CAEPIPE Technical Reference Manual

™
CAEPIPE
Technical Reference Manual
CAEPIPE Technical Reference Manual, Version 10.30, © October 2020, SST Systems, Inc. All
Rights Reserved.
Disclaimer
Please read the following carefully:
This software and this manual have been developed and checked for correctness and accuracy by SST
Systems, Inc. However, no warranty, expressed or implied, is made by SST Systems, Inc., as to the
accuracy and correctness of the manual or the functioning of the software and the accuracy, correctness
and utilization of its calculations. Users must carry out all necessary tests to assure the proper
functioning of the software and the applicability of its results. All information presented by the
software is for review, interpretation, approval and application by a Registered Professional Engineer.
CAEPIPE and CAdvantagE are trademarks of SST Systems, Inc. All other product names mentioned in
this document are trademarks or registered trademarks of their respective companies/holders.
For Info and Support on CAEPIPE, contact:
SST Systems, Inc.

1798 Technology Drive, Ste. 236
San Jose, CA 95110, USA
Tel: (408) 452‐8111
Fax: (408) 452‐8388
[email protected]
[email protected]
www.sstusa.com
Table of Contents
INTRODUCTION......................................................................................................................................... 1
TECHNICAL REFERENCE ............................................................................................................................. 2
ANCHOR.................................................................................................................................................. 3
BALL....................................................................................................................................................... 9
BEAM ................................................................................................................................................... 12
BELLOWS ............................................................................................................................................... 24
BEND.................................................................................................................................................... 26
BRANCH SIF ........................................................................................................................................... 47
BURIED PIPING........................................................................................................................................ 50
COLD SPRING (CUT PIPE) ........................................................................................................................... 65
COMMENT ............................................................................................................................................. 67
COMPRESSOR ......................................................................................................................................... 68
CONCENTRATED MASS .............................................................................................................................. 71
CONSTANT SUPPORT ................................................................................................................................. 72
DATA TYPES............................................................................................................................................ 73
DIRECTION ............................................................................................................................................. 74
DYNAMIC ANALYSIS.................................................................................................................................. 77
DYNAMIC SUSCEPTIBILITY .......................................................................................................................... 83
ELASTIC ELEMENT .................................................................................................................................... 85
EXPANSION JOINTS ................................................................................................................................... 86
FIBER REINFORCED PLASTIC PIPING (FRP) ................................................................................................... 101
FLANGE ............................................................................................................................................... 105
FLANGE AND BOLT STRESSES .................................................................................................................... 112
FORCE ................................................................................................................................................ 124
FORCE SPECTRUM .................................................................................................................................. 125
FROM (NODE) ...................................................................................................................................... 131
GENERIC SUPPORT ................................................................................................................................. 133
GUIDE ................................................................................................................................................ 136
HANGER .............................................................................................................................................. 138
HINGE JOINT......................................................................................................................................... 143
HYDROTEST .......................................................................................................................................... 149
JACKET END CAP ................................................................................................................................... 151
JACKETED PIPING ................................................................................................................................... 152
LIMIT STOP .......................................................................................................................................... 158
LOAD .................................................................................................................................................. 168
LOCAL COORDINATE SYSTEM .................................................................................................................... 172
LOCATION ............................................................................................................................................ 175
LUG EVALUATION .................................................................................................................................. 176
MATERIAL............................................................................................................................................ 185
MISSING MASS CORRECTION .................................................................................................................... 195
MITER BEND ........................................................................................................................................ 197
NODE ................................................................................................................................................. 206
NONLINEARITIES .................................................................................................................................... 207
NOZZLE ............................................................................................................................................... 211
NOZZLE EVALUATION .............................................................................................................................. 221
PRESSURE DESIGN - PIPE & PIPE FITTINGS ................................................................................................... 233
PUMP ................................................................................................................................................. 246
REDUCER ............................................................................................................................................. 256
REFINE NODAL MESH ............................................................................................................................. 263
RELIEF VALVE LOAD ANALYSIS ................................................................................................................... 277
RESTRAINT ........................................................................................................................................... 286
RIGID ELEMENT ..................................................................................................................................... 287
ROD HANGER ....................................................................................................................................... 289
i
Table of Contents
SECTION .............................................................................................................................................. 291

SKEWED RESTRAINT ............................................................................................................................... 295
SLIP JOINT ............................................................................................................................................ 300
SNUBBER ............................................................................................................................................. 303
SPIDER ................................................................................................................................................ 304
STATIC SEISMIC LOAD-ASCE/SEI 7-16....................................................................................................... 305
STATIC SEISMIC LOAD-ANSI A58.1-1988 .................................................................................................. 310
SUPPORTS............................................................................................................................................ 312
TEES ................................................................................................................................................... 315
TIE ROD ............................................................................................................................................... 318
TURBINE .............................................................................................................................................. 320
USER HANGER ....................................................................................................................................... 323
USER SIF ............................................................................................................................................. 336
VALVE................................................................................................................................................. 337
WELD ................................................................................................................................................. 342
WIND LOAD – ANSI A58.1-1982 ............................................................................................................ 343
WIND LOAD – ASCE/SEI 7-16 ................................................................................................................ 344
WIND LOAD – EN 1991-1-4 (2010) ........................................................................................................ 348
ANNEXURE I- DYNAMIC SUSCEPTIBILITY METHOD ................................................................................ 354
ANNEXURE II - NOZZLE STIFFNESS CALCULATIONS ................................................................................ 366
INDEX.................................................................................................................................................... 378
ii
This page is blank
Introduction
Thank you for licensing CAEPIPE (pronounced kay-pipe), the simple yet powerful software
for solving a variety of piping design and stress analysis problems in numerous industries
listed below.
Power(fossil & nuclear) Oil & Gas production (onshore & offshore)
Refinery Chemical & Petrochemical
Fertilizers Pharmaceutical
Sugar & Food Processing Paper & Pulp
Steel / Metal Process Water & Waste Treatment
Aircraft and Aerospace Building Services
Defense Industries Ship Building
CAEPIPE performs linear and nonlinear static and linear dynamic analyses of piping systems
by imposing various loads such as deadweight, thermal, seismic, wind, spectrum, time history
or harmonic, and calculates displacements, forces, moments, stresses, support loads etc.
Further, it checks whether the piping system is piping code and guideline compliant (ASME,
B31, European, Swedish, API 610, etc.) and producesconcise, formatted and easy to
understand reports.
For rapid modeling, CAEPIPE offers you a friendly and productive user interface that
rigorously adheres to Windows standards. Open up to four windows simultaneously to get
feedback on different aspects of the model. Extensive graphical display capabilities allow you
to zoom, pan, rotate the image and see the model from different viewpoints. CAEPIPE uses
the industry standard OpenGL® to render 3D images realistically for easy visualization. As
the model is input and modified, CAEPIPE updates the graphics simultaneously to provide
visual feedback. It animates deflected shapes and mode shapes, and shows color-coded stress
contours, among others.
A true powerhouse in its speed of operation, CAEPIPE uses advanced Windows
programming techniques such as intelligent repainting, scroll box tracking, multithreading,
memory-mapped files for faster data access, among others, to make your job easier and
faster. Every effort is made to keep the program and data file sizes small (e.g., program size is
~2 MB! And a 665-element piping model is 85 KB!).
Many thoughtful and useful details in the program allow you to work more productively. For
example, you can annotate your model with copious comments for enhanced
documentation, or duplicate repetitive inputwith one hotkey combination or rotate sections
of the model with one operation. No unnecessary buttons clutter the toolbar nor are you
forced to use a mouse unnecessarily. The many thoughtful keyboard shortcuts, too, add to
your productivity.
Overall, CAEPIPE stands peerless among the tools available today for piping design and
stress analysis. We invite you to explore the software so that you can make full use of its
capabilities. Our friendly and knowledgeable support engineers are available to assist
you.Should you need to reach them, please email: [email protected].
Two sections make up this manual:
1. Explanation of menus from the different CAEPIPE windows,
2. Appendices with related information.
The manual ends with an index.
1
Technical Reference
Anchor
An Anchor is a support that restrains the pipe movement at a node in the three translational
and the three rotational directions (i.e., restrains the node in all six degrees of freedom). In a
physical piping system, this node may be on an anchor block or a foundation, or a location
where the piping system ties into a wall or a large piece of equipment like a pump.
An Anchor is input by typing “A” in the Data column or selecting Anchor from the Data
Types dialog. A rigid anchor is entered (i.e., an anchor with rigid stiffnesses in all six degrees
of freedom) by default. To change the default, edit the anchor by double-clicking on the
anchor or pressing Ctrl+D in the row that has the anchor. An Anchor dialog is shown next.
Uncheck the “Rigid” checkbox to make the anchor non-rigid (i.e., a flexible anchor), and
enter numerical values for stiffnesses in the six degrees of freedom.
Stiffness
The stiffness for a degree of freedom may be rigid (specified by typing the letter “r” in the
stiffness field), or some value or be left blank. If it is left blank, there is no stiffness in that
degree of freedom, i.e., the pipe is free to move in that degree of freedom. Internally, the
rigid stiffness value is set to 1×1012 (lb./inch) for translational stiffness and 1×1012 (inch-
lb./radian) for rotational stiffness.
Releases for Hanger Selection
These releases apply only during the automatic selection of a hanger by CAEPIPE. If you
checked any of the Release check boxes, the pipe is assumed to be free in that released
degree of freedom during the automatic hanger selection by CAEPIPE. You may release any
combination of degrees of freedom at a hanger node during such automatic hanger selection.
This feature is useful when hangers are located near equipment, where you want the hangers
3
Anchor
to carry most of the weight load and thus reduce the load acting on the nearby equipment.
CAEPIPE, during hot load calculation (preliminary sustained load case) in hanger design,
releases the anchors (if you selected any of the Release checkboxes) so that the weight loads
are taken by the hangers rather than by the anchors (which represent the equipment).
After the hot load calculation, CAEPIPE restores the original values of stiffnesses to the
released anchors before continuing analysis. Release anchors when they are (typically) within
four (4) pipe diameters away from the nearest hangers.
You may release any combination of translations or rotations. Typically, either the vertical
translation or all translations and rotations are released. To release the Anchor in a particular
degree of freedom, check the corresponding checkbox.
.Displacements.
The dialog for Specified Displacements at an Anchor is shown below.
At an anchor, three types of translations and/or rotations in the global X, Y and Z directions
may be specified as listed below.
1. Thermal displacements (up to 10sets can be specified, one each for thermal loads T1
through T10). Applied only to the corresponding Expansion and Operating load cases.
2. Seismic displacement (available for B31.1, B31.9, ASME Section III Class 2, RCC-M and
EN 13480 codes only). Solved as a separate internal load case, the results of which are
added absolutely to static seismic and response spectrum load cases.
3. Settlement (available under ASME B31.1, ASME Section III Class 2, RCC-M and EN
13480 codes only). Applied as a separate load case called Settlement.
You may specify a displacement only if you also specify a corresponding non-zero or rigid
stiffness in that degree of freedom, i.e., the corresponding stiffness should not be left blank.
4
Anchor
Check “Displacements in LCS” if you want to enter anchor movements in the local
coordinate system. These local movements are transformed into the global coordinate system
and displayed in results.
Settlement
For certain piping codes (ASME B31.1, ASME Section III Class 2, RCC-M and EN13480),
an anchor settlement, which is a single non-repeated anchor movement (e.g., due to
settlement of foundation), may be specified. This is applied to the Settlement load case. For
those codes that do not have a provision for settlement (like B31.3), specify the anchor
settlement as a thermal displacement (which tends to be a conservative approach) for one of
the temperature load and define that temperature as equal to reference temperature.
Anchor in Local Coordinate System (LCS)
Check the box “Anchor in LCS” if you want to orient the anchor along a skewed line using
its local coordinate system (LCS), which also aids you in specifying Displacements in LCS.
Notice the naming convention changes (KXX changes to kxx, X changes to x, and so on).
Note:
1. Pressure Thrust (End-cap Force) of Pressure P x Inner Area (A) of pipe is not included
in the Support Loads for Anchors displayed by CAEPIPE at this time. Since
CAEPIPE’s results for numerous problems compare well with the results from other
third-party software, it confirms that the other stress programs are also not including the
Pressure Thrust (End-cap Force) of pipe in the Anchor Loads at this time. Refer to the
Verification Manual supplied with CAEPIPE for comparison of results with other stress
programs.
If you wish to include the effect of Pressure Thrust (End-cap force) due to internal
pressure in your piping on the Anchor loads, then you will have to compute the same
manually (= P x A) and input it as an external force at the Anchor Nodes using the Force
data type available with CAEPIPE. Please choose the option “Add to W+P” in the
Force data type dialog. By doing so, the End-cap force (= P x A) will be included in all
relevant load cases and combinations of CAEPIPE. Of course, when the “None” code is
selected under Options>Analysis> Code, this End-cap force is included in the only case
of “Static”.
2. At an Anchor defined in space to which a pipe support is attached, a “dummy”elementhasto be added.
This additional element should be defined such that the Local Coordinate System forthis element should
be the same as the Local Coordinate System of the attached Anchor.
Example 1: Flexible Anchor
Nodes on most large equipment are modeled as rigid anchors. If you need to specify a non-
rigid (i.e., flexible) anchor (for example at a nozzle to include vessel flexibility), you can input
those stiffnesses by editing the anchor.
Double click on the anchor to show the anchor dialog.
5
Anchor
By default the anchor has all stiffnesses rigid, no releases for hanger selection and no
specified displacements. The stiffness fields are grayed, i.e., non-editable and the Rigid
checkbox is checked. Click on the Rigid checkbox to uncheck it. The stiffness fields now
become editable.
Type in the required stiffness values and press Enter or click on OK. The anchor definition
shown in the next figure is now modified to be a flexible anchor.
Example 2: Rigid Vertical Support with Foundation Settlement

