SAP2000 Version 14.2.

2
Copyright Computers and Structures, Inc., 2010

Release Date: 2010-07-30

SAP2000 Version 14.2.2 is an update to Versions 14.0.0 through 14.2.1. It is available as a full
installation on DVD, or from the ftp as a full installation or a patch.

PLEASE READ THIS FILE!

It contains important information about this new release.

Contents
1.
2.
3.
4.
5.
6.
7.

Installation Instructions for SAP2000 Version 14.2.2

Patching from Versions 14.0.0 through 14.2.1
Installation Instructions for License Manager 8.0.5
File Compatibility with Older Versions
Significant Changes from Previous Versions
New Features in SAP2000 Version 14
Important Note for Users of Version 11 and Earlier

1. Installation Instructions for SAP2000 Version 14.2.2

SAP2000 Version 14.2.2 is available as a full installation from DVD or downloaded from the ftp.
Installing SAP2000 Version 14.2.2 will uninstall Versions 14.0.0 through 14.2.1, but will not
uninstall older versions of SAP2000 (Version 12.0.2 and earlier).
The installation instructions are available in three places:
The SAP2000 Installation Card link on the SAP2000 DVD browser.
A printed document included with the program package.
In the file SAP_Install_Instructions.pdf in the root folder of the DVD.
The instructions include information about the software protection system used and installation
options and instructions. You can choose between a Standalone or Network installation.
Silent installation is also available for the Standalone and Workstation installations. Please contact
CSI Technical Support if you need further information.
A new key driver is available for users who have licenses tied to a USB hardware key. For
Standalone installations the new driver is available in the SAP2000 14\KeyDriver\ folder. For

Network installations the new driver is available in the SAP2000 14 Server \License
Manager\KeyDriver\ folder. If your key is not recognized, please install the new key driver.
Windows User Account Control (UAC)
For recent versions of the Windows operating system (Windows 7, Vista, and XP SP3), User
Account Control (UAC) is enabled by default. When UAC is turned on, you must explicitly give
permission to any program, such as SAP2000 or the installer, that wants to use "Administrative"
privileges. Any program that tries to use Administrative privileges without your permission will be
denied access. This has the following implications:

When installing the Standalone version with UAC enabled, you should expect to be prompted
twice to allow the installation of the following components: SAP2000 14 and
Sap2000TrialLicense. Please allow these to continue so that the installation can complete.
When installing the Workstation version with UAC enabled, you should expect to be prompted
three times to allow the installation of the following components: SAP2000 14 WS,
Sap2000TrialLicense, and [Link]. Please allow these to continue so that the installation
can complete.
When updating an existing installation of SAP2000 v14, if your system has UAC enabled and/or
you have limited access rights, use the following procedure for the first time you run the
Standalone or Workstation version of SAP2000: Right-click on the SAP2000 icon and choose
Run as Administrator (or Run as on Windows XP SP3) and run the program as an
administrator. You need only do this one time. Thereafter, the program can be run by doubleclicking the SAP2000 icon as usual.

2. Patching from Versions 14.0.0 through 14.2.1

Whenever possible, full installation is recommended. However a patch is available that can be
downloaded from the ftp and used to modify an existing SAP2000 v14.0.0 through v14.2.1
installation. If this fails for any reason, you should instead download and install the complete
Version 14.2.2, as described in Section 1 above. The patch cannot be applied to versions prior to
14.0.0.
The patch is a compressed zip file that contains only the files that have changed or been added since
Version 14.0.0. Unzip the contents of the patch into the folder where SAP2000 Version 14.0.0
through 14.2.1 is installed, allowing the new files to overwrite the existing files. You will need
administrative rights to do this.
This patch can be applied to a Standalone or Server installation. No change is needed for a
Workstation installation that references a patched Server installation. All Workstation users should
close SAP2000 before patching a Server installation, or the patch will fail.
When SAP2000 is run for the first time after patching, you may need to Run as Administrator if
UAC is enabled on your system. See section Windows User Account Control (UAC) above for
more information.

3. Installation Instructions for License Manager 8.0.5

The information in this section does not apply if you are using standalone licenses.
For new Network installations, see the System Administrator's Help for more detailed information
about the License Manager and the License Manager Administration program [Link].
This installation contains Sentinel RMS License Manager 8.0.5.
If you are currently running Version 8.0.5 or higher of SentinelLM you do not need to reinstall License Manager.
If you are currently running a version that is older than 8.0.5 then you must install version
8.0.5. Uninstall the older version prior to installing this version. IMPORTANT! All
commuter licenses should be checked in before uninstalling the old version. Note that
Sentinel RMS License Manager 8.0.5 will recognize older licenses for Computers and
Structures products. Please refer to the SAP2000 Installation Card or the
SAP2000_Install_Instructions file for information on how to do this.
If you are serving licenses on later versions of the Windows operating system (Vista,
Windows 7, etc.), you will need to use Sentinel RMS License Manager 8.3.0, which is
available from Computers and Structures, Licensing Department, upon request. This version
cannot be locked to a disk ID, but instead requires a computer ID key. Note that version 8.3.0
is not required when the license server is running on older versions of Windows. It does not
matter what operating system is used to run SAP2000.
To speed up finding a network license when SAP2000 is launched, you can do either or both of the
following:
Create a text file, [Link], and enter the ProgramLevel in a single line. The
ProgramLevel should be one of the following:
BASIC/PLUS/ADVANCED/BASICC/PLUSC/ADVANCEDC/ADVANCEDI
Save this file to the folder where SAP2000 is installed. This file will cause the program to
find the license faster. Note that this file will be included when a new license is sent.
Create a text file, [Link], and enter the network name or IP address of the machine
that is running the License Manager. If you are serving licenses on more than one machine,
enter each name or IP address on a separate line of text. Save this file to the folder where
SAP2000 is installed.
If you experience problems with the license please refer to the appropriate License Trouble
Shooting Guide... located in the SAP2000 program folder.

4. File Compatibility with Older Versions

SAP2000 Version 14.2.2 can open model files (*.SDB) from older versions of SAP2000, as well as
import older SAP2000 database files (*.S2K, *.$2K, *.XLS, and *.MDB.) Note that once you save
or run these models in Version 14.2.2, they will not be usable by older versions of the program, so
you should save them under a new name after opening or importing them in V14.2.2.

