZ 126.377896 1068.323 126.377896 -931.676511 MX -68809.659 87978.697 MY -1936.987 2115.257 MZ 1.878252 9.788509

BOTH ------------------ MIN FX FY -0.972230 -126.377896 -0.972230 -126.377896 2 ================== 1 -----------------FX 7.35E-13 -7.35E-13 FY 9.73E-11 -9.73E-11 FZ 1000.000 -1000.000 MX -72000.000 84000.000 MY 2026.122 -2026.122 MZ -1.35E-12 -7.47E-12 FZ 931.676511 -1068.323 MX -75190.341 80021.303 MY -2115.257 1936.987 MZ -1.878252 -9.788509

SPEC -----------------FX 0.972230 0.972230 FY 126.377896 126.377896 FZ 68.323489 68.323489 MX 3190.341 3978.697 MY 89.135262 89.135262 MZ 1.878252 9.788509

BOTH ------------------ MAX FX 0.972230 0.972230 FY FZ 126.377896 1068.323 126.377896 -931.676511 MX -68809.659 87978.697 MY 2115.257 -1936.987 MZ 1.878252 9.788509

BOTH ------------------ MIN FX FY -0.972230 -126.377896 -0.972230 -126.377896 FZ 931.676511 -1068.323 MX -75190.341 80021.303 MY 1936.987 -2115.257 MZ -1.878252 -9.788509

...Output Edited for Brevity...

Figure 105 Frame Element Joint Forces in the Output File

406

Joint and Element Output Control


424

Chapter XIX The Output Files


S H E L L E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM LOAD JOINT 10 31 25 32 SPEC JOINT 10 31 25 32 COMB JOINT 10 31 25 32 COMB JOINT 10 31 25 32 ELEM LOAD JOINT 31 33 32 34 SPEC JOINT 31 33 32 34 1 ================== 1 -----------------FX 6.28E-27 -1.86E-13 1.49E-27 1.86E-13 FY 9.73E-11 -9.73E-11 9.73E-11 -9.73E-11 FZ 1000.000 -1000.000 1000.000 -1000.000 MX MY -48000.000 9.92E-11 60000.000 -641.113050 -48000.000 -2.86E-10 60000.000 641.113050 MZ 1.31E-12 1.32E-12 -1.31E-12 -1.32E-12

SPEC -----------------FX 0.485709 1.320783 0.484206 1.322272 FY 107.747660 107.748554 107.746818 107.745924 FZ 61.665399 61.665399 61.665399 61.665399 MX 1718.664 2429.634 1718.664 2429.634 MY 6.42E-13 33.426016 1.79E-12 33.426016 MZ 1.818576 1.835782 1.815794 1.831400

BOTH ------------------ MAX FX 0.485709 1.320783 0.484206 1.322272 FY FZ 107.747660 1061.665 107.748554 -938.334601 107.746818 1061.665 107.745924 -938.334601 MX MY -46281.336 9.99E-11 62429.634 -607.687034 -46281.336 -2.85E-10 62429.634 674.539066 MZ 1.818576 1.835782 1.815794 1.831400

BOTH ------------------ MIN FX -0.485709 -1.320783 -0.484206 -1.322272 FY -107.747660 -107.748554 -107.746818 -107.745924 FZ 938.334601 -1061.665 938.334601 -1061.665 MX MY -49718.664 9.86E-11 57570.366 -674.539066 -49718.664 -2.88E-10 57570.366 607.687034 MZ -1.818576 -1.835782 -1.815794 -1.831400

2 ================== 1 -----------------FX 1.86E-13 7.35E-13 -1.86E-13 -7.35E-13 FY 9.73E-11 -9.73E-11 9.73E-11 -9.73E-11 FZ 1000.000 -1000.000 1000.000 -1000.000 MX MY -60000.000 641.113050 72000.000 2026.122 -60000.000 -641.113050 72000.000 -2026.122 MZ -1.32E-12 -1.35E-12 1.32E-12 1.35E-12

SPEC -----------------FX 0.584210 0.235931 0.584210 0.235931 FY 120.062949 120.062949 120.062949 120.062949 FZ 66.239113 66.239113 66.239113 66.239113 MX 2429.634 3190.341 2429.634 3190.341 MY 33.426016 89.135262 33.426016 89.135262 MZ 1.833578 1.878252 1.833578 1.878252

...Output Edited for Brevity...