Assume that you need to model a vertical support on a foundation that has settled using
ASME Section III Class 2 (1980) code for code compliance.
Vertical settlement (–Y) = 6 inches.
First, set the piping code to ASME Section III, which has a provision for Settlement load
case, using the menu Options > Analysis > Code in the Layout window.
Next, create a rigid vertical support at the required node. Press “a” in the Data field to input
a default anchor.
Next, edit the anchor so that it acts as a vertical support only, by modifying the stiffnesses
similar to Example 1 so that only a Rigid stiffness KY in the Y direction remains.
6
Anchor
Now the anchor is modified to act as a Vertical 2-way rigid support. Click on the
Displacements button and type in –6 (inch) for Settlement under Y. You could also input
thermal and seismic displacements if required.
The anchor is now modified to be a rigid vertical support with a specified settlement
displacement.
Example 3: Anchor Release during Hanger Design
7
Anchor
CAEPIPE lets you model equipment nozzles as anchors. Assume that you had one on a
turbine, as shown above, and that you have placed a hanger nearby. The main purpose of
this hanger would be to carry most of the piping weight that would have been imposed on
the nearby turbine nozzle if not for the hanger. To let CAEPIPE do that, you will have to
release all six degrees of freedom of the Anchor during hanger design so that the hanger will
be designed to carry most of the piping weight.
First, enter an anchor for the node and then double click on it to edit it.
Click on the checkboxes for Releases for hanger selection in the required directions (X, Y, Z,
XX, YY, ZZ). The anchor will be released in the specified directions during hanger design.
CAEPIPE restores the anchor to its original state (of no releases) after completing the
preliminary hot load calculation during hanger design. Refer to the section on Hanger Design
Procedure for further details.
8
Ball Joint
A ball joint is a zero-length pipe element that allows rotations about the three orthogonal
global axes (similar to a universal coupling joint in the rear axle of a motor vehicle) while still
allowing the fluid to flow through it. If you do not want rotation in the torsional or (the
two) bending directions, input “Rigid” for the respective stiffnesses. Since the ball joint is a
zero-length element, the “From” and “To” nodes are coincident. Hence, you should leave
the DX, DY and DZ fields in the Layout window blank (CAEPIPE will not let you enter a
value).
A ball joint is input by typing “Ba” or “Ball” in the Type column or selecting Ball joint from
the Element Types dialog.
The Ball joint dialog is shown.
Weight of the ball joint is input in lbf or kgf and NOT its mass. Whenever mass is required
for a calculation as in the case of forming Mass matrix for dynamic analysis, or in calculating
inertia force as (mass x acceleration) for static seismic analysis, CAEPIPE internally
computes the mass to be equal to (weight / g-value).
The rotational stiffnesses, rotation limits and the friction torques are specified independently
in the bending and torsional directions. The torsional direction (local x) is determined by the
preceding element’s local x. If a preceding element is unavailable, the following element is
used to determine the torsional direction. The bending directions (local y and z) are
orthogonal to the torsional direction (local x). Bending friction is determined by a resultant
of frictiontorques in local y and z directions. Similarly, bending rotation limit is determined
by a resultant of rotational limits in local y and z directions.
9
Ball Joint
The stiffnesses, rotation limits and friction torque values are available from the manufacturer
of the ball joints or from their test results. Otherwise, you must use engineering judgment.
The stiffness values may be left blank, in which case CAEPIPE uses a very small value (1 in.-
lb./rad) internally to avoid dividing by zero during internal computation.
A rotation limit of zero (0.0) means that the ball joint cannot rotate (i.e., it is rigid) in that
direction. A rotation limit of “None” or Blank means that rotation is not limited to a finite
value.
When the applied torque is less than the friction torque, there is no rotation. When the
applied torque exceeds the friction torque, rotation is calculated as shown above. When
rotation limit is reached, there is no further rotation irrespective of the applied torque.
When the option “Use friction in dynamic analysis” is selected, for modal analysis,
CAEPIPE uses three different stiffnesses for a ball joint depending on the magnitude of the
applied moment/torque in comparison to the user-specified friction torque. They are as
follows:
Case 1: When the applied moment at the ball joint (for the first operating load case when a
piping code is selected or for the static case when “Piping code=None” is selected) is
less than the friction torque, the friction is not overcome and the ball joint stiffness
is internally set to "rigid”, i.e., 1×1012 (inch-lb./radian).
Case 2: When the applied moment is more than the friction torque, the friction is overcome
and the ball joint starts rotating (with the user-specified rotational stiffness being
applied). This rotational deformation takes place until the user-specified rotational
limit is reached.
So, from the time friction is overcome to the time when the rotational limit is
reached, CAEPIPE internally sets an “equivalent stiffness” for the ball joint as given
below.
If Kb is the user-specified stiffness for the ball joint,
Rotation = (applied moment – friction torque)/ Kb ……… (1)
The “equivalent stiffness” chosen by CAEPIPE is the slope of the straight line from
origin to the point (rotation, applied moment) in the figure given above. In other
words,
10
Ball Joint
Kbe = (applied moment/rotation) …….. (2)

Combining (1) and (2), you get
Kbe = (applied moment) x Kb / (applied moment – friction torque) ……. (3)
Case 3: When the actual rotation reaches the user-specified rotational limit for the first
operating load case/static case, the rotational stiffness for the ball joint is again set to
"rigid”, i.e., 1×1012 (inch-lb./radian).
For modal analysis, CAEPIPE uses “equivalent stiffness” for ball joints as described under
Case 2 above, when the friction torque is exceeded and the computed rotation is yet to reach
the specified rotational limit, as long as the option “Use friction in dynamic analysis” is
selected. When this option is NOT selected, CAEPIPE ignores the nonlinearities of the ball
joint (namely, friction torque and rotational limit) and uses only the user-specified stiffnesses
for modal analysis. That is why, despite very small moments, large rotations may be
computed (i.e., a check against user-specified rotational limit is not performed).
See the topic Expansion Joints for examples.
11
Beam
You can model elaborate structural systems inside CAEPIPE alongside the piping to be
supported. In simple situations, if the structure is much stiffer than the piping is, you may
not need to model the structure at all but simply treat it as rigid (for example: input Rigid for
Stiffness in a vertical Limit stop when simulating a support where pipe could rest on a stiff
beam). But, in cases where you need to account for structural flexibility, use the Beam
element to model structural support systems alongside piping systems.
The material, section and load for a beam are different from those for a pipe. Just as you
would define a material/section/load for a pipe, so too should you define a separate
material/section/load for a beam. Look for Beam Material, Beam Section, and Beam Load
(under Misc menu).
Upon analysis, CAEPIPE reports forces and moments for beam elements.
A beam is input by typing “bea” in the Type column or by selecting “Beam” from the
Element type dialog.
The Beam dialog is shown.
12
Beam
Beta angle
Beta angle is used to define the orientation of a beam’s local axes. See Beam orientation later
in this section.
Beam End Releases
Each end of the beam (From and To ends) can be released to simulate the type of structural
support you want to model. That is, you can use a combination of releases to specify whether
a beam end is fixed, pinned, etc.
Beam material
Before you input a beam element, you must define a beam material, section and load. Select
Beam materials from the Miscellaneous (Misc) menu in the Layout or List window.
A beam material list window is shown. Double click on an empty row to input a new beam
material.
A dialog for inputting beam material is shown.
13
Beam
The material name can be up to five alpha-numeric characters to identify the beam material.
A more complete description can be entered under Description. Enter modulus of elasticity,
Poisson’s ratio (Nu), density of the material and mean coefficient of thermal expansion
between Tref and T1/T2/T3/…./T10 in beam load.
Beam section
Select Beam Sections from the “Misc” menu in the Layout or List window.
14
Beam
A list of beam sections is shown. Double click on an empty row to input a new beam
section.
A dialog to input beam sections is shown.
You can either input the data yourself or click on the AISC Library button for a listing of
different AISC I-beams, channels, tees, etc., that are built into CAEPIPE or click on the
Library button (next to AISC Library) for a listing of different User Defined Beams. Be sure
to verify the properties that are shown in the fields after you select a section from the library.
The name can be up to five alpha-numeric characters to identify the beam section. A more
complete description can be entered in the Description field.
The axial area, major and minor moments of inertia must be input. Input of torsional
constant is optional. If it is not input, it defaults to the sum of major and minor moments of
inertia. Input of shear areas is optional. If they are not input, shear deflection is not included.
15
Beam
Input of depth and width are optional. Presently, they are used only for rendered plots of the
beam.
Dialogs for selecting a beam section from the AISC library are shown below:
The type of the beam section (e.g., I beam, W (Wide Flange)) is selected from this dialog.
Another dialog which shows various available sections for the particular beam section type is
then shown.
After selecting the section, click on OK and the section properties will be entered in the
Beam section dialog.
16
Beam
To create or modify a Beam section library

CAEPIPE offers you flexibility in creating your own Beam section libraries (user-defined
libraries). That way, you do not feel restricted by the offered choices in Beam sections and
can continually keep updating / adding Beam section libraries with your own sections. To
create a library: From the Main window, select File > New and click on Beam Section
Library.
A List window for Beam section is shown.
17
Beam
You can, as before, start typing directly into the fields, or enter properties through a dialog.
The only difference is that sections in the library do not have names whereas those in a
model have names.
After you are done entering sections, you must save to a Beam Section library file by using
the File > Save command.
Give the file a suitable name. The file will be saved with a .bli extension.
Beam load
Select Beam Loads from the Misc menu in the Layout or List window.
18
Beam
A list of beam loads is shown.
Double click on an empty row to input a new beam load through the beam load dialog or
start typing into the fields.
The Load name can be up to five alpha-numeric characters to identify the beam load. You
can enter up to 10 temperatures depending on the preset number of thermal loads. The
additional weight is a uniform weight per unit length added to the weight of the beam. This
could for example be used to add snow load to the beam. Wind load may or may not be
applied to the beam element by using the check box for Wind load in the dialog or typing
“Y” or “N” for Wind load in the List window.
19
Beam
Beam orientation
The Beam orientation is determined by the locations of the “From” and “To” nodes and the
beta angle of the beam element. The local x-axis of the beam is always from the “From”
node to the “To” node. The reference orientation corresponds to beta = 0.0.
A nonzero beta angle (measured from the reference position) rotates the local y- and z-axes
of the beam about the local x-axis of the beam in the counter clockwise direction.
The local coordinate system for beams can be displayed for each beam element through the
List window (Ctrl+L, select Beams, menu View > Show LCS [for Local Coordinate System]).
Global vertical axis is Y
Beam is not Vertical
The local y-axis of the beam lies in the local x - global Y plane (i.e., vertical plane) and is
in the same positive direction as the global Y axis. The local z-axis is the cross product of
the local x and y-axes. Major bending plane is local x-y, that is, Izz = Major moment of
inertia and Iyy = Minor moment of inertia.
Beam is Vertical
The local z-axis of the beam is in the global Z direction. The local y-axis is in global –X
direction. Major bending plane is x-y, i.e., Izz = Major moment of inertia and Iyy =
Minor moment of inertia.
20
Beam
Global vertical axis is Z

Beam is not Vertical
The local z-axis of the beam lies in the local x - global Z plane (i.e., vertical plane) and is
in the same positive direction as the global Z-axis. The local y-axis is the cross product of
the local z and x-axes. Major bending plane is x-z, that is, Iyy = Major moment of inertia
and Izz = Minor moment of inertia.
Beam is Vertical
The local y-axis of the beam is in the global Y direction. The local z-axis is in global –X
direction. Major bending plane is x-z, i.e., Iyy = Major moment of inertia and Izz =
Minor moment of inertia.
21
Beam
Example 1: Pipe Rack using Beams

Here, you see how to use a beam element to construct a pipe rack and connect the beam to
the pipe so that CAEPIPE can account for the rack’s flexibility. The procedure is simple.
First, you need to create a beam material, section and load in addition to pipe material,
section and load.
As the Layout window shows, model the piping (nodes 10 to 40) and the first beam support
(nodes 100 to 140). Then, create the second beam support (nodes 150 to 190) using the
Generate command (under Edit menu in the Layout window). Finally, connect piping at
nodes 20 and 30 to beam nodes 120 and 170 using limit stops.
The “Generate” dialog is shown below:
22
Beam
The “Limit Stop” dialog is shown below:
The graphics is shown below:
Example 2: Base Supported Bend using a Beam

See Example 7 in the Bend section of this manual for modeling a base supported bend using
a beam.
23
Bellows
Bellows expansion joints are flexible elements included in high temperature piping systems to
absorb primarily thermal movement. A Bellows contains one or more convolutions designed
to withstand the internal pressure while still flexible enough to absorb the axial, lateral and
bending deflections. Before use, you should note the critical pressure at which the bellows
becomes unstable. The B31.1 piping code suggests that expansion joints may be employed
only “when piping bends, loops, and offsets are not able to provide adequate flexibility.”
(Para. 11.5, 2010).
Usually manufacturers of these expansion joints publish product catalogs that contain
technical information about the joints you could use in your systems. The EJMA (Expansion
Joint Manufacturers Association) also publishes a standards catalog with guidelines that
“assist users, designers, and others in the selection and application of expansion joints for
safe and reliable piping and vessel installation.”
A Bellows joint is input by typing “bel” in the Type column or selecting “Bellows” from the
Element Types dialog.
The Bellows dialog is shown.
Expansion joints are mainly modeled using the above shown four types of stiffnesses – axial,
bending, torsional and lateral. The required stiffness values, pressure thrust area and weight
should be taken from the manufacturer’s catalog.
24
Bellows
For a rigid stiffness (for example, torsional), enter “r” for Rigid; if highly flexible, enter 1
(lb/in.) as a minimum, to avoid dividing by zero during internal computation.
Axial: Refers to axial extension (as in cryogenic systems) or contraction (as in high-
temperature systems) axially along its centerline while in operation.
Bending (angular): Refers to the bellows bending about its center point on the centerline.
Bending can be in any plane that passes through the centerline.
Lateral: Refers to the direction perpendicular to the centerline of the bellows. The two ends
of the bellows remain parallel to each other while their centerlines are displaced causing an
offset. This direction is also called transverse or parallel offset direction.
Torsional: Usually very stiff, refers to a twisting moment at one bellows end while the other
end either relatively is stationary or twists in the other direction, about the bellows centerline.
The pressure thrust area will impose a thrust load of: (pressure × thrust area), on both nodes
of the bellows. Even if the bellows is tied, it is recommended that the pressure thrust area be
input. The weight is the empty weight. CAEPIPE adds the weight of the contents, insulation
and additional weight to the empty weight.
Weight of the bellows joint is input in lbf or kgf and NOT its mass. Whenever mass is
required for a calculation as in the case of forming Mass matrix for dynamic analysis, or in
calculating inertia force as (mass x acceleration) for static seismic analysis, CAEPIPE
internally computes the mass to be equal to (weight / g-value).
Mean diameter is the “mean” between the outer and inner diameters of any Convolution of
the bellows. Since outer and inner diameters of all convolutions of the bellows are the same,
the Mean diameter is the same for all convolutions of that bellows.
Pipe guides are needed adjacent to the bellows because of its inherent flexibility and the
compressive loading on the adjacent pipes due to the pressure thrust of the joint. Moreover,
proper guiding is necessary to direct thermal movement into the joint and prevent buckling
of the line. Depending on the bellows behavior, you should place the first guide no farther
than four pipe diameters from the joint. Place additional ones appropriately after studying
the nearby deflections and loads.
Also, consider vessel and anchor movements, which may cause a misalignment at the
joint.See the topic on Expansion Joints for examples.
25
Bend
In CAEPIPE, the term Bend refers to all elbows and bends (custom-bent pipes). An elbow
comes prefabricated with a standard bend radius (short or long radius) whereas a bend is
custom- made from bending a straight pipe with a specified bend radius. Geometrically, a
bend is a curved pipe segment which turns at an angle (typically 90° or 45°) from the
direction of the run of the pipe. Some of the items associated with a bend are shown below.
Node 20 is the Bend node, also referred to as the Tangent Intersection Point (TIP). As you
can see from the figure, it is not physically located on the bend. Its only purpose is to define
the bend. CAEPIPE automatically generates the end nodes of the curved portion of the
bend (nodes 20A and 20B), called the near and far ends of the bend. The bend end nodes
(20A and 20B in the figure) may be used to specify data items such as flanges, hangers,
forces, etc.
A bend is input by typing “b” in the Type column or selecting “Bend” from the Element
Types dialog.
26
Bend
If you need to modify an existing bend, double click on it or press Ctrl+T (Edit type) to
bring up the Bend dialog.
Bend Radius
The radius of a bend (measured along the centerline of the bend) can be specified as Long,
Short, or User (defined) by one of the radio buttons for Bend Radius. CAEPIPE has long
and short radii built-in for standard ANSI, JIS, DIN and ISO pipe sizes. For nonstandard
pipe sizes, Long radius is equal to 1.5 times the pipe OD and Short radius equal to the pipe
OD.
Bend Thickness
Input the wall thickness of the bend if different from the preceding pipe’s thickness. If
specified, the Bend Thickness applies only to the curved portion of the bend (node 20A to
node 20B in the figure above).
Bend Material
If the material of the bend is different from that of the preceding pipe, select the Bend
Material from the drop down combo box. The Bend Material, if specified, applies only to the
curved portion of the bend (node 20A to node 20B in the figure above).
Flexibility Factor
This factor is automatically calculated for standard elbows according to the piping code
chosen. If you have your own Flexibility Factor, enter it here instead of the piping code
specified Flexibility Factor. A value of 2.0, for e.g., will mean that the bend is twice as flexible
as a pipe of the same length.
27
Bend
SIFs
These factors are automatically calculated for standard elbows according to the piping code
chosen. If you have your own, specify them here (useful for FRP bends, for example), which
will be used instead of the piping code specified SIFs. If User SIFs are also specified at bend
nodes (A and/or B nodes), they will be used instead of the bend SIFs or code specified SIFs.
Intermediate Nodes
An intermediate node, located in between the ends of the bend, may be required in some
situations to specify data items such as flanges, hangers, forces, etc. You can create an
intermediate node by giving a (new) node number and an angle for it, which is measured
from the near end of the bend (node 20A in figure). Up to two such nodes may be input.
Note that the intermediate nodes 13 and 16 shown below are at angles of 30° and 60°
respectively from node 20A (near end). The intermediate nodes can be used for specifying
data items such as flanges, hangers, forces, etc.
Note:
CAEPIPE will issue a message "Angle is too short" when the User tries to add an
intermediate node at an Angle less than 4.5 degree. Similarly, will issue a message "Angle is
too large" when the user tries to add an intermediate node at an Angle less than 4.5 from the
Far end.
Bend Examples
Some examples follow. They illustrate some common modeling requirements.
Example 1: 90° Bend
Example 2: 45° Bend and Pipe routing along a smooth curve
Example 4: Flanged Bend
Example 5: Reducing Bend
Example 6: Bend Supported by a Hanger
Example 7: Base Supported Bend
Example 8: Circular Piping
28
Bend
To simplify the discussion of bend modeling, it is assumed that the material, section (8”
STD), load and the first node (10) are already defined. It is also assumed that the bend has
long radius (12”) and the cursor is placed in row #3.
 Press Tab in row #3. Node 20 will be automatically assigned and the cursor will
move to the Type column, type “B” (for Bend), Tab to DX, type 2. Enter material,
section, load and press Enter. The cursor moves to the next row (#4).
 Tab to the DY column. The next Node 30 is automatically assigned. In DY column,
type -2 and press Enter. This completes the bend input.
29
Bend
Example 2: 45° Bend and Pipe routing along a smooth curve
move to the Type column. type “B” (for Bend), Tab to DX, type 1’6”. Enter
material, section, load and press Enter. The cursor moves to the next row (#4).
 Tab to the DX column. The next Node 30 is automatically assigned. In the DX
column, type 1, Tab to DY and type -1, then press Enter. This completes the bend
input.
To route a pipe along a smooth curve, you need to split that curve into a number of circular
arc segments. Each circular arc segment is then defined as a bend element by giving the
offsets (DX, DY and DZ) from the previous Point or Bend Tangent Intersection Point (TIP)
to the next TIP, as done for the 450 bend above. You continue this process until the routing
along the smooth curve is completed.
30
Bend