5. Significant Changes from Previous Versions

We have listed only significant changes here. For a complete list of changes, please see the separate
file [Link] in the SAP2000 folder and available by using the Help > Documentation
command in SAP2000. Note that most changes do not affect most users.
Changes from Version 14.2.1 to Version 14.2.2 (issued 07/30/2010)
Enhancements Implemented
Incidents 17754, 17828, 19447 (Analysis): The sign of all results for hyperstatic load cases
has been reversed at the request of several users. There has been no change in the magnitude
of the results, except as may occur due to the changes under Incident 20599. Hyperstatic load
cases calculate the secondary forces due to the support reactions present in another (parent)
load case. With the new sign, the primary response can be obtained by subtracting the
hyperstatic response from the parent load case. Load combinations containing hyperstatic
load cases present in models saved in versions prior to v14.2.2 should be modified by the
user to account for this change in sign.
Incident 22850 (Frame Design): An enhancement was made, adding the Eurocode National
Annex parameters for Eurocode 3-2005 steel frame design and Eurocode 2-2004 concrete
frame design, for Sweden, Finland, and Denmark.
Incident 24495 (Frame Design): Steel-frame design has been implemented for the Australian
design code "AS 4100-1998". This covers the analysis methods and checking of members
subjected to bending, axial compression, axial tension, and combined actions.
Incidents Resolved
Incident 22732 (Modeling): An Incident was resolved in which the interaction surface for
auto-generated steel FEMA hinges was incorrect near the compression limit. The moment
capacity did not decrease with increased magnitude of compressive axial force as would be
expected.
Incident 26180 (Modeling): An Incident was resolved that fixed two problems affecting
built-up cover-plate frame section properties that were imported from database tables
(including .S2K and .$2K files): (1) The section properties (areas, moments of inertia, etc.)
were not properly calculated, but were obtained from the base section alone without cover
plates. This problem also affected built-up cover-plate sections defined through the Open
API. Properties would be correctly recalculated if the frame section property was viewed in
the graphical user interface and the OK button was clicked on the form after viewing. (2) The
values of the fy overwrites, when specified, were truncated to the nearest whole value (e.g.,
56.7 would be truncated to 56). Neither of these problems would occur if the frame section
property was directly defined in the graphical user interface.
Incident 26233 (Modeling): An incident was resolved where the width values for
HSS16x4x3/16 and HSS14x6x5/8 sections in the [Link] file were incorrect. Section
properties (i.e. area, moment of inertia, etc) were correct. Design of members using these
section sizes will have used the incorrect width values and should be rechecked.
Incident 26373 (Modeling): An Incident was resolved in which the built-up coverplate frame
section property was not properly accounting for different material properties when
calculating certain section properties: (1) The minor moment of inertia I22, minor section

modulus S22, and minor radius of gyration r22 were not corrected for the ratio of E values.
(2) The torsional constant J and the minor shear area As3 were not correctly considering the
ratio of G values. (3) The plastic moduli Z22 and Z33 were not correctly considering the
ratio of fy values. (4) However, the area A, major moment of inertia I33, major section
modulus S33, major radius of gyration r33, and major shear area As2 were all correct.
Incident 26126 (Section Designer): An incident was resolved in which the mass and weight
calculated for certain Section Designer sections could be incorrect, even negative. This
happened in rare cases where overlapping objects were drawn of different materials or an
opening was drawn in a particular order. There was no error in the normal case where
included objects were drawn after the containing object was drawn.
Incidents 24730 (Bridge Modeler): An Incident was resolved for the bridge modeler in
which, for steel-girder or precast-concrete-girder deck sections, setting the fillet dimension f1
and the corresponding overhang L3 both to zero, or setting the fillet dimension f2 and the
corresponding overhang L4 both to zero, could result in an invalid linked model being
created for the top deck slab. In some cases the edge would be disconnected from rest of the
deck, and in other cases duplicated elements would be created that made the deck overly
stiff. Now restrictions are placed on the overhangs so that they remain outside the exterior
girder widths.
Incidents 24783 (Bridge Modeler): An Incident was resolved in which bridge temperature
loading on steel I-girder bridge deck sections was not being applied during analysis in any
bridge span that had skewed supports and/or non-prismatic girder sections when the steel
girders were modeled as area objects. This did not affect spine models or area models of the
deck when the steel girders were modeled as frames.
Incident 21895 (Analysis): An Incident was resolved in which the P-delta moments
calculated for link elements sometimes included the effect of joint rotations. This did not
affect the results for two-joint links with non-zero length when the P-delta moment was
carried with a shear couple (the default case). It did affect the following cases: (1) One-joint
link supports that carried some or all of the P-delta moment at the j-end (the default is for
half of the moment to be at the j-end). (2) Zero-length two-joint elements, which always
carry the P-delta moment at one or both ends. (3) Two-joint link elements with non-zero
length if the advanced P-delta parameters were specified to carry some or all of the P-delta
moment as moments rather than a shear couple.
Incident 23251 (Analysis): An Incident was resolved in which creep for transverse shear
strain was not being correctly calculated for the bending behavior of thick plates and shells,
causing time-dependent staged-construction analyses to give obviously inaccurate results or
to fail to converge. Membrane (in-plane) behavior was not affected. The thin plate/shell
formulation for bending was not affected. The problem was introduced with version 14.2.0,
and did not affect prior versions.
Incident 24373 (Analysis): An Incident was resolved in which temperature and membranestrain loading could be significantly in error when a shell element was warped, i.e., it did not
lie in a flat plane. The induced displacements, or stresses in a restrained case, could have
been significantly too large. This affected the thin and thick homogenous shell elements, but
not the layered shell. Gravity, self-weight, uniform, temperature-gradient (through the
thickness), and curvature-strain loads were not affected. Surface-pressure loads acting on the
edges of the element were affected, but not the more common case of surface-pressure
loading on the top or bottom faces. This issue was introduced in v14.0.0 and did not affect
previous versions.