Figure 106 Shell Element Joint Forces in the Output File Joint and Element Output Control
425

407

SAP2000 Analysis Reference


P L A N E E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM LOAD JOINT 1 2 3 6 7 8 11 12 13 SPEC JOINT 1 2 3 6 7 8 11 12 13 ELEM LOAD JOINT 16 17 18 21 22 23 26 27 28 1 ================== 1 -----------------FX .000000 .000000 0.000108 .000000 .000000 3.09E-18 .000000 .000000 -0.000108 FY FZ -4.04E-11 167.000000 6.17E-11 -1.03E-10 2000.000 -127.672098 -1.08E-10 666.000000 6.90E-11 3.92E-10 4.15E-11 -744.655804 -1.78E-11 167.000000 1.43E-11 -1.35E-10 -2000.000 -127.672098 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

SPEC -----------------FX .000000 .000000 3.00E-06 .000000 .000000 5.36E-07 .000000 .000000 3.00E-06 FY 2.095103 6.736706 45.955743 6.792403 21.085504 24.347802 2.095289 6.737990 45.955810 FZ 1.960365 4.555885 3.475362 6.514757 14.519078 26.751296 1.960390 4.555878 3.475244 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

2 ================== 1 -----------------FX .000000 .000000 -0.000108 .000000 .000000 -3.09E-18 .000000 .000000 0.000108 FY FZ -4.04E-11 167.000000 6.17E-11 -1.03E-10 2000.000 -127.672098 -1.08E-10 666.000000 6.90E-11 3.92E-10 4.15E-11 -744.655804 -1.78E-11 167.000000 1.43E-11 -1.35E-10 -2000.000 -127.672098 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

...Output Edited for Brevity...

Figure 107 Plane Element Joint Forces in the Output File

408

Joint and Element Output Control


426

Chapter XIX The Output Files


A S O L I D E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM 1 ==================

LOAD INTPRES -----------------JOINT 1 2 3 12 13 14 23 24 25 FX 2.93E-14 -4.98E-15 -0.035999 -5.95E-14 -1.96E-13 -0.143192 -1.69E-14 -4.04E-14 -0.035999 FY -0.060041 0.084594 -0.056720 -9.15E-15 -9.98E-14 1.38E-14 0.060041 -0.084594 0.056720 FZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

LOAD LOGTEMP -----------------JOINT 1 2 3 12 13 14 23 24 25 ELEM FX 4.55E-13 1.36E-12 -0.746624 -3.64E-12 -3.75E-12 -3.173884 -1.36E-12 1.34E-12 -0.746624 FY 13.546107 16.192718 12.833193 3.47E-13 -5.54E-13 3.53E-12 -13.546107 -16.192718 -12.833193 FZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

2 ==================

LOAD INTPRES -----------------JOINT 3 4 5 14 15 16 25 26 27 FX 0.035999 -3.80E-14 -0.028017 0.143192 -1.44E-13 -0.110260 0.035999 -9.38E-15 -0.028017 FY -0.099947 0.138916 -0.093595 -3.29E-14 8.16E-14 1.34E-13 0.099947 -0.138916 0.093595 FZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

...Output Edited for Brevity...

Figure 108 Asolid Element Joint Forces in the Output File Joint and Element Output Control
427

409

SAP2000 Analysis Reference


S O L I D E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM LOAD JOINT 3 18 4 19 8 23 9 24 SPEC JOINT 3 18 4 19 8 23 9 24 ELEM LOAD JOINT 8 23 9 24 13 28 14 29 1 ================== 1 -----------------FX 5.27E-12 3.77E-12 2.19E-11 -2.39E-12 -12.997792 12.997792 2.835433 -2.835433 FY -2000.000 -2000.000 3000.000 3000.000 -1755.081 -1755.081 755.081180 755.081180 FZ 127.672098 127.672098 -249.620408 -249.620408 372.327902 372.327902 -250.379592 -250.379592 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