A 180° bend or U-bend, is often used in an expansion loop to relieve thermal stresses in the
piping system. It is modeled as two 90° bends back-to-back.
move to the Type column. type “B” (for Bend), Tab to DY, type –1’6”. Enter
material, section, load and press Enter. The cursor moves to the next row (#4).
 Press Tab. Node 30 will be automatically assigned and the cursor will move to the
Type column. type “B” (for Bend), Tab to DX, type 2. (DX is 2’ because 8” std long
radius bend has 12” radius and since these two bends are back to back, DX = 2R).
Press Enter and the cursor moves to the next row (#5).
 Tab to the DY column. The next Node 40 is automatically assigned. In DY column,
type 1’6”, then press Enter. This completes the bend input.
31
Bend
Example 4: Flanged Bend

Bends are often connected to the adjacent pipe sections with flanges. A flange may exist on
one or both sides of the bend. Flange weight may have a significant effect on the pipe
stresses. Also, the stress intensification and flexibility factors for a bend will decrease if one
or both of the ends are flanged.
Model the bend as in Example 1. Then input flanges at nodes 20A and 20B. Since these are
internally generated nodes, i.e., they do not normally appear in the Layout window, it is
necessary to specify input at these nodes using the Location type. To input the flange at node
20A, in row #5, type 20A for Node, Tab to Type column and type “L” for Location. This
opens the Data Types dialog.
32
Bend
Select Flange as the data type and click on OK. This opens the Flange dialog.
Select the Type of the flange from the drop-down combo box, e.g., Single welded slip-on
flange. To get the weight of the flange, click on the Library button.
Select the pressure rating for the flange (e.g., 600) and press Enter. The weight of the flange
is automatically entered in the Flange dialog. Press Enter again to input the flange.
Repeat the same procedure for the flange at node 20B.
33
Bend
The rendered graphics is shown below:
34
Bend
Example 5: Reducing Bend

CAEPIPE does not have a reducing bend element. A reducing bend may be modeled using
an average OD (outside diameter) and average thickness of the large and small ends of the
bend. The bend radius of the reducing bend should be input as user bend radius. The Stress
Intensification Factor (SIF) of the reducing bend, if available, should be input as Bend SIF.
 The 8” std pipe (OD = 8.625”, Thk = 0.322”) with Name = 8 is already defined.
Now define a 4” std pipe (OD = 4.5”, Thk = 0.237”) with Name = 4.
The average OD of the two sections is (8.625 + 4.5) / 2 = 6.5625” and the average
Thickness is (0.322 + 0.237) / 2 = 0.2795”.
Define a “Non Std” section with Name = AVG, OD = 6.5625” and Thickness = 0.2795”.
The list of sections is shown below.
 Note that the section specified on the Bend row in the Layout window applies to the
curved portion of the Bend (between the A and B nodes) as well as to the straight
portion from the preceding node to the A node. In this case, we want to assign the
section “AVG” only to the curved portion and assign the section “8” to the straight
portion. This can be done by defining an additional node that is coincident with the
A node thus making the straight portion of the bend zero length.
 In row #2, the first node (10) is already defined and the cursor is placed in row #3.
Type 15 for Node, Tab to DY, type -8”. Enter material (1), section (8), load (1) and
press Enter. The cursor moves to the next row (#4).
 Press Tab. Node 20 is automatically assigned and the cursor will move to the Type
column, type “B” (for Bend), Double click in the Type column to edit the Bend.
Click on the User Bend Radius button and type 16 for bend radius. Press Enter to
modify the Bend and return to the Layout window. Tab to DY, type -1’4”. Tab to the
35
Bend
section column and type “AVG”. Then press Enter. The material and load are copied
from the previous row and the cursor moves to the next row(#5).
 Tab to the DX column. The next Node 30 is automatically assigned. In DX column,
type 2’. Tab to section column, type 4 and then press Enter. This completes the
reducing bend input.
The Layout window is shown below:
36
Bend
Example 6: Bend Supported by a Hanger

Assume that the bend is supported in the middle by a hanger.
Model the bend as in Example 1. The hanger is input at node 15, which is in the middle of
the bend. Node 15 is created as an intermediate node on the bend as follows:
Double click on the bend (in the type column of the Layout window) to edit it. The bend
dialog is shown.
Under intermediate nodes, type 15 for node and 45 for its angle, then click on OK. This
creates an intermediate node 15 at 45° from the node 20A (near end of the bend) as shown
in the figure above.
37
Bend
Since node 15 does not show in the Layout window, it is necessary to specify data input at
this node using the Location type. To input the hanger at this node, in row #5, type 15 for
Node, Tab to Type column and type “L” for Location. This opens the Data Types dialog.
Click on Hanger, the hanger dialog is shown.
Click on OK to accept the default hanger and a hanger is entered at node 15.
The rendered graphics is shown next.
38
Bend
Example 7: Base Supported Bend

Two examples of base supported bends are shown below. In the figure on the left, the
support is modeled using a rigid vertical restraint. In the figure on the right the support is
modeled using a beam element.
Vertical Restraint Support

 Model the bend at node 20 as before.
 To put a vertical restraint at node 20B, type 20B for node and “L” for Location. This
will open the Data types dialog.
39
Bend
Double click on Restraint. This will open the Restraint dialog.
 Click on the Vertical button to check the Y or Z restraint (depending on the vertical
axis) and click on OK.
40
Bend
Beam Support
 Model the bend at node 20 as before.
 Create a beam material, section and load as described earlier under the Beam section
in the Technical Reference manual.
 Input a beam element from node 20B to node 100.
Type 20B in the Node column and “f” (for From) in the Type column to create a starting
point. Press Enter to move to the next row.
Type 100 in the Node column and “bea” (for Beam) in the Type column. In the DY
column, type the beam length with a negative sign (since the beam is going downward
from node 20B to node 100). Type the beam material, beam section and beam load names
in the Matl, Sect and Load columns. In the Data column type “a” to input an Anchor.
41
Bend
Example 8: Circular Piping

Circular piping around the Nuclear Containment or Tank can be modeled using a series of
Bends. Depending upon the number of nodes required to represent the layout and its
supports, one can decide the required bend angle.
In this example, Bends with 22.5degrees are chosen to represent the layout. The center line
diameter of the Circular pipe layout is 96” and the Section property is 3” Schedule 40.
Origin of the Circular piping is at absolute coordinates Xo=150”, Yo=30” and Zo=0.0”.
Vertical Axis is chosen as “Z”.
With reference to the Origin, absolute coordinate for Node 10 is
X' =150”+Radius (96”) = 246”; Y'=30” and Z' =0.0”.
Bend Angle chosen (A) = 22.5 deg.
Distance between Origin and Tangent Intersection Point (TIP) = R' = Radius / COS(A/2) =
96/COS(11.25) = 97.88”.
Using the above information, absolute coordinates and the offset distances for TIP nodes are
computed and presented in the Table below.
Two layouts are generated.
The first layout is generated using absolute coordinates and hence under the Node column
you will see a “*” after the Node number in CAEPIPE layout, meaning that DX, DY and
DZ values are not offsets, instead they are absolute coordinates for the Node being input..
The second layout is generated using the Offset Distances listed in the table below.
42
Bend
Ring Radius = Bend Radius (in) 96 Bend Angle 22.5

Xo (in) Yo (in) Zo (in)
Absolute Coordinates of Circular Ring Center 150 30 0
X' (in) Y' (in) Z' (in)
Absolute Coordinates for Node 10 246 30 0
R' = Absolute Coordinates Offset Distances

Angle (B) @ Distance
Tangent between
Intersection Origin and X' = Xo + Y' = Yo +
TIP Point (TIP) TIP [COS(B) x R'] [SIN(B) x R'] DX DY
Node (Deg) (in) (in) (in) (in) (in)
20 11.25 97.88 246.00 49.10 0.00 19.10
30 33.75 97.88 231.38 84.38 -14.62 35.28
40 56.25 97.88 204.38 111.38 -27.01 27.01
50 78.75 97.88 169.10 126.00 -35.28 14.62
60 101.25 97.88 130.90 126.00 -38.19 0.00
70 123.75 97.88 95.62 111.38 -35.28 -14.62
80 146.25 97.88 68.62 84.38 -27.01 -27.01
90 168.75 97.88 54.00 49.10 -14.62 -35.28
100 191.25 97.88 54.00 10.90 0.00 -38.19
110 213.75 97.88 68.62 -24.38 14.62 -35.28
120 236.25 97.88 95.62 -51.38 27.01 -27.01
130 258.75 97.88 130.90 -66.00 35.28 -14.62
140 281.25 97.88 169.10 -66.00 38.19 0.00
150 303.75 97.88 204.38 -51.38 35.28 14.62
160 326.25 97.88 231.38 -24.38 27.01 27.01
170 348.75 97.88 246.00 10.90 14.62 35.28
43
Bend
Layout using Absolute Coordinates
44
Bend
Layout using Offset Distances
45
Bend
As an example, shown below are the snap shots to split the Bend at Node 30 with an
intermediate Node 35 and to place a Limit Stop at that intermediate Node 35.
46
Branch SIF
A tee is modeled using three pipes that come together at a node, which should be designated
as a “Tee” using the Branch SIF data type. If not, then CAEPIPE cannot calculate a code-
specified SIF for the tee to use in stress calculations.
A Stress Intensification Factor (SIF) type for a tee can be input by typing “br” in the Data
column or selecting “Branch SIF“ from the Data types dialog.
The Branch SIF dialog is shown.
The type of the branch SIF can be selected from the Type drop-down combo box.
Depending on the piping code selected, different types of branch SIFs may be available.
Typical branch SIF types (for B31.1 piping code) are shown below.
A few branch SIFs may need additional input; for example, in the case of a reinforced
fabricated tee, a pad thickness is required.
47
Branch SIF
The field “No. of Flanges or Rigids at Run Pipe ends…” shown above is available only when
the Option “Use B31J for SIFs and Flexibility Factors” is turned ON through Layout
window > Options > Analysis for B31.x codes. This field can be blank or entered as 1 or 2.
When entered as either 1 or 2, CAEPIPE will multiply the flexibility factors for the Branch
by a factor ‘c’ provided in “Table 1-3 – Flanged End Corrections” of ASME B31J-2017. For
more details, see the section titled “ASME B31J-2017” from Code Compliance Manual.
CAEPIPE differentiates between a header (run) and a branch line based on their ODs. So,
when CAEPIPE finds two lines with ODs of 8 inches and 6 inches coming together at a
node, it designates the 8 inch line as the header (or main) line with the 6 inch line designated
as the branch line (this is also how you would model a reducing tee).
When the header and the branch lines have the same ODs, reduce the branch OD slightly so
that the header and the branch lines are properly designated (e.g., OD header = 168.4 mm,
ODbranch = 168.3 mm).
Examples:
Say you want to model a reducing tee as shown in the provided “Sample.mod”. It’s an 8”x6”
reducing tee. First, we need to model the three pipes – two for the run (nodes 20 to 30 and
30 to 40), one for the branch (nodes 30 to 60), which will be assigned the 6” section. Each
pipe section can have its own OD/Thickness. Lastly, designate the common node as a (Butt)
Welding Tee.
48
Branch SIF
The Tee is as shown at node 30 in the graphics window below.
A “latrolet” or a lateral tee (commonly, the branch is at a 45° angle from the header) or a Y-
joint is modeled like a regular tee (as above). But, the SIF for this joint needs to be input by
you after consultation with the tee/joint manufacturer. Use a “User-SIF” data type at the
common node to specify the SIF. See topic on Tees for more information.
49
Buried Piping
Soil in Buried piping analysis is modeled by using bilinear restraints with an initial stiffness
and an ultimate load. After the ultimate load is reached, the displacement continues without
any further increase in load, i.e., the yield stiffness is zero. The initial stiffness is calculated by
dividing the ultimate load by the yield displacement which is assumed to be D/25 where D is
the outside diameter of the pipe.
Soil modeling is based on Winkler’s soil model of infinite, closely spaced elastic springs. Soil
stiffness is calculated for all three directions at each node. The pressure value in the load is
suitably modified to consider the effect of static overburden soil pressure. The model is
analyzed for operating (W+P1+T1) condition and the displacements in the three directions
are noted. A check is made for whether skin friction is mobilized and the soil has attained the
yield state. If true, then the spring is released in that direction, indicating that soil no longer
offers resistance in that direction. This modified model is again analyzed and checked for the
yield stage. The iterative process is continued until the percentage difference between
displacements at each node for two successive iterations is less than 1%. The final stiffness is
the true resistance offered by the soil to the pipe.
General Procedure to model buried piping
1. First, define soils using the command Misc > Soils in the Layout or List window.
2. Next, tie these defined soils with pipe sections (Ctrl+Shft+S to list Sections, double click
on an empty row, you will see the field Soil in the bottom right corner. Pick the soil
name from the drop-down combo box).
3. Use this modified section for each element on the Layout window that is buried with this
soil around it.
4. Discretize long sections of buried piping (Refine Nodal Mesh) through Layout window
> Edit > Refine Nodal Mesh > Buried Piping.
It is at the bends, elbows, and branch connections that the highest stresses are found in
buried piping subjected to thermal expansion. These stresses are due to the soil forces
that bear against the transverse runs. The stresses are proportional to the amount of soil
deformation at the elbows or branch connections. Hence, piping elements adjoining to
bends, elbows and branch connections are to be discretized in the stress model.
In addition, to best simulate Winkler’s soil model, it is recommended to discretize even
the remaining long straight buried pipe sections in the stress layout
The details of such discretization are explained below.
Refinement of Nodal Mesh for Buried Piping
Modulus of Subgrade Reaction (k)
This factor k defines the resistance of the soil or backfill to pipe movement due to the
bearing pressure at the pipe/soil interface. Several methods for calculating modulus of
subgrade reaction (k) have been developed in recent years. As per Trautmann, C.H., and
O’Rourke, T.D., “Lateral Force-Displacement Response of Buried Pipes,” Journal of
Geotechnical Engineering, ASCE, Vol. 111, No. 9 Sep 1985, pp. 1077-1092, the modulus of
subgrade reaction, k, can be calculated as per Eq. (2) in Appendix VII of ASME B31.1-2018
code.
k  Ck N h wD
50
Buried Piping
where,
Ck = a dimensionless factor for estimating horizontal stiffness of compacted backfill. C k may
be estimated at 20 for loose soil, 30 for medium soil, and 80 for dense or compacted soil. In
the current version of CAEPIPE, the value of Ck is internally set as 80 for both cohesive and cohesionless
soil.
D = pipe outside diameter
w = soil density
Nh = a dimensionless horizontal force factor from Fig. 8 of above stated technical paper. For
a typical value where the soil internal friction angle is 30 deg. the curve from Fig. 8 may be
approximated by a straight line defined by
Nh = 0.285H/D + 4.3
where
H = the depth of pipe below grade at the pipe centerline
Influence Length (Lk)
The influence length is defined as the portion of a transverse pipe run which is deflected or
“influenced” by pipe thermal expansion along the axis of the longitudinal run.
From Hetenyi’s theory, (Beams on Elastic Foundation, The University of Michigan Press, Ann Arbor,
Michigan 1967) (also, see Section VII-3.3.2 of Appendix VII of ASME B31.1-2018 code)
3
Lk 
4
where,
1/ 4
 k 
Pipe / Soil System Characteristics =   
 4 EI 
E = modulus of elasticity of pipe at reference temperature
I = moment of inertia of pipe cross section
k = modulus of subgrade reaction of soil as detailed above.
Implementation in CAEPIPE
As stated earlier, it is at the bends, elbows, and branch connections that the highest stresses
are found in buried piping subjected to thermal expansion of the pipe. These stresses are due
to the soil forces that bear against the transverse runs. The stresses are proportional to the
amount of soil deformation at the elbows or branch connections. Hence, piping elements
adjoining to bends, elbows and branch connections are to be discretized in the stress model.
In addition, to best simulate Winkler’s soil model, it is recommended to discretize even the
remaining long straight buried pipe sections in the stress layout
This can be performed through Layout window > Edit > Refine Nodal Mesh > Buried
Piping.
51
Buried Piping
When the command is selected, CAEPIPE will refine the piping layout as detailed below.
1. Calculate modulus of subgrade reaction (k) as detailed above. While calculating k, the
value of Ck is taken as 80 for both cohesive and cohesionless soil.
2. Calculate influence length (Lk) for the element that is fully buried.
3. If the length of the pipe element near bend / elbow / branch connection is greater
than or equal to the influence length (L k), then the pipe element will be split into a
number of short elements with length of each short element being equal to 2 x OD
of that pipe section until the Influence length (L k).
4. On the other hand, if the length of the pipe element near bend / elbow / branch
connection is less than the influence length (L k) and greater than 2 x OD of the pipe,
then the pipe element will be split into a number of short elements with length of
each short element being equal to 2 x OD of that pipe section.
5. If any buried straight pipe element is longer than the influence length (Lk), then the
straight pipe element will be split into a number of equal length elements, where the
number of such equal length elements will be computed as [(int)(Original Straight
Pipe length/Influence Length) + 1]. For example, if the total length of straight pipe
element is equal to 1843” and influence length is 400”, then the straight pipe will be
split into 5 equal length elements [= (int)(1843/400) + 1 = ((int)(4.61) + 1) = (4 +
1)] with each element having a length of 368.6” [= 1843/5].
Note:
While refining the layout, the new node number will be generated by adding the node
increment specified (through Layout Window > Options > Node increment) to the available
free node number. Hence, set the node increment value as required before refining the
buried piping layout.
52
Buried Piping
It is possible to specify different soil characteristics for different portions of the pipe model.
Here is how.
1. Define different soils using the command Misc > Soils.
2. Associate each soil type with those sections that are buried in that soil.
3. Model the buried layout using the different sections for different buried portions.
Ground Level
Ground level for soil is the height of the soil surface from the global origin (height could be
positive or negative). It is NOT a measure of the depth of the pipe’s centerline.
In the figure, the height of the soil surface is 3 feet above the global origin. Pipe node 10
[model origin] is defined at (0,-5,0). So, the pipe is buried 8’ (3’ - [-5’]) deep into the soil.
Define similarly for the other soil.
The pipe centerline is calculated by CAEPIPE from the given data
Two Soil types