Incident 25328 (Analysis): An Incident was resolved that fixed several problems with the
results for hyperstatic load cases. These tended to particularly affect bridge models: (1) The
support reactions that constitute the loading in a hyperstatic load case were not being
converted from joint local coordinates to the global coordinate system. This affected any
support having joint local axes, as is common in many bridge models. (2) The small internal
stiffnesses used to stabilize the structure during a hyperstatic analysis were too large in some
models, thereby reducing the hyperstatic response. This tended to affect models where large
stiffness values were used for supports, as is common in bridge models. (3) Joint reactions
and base reactions reported for hyperstatic load cases were and are not meaningful, and are
now set to zero. (4) Note that the sign of the results for hyperstatic load cases has been
reversed at the request of several users. This latter change is further described under Incident
17828.
Incident 23582 (Frame Design): An incident was resolved for steel-frame design in which
the design details sheet accessed by right-button click for auto-select sections was presenting
design results for the current analysis section instead of the current design section when the
two sections were of different sizes. The tables accessed directly from the right-button click
results also had the same problem. However, database tables accessed from Display > Show
Tables were correct, as were the tables for export and for printing. This problem was
introduced in version 14.0.0.
Incident 24384 (Frame Design): An Incident was resolved for steel frame design using code
Eurocode 3-2005 in which the calculated values of [Link] and [Link] were not correct.
These were being computed as if the slenderness (Kl/r) was too small, which led to using full
plastic capacity. This could have resulted in an unconservative design.
Incident 26649 (Frame Design): An incident was resolved where large demand/capacity
ratios were being reported for the Eurocode 3-2005 steel frame design when using Method 1
(Annex A) for computation of the K factors. This could occur on members with end moments
and would result in overly conservative results.
Incidents 23401 (Bridge Design): An Incident was resolved for bridge design of multi-cell
prestressed concrete box girder superstructures where the live-load distribution factors
(LLDF) calculated according to code equations were incorrect as follows: (1) The factor for
interior beam moment for two or more lanes loaded was too small by a factor of up to 4.18,
and (2) The skew correction factors for interior and exterior beam moment and shear were
incorrectly calculated for any non-zero skew angle. Flexure and shear design checks using
code-based LLDF for multi-cell box girder superstructures that were run with versions 14.0.0
to 14.2.0 should be re-checked with the new version. Design checks for composite precastconcrete I- and U-girder superstructures were not affected. Design checks that did not use
code based LLDF were not affected. Design checks for the segmental (non-multicell)
concrete box girder do not use LLDF and are not affected.
Incident 23700 (Bridge Design): An Incident was resolved for bridge superstructure design
of multicell concrete box-girder deck sections and precast-concrete girder deck sections in
which the stress-check results were incorrect when the design check was performed while the
current length units shown in the graphical user interface were different from the database
length units (the database units are the units chosen when the model was created). The effect
of the different length units was to affect the location of the points used to calculate the
stresses in the section. This did not affect any other design checks, nor did it affect the stress
or principal-stress checks for the non-multicell box girder.

Incident 23703 (Bridge Design): An Incident was resolved for bridge design of precastconcrete girder deck sections in which the stress-check results for the girders were incorrect
if the concrete material used for a girder had a different modulus of elasticity (Egirder) from
that for the slab (Eslab). The reported girder stresses were incorrectly multiplied by the
square of the ratio Egirder/Eslab when the live-load distribution factors (LLDF) are
calculated from analysis (SAP), and were incorrectly multiplied by the ratio Egirder/Eslab
otherwise. For the common case where EGirder > Eslab, the reported stress values were too
large. Design stresses reported for the slab were correct for both cases.
Incident 23816 and 25220 (Results Display and Output): An Incident was resolved where
section cuts specified using a quadrilateral cutting plane were reporting the results based on
the nodes nearest to the cut and were ignoring the side specification. This could result in
incorrect values being reported. This issue was inadvertently introduced in v14.2.0 and did
not affect prior versions. The section cut algorithm for quadrilateral cutting planes has also
been enhanced to capture discontinuities. Using a cutting plane at a discontinuity will now
report different results based on which "side" is chosen. This is achieved by moving the
cutting plane back by a tolerance if the positive "side" is chosen and forward by a tolerance if
the negative "side" is chosen. The automerge tolerance specified for the model is used for
this purpose.
Incident 24909 (Results Display and Output): An Incident was resolved in which the bridge
response (superstructure forces, moments, and stresses) for steel I-girder deck sections could
be incorrect in the case where all of the following are true: (1) The steel girders are
nonprismatic, (2) The steel girders are modeled as areas, and (3) The bridge object is updated
as a linked model more than once without changing geometry. When a duplicate linked
model was created, the elements were being replaced with identical elements, but the new
elements did not get added to the section cuts used to calculate bridge response.
Other minor Incidents as detailed in [Link].

Changes from Version 14.2.0 to Version 14.2.1 (issued 03/31/2010)

Note: Version 14.2.1 is identical to previously released version 14.2.0, except for minor changes
made strictly for the Chinese release that do not affect any other users.
Incidents Resolved
Incident 20270 (Bridge Design): An Incident was resolved in which the superstructure bridge
design check for flexure using the JTG design code would not run.
Other minor Incidents as detailed in [Link].
Changes from Version 14.1.0 to Version 14.2.0 (issued 03/22/2010)
Enhancements Implemented
Incident 22674 (Analysis): The CQC3 method of directional combination has been added as
an option for response-spectrum load cases. When identical spectra are specified with
different scale factors for loading in the two horizontal directions, this method determines the
critical angle in the horizontal plane for which this loading will produce the maximum
response, and reports that response. If the scale factors are identical, this reduces to the SRSS
method of directional combination, which is independent of the angle of loading. The CQC3

directional combination can be used in conjunction with any method of modal combination,
and both the amplified and the rigid response are considered. The CQC3 methodology can
also be applied using different spectra in the two horizontal directions, and a critical response
will be calculated, however the user must carefully interpret the meaning of the results.
Incident 19436 (Frame Design): Eurocode 8 seismic provisions have been comprehensively
implemented for concrete frame design using code Eurocode 2-2004.
Incident 19437 (Frame Design): Eurocode 8 seismic provisions have been comprehensively
implemented for steel frame design using code Eurocode 3-2005.
Incident 20661 (Frame Design): The Singapore National Annex to Eurocode 2 (NA to SS EN
1992-1-1 : 2008) has been implemented for concrete frame design using Eurocode 2-2004.
Incident 20800 (Frame Design): Concrete frame design using code "ACI 318-08" has been
implemented.
Incident 22232 (External Import/Export): A new option to export a Perform3D Structure has
been added. Most geometry, material and section properties, and loads are exported for frame
and shell elements. Corresponding Perform3D members are created depending on the
properties and assignments present on the individual SAP2000 members. Detailed hinge
properties and other Perform3D-specific features must be defined in Perform3D. Restraints,
diaphragm constraints, groups, and basic load cases are also exported. The export option is
available after the SAP2000 analysis model has been created.
Other minor enhancements as detailed in [Link].