SPEC -----------------FX 0.109772 0.109772 0.141037 0.141037 0.570985 0.570985 0.174693 0.174693 FY 47.939171 47.939171 91.193957 91.193957 59.100845 59.100845 20.588032 20.588032 FZ 5.993277 5.993277 11.462987 11.462987 16.591391 16.591391 11.247946 11.247946 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

2 ================== 1 -----------------FX FY FZ 12.997792 1755.081 372.327902 -12.997792 1755.081 372.327902 -2.835433 -755.081180 -250.379592 2.835433 -755.081180 -250.379592 -1.28E-11 2000.000 127.672098 2.42E-11 2000.000 127.672098 -1.25E-11 -3000.000 -249.620408 -2.08E-11 -3000.000 -249.620408 MX .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MY .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000 MZ .000000 .000000 .000000 .000000 .000000 .000000 .000000 .000000

...Output Edited for Brevity...

Figure 109 Solid Element Joint Forces in the Output File

410

Joint and Element Output Control


428

Chapter XIX The Output Files


N L L I N K E L E M E N T J O I N T F O R C E S

FORCES AND MOMENTS ACTING ON ELEMENTS, IN GLOBAL COORDINATES ELEM MODE JOINT 8 15 MODE JOINT 8 15 ELEM MODE JOINT 5 14 MODE JOINT 5 14 ELEM MODE JOINT 2 13 MODE JOINT 2 13 N1 ================== 1 -----------------FX 7339.063 -7339.063 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

2 -----------------FX 15370.177 -15370.177 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

N2 ================== 1 -----------------FX 5608.591 -5608.591 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

2 -----------------FX -11274.612 11274.612 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

N3 ================== 1 -----------------FX 2499.958 -2499.958 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

2 -----------------FX -17122.498 17122.498 FY .000000 .000000 FZ .000000 .000000 MX .000000 .000000 MY .000000 .000000 MZ .000000 .000000

...Output Edited for Brevity...

Figure 110 Nllink Element Joint Forces in the Output File Joint and Element Output Control
429

411

SAP2000 Analysis Reference The results for Moving Loads and Combos that contain Moving Loads may not be available for all joints and elements. When Moving Load results are printed, the extreme values of each individual response quantity are always given without correspondence, even if corresponding values were calculated. The selection of analysis results for the output file has no affect on the availability of results in the SAP2000 graphical user interface. See Topic Analysis Cases (page 292) in Chapter Static and Dynamic Analysis for more information.

Joints
For each analysis case, you may select the types of joint response to be output using the parameter jtypes, which can take on one or more of the following values: DISP: Displacements APPL: Applied and inertial loads REAC: Restraint, constraint, spring, and Nllink Forces The selected results are printed for all applicable joints. For more information: See Topic Displacement Output (page 100) in Chapter Joints and Degrees of Freedom. See Topic Force Output (page 102) in Chapter Joints and Degrees of Freedom.

Frame and Nllink Elements


For each analysis case, you may select the types of Frame and/or Nllink element response to be output using the parameter frtypes, which can take on one or more of the following values: FORCE: Internal forces (and deformations for the Nllink element) JOINTF: Element joint forces The selected results are printed for all Frame and/or Nllink elements. For more information:

412

Joint and Element Output Control


430

Chapter XIX The Output Files See Topic Internal Force Output (page 176) in Chapter The Frame Element. See Topic Internal Force and Deformation Output (page 275) in Chapter The Nllink Element. See Topic Element Joint Force Output (page 402) in this chapter.

Shell Elements
For each analysis case, you may select the types of Shell element response to be output using the parameter shtypes, which can take on one or more of the following values: FORCE: Internal forces STRESS: Stresses JOINTF: Element joint forces The selected results are printed for all Shell elements. For more information: See Topic Internal Force and Stress Output (page 198) in Chapter The Shell Element. See Topic Element Joint Force Output (page 402) in this chapter.

Plane, Asolid, and Solid Elements


For each analysis case, you may select the types of Plane, Asolid, and/or Solid element response to be output using the parameter pltypes, which can take on one or more of the following values: STRESS: Stresses JOINTF: Element joint forces The selected results are printed for all Plane, Asolid, and/or Solid elements. For more information: See Topic Stress Output (page 215) in Chapter The Plane Element. See Topic Stress Output (page 230) in Chapter The Asolid Element. See Topic Stress Output (page 240) in Chapter The Solid Element.