Two types of soils can be defined - Cohesive and Cohesionless. Soil density and Ground
level are input for both cohesive and cohesionless soils. The Ground level is used to calculate
the depth of the buried section. For cohesive soil, Strength is the un-drained cohesive
strength (Cs). For cohesionless soil, Delta () is the angle of friction between soil and pipe,
and Ks is the Coefficient of horizontal soil stress. See the nomenclature below for more
information.
Highlight buried sections of the model in graphics
If your model contains sections that are above ground and buried, then you can selectively
see only the buried sections of piping in CAEPIPE graphics by highlighting the section that
is tied to the soil. Use the Highlight feature under the Section List window and place
highlight on the buried piping section (see Highlight under List window>View menu, or
press Ctrl+H). The Graphics window should highlight only that portion of the model that is
using that specific section/soil.
53
Buried Piping
Nomenclature
When the option “Include Insulation Thickness” in “Soil” input is turned OFF, then
D = Outside diameter of the pipe
When the option “Include Insulation Thickness” in “Soil” input is turned ON, then
D = Outside diameter of pipe + (2.0 x Insulation Thickness)
Ks = Coefficient of horizontal soil stress, which depends on the relative density and state of
consolidation of soil. Ks is empirical in nature and may be estimated from Nq/50. Ks
can vary depending on the compaction of the soil from 0.25 (for loose soil) to 1.0
(really compacted soil).
(0.107311 )
Nq = Bearing capacity factor = 0.98414e
 =  + 5°
  angle of friction between soil and pipe
Normal values for delta ranges between 25° – 45° (for sand).
Clean granular sand is 30° . With a mix of silt in it, the angle is 25°
Sp = soil pressure = soil density × depth
Cs = Undrained cohesive strength (input for cohesive soil), (Cs in kN/m2) ≥1.0
(-0.00833699 Cs)
Af = Adhesion factor = 1.7012775 e
kp = Coefficient of passive earth pressure
= (1 + sin ) / (1 – sin )
bottom depth = depth + D/2
top depth = depth – D/2
Nr = (Nq – 1.0) tan (1.4 )
dq = dr = 1.0 + 0.1 tan (/4 + /2) × depth / D, for > 10° , otherwise dq = dr = 1.0
Calculation of Ultimate Loads
The ultimate loads (per unit length of pipe for axial and transverse directions and per unit
projected length of pipe for vertical direction) are calculated as shown below.
Different equations are used for cohesive (clayey) and cohesionless (sandy) soils.
Axial direction
Cohesive soil: Axial load =  × D × Af × Cs
Cohesionless soil: Axial load = × D × Ks × Sp × tan 
Transverse direction
Cohesive soil: Transverse load = D × (2 Cs + Sp + 1.5 Cs × depth / D)
Cohesionless soil: Transverse load = kp × kp × Sp × D
Vertically downward direction
Cohesive soil: Downward load = D × (5.7182 Cs + Soil density × bottom depth)
54
Buried Piping
Cohesionless soil: Downward load = D × (Soil density × bottom depth × Nq × dq

+ 0.5 Soil density × D × Nr × dr)
Vertically upward direction
Cohesive soil: Upward load = D × Soil density × top depth + 2 Cs × top depth
Cohesionless soil: Upward load = D × Soil density × top depth
Buried Piping Example
Ultimate Loads and Stiffnesses computed by CAEPIPE for this example are verified later in
this section.
Example data:
A 12” Std pipe 6’ long is buried, 3’ in cohesionless and 3’ in cohesive soils with No
Insulation properties
Soil properties are as follows:
Cohesionless (Name of soil: S1, associated with pipe section 12A):
Density = 120 lbf / ft3
Delta () = 20°
Ks = 0.29 (calculated from Nq/50, where Nq = 14.394)
Ground level = 3’
Cohesive (Name of soil: S2, associated with pipe section 12B):
Density = 150 lbf / ft3
Strength = 100 psi
Ground level = -1’
1. Define soils using the command Misc > Soils.
55
Buried Piping
A List window for soils will be displayed. Double click on an empty row to define a new soil.
For our example, define two soils - one cohesionlessand the other cohesive with properties
as shown in the following dialogs.
Dialog for cohesionless soil:
Dialog for cohesive soil:
After you have defined the soils, you should see the two soils listed in the List window.
56
Buried Piping
2. Define pipe sections and then associate the soils with these sections.
Define two pipe sections, both 12”/STD pipe sections (name them 12A and 12B), and select
the correct soil in the pipe section dialog box using the Soil drop-down combo box.
Soil S1 is associated with section 12A:
Soil S2 is associated with section 12B:
57
Buried Piping
3. Define the layout from 10 to 20 to 30; the first pipe element from 10 to 20 uses section
12A (Cohesionless soil type S1), and the next pipe element 20 to 30 uses section 12B
(Cohesive soil type S2). Check Operating load case under Loads menu > Load cases for
analysis.
Save the model and analyze. Choose yes to view the results. From the Results dialog, pick
Soil Restraints. The soil loads and soil stiffnessesin different directions as computed by
CAEPIPE are now shown, which are verified below in this section.
Example Verification
Verification of cohesionless restraints (for pipe element 10 to 20)
depth = 3’ - (-5’) = 8’ (since the pipe centerline is at -5’ and ground level is at 3’).
Sp = 120 lb/ft3 × 8 ft = 960 lb / ft2 = 6.6667 lb/in2
Axial direction
Axial load = × D × Ks × Sp × tan 
×12.75×0.29×6.6667×tan (20)
= 28.1861 lb/in
= 1014.7 lb (for 36”, length of pipe) (CAEPIPE: 1014.7)
Assuming yield displacement = D/25,
Axial stiffness = 25 × 1014.7 / 12.75 = 1989.6 (lb /in) (CAEPIPE: 1989.6)
58
Buried Piping
 = + 5° = 20° + 5° = 25°
kp = Coefficient of passive earth pressure
= (1+ sin ) / (1 - sin )
= 2.4639
Transverse load = kp × kp × Sp × D
= 2.4639 × 2.4639 × 6.6667 × 12.75
= 516.0239 lb /in
= 18576.88 lb (for 36”) (CAEPIPE: 18577)
Transverse stiffness = 25 × 18576.88 / 12.75
= 36425 lb / in (CAEPIPE: 36425)
Vertically downward direction
bottom depth = 96” + 12.75”/2 = 102.375”
(0.107311 )
Nq = Bearing capacity factor = 0.98414 e = 14.39366
Nr = (Nq – 1.0) × tan (1.4 ) = 9.37834
since > 10°
dq = dr = 1.0 + 0.1 × tan(/4 + /2) × depth / D = 2.18188
Downward load = D × (Soil density × bottom depth × Nq × dq + 0.5 × Soil density × D ×
Nr × dr)
= 12.75 × ((120/1728) × 102.375” × 14.39366 × 2.18188 + 0.5 × (120/1728) × 12.75”
×9.37834×2.18188)
= 12.75 × 232.3305 lb / in
= 106639.7 lb (for 36”) (CAEPIPE: 106640 lb)
Downward stiffness = 25 × 106639.7 / 12.75
= 209097 lb/in (CAEPIPE: 209098)
Vertically Upward Direction
top depth = 96” – 12.75”/2 = 89.625”
Upward load = D × Soil density × top depth
= 12.75” × (120/1728) × 89.625
= 79.35547 lb / in
= 2856.7968 lb (for 36”) (CAEPIPE: 2856.8)
Upward stiffness = 25 × 2856.7968 / 12.75 = 5601.56 lb/in (CAEPIPE: 5601.6)
Verification of cohesive restraints (for pipe element 20 to 30)
depth = –1’ – (–5’) = 4’ (since the pipe centerline is at -5’ and ground level is at -1’).
Sp = 150 lbf/ft3 × 4 ft = 600 lb / ft2 = 4.16667 lb/in2
D = 12.75” + (2 x 2.0) = 16.75” (as insulation thickness is defined as 2” andthe option
“Include Insulation Thickness” is turned ON in Soil S2 input).
59
Buried Piping
Axial direction
Soil strength = Cs = 100 psi = 100 × 6.89476 KN/m2 = 689.476 KN/m2
(-0.00833699 Cs)
Af = Adhesion factor = 1.7012775 e [where Cs should be in KN/m2]
= 5.424795E-3
Axial load = × D × Af × Cs
×16.75”×5.424795E-3×100 psi
= 28.54 lb / in
= 1027.66 lb (for 36”) (CAEPIPE: 1027.7)
Axial stiffness = 25 × 1027.66 / 16.75 = 1533.82 lb /in (CAEPIPE: 1533.8 lb/in)
Transverse load = D × (2 Cs + Sp + 1.5 Cs × depth / D)
= 16.75” × [(2.0×100 + 4.166667 + (1.5×100×48/16.75)]
= 10619.79 lb/in
= 382312.5 lb for 36” (CAEPIPE: 382313)
Transverse stiffness = 25 ×382312.5 / 16.75
= 570615.67 lb /in (CAEPIPE: 570616)
Vertically Downward direction
bottom depth = 48” + 16.75”/2 = 56.375”
Downward load = D × (5.7182 Cs + Soil density × bottom depth)
= 16.75” × [(5.7182×100) + ((150/1728) × 56.375”)]
= 9659.95 lb/in
= 347758.34 lb for 36” (CAEPIPE: 347758)
Downward stiffness = 25 ×347758.34 / 16.75
= 519042.30 lb / in (CAEPIPE: 519042)
Vertically Upward Direction
top depth = 48” – 16.75”/2 = 39.625”
Upward load = D × Soil density × top depth + 2 Cs × top depth
= 16.75” × (150/1728) × 39.625” + (2×100×39.625”)
= 7982.61 lb / in
= 287374.12 lb for 36” (CAEPIPE: 287374)
Upward stiffness = 25 ×287374.12 / 16.75
= 428916.60 lb / in (CAEPIPE: 428917)
References
1. Tomlinson, M. J., Pile Design and Construction Practice. Fourth Edition. London: E &
FN Spon, 1994.
2. Fleming, W.G.K., et al. Piling Engineering. Second Edition. Blackie Academic and
Professional. (Chapters 4 and 5).
60
Buried Piping
Discretization Example
61
Buried Piping
Soil characteristics
Soil density, w = 130 lbf/ft3 = 0.075 lb/in3
Pipe depth below grade, H = 12 ft (144 in)
Type of backfill, dense sand (cohesionless soil)
Ck = 80
Calculation of Modulus of subgrade reaction (k)
Nh = 0.285H/D + 4.3
Nh = (0.285 x 144 / 12.75) + 4.3 = 7.518
k  Ck N h wD = 80 x 7.518 x 0.075 x 12.75 = 575.127 psi

Calculation of Influence Length (Lk)
Moment of inertia, I = 279.3 in4
Modulus of elasticity, E = 27.9 x 106 psi
3
Lk 
4
1/ 4
 k 
 