Incidents Resolved
Note: We have performed a thorough audit of the program, and have resolved all known
Incidents for which we have been able to determine that the program is not behaving as intended.
Incidents 18802 and 19448 (Modeling): An Incident was resolved in which there could be a
small error in the calculated section properties for the precast U-girder frame section.
Approximately half of the left-side upper flange was omitted from the shape if there was
flange present.
Incident 20852 (Section Designer): An Incident was resolved in which the PMM surface
calculated in Section Designer for a concrete-filled pipe section was not always symmetrical,
and did not always close properly at the bottom (pure-tension) side of the surface.
Incident 17520 (Bridge Modeler): An Incident was resolved for the Bridge Modeler in which
the weight of cross diaphragms was not being included in bridge objects that were updated as
spine (frame object) models. This affected self-weight and gravity loads. The weight of the
diaphragms was being included in bridge objects updated as area and solid objects. The mass
of the diaphragms was being included for inertial purposes in all linked models, including
spine models. The weight of the cross diaphragms in spine models is now being modeled as
zero length link elements at the cross-diaphragm locations.
Incidents 18659 and 19235 (Bridge Modeler): An Incident was resolved in which the frame
section properties generated for the bridge superstructure in spine models sometimes had a
zero value for the torsional constant J. This occured when the section was large or detailed,
and memory limitations prevented calculation of the constant J from completing. This
problem was introduced in SAP2000 V14.1.0 when more accurate meshing for torsion was
implemented. Now a more efficient method is utilized that achieves greater speed but uses
less memory. The value of J is calculated when the analysis model is generated. This means

that J values for bridge deck sections may be reported in database tables as zero prior to
running the analysis. After the analysis, the correct values will be shown in the tables.
Incidents 18804 and others (Bridge Modeler): An Incident was resolved for the Bridge
Modeler in which an error was generated when updating the linked model for a steel I-girder
deck section with girders modeled as areas and having beam-type cross diaphragms.
Incident 12794 (Loading): An Incident was resolved in which projected loads on frame
objects defined in alternate coordinate systems were not correctly applied or displayed.
Loads were transformed to the global system, then projected on the element using global load
components. This could result in the load being applied with a somewhat different direction
and magnitude than intended. Loads applied in the global or element local coordinates
systems were not affected. Non-projected loads applied in alternate coordinate systems were
not affected. This error was introduced in v12.0.0. Models having this type of loading should
be re-verified.
Incident 18310 (Loading): An Incident was resolved in which uniform loads applied to shell
elements in the Global or other fixed coordinate systems were not always being properly
rotated into the element local system, possibly resulting in the load being applied in the
wrong direction. This did not affect self-weight, gravity, or pressure loads, and it did not
affect uniform loads applied directly in the element local coordinate system. This error was
introduced in v12.0.0. Models having this type of loading should be re-verified.
Incident 18572 (Loading): An incident was resolved where the uniformly distributed shell
loads were not being fully applied in certain rare cases. This only occurred when the shell
uniform load was used for the generation of notional loads and the area object was further
meshed. The first element of the meshed area object was loaded, but not the remaining
elements for that object. This affected Versions 12.0.0 to 14.1.0. Models satisfying these
conditions should be reevaluated.
Incident 18583 (Loading): An Incident was resolved in which joint displacement loads
assigned to a load pattern were not applied in a given load case if that load pattern was listed
after any acceleration load in the list of applied load patterns for that load case. Displacement
loads were correctly applied in load patterns listed before any acceleration load, or if there
were no acceleration loads applied in the load case.
Incident 20091 (Loading): An Incident was resolved in which deleting a load pattern could,
in rare cases, delete load assignments to objects that are part of another load pattern. This
only affected deformation and target-force loads that were assigned to frame/cable/tendon
objects, and only for assignments to load patterns that were defined after the deleted load
pattern. This did not affect any other type of object, or any other type of force, strain, or
temperature load.
Incidents 16145 and others (Loading): An Incident was resolved in which tendons that are
modeled as loads did not always properly load the structure where one end of a discretized
tendon segment was not inside any bounding box. This could create unbalanced loads,
causing large displacements and/or reactions. Now, when any part of a discretized tendon
segment falls inside any bounding box, self-equilibrating loads are applied to the structure.
No load is applied for any discretized segment that lies entirely outside all bounding boxes.
Incident 16015 (Analysis): An Incident was resolved in which the structural period of a
model could change when the local axes of a joint that was connected to a link element with
fixed degrees-of-freedom are rotated. Rotating the joint local axes from the global system
couples the constraint equations, and that coupling was not always properly handled. This
has been corrected. However, the period may still be affected by rotating joint local axes if