Joint and Element Output Control


431

413

SAP2000 Analysis Reference See Topic Element Joint Force Output (page 402) in this chapter.

Pagination Control
Pagination control as described in this topic is available only by using the input data text file. You may skip this topic if you are preparing your model using the SAP2000 graphical user interface. The input echo (.EKO) and the results output (.OUT) files are both paginated. A page header is printed at the top of each page that includes the company name, the problem title, the filename, and the page number. A sample page header is shown below:
PROGRAM SAP2000 - VERSION N6.10 NONLINEAR VERSION YOUR PROBLEM TITLE GOES HERE FILE:FRAME.OUT PAGE 1

The analysis log (.LOG) file is not paginated. A single page header appears at the beginning of the file. You may use the parameters page and lines to control pagination as described below.

Pagination by Lines
If you want a header at the top of every physical printed page, specify page = LINES and specify the maximum number of lines per page using the parameter lines. The default value for lines is 59, which is appropriate for many laser printers. A new page will be started at each new section heading and each time the specified number of lines per page is reached. Pagination by lines is most appropriate for printed output files.

Pagination by Sections
For easier viewing of the output files with an editor or word-processor, specify page=SECTIONS. A new page will be started only at each new section heading. The prevents page headers from interrupting the various output sections. The parameter lines is ignored for pagination by sections.

414

Pagination Control
432

C h a p t e r XX

References
AASHTO, 1996 a Standard Specifications for Highways Bridges, Sixteenth Edition, The American Association of State Highway and Transportation Officials, Inc., Washington, D.C. AASHTO, 1996 b LRFD Bridge Design Specifications, Customary U.S. Units, 1996 Interim Revisions, The American Association of State Highway and Transportation Officials, Inc., Washington, D.C. ACI, 1995 Building Code Requirements for Structural Concrete (ACI 318-95) and Commentary (ACI 318R-95), American Concrete Institute, Farmington Hills, Mich. AISC, 1994 Manual of Steel Construction, Load & Resistance Factor Design, 2nd Edition, American Institute of Steel Construction, Chicago, Ill.

415
433

SAP2000 Analysis Reference K. J. Bathe, 1982 Finite Element Procedures in Engineering Analysis, Prentice-Hall, Englewood Cliffs, N.J. K. J. Bathe and E. L. Wilson, 1976 Numerical Methods in Finite Element Analysis, Prentice-Hall, Englewood Cliffs, N.J. K. J. Bathe, E. L. Wilson, and F. E. Peterson, 1974 SAP IV A Structural Analysis Program for Static and Dynamic Response of Linear Systems, Report No. EERC 73-11, Earthquake Engineering Research Center, University of California, Berkeley. J. L. Batoz and M. B. Tahar, 1982 Evaluation of a New Quadrilateral Thin Plate Bending Element, International Journal for Numerical Methods in Engineering, Vol. 18, pp. 16551677. Caltrans, 1995 Bridge Design Specifications Manual, as amended to December 31, 1995, State of California, Department of Transportation, Sacramento, Calif. R. D. Cook, D. S. Malkus, and M. E. Plesha, 1989 Concepts and Applications of Finite Element Analysis, 3rd Edition, John Wiley & Sons, New York, N.Y. R. D. Cook and W. C. Young, 1985 Advanced Mechanics of Materials, Macmillan, New York, N.Y. A. K. Gupta, 1990 Response Spectrum Method in Seismic Analysis and Design of Structures, Blackwell Scientific Publications, Cambridge, Mass. J. P. Hollings and E. L. Wilson, 1977 39 Node Isoparametric Planar or Axisymmetric Finite Element, Report No. UC SESM 78-3, Division of Structural Engineering and Structural Mechanics, University of California, Berkeley.