Pipe / Soil System Characteristics =  4 EI  = [575.127 / (4 x 27.9 x 106 x 279.3)] 1/4 =
0.01165
Influence Length (Lk) = 3 x 3.14 / (4 x 0.01165) = 202.145 in
As the lengths of pipe elements near the bends and branch connection are greater than the
influence length (Lk = 202.145 in), the pipe elements near the bends and branch connection
are split into a number of short elements with length of each short element being equal to 2
x OD = 2 x 12.75 = 25.5 in until the influence length (Lk). See figures given below for
details.
62
Buried Piping
63
Buried Piping
64
Cold Spring (Cut Pipe)
Cold spring (cut short or cut long) is used to reduce thermal forces on equipment connected
to the piping system. When lengths of pipes are cut short or extended by design, they are
pulled together or pushed apart to join them during installation, giving rise to a “cold-
sprung” system.
Such an installation process (cold condition) obviously introduces stresses, which are relieved
when the system starts up (hot condition). Note however, that the piping codes do not allow
credit for any reduction in stresses due to cold spring since the displacement range is
unaffected (similar to self-springing. See B31.1 para. 119.2 fordetails). But, codes allow
reduction in support loads due to cold spring (which can be helpful at the equipment).
This feature should be used only with a proper understanding of the implications.
Cold spring for a straight pipe is input by typing “c” in the Type column or selecting “Cut
pipe“ from the Element Types dialog.
The Cut pipe dialog is shown.
Select “Cut short” or “Cut long” using the radio buttons. The amount of cut (short or long)
should be positive.
Since the piping codes do not allow credit for cold spring in stress calculations, a cold spring
is used in additional sustained and operating load cases (designated “Cold Spring (W+P),
Cold Spring (W+P1+T1)” etc.) which are not used in stress calculations but are used for
support loads and rotating equipment reports.
Cold Spring load cases appear in the Loads menu (under Load cases) after a cold spring (Cut
pipe element) is input into the model. The Load cases menu is shown next:
65
Cold Spring (Cut Pipe)
For analysis, select the desired Cold Spring load cases from those shown. The built-in
Hanger selection procedure does not consider the cold spring since the selection is based on
the first Operating (W+P1+T1) load case. However, if Cold Spring is used, the hanger loads
for the Cold spring load cases [for example, Cold Spring (W+P1+T1)] will include the effect
of the Cold spring.
For an example on Cold Spring modeling and its use in reducing Anchor loads for operating
load case(s), refer http://www.sstusa.com/caepipe-tutorials.php.
66
Comment
Just as a computer programmer benefits immensely from clear documentation about the
program, so too will an engineer benefit from clear notes, design decisions and comments
about a piping system. Use the Comment feature to write as many notes and comments as
required anywhere in the CAEPIPE Layout window. They can be printed along with the
layout data.
Two ways for putting in a comment:
1. Simply type “c” first (in the Node column) on an empty row, or
2. On an empty row, select “Comment” from the Element Types dialog (Ctrl+Shft+T).
Use menu Edit > Insert (Ctrl+Ins) to insert an empty row between two existing rows of
data.
Rows 11, 12, 15 and 17 are the comment lines (highlighted with a light green background).
67
Compressor
Pumps, compressors and turbines in CAEPIPE, referred to as rotating equipment, are each
governed by an industry publication - API (American Petroleum Institute) publishes an API
610 for pumps, ANSI (American National Standards Institute) publishes an ANSI/HI 9.6.2
for Rotodynamic Pumps, API 617 for compressors and NEMA (National Electrical
Manufacturers Association) publishes the NEMA SM-23 for turbines. These publications
provide guidelines for evaluating nozzles connected to equipment among other technical
information including the items relevant to piping stress analysis – criteria for piping design
and a table of allowable loads.
Modeling the equipment is straightforward since it is assumed rigid (relative to connected
piping) and modeled only through its end points (connection nozzles).
1. In your model, anchor all the nozzles (on the equipment) that need to be included in the
analysis.
2. Specify these anchored nodes during the respective equipment definition via Misc. menu
> Pumps/Compressors/Turbines in the Layout window.
CAEPIPE does not require you to model all the nozzles nor their connected piping. For
example, you may model simply one inlet nozzle of a pump with its piping. Or, you may
model one pump with both nozzles (with no connected piping) and impose external forces
on them (if you have that data). Further, there is no need to connect the two anchors of the
equipment with a rigid massless element like required in some archaic methods.
A compressor (like a turbine or a pump) is input by selecting “Compressors” from the Misc
menu in the Layout or List window. Upon analysis, an API 617 compressor compliance
report is produced. See the section titled “Rotating Equipment Qualification” from Code
Compliance Manual for related information.
68
Compressor
Once you see the Compressor List window, double click on an empty row for the
Compressor dialog and enter the required information.
A short description to identify the compressor may be entered for Description. The nozzle
nodes must be anchors and the shaft axis must be in the horizontal plane. Some of the
nozzle nodes may be left blank if they are not considered as a part of the piping system being
analyzed (e.g., extraction nodes).
69
Compressor
If you have input multiple temperatures, corresponding reports for additional operating load
cases are shown.
70
Concentrated Mass
A concentrated mass is input by typing “conc” in the Data column or selecting “Conc.
Mass” from the Data Types dialog.
The Concentrated Mass dialog is shown.
The weight of the concentrated mass should be input for Weight. Weight is to be input in lbf
or kgf and NOT in mass units. Whenever mass is required for a calculation as in the case of
forming Mass matrix for dynamic analysis, or in calculating inertia force as (mass x
acceleration) for static seismic analysis, CAEPIPE internally computes the mass to be equal
to (weight / g-value).
The concentrated mass is located at the offset (DX, DY, DZ) from the node. Deadweight,
seismic and dynamic loads due to concentrated masses are applied to the model.
71
Constant Support
A constant support hanger exerts a constant vertical supporting force on the piping,
irrespective of whether the pipe is in hot or cold condition. It is equivalent to a weight
pulling up the pipe through a pulley, where the same upward force is exerted on the pipe
irrespective of the position of the pipe. The constant support load is automatically calculated
by CAEPIPE. To analyze an existing constant support with a known load, input it as a user
hanger with a zero spring rate.
A constant support is input by typing “cons” in the Data column or selecting “Constant
Support” from the Data Types dialog.
The Constant Support dialog is shown.
Tag
Tag can be 14characters long. Tags are useful in identifying a support while modeling,
reviewing of reports and in field erection. Tag Name entered in this field is shown in all
reports.
Number of Hangers
The number of hangers is the number of separate hangers connected in parallel at this node.
Connected to Node
By default the hanger is connected to a fixed ground point which is not a part of the piping
system. A hanger can be connected to another node in the piping system by entering the
node number in the “Connected to node” field. This node must be above the hanger node.
72
Data Types
Items such as anchors, hangers and external forces, which are defined at nodes, are input in
the Data column as Data types; different from inline elements such as pipes, bends and
valves that connect nodes, and are input in the Type column as Element types.
The Data items can be selected from the Data Types dialog which is opened when you click
on the Data header in the Layout window.
You may also use the command: Misc > Data types,
or press Ctrl+Shift+D to open the Data Types dialog.
You can select the data type by clicking on the radio button or pressing the underlined letter
of the item, e.g., press “f” for Flange, Force or Force spectrum load. Or, you may simply
start typing the first few letters of the item in the Data column. (For example, typing “fo”
automatically opens a Force data type dialog).
73
Direction
Direction is required for several items such as Pump, Compressor, Turbine, Nozzle (for
vessel axis), Limit Stop, Skewed restraint, Elastic element and Hinge joint.
The axis or the orientation of an item (listed above), is called the direction vector which is
described in terms of the vector’s global X, Y and Z components.
The angles the vector makes with the X, Y and Z axes are called Direction angles, whose
cosines are called Direction cosines (or global X, Y and Z components used in CAEPIPE).
There are two methods of computing the X, Y and Z components.
First method: When you know the direction angles (see examples 1, 2 and 3).
Second method: When you know the coordinates of the end points of the vector (see
example 4).
Example 1: Vertical Vessel
Assume a vertical vessel with axis in the Y direction, and , ,  as the direction angles the
axis of the vessel makes with global X, Y and Z axes.
The angles are  = 90° ,  = 0° (since axis is parallel to Y axis) and  = 90° .
So, the direction cosines or X, Y and Z components are
X comp = cos ( = 90° ) = 0,
Y comp = cos ( = 0° ) = 1,
Z comp = cos ( = 90° ) = 0.
For Z vertical: X comp = 0, Y comp = 0 and Z comp = 1.
Example 2: Limit Stop at 45° from the X-axis in the X-Y plane
For a limit stop whose axis is oriented at 45° from the X-axis in the X-Y plane, the angles are
 = 45°,  = 45° and  = 90°.
So, the direction cosines or X, Y and Z components are
X comp = cos ( = 45° ) = 0.70711,
Y comp = cos ( = 45° ) = 0.70711,
Z comp = cos ( = 90° ) = 0.0
74
Direction
Example 3: Hinge axis at 30° from the Z-axis in the Y-Z plane
From the above figure, we have the angles α = 90°, β = 60° and γ = 30°. Assuming L = 1
(or any length), the direction cosines or X, Y and Z components are
X comp = cos (α = 90° ) = 0.0,
Y comp = cos (β = 60° ) = 0.5,
Z comp = cos (γ = 30° ) = 0.866
Example 4: Skewed Support
Assume that we have a skewed support along P1P2 (which is the direction vector) shown in
the figure above, Assume that the coordinates of these two points are P1 = (12’,12’,12’) and
P2 =(15’,16’,14’).
Let us calculate this vector’s global X, Y and Z components. There are two methods here
75
Direction
Short method
X comp = (X2 − X1) = (15 − 12) = 3
Y comp = (Y2 − Y1) = (16 − 12) = 4
Z comp = (Z2 − Z1) = (14 − 12) = 2
Long method
First, let us calculate the length of the vector, L.
𝐿 = √(X2 − 𝑋1)2 + (Y2 − 𝑌1)2 + (Z2 − 𝑍1)2 = 5.385′

The angles α, β and γ which the vector makes with the global X, Y and Z axes are called the
Direction angles of the vector; The cosines of these angles are called Direction cosines.
X 2 − X1 Y2 − 𝑌1 Z2 − 𝑍 1
cos 𝛼 = , cos 𝛽 = , cos 𝛾 = ,
𝐿 𝐿 𝐿
The direction cosines are
X comp = cos α = 0.55709, Y comp = cos β = 0.74278, Z comp = cos γ = 0.37139
For information, the direction angles are α = 56°8’0”, β = 42°1’0” and γ = 68°11’0”.
To verify the results, the sum of the squares of the direction cosines must be 1.0. Thus,
cos2α + cos2β + cos2γ = 0.5570922 + 0.7427822 + 0.3713922 = 1.0
76
Dynamic Analysis
Modal Analysis
The equations of motion for an undamped lumped mass system may be written as:
[𝑀]{𝑢̈ } + [𝐾 ]{𝑢} = {𝐹 (𝑡)} (1)
Where [𝑀]= diagonal mass matrix
{𝑢} = displacement vector
{𝑢̈ } = acceleration vector
[𝐾 ] = stiffness matrix
{𝐹 (𝑡)} = applied dynamic force vector
If the system is vibrating in a normal mode (i.e., free not forced vibration), we may make the
substitutions
{𝑢} = {𝑎𝑛 } sin 𝜔𝑛 𝑡
{𝑢̈ } = −𝜔𝑛2 {𝑎𝑛 } sin 𝜔𝑛 𝑡
{𝐹 (𝑡)} = 0
to obtain
−𝜔𝑛2 [𝑀]{𝑎𝑛 } + [𝐾 ]{𝑎𝑛 } = 0
or
[𝐾 ]{𝑎𝑛 } = 𝜔𝑛2 [𝑀]{𝑎𝑛 } (2)
where {𝑎𝑛 } is the vector of modal displacements of the n𝑡ℎ mode (eigenvector).
Thus we have an eigenvalue (characteristic value) problem, and the roots of equation (2) are
the eigenvalues (characteristic numbers), which are equal to the squares of the natural
frequencies of the modes.
In CAEPIPE, the eigenvalue problem is solved using a determinant search technique. The
solution algorithm combines triangular factorization and vector inverse iteration in an
optimum manner to calculate the required eigenvalues and eigenvectors. These are obtained
in sequence starting from the lowest eigen-pair [𝜔12 ,{𝑎1 }]. An efficient accelerated secant
procedure which operates on the characteristic polynomial
𝑝(𝜔2 ) = 𝑑𝑒𝑡([𝐾 ] − 𝜔2 [𝑀])
is used to obtain a shift near the next unknown eigenvalue. The eigenvalue separation
theorem (Sturm sequence property) is used in this iteration. Each determinant evaluation
requires a triangular factorization of the matrix([𝐾 ] − 𝜔2 [𝑀 ]). Once a shift near the
unknown eigenvalue has been obtained, inverse iteration is used to calculate the eigenvector.
The eigenvalue is obtained by adding the Rayleigh quotient correction to the shift value. The
eigenvector {𝑎𝑛 }, has an arbitrary magnitude and represents the characteristic shape of that
mode.
77
Dynamic Analysis
Orthogonality
For any two roots corresponding to the n thand mthmodes, we may write equation (2) as
which is the orthogonality condition for eigenvectors.
[𝐾]{𝑎𝑛 } = 𝜔𝑛2 [𝑀]{𝑎𝑛 } (3)
2
[𝐾]{𝑎𝑚 } = 𝜔𝑚 [𝑀]{𝑎𝑚 } (4)
If we postmultiply the transpose of (3) by {𝑎𝑚 }, we obtain
([𝐾]{𝑎𝑛 })𝑇 {𝑎𝑚 } = (𝜔2𝑛 [𝑀]{𝑎𝑛 })𝑇 {𝑎𝑚 }
or (5)
{𝑎𝑛 }𝑇 [𝐾]𝑇 {𝑎𝑚 } = 𝜔𝑛2 {𝑎𝑛 }𝑇 [𝑀]𝑇 {𝑎𝑚 }
Premultiplying (4) by {an }T , {an }T [K]{am } = ω2m {an }T [M]{am } (6)

Since [M] is a diagonal matrix, [M] = [M]T. Also, since [K] is a symmetric matrix, [K] = [K]T .
The left sides of equations (5) and (6) are therefore equal.
Subtracting (6) from (5),
(𝜔𝑛2 − 𝜔𝑚
2 ){
𝑎𝑛 }𝑇 [𝑀]{𝑎𝑚 } = 0 (7)
Since ωn ≠ ωm ,
{𝑎𝑛 }𝑇 [𝑀]{𝑎𝑚 } = 0 (8)
which is the orthogonality condition for eigenvectors.

Modal Equations
Since the eigenvectors (modal displacements) may be given any amplitude, it is convenient to
replace {𝑎𝑛 } by {∅𝑛 } such that
{∅𝑛 }𝑇 [𝑀]{∅𝑛 } = 1 (9)
The eigenvectors are evaluated so as to satisfy equation (9) and at the same time keep the
displacements in the same proportion as those in {𝑎𝑛 }. The eigenvectors are then said to be
normalized. Note that equation (7) is still satisfied since, if n = m,𝜔𝑛2 − 𝜔𝑚
2
= 0, and the
remaining terms may be given any desired value.
Equation (2) now may be written for the n th mode as
[𝐾]{∅𝑛 } = 𝜔𝑛2 [𝑀]{∅𝑛 }
Let [Φ]be a square matrix containing all normalized eigenvectors such that the n th column is
the normalized eigenvector for the n th mode. We can therefore write the matrix equation so
as to include all modes as follows:
[𝐾 ]{Φ} = [𝑀][Φ][𝜔2 ] (10)
𝑛
Where [𝜔𝑛2 ]
is a diagonal matrix of eigenvalues. We now premultiply both sides of (10) by
[Φ]Tto obtain
[𝛷]𝑇 [𝐾][𝛷] = [𝛷]𝑇 [𝑀][𝛷][𝜔𝑛2 ] (11)
78
Dynamic Analysis
It may be shown that

[𝛷]𝑇 [𝑀][𝛷] = [𝐼] (12)
where [𝐼] is the unit diagonal matrix. Equation (12) can easily be verified by expansion and
follows from the orthogonality condition and the fact that [Φ] has been normalized.
Equation (11) therefore can be written as
[𝛷]𝑇 [𝐾][𝛷] = [𝜔𝑛2 ] (13)
Returning now to the equation of motion (1),
let {u} = [Φ]{An }
and {ü } = [Φ]{Än } (14)
Where {𝐴𝑛 } is the modal amplitude of the n th mode. This merely states that the true
modal displacements equal the characteristic displacements (eigenvector displacements)
times the modal amplitude determined by the response calculations and, further that the total
displacements are linear combinations of the modal values. If we now pre-multiply equation
(1) by [Φ]Tand substitute equations (14), we obtain
[𝛷]𝑇 [𝑀][𝛷]{𝐴̈𝑛 } + [𝛷]𝑇 [𝐾][𝛷]{𝐴𝑛 } = [𝛷]𝑇 {𝐹(𝑡)} (15)
Substituting from equations (12) and (13) in equation (15),

{𝐴̈𝑛 } + [𝜔𝑛2 ]{𝐴𝑛 } = [Φ]T {𝐹 (𝑡)} (16)
which represents the modal equations of motion.