fewer than all six link degrees-of-freedom are fixed, due to loss of some off-diagonal mass
terms. This can be avoided by using stiff rather than fixed link degrees-of-freedom in that
case.
Incident 16264 (Analysis): An Incident was resolved for influence-based moving load
analysis in which the reactions at a restrained joint did not include the influence of loads
applied directly at that joint. This only affected models in which the loaded surface was
modeled with shell or solid elements and the restraints were connected directly to the loaded
surface, hence most models created by the Bridge Modeler were not affected. Only the
restraint reactions themselves were affected. All other response quantities were correct,
including displacements, forces, stresses, reactions at spring supports, and reactions at
restraints due to loads not at the restraint itself.
Incident 17649 (Analysis): An Incident was resolved in which joint displacements for a
moving load case are reported incorrectly if the joint has a rotated local coordinate system.
This has no effect on joints that have their local coordinate system equal to the global system,
which is the usual and default case.
Incident 18365 (Analysis): An Incident was resolved in which joint reactions for an
influence-based moving load case did not include any contribution from connected solid or
plane elements. Force transmitted to the restraints from frame, shell, and link elements were
correct.
Incident 18949 (Analysis): An Incident was resolved in which thermal loading of solid
elements in nonlinear static and nonlinear direct-integration time-history load cases did not
produce the correct stresses. The response of the rest of the structure was correct, including
displacements, reactions, and forces/stresses in other elements. Only the stresses in the
thermally loaded solids were incorrectly reported.
Incident 19433 (Analysis): An Incident was partially resolved in which frame PMM hinges
sometimes exhibited excessive hardening when the stress point was near to the tension region
of the interaction surface. Further improvements will be made in a future release. Note that it
is not expected that the moment-rotation behavior will exactly follow the backbone curve
when the load point on the PMM surface moves from the point of initial yield.
Incident 19522 (Analysis): An Incident was resolved for the Bridge Modeler in which the
reported superstructure forces, moments, and stresses were, in certain cases, much too large
at the abutments. This could occur when edge constraints were created at the abutments to
handle mismatched meshes between the deck and the diaphragms. This was a localized
reporting error only and did not affect the behavior or results in the rest of the structure.
Incident 22452 (Analysis): An Incident was resolved in which the effect of temperature and
strain load was not being included in linear buckling analysis for shell and solid elements,
and the effect of temperature load was not being included in the nonlinear P-delta effects for
solid elements.
Incidents 14485 and 18785 (Analysis): An Incident was resolved in which the incremental Pdelta effect in frame elements was not always correct for nonlinear static and nonlinear
direct-integration time-history load cases. The error could be significant in an increment
when the axial load was changing and large lateral displacements were already acting on the
element. This was of primary concern for nonlinear time-history load cases, and for some
nonlinear static cases where many increments were used. Most nonlinear static load cases
used for design apply the load in one step and were not affected. Linear load cases using the
stiffness from a nonlinear P-delta case were not affected. Large-displacement analysis was

not affected. See Incident 21284 for a related issue for shell elements. Solid, link, and other
elements were not affected.
Incidents 17292 et al (Analysis): An Incident was resolved in which the mass value, when
computed from loads used as the mass source, had double the mass value at the third joint in
triangular area elements (shell, plane, and asolid). This would lead to 33% extra total mass
for each triangular element, and the acceleration loads were correspondingly increased.
Similarly, when solid elements were collapsed to create 5, 6, or 7-noded elements, the mass
values at the collapsed joints were duplicated when mass was computed from loads. In the
most common, default case where mass was computed from material mass density, there was
no duplication of mass or acceleration load.
Incidents 17401 (Analysis): An Incident was resolved in which load applied in a userdefined coordinate system is zero if that coordinate system is identical to the global
coordinate system. This does not affect the majority of cases where load is applied in element
local coordinate systems, the global system, or user-defined systems that are different from
global.
Incidents 18808 et al (Analysis): An Incident was resolved in which the base reactions were
sometimes incorrect when restrained joints were present with local coordinates different from
the global directions.
Incident 21284 (Analysis): An Incident was resolved in which the incremental P-delta effect
in shell elements was not always correct for nonlinear static and nonlinear direct-integration
time-history load cases. The error could be significant in an increment when the axial load
was changing and large lateral displacements were already acting on the element. This was of
primary concern for nonlinear time-history load cases, and for some nonlinear static cases
where many increments were used. Most nonlinear static load cases used for design apply the
load in one step and were not affected. Linear load cases using the stiffness from a nonlinear
P-delta case were not affected. Large-displacement analysis was not affected. See Incident
14485 for a related issue for frame elements. Solid, link, and other elements were not
affected.
Incident 12843 (Frame Design): An Incident was resolved for steel frame design using code
BS5950 2000 and "BS5950 90" in which the incorrect strut curve was being used for the
axial-capacity calculation for an H-section (an I-Section with depth/width < 1.2
(BS5959:2000 1.3.23, BS5950:1990 1.2.22)). H-sections and I-sections are assigned with
different sets (major and minor) of strut curves (BS5959:2000 Table 23, BS5950:1990 Table
25), and hence, different Robertson constants (BS5959 App. C) are used for the axial
capacity calculation. Capacity of H-sections were taken unconservatively higher. I-sections
were correct.
Incident 17250 (Frame Design): An Incident was resolved for steel frame design in which
the deflection reported for cantilevered members was not the tip deflection, but rather was
the in-span deflection relative to the chord, the same as for members supported at both ends.
Now the effect of the rotation at the cantilever support is considered. Only the results of
cantilever beams were affected. All steel frame design codes are affected by this change.
Incident 18840 (Frame Design): An Incident was resolved for steel frame design using code
AISC 360-05/IBC2006 in which the stress-based interaction equation H2-1 was not being
used as required for sections which are not doubly symmetric and for which Iyc/Iy is outside
of the range 0.1 to 0.9. This only affects the singly symmetric I-sections with Iyc/Iy outside
of the range 0.1 to 0.9, all T-sections, all double-angle sections with Iyc/Iy outside of the
range, and all equal-legged angle sections. For singly symmetric I-sections, the old