416
434

Chapter XX References A. Ibrahimbegovic and E. L. Wilson, 1989 Simple Numerical Algorithms for the Mode Superposition Analysis of Linear Structural Systems with Nonproportional Damping, Computers and Structures, Vol. 33, No. 2, pp. 523531. A. Ibrahimbegovic and E. L. Wilson, 1991 A Unified Formulation for Triangular and Quadrilateral Flat Shell Finite Elements with Six Nodal Degrees of Freedom, Communications in Applied Numerical Methods, Vol. 7, pp. 19. L. E. Malvern, 1969 Introduction to the Mechanics of a Continuous Medium, Prentice-Hall, Englewood Cliffs, N.J. S. Nagarajaiah, A. M. Reinhorn, and M. C. Constantinou, 1991 3D-Basis: Nonlinear Dynamic Analysis of Three-Dimensional Base Isolated Structures: Part II, Technical Report NCEER-91-0005, National Center for Earthquake Engineering Research, State University of New York at Buffalo, Buffalo, N. Y. Y. J. Park, Y. K. Wen, and A. H-S. Ang, 1986 Random Vibration of Hysteretic Systems under Bi-Directional Ground Motions, Earthquake Engineering and Structural Dynamics, Vol. 14. R. J. Roark and W. C. Young, 1975 Formulas for Stress and Strain. 5th Edition, McGraw-Hill, New York, N.Y. R. L. Taylor and J. C. Simo, 1985 Bending and Membrane Elements for Analysis of Thick and Thin Shells, Proceedings of the NUMEETA 1985 Conference, Swansea, Wales. K. Terzaghi and R. B. Peck, 1967 Soil Mechanics in Engineering Practice, 2nd Edition, John Wiley & Sons, New York, N.Y. S. Timoshenko and S. Woinowsky-Krieger, 1959 Theory of Plates and Shells, 2nd Edition, McGraw-Hill, New York, N.Y.

417
435

SAP2000 Analysis Reference Y. K. Wen, 1976 Method for Random Vibration of Hysteretic Systems, Journal of the Engineering Mechanics Division, ASCE, Vol. 102, No. EM2. D. W. White and J. F. Hajjar, 1991 Application of Second-Order Elastic Analysis in LRFD: Research to Practice, Engineering Journal, ACI, Vol. 28, No. 4, pp. 133148. E. L. Wilson, 1970 SAP A General Structural Analysis Program, Report No. UC SESM 70-20, Structural Engineering Laboratory, University of California, Berkeley. E. L. Wilson, 1972 SOLID SAP A Static Analysis Program for Three Dimensional Solid Structures, Report No. UC SESM 71-19, Structural Engineering Laboratory, University of California, Berkeley. E. L. Wilson, 1985 A New Method of Dynamic Analysis for Linear and Non-Linear Systems, Finite Elements in Analysis and Design, Vol. 1, pp. 2123. E. L. Wilson, 1993 An Efficient Computational Method for the Base Isolation and Energy Dissipation Analysis of Structural Systems, ATC17-1, Proceedings of the Seminar on Seismic Isolation, Passive Energy Dissipation, and Active Control, Applied Technology Council, Redwood City, Calif. E. L. Wilson, 1997 Three Dimensional Dynamic Analysis of Structures with Emphasis on Earthquake Engineering, Computers and Structures, Inc., Berkeley, Calif. E. L. Wilson and M. R. Button, 1982 Three Dimensional Dynamic Analysis for Multicomponent Earthquake Spectra, Earthquake Engineering and Structural Dynamics, Vol. 10. E. L. Wilson, A. Der Kiureghian, and E. P. Bayo, 1981 A Replacement for the SRSS Method in Seismic Analysis, Earthquake Engineering and Structural Dynamics, Vol. 9.

418
436

Chapter XX References E. L. Wilson and I. J. Tetsuji, 1983 An Eigensolution Strategy for Large Systems, Computers and Structures, Vol. 16. E. L. Wilson, M. W. Yuan, and J. M. Dickens, 1982 Dynamic Analysis by Direct Superposition of Ritz Vectors, Earthquake Engineering and Structural Dynamics, Vol. 10, pp. 813823. V. Zayas and S. Low, 1990 A Simple Pendulum Technique for Achieving Seismic Isolation, Earthquake Spectra, Vol. 6, No. 2. O. C. Zienkiewicz and R. L. Taylor, 1989 The Finite Element Method, 4th Edition, Vol. 1, McGraw-Hill, London. O. C. Zienkiewicz and R. L. Taylor, 1991 The Finite Element Method, 4th Edition, Vol. 2, McGraw-Hill, London.

419
437

438

You might also like