Support Motion
Solutions for support motion may be obtained if {𝐹 (𝑡)}is replaced by−𝑢̈ 𝑠 (𝑡){𝑀}where
𝑢̈ 𝑠 (𝑡)is the prescribed support acceleration. Thus the modal equations of motion may be
written as
{𝐴̈𝑛 } + [𝜔𝑛2 ]{𝐴𝑛 } = −𝑢̈ 𝑠 (𝑡)[Φ]T {𝑀} (17)
Where 𝐴𝑛 is the relative modal displacement for the nth mode with respect to the support.
The participation factors for the modes are given by
{Γ𝑛 } = [Φ]T {𝑀} (18)
Then the modal amplitude for the n th mode is given by

𝐴𝑛 = Γ𝑛 𝑢𝑛0 (19)
where u0n is the response of a single degree of freedom system having circular frequencyωn .
Using equations (14) and (19), the displacements are given by
{𝑢} = [Φ]{𝐴𝑛 } = [Φ]{Γ𝑛 𝑢𝑛0 } (20)
79
Dynamic Analysis
Effective Modal Mass

Effective modal mass is defined as the part of the total mass responding to the dynamic
loading in each mode. When the participation factor is calculated using normalized
eigenvectors as in equation (18), the effective modal mass for the n th mode is simply the
square of the normalized participation factor,
𝑀𝑛 = Γn2
Effective modal mass is useful to verify if all the significant modes of vibration are included
in the dynamic analysis by comparing the total effective modal mass with the total actual
mass.
Response Spectrum
The concept of response spectrum, in recent years has gained wide acceptance in structural
dynamics analysis, particularly in seismic design. Stated briefly, the response spectrum is a
plot of the maximum response (maximum displacement, velocity, acceleration or any other
quantity of interest), to a specified loading for all possible single degree-of-freedom systems.
The abscissa of the spectrum is the natural frequency (or period) of the system, and the
ordinate, the maximum response.
In general, response spectra are prepared by calculating the response to a specified excitation
of single degree-of-freedom systems with various amounts of damping. Numerical
integration with short time steps is used to calculate the response of the system. The step-by-
step process is continued until the total earthquake record is completed and becomes the
response of the system to that excitation. Changing the parameters of the system to change
the natural frequency, the process is repeated and a new maximum response is recorded.
This process is repeated until all frequencies of interest have been covered and the results
plotted. CAEPIPE provides fourteen (14) response spectra for your convenience. Refer to
Appendix B titled “Response Spectrum Libraries” in CAEPIPE User’s Manual for details.
Since the response spectra give only maximum response, only the maximum values for each
mode are calculated and then superimposed (modal combination) to give total response. A
conservative upper bound for the total response may be obtained by adding the absolute
values of the maximum modal components (absolute sum). However this is excessively
conservative and a more probable value of the maximum response is the square root of the
sum of squares (SRSS) of the modal maxima.
To calculate response of the piping system, for each natural frequency of the piping system,
the input spectrum is interpolated (linearly or logarithmically). The interpolated spectrum
values are then combined for the X, Y and Z directions (direction sum) either as absolute
sum or SRSS sum to give the maximum response of a single degree-of-freedom system:
0
𝑢max at that frequency.
From equation (20), the maximum displacement vector for the nth mode can be calculated
from the maximum response of a single degree-of-freedom system,
{𝑢𝑛 }𝑚𝑎𝑥 = {ϕn }Γn 𝑢𝑛0 𝑚𝑎𝑥
The maximum values of element and support load forces per mode are calculated from the
maximum displacements calculated per mode as above using the stiffness properties of the
structure.
80
Dynamic Analysis
The total response (displacements and forces) is calculated by superimposing the modal
responses according to the specified mode sum method which can be absolute sum, square
root of sum of squares (SRSS) or closely spaced (10%) modes method.
Closely Spaced Modes
Studies have shown that SRSS procedure for combining modes can significantly
underestimate the true response in certain cases in which some of the natural frequencies of
a structural system are closely spaced. The ten percent method is one of NRC approved
methods (Based on NRC Guide 1.92) for addressing this problem.
𝑅 = √ ∑ 𝑅𝑛2 + 2 ∑|𝑅𝑖 𝑅𝑗 |
𝑛=1
where 𝑅 = Total (combined) response

𝑅𝑛 = Peak value of the response due to the nth mode
𝑁 = Number of significant modes
The second summation is to be done on all i and j modes whose frequencies are closely
spaced to each other. Let 𝜔𝑖 and 𝜔𝑗 be the frequencies of the ith and jth modes. The modes are
closely spaced if:
𝜔𝑗 − 𝜔𝑖
≤ 0.1 and 1 ≤ 𝑖 ≤ 𝑗 ≤ 𝑁
𝜔𝑖
Time History
Time history analysis requires the solution to the equations
[𝑀 ]{𝑢̈ } + [𝐶 ]{𝑢̇ } + [𝐾 ]{𝑢} = {𝐹 (𝑡)} (21)
where [𝑀] = diagonal mass matrix
[𝐶 ] = damping matrix
{𝑢̇ } = velocity vector
{𝐹 (𝑡)} = applied dynamic force vector
81
Dynamic Analysis
The time history analysis is carried out using mode superposition method. It is assumed that
the structural response can be described adequately by the p lowest vibration modes out of
the total possible n vibration modes and p < n. Using the transformation𝑢 = Φ𝑋, where the
columns in Φ are the p mass normalized eigenvectors, equation (21) can be written as
𝑋̈ + ∆𝑋̇ + Ω2 X = Φ𝑇 𝐹 (22)
where ∆ = diag (2𝜔𝑖 𝜉𝑖 )
Ω2 = diag (𝜔𝑖2 )
In equation (22), it is assumed that the damping matrix [𝐶 ]satisfies the modal
orthogonalitycondition
{𝜙𝑖 }𝑇 [𝐶 ]{𝜙𝑗 } = 0 (𝑖 ≠ 𝑗 )
Equation (22) therefore represents p uncoupled second order differential equations. These
are solved using the Wilson 𝜃 method, which is an unconditionally stable step-by-step
integration scheme. The same time step is used in the integration of all equations to simplify
the calculations.
Harmonic Analysis
A harmonic analysis is performed to determine the response of a piping system to sinusoidal
loads. Harmonic forces can arise from unbalanced rotating equipment, acoustic vibrations
caused by reciprocating equipment, flow impedance, and other sources. These forces can be
damaging to a piping system if their frequency is close to the piping system’s natural
frequency, thereby introducing resonant conditions. The equation of dynamic equilibrium
associated with the response of the structure subjected to harmonic forces is:
[𝑀]{𝑢̈ } + [𝐶 ]{𝑢̇ } + [𝐾 ]{𝑢} = sin(𝜔𝑡)𝐹 (23)
where [𝑀] = diagonal mass matrix
[𝐶 ] = damping matrix
{𝑢̇ } = velocity vector
𝜔 = frequency of the applied force
𝑡 = time
𝐹 = maximum magnitude of the applied force
It is feasible that multiple harmonic loads may be applied simultaneously at different
locations of a piping system. More complex forms of vibration, such as those caused by the
fluid flow, may be considered as superposition of several simple harmonics, each with its
own frequency, magnitude, and phase.
82
Dynamic Analysis
A harmonic analysis uses the results from the modal analysis to obtain a solution. A single
damping factor is used for all modes.
First, the maximum response for each harmonic load is obtained separately. Then, the total
response for multiple simultaneous harmonic loads is determined by combining the
individual responses. The combination method may be specified as the Root Mean Square
(RMS) or Absolute Sum. Even in the case of a system with a single harmonic load, the said
combination is always carried out, so that the resulting solution becomes an “unsigned” case.
For an unsigned case, the actual values for displacements, element forces and moments, etc.
computed internally by CAEPIPE prior to such combination can be +ve or -ve for the
dynamic event. After the combination, the resulting values become “unsigned”.
Dynamic Susceptibility
Dynamic Susceptibility feature is a screening tool for potentially large alternating stresses.
The dynamic stresses are the dynamic bending stresses associated with vibration in a natural
mode. In other words, the modal analysis result has been generalized to include the
alternating bending stresses associated with the vibration in a natural mode. The dynamic
susceptibility for any mode is the ratio of the maximum alternating bending stress to the
maximum vibration velocity. This “susceptibility ratio” provides an indicator of the
susceptibility of the system to large dynamic stresses. Also, the associated animated mode
shapes include color-spot-markers identifying the respective locations of maximum vibration
velocity and maximum dynamic bending stress. The susceptibility ratio and the graphics
feature provide incisive insights into the reasons for high susceptibility and how to make
improvements.
The “Modal Analysis” output load case in CAEPIPE has been enhanced. In addition to the
modal frequencies and mode shapes, you will see two new results items called “dynamic
stresses” and “dynamic susceptibility.”
In case you do not see these two items in the results dialog, you need to activate this feature
by defining an environment variable. See Annexure I for a detailed discussion.
First method:
An environment variable “HARTLEN” needs to be declared under My Computer >
Properties > Environment > Variable (HARTLEN), and its Value set to (YES). Please check
with your System Admin because different versions of Windows have slightly different
methods of doing it.
83
Dynamic Analysis
Second method:
Open the MS-DOS Command Prompt. Type “SET HARTLEN=YES” (enter), change
directory (using CD command) to where CAEPIPE program files are located, start
CAEPIPE.EXE.
Upon (modal) analysis, the Results dialog will display the required results (dynamic stresses
and dynamic susceptibility).
84
Elastic Element
The elastic element is a general 6 × 6 stiffness matrix, with nonzero diagonal terms and zero
off-diagonal terms. Use this element to model the stiffness of a component unavailable in
CAEPIPE.
An elastic element is input by typing “e” in the Type column or selecting “Elastic element”
from the Element Types dialog.
The Elastic element dialog is shown.
The stiffnesses are in the local coordinate system defined by the directions of the local x-, y-
and z-axes. As done for all other element types, the positive local x-axis for the elastic
element is along the element from the “from” node to the “to” node. The local y-axis should
be perpendicular to the local x-axis (i.e., their dot product should be zero). The local z-axis is
internally calculated as cross product of the local x- and y-axes.
The elastic element is not subjected to any sustained or thermal expansion loads.
85
Expansion Joints
CAEPIPE provides the following types of expansion joints:

1. Ball joint,
2. Bellows,
3. Hinge joint and
4. Slip joint
Using these types, you can also model tied bellows, a gimbal, a dual gimbal, a pressure-
balanced elbow and a tee, a slotted hinge joint and a universal joint among other complex
arrangements.
Before selecting the types and locations for the expansion joints, you must study a piping
system for the direction and magnitude of the thermal movements to be absorbed,
availability of support structures for anchoring, and guiding of the piping. EJMA and
manufacturer catalogs contain technical information that can guide you through this process.
CAEPIPE becomes an ideal “what-if” tool for such rapid studies.
This topic will show how to model the following types of joints: Tied bellows, hinged
bellows, gimbal, universal hinged bellows and pressure balanced joints. Also see discussion
on Bellows.
Example 1: Tied Bellow (without gaps)
Whenever a bellow is present in a piping system, the equipment nozzle/piping support
adjacent to the bellow will experience a pressure thrust force (=pressure thrust area x
pressure) generated by the bellow during normal operation. Tie rods can be added to the
bellow in order to fully absorb such pressure thrust force, while still allowing the bellow to
laterally deflect (i.e., allowing lateral displacement and lateral rotation).
In the example shown below, four tie rods are attached to the bellow without any “gaps” on
tie rods on either side of the bellow. Because there are no “gaps”, the tie rods offer the same
stiffness under both tension and compression (as long as the compression is not large
enough to buckle the tie rods). In order to determine the axial force carried by each tie rod,
pressure thrust area for the bellow must be input. One way of modeling the tie rods is to
lump all four tie rods into a single tie rod along the bellow center line (with tension stiffness
= compression stiffness = n x stiffness of each tie rod = n x EA/L, where ‘n’ is the number
of tie rods, E is the Young’s Modulus of the tie rod material, A and L are the cross-sectional
area and length of each tie rod).
86
Expansion Joints
In the example shown above, the properties of the Tied Bellow are as follows.
Note:
Weight is to be input in lbf or kgf and NOT in mass units. Whenever mass is required for a
calculation as in the case of forming Mass matrix for dynamic analysis, or in calculating
For Bending stiffness of the bellow, the following two options are provided.
Option 1: Input the Bending stiffness as specified by the manufacturer or as reasonably
determined from industry standards such as EJMA. If a non-zero value for Bending stiffness
is input, then leave the “Mean diameter” field blank or zero.
Option 2: If a non-zero value for Bending stiffness is not input as per Option 1 above and
is left blank, then input the actual non-zero value for “Mean diameter”, in which case
CAEPIPE will internally calculate the Bending stiffness for the bellow based on the Mean
diameter and other inputs provided for that bellow. In this case, the Mean diameter is the
“mean” between the outer and inner diameters of any Convolution of the bellow. Since
outer and inner diameters of all convolutions of the bellow are the same, the Mean diameter
is the same for all convolutions of that bellow.
Among the above two options, Option 1 is recommended if you are able to specify a realistic
non-zero value for the Bending stiffness of the bellow.
Tie Rods properties
No. of Tie Rods (n) = 4 Nos.
Diameter of Tie Rod (D) = 3/4”
Length of Tie Rod (L) = 12”
Young’s Modulus of Tie Rod (E) = 29.9E+6 psi
87
Expansion Joints
Stiffness of Tie Rods = n x AE/L = 4 x (π/4) x 0.752 x 29.9E+6 / (12”) = 4.403E+6 lb/in
Accordingly, for Tie Rods, Tension Stiffness = Compression Stiffness = 4.403E+6 lb/in.
Example 2: Tied bellow with free compression

The model shown below has a tied bellow between Nodes 80 & 90. Tie Rod is defined with
the same tension stiffness and compression stiffness of 6.848E+06 lb/in (equals to
combined axial stiffness of 4 Nos. of 1.25” dia. tie rods). However, gaps are set differently in
the tension and compression directions, namely 0.0” in the tension direction and 2.0” in the
compression direction (assuming 2.0” as the maximum compression permitted by the
manufacturer). This allows the bellow to compress freely up to 2.0” and at the same time
restricts the bellow from extension. Beyond 2.0” of compression, compression stiffness of
tie rods will come into play.
88
Expansion Joints
89
Expansion Joints
From“Flex. Joint” displacements results of CAEPIPE, it is observed that the deflection for
bellow between Nodes 80 and 90 is +0.003” for Sustained Case and -1.359” for Expansion
load case (which is less than the compression gap of 2.0” provided). Please observe that the
bellow compresses for the Expansion load in this model as the bellow is in between two
anchors. This confirms that the modeling of Tied bellow with 0.0” gap for tension and 2.0”
gap for compression directions produces the expected results.
90
Expansion Joints
Example 3: Hinged Bellow

A hinged expansion joint contains one bellow and is designed to permite angular rotation in
one plane only, by the use of a pair of pins through hinge plates attached to the expansion
joint ends. The hinges and hinge pins must be designed to restrain the thrust of the
expansion joint due to internal pressure and extraneous forces, where applicable. See Figure
shown below.
The sample model shown below has a Tied bellow between Nodes 30 and 40. The
stiffnesses of the bellow in Axial = 2088 lb/in, Bending = 418 in-lb/deg, Torsion = 100000
in-lb/deg (in case of unavailability of data, set the Torsional stiffness of the bellow to be the
same as the torsional stiffness of equivalent pipe), and Lateral = 34655 lb/in. The stiffnesses
of the hinge plates are assumed to be “Rigid” in this example. Accordingly, to connect the
Bellow Nodes 30 and 40 to Hinge plates, four (4) weightless “Rigid” elements are defined
connecting the Nodes 30-70, 30-110, 40-90 and 40-140 with each one having its length as 9”
(as the OD of the Flange is indicated as 18” in hinged bellow catalog referred). In addition,
four (4) more weightless “Rigid” elements were defined connecting the Nodes 70-80, 81-90,
110-120 and 121-140 and two (2) hinges connecting nodes 80-81 and 120-121.
91
Expansion Joints
Now from the displacements results of CAEPIPE for Expansion load case, it is observed
that the rotation at Node 40 is much larger than the rotation at Node 30 in YY direction. In
other words, the hinges at Nodes 80 and 120 are allowing the two ends of the bellow to
bend. This in effect confirms that the modeling of hinged bellow as shown in this model
produces the expected results.
92
Expansion Joints
93
Expansion Joints
Example 4: Gimbal Bellow