implementation was slightly unconservative with a maximum factor of 8/9. For T-sections,
double-angle sections, and equal-legged angle sections, old implementation was sometimes
unconservative with a maximum factor of 8/9, sometimes conservative, and sometimes
matching.
Incident 18969 (Frame Design): An Incident was resolved for concrete frame design using
the Eurocode 2-2004 design code in which the alpha-cc parameter specified in the design
preferences was not working. The alpha-cc was not connected to the interaction surface
generation.
Incident 19210 (Frame Design): An Incident was resolved for steel frame design using the
Eurocode 3-2005 design code in which the Nc,Rd values were incorrectly computed. The
calculation of the non-dimensional slenderness (lambda_bar) was previously using
(pi)^2*E*I/L^2 rather than Cl. [Link](1) from the code.
Incident 19303 (Frame Design): An Incident was resolved for concrete column design using
any design code in which the PMM interaction surface used for design was not being updated
when design preferences were changed after the first time design was performed following
an analysis. Unlocking the model and re-running the analysis would update the PMM surface
correctly. Now the PMM surface is updated every time a design preference is changed, even
without re-running the analysis.
Incident 19304 (Frame Design): An Incident was resolved for concrete frame design using
the Eurocode 2-2004 code in which the calculated imperfection inclinations (theta) were
not correct if the base length unit for the model is anything other than meter. Since the
alpha_h parameter is always taken in between 2/3 and 1, the effect of this error was limited.
If the base length unit was mm, cm, or inch, the resulting theta was unconservative.
Incident 19528 (Frame Design): An Incident was resolved for steel frame design using the
"Chinese 2002" code in which the design was still using the gross area for design under
clause 5.2.1 even when the "Gross Area / Net Area" design overwrite was specified. This has
been corrected so that now the net area is used when this ratio is specified.
Incident 20248 (Frame Design): An Incident was resolved for steel frame design using
"Eurocode 3-2005" in which the calculation of the interaction ratios could be incorrect in the
case where lateral torsional buckling governs per equations 6.61 or 6.62. The value of lamLT
was being taken as zero for these equations. The previously reported result could be
unconservative. For other capacity checks, the correct value of lamLT was being used.
Incident 22646 (Frame Design): An Incident was resolved that corrected two minor issues
for steel frame design using "Eurocode 3-2005": (1) The Kl/r for angle sections was
calculated along the axes parallel to the legs of the angle rather than the principal axes. This
may unconservatively affect the axial capacity and slenderness check in some cases. (2)
Members with very small axial load were being treated as being of Class 4, and therefore not
checked or designed. Now such members may be treated as being of Class 1 if the small axial
force is compressive, and therefore they will be checked or designed.
Incident 18799 (Bridge Design): An Incident was resolved for bridge superstructure design
in which the vertical location of the three stress points at the top of prestressed concrete Igirders and U-girders was in error by an amount equal to the top flange thickness. This only
affected the case where the spine model was used.
Incident 18889 (Bridge Design): An Incident was resolved in which the tabulated stress
results for the "AASHTO STD 2002" bridge design stress check for prestressed concrete box
girder superstructures were not being scaled by the corresponding demand-set limit factor.
For each demand set, the reported stress should be divided by the limit factor so that the

controlling stress can be compared between the various demand sets to determine the
controlling value. This was properly done during the design check and for the plotted results.
Only the tabulated results were not being scaled.
Incident 19494 (Bridge Design): An Incident was resolved for bridge design using the
AASHTO stress check of prestressed concrete box-girder or precast composite-girder
superstructures in which the reported stress values were not scaled for the live load
distribution factor (LLDF). This affected that portion of the stress coming from the live load,
not the portion of the stress from dead or other types of loads.
Incident 21013 (Bridge Design): An Incident was resolved in which the stresses reported for
bridge design stress-checks may be incorrect for unsymmetrical girders when the LLDF
(live-load distribution factors) are computed by analysis. The effect of the product of inertia
I23 was not being included. This affected the slabs of precast girder composite sections, as
well as the exterior girders of multi-cell concrete box sections. This did not affect stress
checks when LLDF were code-calculated, user-specified, or uniform. This also did not affect
the concrete box stress-checks when LLDF are not used.
Incident 21014 (Bridge Design): An Incident was resolved in which the stresses reported for
bridge design stress-checks may be incorrect for precast-composite and multi-celled box
girders when using LLDF (Live-Load Distribution Factors). The moments used to compute
the stresses were being computed at the girder output point rather than at the centroid of the
girder where the moments of inertia were computed. This also did not affect the concrete box
stress-checks when LLDF were not being used.
Incidents 21008 et al (Bridge Design): An Incident was resolved for bridge seismic design in
which the automated determination of the SDC (seismic design category) was not always
correct for AASHTO2007, IBC2006, and NCHRP2007 response-spectrum curves. The value
S1 was being used to determine the SDC rather than the correct value, SD1 = Fv * S1. For
other codes and user-defined functions, the correct value of SD1 was being obtained directly
from the curve, so there was no error.
Incidents 10962 et al (Results Display and Output): An Incident was resolved in which loads
applied to joints, solids, links, plane, and asolid elements were not included in the restraint
reactions at joints connected directly to those elements for nonlinear static and nonlinear
direct-integration time-history analysis. Note that this only applies to the load applied
directly to those elements connected to the restraints, not to load applied elsewhere in the
structure. This did not affect frame or shell elements. This did not affect the reactions
reported at spring supports or one-joint links. This was a reporting issue for restraint
reactions and base reactions only and did not affect displacements, element stresses or forces,
or any other response quantity. Because of these changes, the base shear reaction reported for
the nonlinear direct-integration time-history load case in verification Example 6-010, "SUNY
Buffalo Eight-Story Building with Rubber Isolators", differs by 0.39005% from the
published results for Version 14.0.0. This is a slight improvement since the mass of the
isolators is now accounted in the reaction, and is considered acceptable.
Incident 13732 (Results Display and Output): An Incident was resolved in which load
combinations of the additive type did not give correct results when multi-step static load
cases were included. This did not affect other types of load cases and other types of load
combinations, such as the enveloping type.
Incidents 17479 and 17605 (Results Display and Output): An Incident was resolved in which
the element joint forces reported in database tables were always given in the global
coordinate system rather than in the joint local coordinate system.

Incident 18568 (Results Display and Output): An Incident was resolved in which the material
take-off for frame section properties was not always correct when there were cables present
in the model.
Incident 18892 (Results Display and Output): An Incident was resolved in which plot
functions for shell, plane and solid stresses were not correctly accounting for non-zero offsets
specified in the plot-function definition. Note that if offsets are specified that fall outside of
the object, results will be plotted for the nearest point element that is part of the analysis
model for that object.
Incident 19306 (Results Display and Output): An Incident was resolved in which the stresses
displayed for the bridge superstructure were incorrect when the entire section was not present
during staged construction. Now stresses are reported separately for the beam and slab of
composite deck sections (using steel or precast-concrete beams), but not for the composite
girder (combined beam plus slab). Furthermore, these beam and slab stresses are computed
on a girder-by-girder basis rather than for the entire deck section. Stresses for the beam and
slab of composite sections are only available when the linked model is udpated as areas or
solids, not as a spine model. For integral deck sections (concrete box girder, flat slab, teebeam), stresses are computed on the entire section or the entire girder, and do not account for
partial sections during staged construction.
Incident 22110 (Results Display and Output): An incident was resolved where the vertical
coordinate of the girder force output points was not correctly calculated for steel I and U
girder bridges based on the user input. This affected the moment M3 and torsion T reported
for individual composite girders using the command Display > Show Bridge Forces/Stresses.
It did not affect the moments for the overall section, or any other force or moment for the
individual girders. This was a reporting issue only. Design results were not affected.
Incident 22202 (Open API): An Incident was resolved in which the following OpenAPI
functions in CPointObj were not converting length from current units to model database
units: AddCartesian, AddCylindrical, and AddSpherical. The given lengths were being
assumed to be in database units. Now they are treated as being current units and are
converted to database units.
Other minor Incidents as detailed in [Link].

Changes from Version 14.0.0 to Version 14.1.0 (issued 07/29/2009)

Enhancements Implemented
Incident 17537 (User Interface and Display): The Rotate 3D View command has been
enhanced to rotate about the center coordinate of the current view, accounting for zoom and
pan. Previously the rotation was always the center of the structure for perspective views, and
about the origin for orthographic views.
Incident 18152 (Modeling): Automated wind lateral loading for Australia and New Zealand
has been implemented according to the AS/NZS 1170.2 2002 code.
Incident 18523 (Bridge Modeler): The U-girder frame section can now be used to model
steel tubs in the Bridge Modeler. When a steel material is chosen for the frame section, the
generated linked bridge model will treat the alignment of the top and bottom flanges
similarly to how they are treated for steel I-girders: The haunch distance is measured to the
bottom of the top steel flange, rather than to the top of the top flange as is the case for
concrete U-girders. One or two bearings will be created for each steel U-girder depending on

the definition of the frame section. A single bearing is used for concrete sections. Significant
further specializations for steel U-girders are under current development.
Incident 17972 (Design): AASHTO LRFD 1997 concrete frame design, which had been
removed with Version 12, has now been reinstated after updating it to AASHTO LRFD
2007.
Incident 18008 (Design): Steel frame design has been added for NORSOK N-004,
including punching checks. The design of non-tubular sections is in accordance with
Eurocode 3-2005 with the Norwegian National Annex. The SAP2000 Offshore add-on
license is required to use this feature.
Incident 17922 (Bridge Design): Detailed design checking has been implemented for multicell prestressed concrete box-girder bridge superstructures according to the AASHTO
LRFD 2007 code. Separate design checks are available for stress, flexural capacity, and
shear capacity using MCFT (modified compression field theory). Live-load distribution
factors can be automatically calculated using code formulae, specified by the user, or
determined from detailed 3-D live-load analysis. Results are reported for each girder, and
include plots of stress, moment demand and capacity, shear demand and capacity, and shearrebar requirements. Detailed tables showing all results and intermediate values are available
for display, printing, and export to Excel or Access. These new checks are in addition to the
whole-section checks that were previously released.
Incident 18280 (Database Tables): The Floating Point Number Format and Units
settings have both been removed from the database table format file. Now all displayed
tables, printed tables, and tables in formatted reports will use the corresponding settings that
are controlled using the menu command Options > Set Program Default Display Units. This
makes it easier to format the tables and provides more consistent formatting.
Incident 18155 (External Import/Export): The capability to export a model to SAFE V12 has
been implemented. A single global-Z elevation is chosen, and all objects connected to joints
within a tolerance of that elevation are exported. Linear static and response-spectrum load
cases and associated data are also exported.
Other minor enhancements as detailed in [Link].

Incidents Resolved

Incident 18157 (User Interface and Display): An Incident was resolved in which the license
was sometimes lost during creation of the analysis model for certain types of models. This
only affected network licenses.
Incident 18376 (Modeling): An Incident was resolved in which some of the Uniform to
Frame loading was lost when assigned to an area object that could not be meshed by cookiecutting into 3- and 4-noded areas the using existing frame objects. Now a warning message
is given to the user if this problem is detected while assigning the load or while creating the
analysis model. If the warning is issued when creating the analysis model, the user is given
the option of either converting the load to a uniform area load, in which case no load is lost,
or canceling the creation of the analysis-model.
Incident 15873, et al (Analysis): An Incident was resolved for bridge moving-load cases in
which the vehicle axle loads were not being restricted to remain within the lane-edge
distances specified for the vehicle, but instead were allowed to move to the edge of the lane.
The effect of this was conservative.

Incident 18277 (Analysis): An Incident was resolved in which the P-delta force in frame
elements was assumed to be constant over the element using the value at the I-end of the
element rather than at the center of the element. The P-delta force is used for nonlinear static
analysis, nonlinear direct-integration analysis, and buckling analysis. This means that
different P-delta effects or buckling loads could be calculated when the local 1 axis is
reversed (i.e., by switching the ends I and J) for the case where there is significant axial load
applied to the element itself compared with the total axial force being carried by the element.
Incident 16491 (Design): An Incident was resolved for steel frame design in which the panelzone design shear force used to determine the doubler plate thickness was sometimes
calculated incorrectly. This design shear force is the difference |Vb|-|Vc|, where Vb is the
shear force from the beam flanges connecting to the joint, and Vc is the shear force from the
column above the joint. Vb is calculated as the larger of the capacity moment or the factored
moment from the beams, divided by their mean flange distance. Normally the capacity
moment governs, and this was being calculated correctly. However, when the factored
moment governed, the design could be unconservative, since the moments from the two
beams were added algebraically, but should have been added using their absolute values. The
affected codes are AISC-ASD 01, AISC-LRFD99, UBC97-ASD, UBC97-LRFD,
CAN/CSA-S16-01.
Incident 18147 (Design): An Incident was resolved for steel frame design using codes
Eurocode 3-1993 and Eurocode 3-2005 in which the section classification was
sometimes incorrect for sections in pure compression.
Incident 18347 (Design): An Incident was resolved in which superimposed dead load
patterns were not being included in auto-generated strength design load combinations for all
types of design and all design codes.
Incident 18402 (Design): An Incident was resolved for concrete frame design using code
ACI 318-05/IBC 2003 in which column shear forces were not being properly amplified
when Sway Ordinary frame design procedure was chosen and the Seismic Design Category
was greater than or equal to B. Now the shear design of columns will be performed in this
case according to the Sway Intermediate Frame procedure. In addition, for such cases
(Ductility=OMF, SDC >= B), the shear design of beams was previously being performed as
beams in Sway Intermediate Frame. Now such beams are designed as Sway Ordinary beams
irrespective of SDC.
Incident 18403 (Design): Steel frame joint design has been improved to now consider
whether the joint being designed is at the top of the column or not. Previously the joint was
always assumed to have a column above it, which could lead to an unconservative design for
the topmost joint where there is less restraint. The affected codes are AISC36005/IBC2006, AISC-ASD 01, AISC-LRFD99, UBC97-ASD, UBC97-LRFD,
CAN/CSA-S16-01.
Incident 18404 (Design): An Incident was resolved for steel frame design in which the panelzone design shear force used to determine the doubler plate thickness was sometimes
calculated incorrectly. This design shear force is the difference |Vb|-|Vc|, where Vb is the
shear force from the beam flanges connecting to the joint, and Vc is the shear force from the
column above the joint. Previously Vc was taken as the shear force from the column below
rather than above the joint. This error could be slightly unconservative, since Vc is usually
much smaller than Vb. The affected codes are AISC360-05/IBC2006, AISC-ASD 01,
AISC-LRFD99, UBC97-ASD, UBC97-LRFD, CAN/CSAS16-01.

Incident 18408 (Design): An Incident was resolved in which the D/C ratio calculated for
design could be incorrect for unsymmetrical Section Designer sections due to an error in
interpolating the capacity from P-M-M surface. Sections which are symmetrical for major
(M3) bending were not affected.
Incident 18457 (Design): An Incident was resolved in which the punching shear overwrites
were being ignored for steel frame design using the API design codes.
Incident 18498 (Design): An Incident was resolved for steel frame design using code
Eurocode 3-2005 in which the buckling moment capacity was not being considered in the
PMM interaction equation.
Incident 17910 (Bridge Design): An Incident was resolved for bridge design of precast-girder
superstructures in which the live-load distribution factors calculated from code equations
sometimes gave a zero value for the interior girder of a three-girder section.
Incident 17945 (Bridge Design): An Incident was resolved for the bridge superstructure shear
check of precast girder superstructures in which the correct controlling results for shear rebar
were not always presented in the output tables. Results were presented for the correct
demand set and load combination, but not always for the correct correspondence set, e.g.,
maximum or minimum moment with maximum or minimum shear. The plotted results were
shown for the correct controlling case.
Other minor Incidents as detailed in [Link].

6. New Features in SAP2000 Version 14

SAP2000 Version 14.0.0 is a major new release. Significant new features added or enhanced since
Version 12 include the following:
Nonlinear and Pushover
A nonlinear, layered shell element using a directional material model has been added for
pushover analysis of shear-wall structures and similar applications. A Quick-Start option is
provided for easy modeling of reinforced concrete sections.
Plot functions have been added for shell layer stresses.
Frame hinges for Section Designer sections To be Designed have been enhanced to use the
designed amount of rebar when available.
Section Designer has been enhanced for the display of fiber-model PMM surface.
The default material properties for concrete have been modified to improve convergence
behavior.
The tangent stiffness used for the iteration of fiber hinges and multi-linear links has been
changed to improve convergence.
Dynamics
Material-based damping is now available for linear and nonlinear direct-integration timehistory analysis.
Material-based damping is now available for steady-state and PSD analysis.
Stiffness-proportional damping now uses initial stiffness instead of tangent stiffness to
improve the consistency of results and convergence behavior.

The rigid-response calculation in response-spectrum analysis has been enhanced for NRC
and general use.
Base reactions for response-spectrum and modal time-history analysis have been improved
for springs and grounded link supports to better capture missing-mass effects.
Base reactions no longer include constraint forces at restraints, for consistency.

Bridge Modeling and Design

AASHTO LRFD 2007 superstructure design for precast concrete composite sections has
been implemented. Checks include: stress, flexure, and shear (using MCFT).
Fully automated bridge design check per AASHTO Guide Specifications for LRFD Seismic
Bridge Design 2009 has been implemented, including pushover analysis when required.
Automated handling of secondary prestress force has been implemented for the AASHTO
LRFD 2007 superstructure flexural design check for prestressed concrete box girder sections.
The AASTHO/USGS 2007 response-spectrum function has been added.
Variable girder spacing along the length of the bridge superstructure is now available.
Variable reference-point location along the length of the bridge superstructure is now
available.
Alignment of shell local axes in generated bridge models has been enhanced.
The longitudinal discretization of bridge models is now more uniform in complex models.
Tendons modeled as elements now allow elastic, creep, shrinkage, and steel-relaxation losses
to be directly specified.
The displacements for constrained joints in staged construction are no longer updated for
deflection until they are actually added to the model.
Design
Eurocode-2 2004 concrete frame design added, without seismic provisions.
Eurocode-3 2005 steel frame design added, without seismic provisions.
Australian concrete frame design per code AS 3600-2001 added, including seismic
provisions.
Multiple enhancements made for Indian concrete frame design per code IS 456-2000.
Multiple enhancements made for Chinese 2002 concrete frame design.
Multiple enhancements made for Chinese 2002 steel frame design.
Auto-lateral loads added for Eurocode: Wind, Seismic, and Response-spectrum.
Auto-lateral loads added for Australia: Seismic and Response-spectrum.
Auto-lateral loads added for New Zealand: Seismic and Response-spectrum.
General
Multiple enhancements have been made to the graphical display for clarity and speed.
Multi-stepped static analysis of Bridge Live loads acting on frame elements is now available
in Plus/Advanced levels without the Bridge license for modeling crane loads, footfall, etc.
Single-stage construction cases starting from zero are now available in Plus/Advanced levels
without the Staged-construction license for modeling separate gravity and lateral
configurations, different support conditions, etc.
The import of StruCAD*3D data files is now available.
New and updated API functions have been implemented.

More detail on these and other enhancements can be found in the separate file [Link] in
the SAP2000 folder and available by using the Help > Documentation command in SAP2000.

7. Important Note for Users of Version 11 and Earlier

If you are upgrading from Version 11 or earlier, you should be aware that there was a significant
change in nomenclature regarding loading starting with Version 12:

The term load pattern replaces load case of older versions

The term load case replaces analysis case of older versions
The term load combination has not changed

These changes are being made consistently in all future releases of CSI products. These are simply
changes in terminology; the concepts and behavior remain the same, and models from Version 11
and earlier will open in Version 14 without any action required by you.