A gimbal expansion joint is designed to permit angular rotation in any plane by the use of
two pairs of hinges affixed to a common floating gimbal ring. The gimbal ring, hinges and
pins are designed to restrain the thrust of the expansion joint due to internal pressure and
extraneous forces, where applicable.
In this sample model, the Gimbal is simulated by connecting the Bellow Nodes 30 & 40
using two “massless” Rigid Elements and one Ball Joint (i.e., a Rigid Element from Nodes 30
to 70 followed by a Ball Joint connecting Nodes 70 & 80 and another Rigid Element from
Nodes 80 to 40). All the stiffnesses of the Ball Joint are made as “Rigid” excepting the
Bending Stiffness. The Bending Stiffness (the same applied in both “local y” and “local z”
directions) is defined as “1” in-lb/deg. In addition, weight of this ball joint is left blank (i.e.,
equal to 0.0).
94
Expansion Joints
95
Expansion Joints
As expected, the “Displacements” results for the bellow displayed in CAEPIPE have a
sudden change in XX and ZZ rotations, confirming the fact that the Gimbal is getting
rotated in the two orthogonal directions due to the deformation of the two orthogonal lines.
96
Expansion Joints
Example 5: Universal Hinged Expansion Joints

Universal Hinged Expansion Joints have two bellows separated by a pipe spool with overall
length restrained by hinge hardware designed to contain pressure thrust. A hinged universal
expansion joint accepts large lateral movements in a single plane with very low spring forces.
This sample model simulates the Universal Hinged Expansion Joints with two Tie Rods
using the CAEPIPE's Tie Rod elements. The advantages of this model are (a) stiffness of the
tie rods can be input explicitly (in this case, stiffness corresponding to 1" dia tie rod is input)
and (b) gaps can be specified to simulate slotted holes.
In this sample model, the Universal Hinged Expansion Joint is simulated by connecting the
Bellow Nodes 30 & 60 using Tie Rods and “massless” Rigid Elements, namely four
“massless” Rigid Elements connecting Nodes 30-100, 30-220, 60-180 and 60-270; two Tie
Rods connecting Nodes 100-180 and 220-270 and four hinges connecting Nodes 140-150,
160-170, 230-240 and 250-260. See snap shots shown below for details.
97
Expansion Joints
98
Expansion Joints
Example 6: Pressure Balanced Elbow Expansion Joint

Pressure Balanced Elbow Expansion Joints can consist of a single or double bellows in the
flow section, and a balancing bellow of equal area on the back side of the elbow. Tie rods
attach the outboard end of the balancing bellow to the outboard end of the flow bellows.
Under pressure, the tie rods are loaded with the pressure thrust force. If the flow bellows
compresses in service, the balancing bellow extends by the same amount without exposing
the adjacent anchors to pressure thrust forces. However, the spring forces associated with
bellows movements are imposed on the adjacent equipment. A pressure balanced elbow type
expansion joint can accept axial compression, axial extension, lateral movements and
very limited angular motion.
The sample model shown below simulates the Pressure Balanced Elbow Expansion Joint
with Four Tie Rods using the CAEPIPE's Tie Rod elements. The stiffness of the tie rods can
be input explicitly (in this case, stiffness corresponding to 1" dia tie rod is input). See snap
shots below for details.
99
Expansion Joints
100
Fiber Reinforced Plastic Piping (FRP)
FRP piping has gained wide acceptance in many industries due to its lightweight nature,
superior corrosion resistance, temperature capabilities and mechanical strength. Several
manufacturers produce different types of FRP pipes and fittings and provide technical
assistance to their customers on design matters through installation. You can model FRP
materials in CAEPIPE and have it calculate deflections, forces, moments and stresses.
To define the FRP material, click on “Matl” in the header row in the Layout window.
In the Material List window that is shown, double click on an empty row to input a new
material or on a material description to edit the material properties.
The Material dialog will be shown.
101
The material name can be up to five alpha-numeric characters. Enter description and density.
You need to select “FR: Fiber Reinf. Plastic (FRP)” from the Type drop-down combo box
before you click on the Properties button. Poisson's ratio (Nu) is a measure of the Poisson
effect, the phenomenon in which a material tends to expand in directions perpendicular to
the direction of compression. Conversely, if the material is stretched rather than compressed,
it usually tends to contract in the directions transverse to the direction of stretching.
When you click on the Properties button, you are shown the table below where you enter
temperature-dependent properties. Additionally, you can define the Axial and Torsional
allowable stresses so that CAEPIPE can use them to compare with calculated stresses under
the FRP "Sorted Stresses” results.
FRP Material Moduli

CAEPIPE requires three moduli for a FRP material:
Axial or Longitudinal (this is the most important one)

Hoop (used in Bourdon effect calculations). If this modulus is not available, use axial
modulus.
Shear or Torsional. If this modulus is not available, use engineering judgment in specifying
1/2 of axial modulus or a similar value. Note that a high modulus will result in high stresses,
and a low modulus will result in high deflections.
For FRP bends, a Flexibility factor of 1.0 is used unless you override it by specifying a
Flexibility factor inside the bend dialog. Also for FRP bends, CAEPIPE uses a default SIF of
2.3. You can override this too by specifying User-SIFs at the bend end nodes (A and B
nodes).
Stiffness matrix
The stiffness matrix for an FRP material is formulated in the following manner:
The stiffness matrix for a pipe is calculated using the following terms:
Axial term = L / EA
Shear term = shape factor x L / GA
Bending term = L / EI
Torsion term = L / 2GI
102
where L = length, A = area, I = moment of inertia

E = Elastic modulus, G = shear modulus
For an isotropic material,

G = E / 2(1 + 𝜈)
where 𝜈 = Poisson’s ratio,
For a FRP material, however,
E = axial modulus and G is independently specified (i.e., it is not calculated using E and 𝜈).
The hoop modulus and FRP Poisson’s ratio are only used in Bourdon effect calculation
where,
Poisson’s ratio used = FRP Poisson’s ratio input x (axial modulus / hoop modulus)
Results
CAEPIPE calculates deflections, forces, moments and stresses. Each item can be seen under
the respective title in Results. FRP element stresses can be seen, sorted or unsorted. These
FRP stresses are computed as per the formulae given in Section titled “Piping Code
Compliance” in the Code Compliance Manual.
103
Axial and Torsional allowables may be entered under material properties so that they can be
used to compare against calculated stresses in “Sorted FRP Stresses.” Forces, Stresses and
Sorted stresses for FRP piping may be printed to a .CSV file (spreadsheet-compatible).
CAEPIPE renders FRP piping in golden color.
104
Flange
A flange is a method of connecting pipes, valves, pumps and other equipment to form a
piping system. It also provides easy access for cleaning, inspection or modification. Flanges
are usually welded or screwed. Flanged joints are made by bolting together two flanges with a
gasket between them to provide a seal. The material of a flange, is basically set during the
choice of the pipe, in most cases, a flange is of the same material as the pipe. There are many
different flange standards being followed worldwide. To allow easy functionality and inter-
changeability, these are designed to have standardized dimensions. Common world standards
include ASA/ANSI/ASME (USA), PN/DIN (European), BS10 (British/Australian), and
JIS/KS (Japanese/Korean).
Gasket
A flange is input by typing “fl” in the Data column or selecting “Flange” from the Data
Types dialog. If flanges are located at the bend end nodes (A, B nodes), or jacket bend nodes
(C, D nodes), the bend flexibility and SIF are internally modified in CAEPIPE.
The Flange dialog is shown.
105
Flange
Several flange types are available – weld neck, socket welded, threaded, lap joint, etc. Use the
Type drop-down combo box to select one.
Weight
The weight you provide should be the total weight of flanges, i.e., if there are two flanges the
weight should be the weight of two flanges.Weight is to be input in lbf or kgf and NOT in
mass units. Whenever mass is required for a calculation as in the case of forming Mass
matrix for dynamic analysis, or in calculating inertia force as (mass x acceleration) for static
seismic analysis, CAEPIPE internally computes the mass to be equal to (weight / g-value).
Gasket Diameter
The gasket diameter is used in calculating equivalent flange pressure in the flange report.
As stated in Section titled “Flange Report” below, Gasket Diameter (G) required is the
“diameter at location of gasket load reaction”. This Gasket Diameter (G) can be calculated as
detailed in “Flange Report” section.
Allowable Pressure
CAEPIPE provides an approximation for the tendency of a flange to leak by calculating an
“equivalent flange pressure” and comparing it to the (user-input) allowable pressure for the
flange in the flange report. Often, the allowable pressure may be conservatively set to the
flange rating. The allowable pressure can be taken from B16.5 (or a similar standard) for the
flange class, material, pressure and temperature.
The temperature-pressure ratings provided in ASME/ANSI B16.5 are computed using the
formula given in para. D2.1 of Annex D of ASME/ANSI B16.5 (given below). The values
thus obtained are listed in a tabular form for all materials at different flange ratings.
PT = (Pr x SI) / 8750 <= Pc
where,
Pc = Ceiling pressure as specified in D3 of Annex D at temperature.
PT = Rated working pressure in psig for specified material at temperature.
Pr = Pressure rating as per Class in Psig.
SI = Selected stress in Psig for specified material at temperature.
106
Flange
For example, from the values shown in Table 2.-1.1 of ASME B16.5 (1998) (shown above),
for a 300# carbon steel flange with Material A105, the allowable pressure is 740 psig at
100°F. This is calculated as detailed below.
Allowable Stress = Minimum (60% of Min. Yield, 1.25 x Allowable Stress at Temperature) as
per para. D2.2 of Annex D of ASME B16.5 (1998).
Accordingly, Allowable Stress (SI) from ASME Boiler and Pressure Vessel Code, Section II,
Part D for A105 at Temperature 100 deg. F = Minimum(60% of Min. Yield, 1.25 x
Allowable Stress at Temp) = Minimum(60% of 36000, 1.25 x 20000) = Minimum(21600,
25000) = 21600 psi
Pressure rating (Pr) = 300 #
So, rated working pressure (PT) = (21600 x 300) / 8750 = 740 psi (same as the value shown
in the snapshot).
107
Flange
Flange Library
You may access the flange library by clicking on the Library button of the flange dialog. The
default weight in the Flange Library is for two Flanges. However, each library dialog has an
option to include weight for a single flange. See checkbox in the Flange library dialog below
to include weight for a single flange.
ANSI Library
European Library
The default weight in the library is the weight of two weld neck flanges (including bolts).
Flange Report
108
Flange
CAEPIPE lists every flange in a model in the flange report. The “Flange Pressure” is an
equivalent pressure calculated from the actual pressure in the piping element, the bending
moment and the axial force on the flange from the operating case(s), as follows:
16 × 𝐵𝑒𝑛𝑑𝑖𝑛𝑔𝑀𝑜𝑚𝑒𝑛𝑡 4 × 𝐴𝑥𝑖𝑎𝑙𝐹𝑜𝑟𝑐𝑒
𝐹𝑙𝑎𝑛𝑔𝑒 𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 = 𝑃𝑟𝑒𝑠𝑠𝑢𝑟𝑒 + +
𝜋 × 𝐺𝑎𝑠𝑘𝑒𝑡𝑑𝑖𝑎3 𝜋 × 𝐺𝑎𝑠𝑘𝑒𝑡𝑑𝑖𝑎2
The Gasket diameter is at the gasket loading location (if it is not input, it is conservatively
assumed to be the internal diameter of the pipe).“Gasket loading location” is the location
where the gasket load reaction is acting. As per ASME Sec. VIII Division 1, “G” is the
diameter at location of gasket load reaction and can be calculated as follows. See snap shots
shown below for details.
G = mean diameter of gasket contact face, inches, when b 0<= ¼ inches or G = outside
diameter of gasket contact face - 2b, inches, when b0> ¼ inches.
where,
b = effective gasket or joint-contact-surface seating width, inches.
b0 = basic gasket seating width, inches.
109
Flange
In the absence of seating width data, G can be taken as the mean diameter[= (ID + OD) /
2]. For example, as per the snap shot shown below, the mean diameter G for the "Ring
Gaskets for 250/300# ANSI Pipe Flanges" for 8” is 10.375” [= (8.625” + 12.125”) / 2].
The above equation for equivalent flange pressure is based on the article NC3658.1 of
ASME Section III Class 2, 1992 or later editions, around page 176. The computed equivalent
flange pressure is compared with the flange allowable pressure.
If you have input more than one temperature load, the flange equivalent pressure is
calculated for all the applicable operating load cases, the worst of which is reported in the
Flange report.
A flange report is generated even when no piping code is chosen. The flange report is shown
in the results.
110
Flange
Suggestions for dealing with high equivalent flange pressure to allowable ratios
The Flange report in the CAEPIPE results window shows the loads at each flange location
for the worst operating load case (W+P+T).
The “equivalent” flange pressure is the sum of three terms from the flange equation as
shown above. The last column in the Flange report shows a ratio of this equivalent flange
pressure to a user-input allowable pressure. This ratio is flagged in red when greater than 1.0.
Ensure that you input an allowable pressure for the flange by looking up B16.5 or a
similar code (as a function of design temperature and pressure).
Since a flange is unlikely to fail by collapse, the key idea of the flange report is to
“quantify” the tendency of a flange to leak its contents. Engineering judgment will
play an important role in interpreting this report.
If the ratio of equivalent flange pressure to allowable pressure is flagged in red, then
try to reduce the bending moment at that flange location. Be sure to examine all load
cases, but frequently the excessive moments come from the expansion case. If so,
consider introducing loops, bends and offsets as required to reduce the bending
moments at flange locations for expansion load case(s).If it is practical, move those
flanges with high (flange pressure-to-allowable) ratios to piping locations where the
bending moments are less.
111
Flange Qualification Module - Flange and Bolt Stresses
Flange joints are essential components in all pressurized systems; they are also one of the
most complex. Many factors are involved in determining the successful design and operation
of a bolted flange joint service, namely, the interaction between the bolting, flange, and
gasket as well as important non-linear variables such as friction and gasket properties. The
Pressure Vessel and Piping Codes were developed with safety in mind; they provide a
method for sizing the flange and bolts to be structurally adequate for the specified design
conditions.
The Flange Qualification module implemented in CAEPIPE addresses the design rules
contained in the ASME Section VIII, Division 1, Appendix 2 on bolted flange connections
with gaskets.
These design rules will help you to obtain better insight into a flange joint's tendency to leak,
beyond that provided by the rudimentary (yet indicative) flange report produced by a piping
analysis, as seen in the previous section. You can examine the flange and bolt stresses arising
from the bolt tightening loads required for a leakage-free joint.
The Flange Qualification module assumes that you already have flanges and gaskets picked
out for your system and performed a piping flexibility analysis of that system with
CAEPIPE, which will have produced a flange reportas shown below. The previous section
titled “Flange” provides details on this flange report.
Note that this Flange Qualification module to calculate flange and bolt stresses is separate
from a piping stress model file and can be accessed from File Menu > Open/New command.
TheFlange Qualification module performs three (3) qualifications namely,
1. Equivalent Flange Pressure as per NC3658.1 of ASME Section III Class 2,
2. Flange Stresses for Operating Case as per Appendix 2 of ASME Section VIII Division 1,
and
3. Flange Stresses for Gasket Seating Case as per Appendix 2 of ASME Section VIII
Division 1.
Out of the three (3) qualifications listed above, Axial Force and Bending Moment entered in
“Flange Qualification” module are used ONLY in calculating the Equivalent Flange Pressure
as per NC3658.1 of ASME Section III Class 2. The same results are also shown in “Flange
report” under CAEPIPE Results. The equation involved in calculating Equivalent Flange
Pressure is given in the Section above titled “Flange”.
On the other hand, Flange Stresses and Flange Rigidity Factors computed as per ASME
Section VIII Division 1 Appendix 2 are independent of Axial Force and Bending Moment as
112
observed from the detailed write-up given in the Section titled “Flange Qualification” of
CAEPIPE Code Compliance Manual.
Because this module accepts axial load and bending moment at a flange as input among many
others, you will need to first create in CAEPIPE your pipe stress model that includes flanges
(which you need to validate) and generate a Flange Report as shown above. Such a report will
contain the information you can now use in the Flange Qualification module to calculate
flange and bolt stresses.
When you first create a new Flange Qualification file, it comes populated with default values
for a sample flange (see example 1 later on in this topic).
113
Double-clicking anywhere in the previous screen (or Edit menu > Edit (Ctrl+E)) opens a
dialog with input fields (with default values) you can edit. You will need to enter all of your
flange data in this dialog. The different parameters you see here are explained in detail in the
Section titled “Flange Qualification” of the Code Compliance manual.
114
Required flange input information is organized into three Property tabs – Flange Details,
Bolt and Gasket Details, and Load Data, the last of which accepts data from a piping model’s
Flange Report. Once all the data is input, save the model (Flange Qualification filenames will
have a .flg extension). Now, select File menu > Analyze to calculate flange stresses, which
will be shown right below the input information.
115
There are three main sections in the shown results:

Flange Equivalent Pressure (same as the one shown in piping model results > Flange
Report),
Flange stresses in the Operating condition, and
Flange stresses in the Gasket seating condition
116
Flange Qualification Module Menus

Use these numbers in accordance with the ASME Section VIII publication and your
engineering judgment to qualify these flanges for use in your piping systems.
File Menu
.Analyze.
Analyze command calculates the flange and bolt stresses and compares them to the input
allowable stresses.
Print.
You can print a Flange Report by using the Print command. You can also preview the report
by clicking the Preview button on the print dialog.
117
Edit Menu
.Edit.
You can edit the Flange, Bolt, and Gasket Details, as well as Load Data by clicking the Edit
command.
118
Flange Qualification Window

You can edit the Flange, Bolt, and Gasket Details, as well as Load Data by clicking the Edit
command.
Flange Details Tab
Modify the flange details using this window.
119
Bolt and Gasket Details Tab

Modify the bolt and gasket details using this window.
120
Load Data Tab

Modify the load data using this window.
Options Menu
.Units.
See Units in the Layout Window Options Menu section of the CAEPIPE User’s Manual.
.Font.
See Font in the Layout Window Options Menu section of the CAEPIPE User’s Manual.
121
Sample Problem
Problem 1:
(Example on page 19 in Chapter 40 “Bolted-Flange Joints and Connections” by William J.
Koves on “Companion Guide to the ASME Boiler & Pressure Vessel Code” by K .R. Rao
[2001], American Society of Mechanical Engineers, U.S.)
Flange Details:
Flange Type : Integral Flanges
Flange Outside Diameter [A] = 39.125 (inch)
Flange Inside Diameter [B] = 32 (inch)
Flange Thickness [t] = 2 (inch)
Small End Hub Thickness [g0] = 0.5 (inch)
Large End Hub Thickness [g1] = 1.125 (inch)
Hub Length [h] = 2.75 (inch)
All. Stress @ Design Temp [sf] = 19600 (psi)
All. Stress @ Ref. Temp [sfa] = 20000 (psi)
Modulus @ Design Temp [E] = 2.7E+7 (psi)
Modulus @ Ref. Temp [Ea] = 2.92E+7 (psi)
Bolting Information:
Bolt Circle Diameter = 37 (inch)
Number of Bolts = 36
Bolt Diameter = 1 (inch)
All. Stress @ Ref. Temp [sa] = 25000 (psi)
All. Stress @ Design Temp [sb] = 25000 (psi)
Gasket Information:
Gasket Outside Diameter = 35.5 (inch)
Gasket Inner Diameter = 33.5 (inch)
Leak Pressure Ratio [m] = 3.00
Gasket Seating Stress [y] = 10000 (psi)
Facing Sketch = 1
Facing Column = 1
Load Data:
Design Pressure = 414 (psi)
Design Temperature = 500 (F)
Axial Force = 1000 (lb)
Bending Moment = 200 (ft-lb)
Allowable Pressure = 665 (psi)
122
Comparison of Results
Flange Stresses Text Book Results CAEPIPE CAESAR II
(psi) (psi) (psi)
Operating condition
Longitudinal Hub (SH) 24150 24152 24227
Radial Flange (SR) 11590 11590 11636
Tangential Flange (ST) 7230 7232 7205
0.5(SH + SR) 17870 17871 17932
0.5(SH + ST) 15690 15692 15716
Gasket Seating Condition
Longitudinal Hub (SH) 15270 15269 15292
Radial Flange (SR) 7330 7327 7345
Tangential Flange (ST) 4570 4572 4547
0.5(SH + SR) 11300 11298 11318
0.5(SH + ST) 9900 9921 9920
Legend for the different parameters and more examples are given in Section titled “Flange
Qualification” in the Code Compliance Manual.
123
Force
External forces and moments on the piping system in the global X, Y and Z directions may
be input at any location. Type “fo” in the Data column or select “Force” from the Data
Types dialog.
The Force dialog is shown.
If you select “Add to W+P”, the specified forces and moments are applied to the sustained
and operating load cases.
If you select “Add to SEISMIC”, the specified forces are applied to the Static Seismic Load
case. Moments cannot be defined for the Static Seismic Load case and hence moment input
fields are disabled.
Force defined in Global X direction (FX) will be included only with x’g solution when x’g
acceleration input is non-zero. Similarly, forces defined in Global Y (FY) and Global Z (FZ)
directions will be included in y’g and z’g solutions respectively, when y’g and z’g accelerations
are input as non-zero values.
See Section titled “Static Seismic Load” from CAEPIPE User’s Manual for further details on
how CAEPIPE performs Static Seismic Load analysis.
If you select any of the thermal cases (“Add to T1”, “Add to T2”, “Add to T3”, all the way
up to “Add to T10”), the specified forces and moments are applied to the selected thermal
load case (T1/T2/T3/…/T10) and its operating load case counterpart (i.e.,T1 and
W+P1+T1, or T2 and W+P2+T2, or T3 and W+P3+T3 or T4 and W+P4+T4 or T5 and
W+P5+T5 or T6 and W+P6+T6 or T7 and W+P7+T7 or T8 and W+P8+T8 or T9 and
W+P9+T9 or T10 and W+P10+T10).
124
Force Spectrum
Force spectrum analyses (different from harmonic analyses) are generally performed to
determine the response of the piping system to short-duration impulsive loads such as fluid
hammer, safety relief valve (SRV) and slug flow loads. For an actual short-duration impulsive
dynamic load exerted on a piping system, a fluid transient analysis is first carried out in order
to arrive at the “time-history loads” (i.e., force vs. time) acting in the three global directions
(namely global X, Y and Z) at all affected points in the piping system. The time-history load
sets so computed are then applied, one time-history load set at a time, on a single degree-of-
freedomspring-mass system with a pre-set natural frequency, to determine the maximum
dynamic response of this single degree-of-freedomsystem with that natural frequency. Such
dynamic analysis for that time-history load is repeated on the same single degree-of-freedom
system with different pre-set natural frequencies. The force spectrum for that time-history
load would then be a table of maximum dynamic response computed for the single degree-
of-freedom system with different natural frequencies. In other words, the force spectrum is a
table of force spectral values vs frequencies that captures the maximum intensity and
frequency content of that time-history load. Similarly, force spectrum tables are determined
for all other time-history load sets. The above force spectrum tables (i.e., maximum dynamic
force vs frequency) are then applied as inputs at the respective piping nodes of the
CAEPIPE stress model to compute displacements, forces and stresses.
For any piping system, there are as many natural modes of vibrations as the number of
dynamic degrees of freedom for that system. The force spectral value corresponding to a
natural frequency of the piping system is arrived at by interpolating the force spectrum vs
frequency table as determined above. For better understanding, as an example, refer to the
graph shown next as well as the natural frequencies computed for a piping system at 10 Hz,
14 Hz, 21 Hz, 29 Hz and 33.8 Hz.
125
Force Spectrum
From the above graph, to arrive at a force value corresponding to a natural frequency of 14
Hz, CAEPIPE interpolates the force spectral values between 13 and 15 Hz. Similarly, to
arrive at a force value corresponding to a natural frequency of 21 Hz, CAEPIPE interpolates
the force spectral values between 20 Hz & 25 Hz. Since force spectral values above 25 Hz are
not defined in the graph shown above, CAEPIPE arrives at a force value of 1650 lb. (i.e., the
spectral value corresponding to the maximum frequency of 25 Hz in the above plot) even for
natural frequencies of 29 and 33.8 Hz. Similarly, CAEPIPE arrives at a force value of 900 lb.
for a natural frequency of 10 Hz (i.e., the spectral value corresponding to the minimum
frequency of 13 Hz in the above plot).
If only one set of force versus frequency is input (for example, 1000 lb. at 14 Hz) in the force
spectrum table for your model, CAEPIPE applies the same force (1000 lb.) for all natural
frequencies computed for that piping system. Note that the displacement produced at a node
will remain unchanged even when the sole frequency in the force spectrum table is changed
from 14 Hz to any other frequency.
Here, the results of the modal analysis are used with force spectrum loads to calculate the
response (displacements, support loads and stresses) of the piping system. It is often used in
place of a time-history analysis to determine the response of the piping system to sudden
impulsive loads such as water hammer, relief valve and slug flow. The force spectrum is a
table of spectral values versus frequencies that captures the intensity and frequency content
of the time-history loads. It is a table of Dynamic Load Factors (DLF) versus natural
frequencies. DLF is the ratio of the maximum dynamic displacement divided by the
maximum static displacement. Note that Force spectrum is a non-dimensional function
(since it is a ratio) defining a curve representing force versus frequency. The actual force
spectrum load at a node is defined using this force spectrum in addition to the direction (FX,
FY, FZ, MX, MY, MZ), units (lb, N, kg, ft-lb, in-lb, Nm, kg-m) and a scale factor.
The Force spectrums are input from the Layout or List menu: Misc > Force spectrums.
126
Force Spectrum
The Force spectrum list appears.
Enter a name for the force spectrum and spectrum values versus frequencies table. In
addition to inputting the force spectrum directly, it can also be read from a text file or
converted from a previously defined time function.
127
Force Spectrum
To read a force spectrum from a text file:

use the List menu: File > Read force spectrum.
The text file should be in the following format:

Name (up to 31 characters)
Frequency (Hz) Spectrum value
. .
. .
. .
The frequencies can be in any order. They will be sorted in ascending order after reading
from the file.
To convert a previously defined time function to force spectrum:
use the List menu: File > Convert time function.
The Convert Time Function dialog appears.
128
Force Spectrum
Select the time function to convert from the Time function name drop down combo box.
Then input the Force spectrum name (defaults to the Time function name), Maximum
frequency, Number of frequencies and the Damping. When you press Enter or click on OK,
the time function will be converted to a force spectrum and entered into the force spectrum
list.
The time function is converted to a force spectrum by solving the dynamic equation of
motion for a damped single spring mass system with the time function as a forcing function
at each frequency using Duhamel’s integral and dividing the absolute maximum dynamic
displacement by the static displacement.
Force Spectrum Load
The force spectrum loads are applied at nodes (in Data column in Layout window). At least
one force spectrum must be defined before a force spectrum load at a node can be input.
To apply the force spectrum load at a node click on the Data heading or press Ctrl+Shift+D
for Data Types dialog.
Select “Force Sp. Load” as the data type and click on OK. This opens the Force Spectrum
Load dialog.
Select the direction, units and force spectrum using the drop down combo boxes and input
appropriate scale factor. The scale factor can be a scalar value, which, when multiplied by the
non-dimensional force spectrum, will give the actual magnitudes of the force versus
frequency in the global direction and unit selected in the above dialog. Then click on OK to
enter the force spectrum load at that node.
129
Force Spectrum
Input force spectrum loads at other nodes similarly. Then select the force spectrum load case
for analysis using the Layout menu: Loads > Load cases.
Note that Modal analysis and Sustained (W+P) load cases are automatically selected when
you select Force spectrum load case. The force spectrum load case is analyzed as an
Occasional load.
A Relief Valve Analysis may be performed by first obtaining the data about valve opening
and subsequent behavior, as a force versus time history profile. Enter the profile as a time
function. See under Time History Load for how to.
Force
Time
Valve opening time Valve closing time
(a) (b)
Then under Force Spectrum, use “Convert time function” to convert the force-time history
profile into a Force-Spectrum. Input loads and analyze. See the topic Relief Valve Load
Analysis for more details.
130
From (Node)
“From” is a special Element Type used to start a new (branch) line. For a new node number,
any values you type in under the DX, DY and DZ columns are taken to be coordinates (and
not offsets). If you use “From” for an existing node number, then you do not have to specify
values for the DX, DY and DZ fields.
If you specify a new node number other than the global origin [that has coordinates (0, 0, 0)],
then you must specify the coordinates for the new node using “From” and DX, DY and DZ
fields. If you do not, then the new node number will have the same location as the node that
is the global origin.
“From” is input by typing “f” in the Type column or selecting “From” from the Element
Types dialog.
The first node of a model is always a fixed “From” node since you have to start the model
from a point. The DX, DY, DZ fields for this node may be left blank to mean the global
origin (0,0,0) or global coordinates may be specified in the DX, DY and DZ fields to have a
nonzero reference point for the model.
Values specified for DX, DY and DZ for any other node other than a “From” node are
interpreted as offsets (not as coordinates) unless the node is suffixed with an asterisk “*”. See
the topic “Node” for more information.
It is helpful to create a model starting from a reference node located using its global
coordinates and continue in an orderly manner from there, especially when you plan to
merge (see File menu) two models later.
131
From (Node)
Example:
Assume you had two separate models you wanted to combine later.
First model: In this first model, you would start modeling from node 10 up to node 95.
Since Node 100 in above is a Tangent-Intersection-Point (TIP), it is not a good idea to create
a model upto TIP. That means, the bend at Node 100 would not know which way it should
turn!!
Second model: You would model from node 95 up to node 130, which is a better way to
model rather than modeling from node 130 to node 95, which would complicate the
“Merge” process.
132
Generic Support
A generic data type called “Generic support” is available to define a complex support fitting
for which the support stiffnesses are obtained from an external source, using a reduced 6x6
stiffness matrix. An example would be a support assembly as shown below.
Interface Nodes (at

center of circular
clamps)
A reduced 6x6 stiffness matrix representative of this support at a single interface node (as
shown in figure above) will first need to be arrived at using any general purpose FEA
software package (such as ANSYS, NASTRAN, etc.) Then, those stiffness values
representing the support assembly can then be input into CAEPIPE model for further
analysis.
“Generic Support” is input by typing “ge” in the Data column or selecting “Generic
Support” from the Data Types dialog.
Since the reduced 6x6 stiffness matrix representing the generic support is always a symmetric
matrix, only the upper triangular stiffness values need to be input, with the diagonal terms
having positive non-zero values. Diagonal terms are never zero or left blank. Off-diagonal
terms can be zero, positive or negative values.
133
Generic Support
Units for Groups A and C are controlled centrally through the Units command (Ctrl+U in
Layout window). Units for Group B need to be set in this dialog.
The Stiffnesses in the Groups A, B and C are to be computed in Global Coordinate System
and entered.
The fields in Group A are translational stiffnesses. The fields in Group C are rotational
stiffnesses.
The fields in Group B are the coupling stiffnesses. For example, a lateral force at the end of a
Cantilever beam produces not only the displacement in the direction of that force, but also a
rotation. Similarly, a moment at the end of a Cantilever beam produces not only the rotation
in the direction of that moment, but also a lateral displacement.
Each stiffness field in the dialog is editable. The default “rigid” stiffness is shown for all the
diagonal terms. If you have a stiffness value for any of these including the off-diagonals, enter
them here (ensure units integrity). Graphically, the support is shown as a solid block (at node
330 in the next figure).
134
Generic Support
Displacements, support loads and support load summaries are shown for this support type.
135
Guide
A guide is used in the field to control or direct pipe movement. Likewise, its CAEPIPE
equivalent allows axial movement while restraining the pipe against lateral translations (but
not rotations). A guide restricts the translational movement normal to its axis, i.e.,
displacements are restrained in the local y and z directions of the element to which the guide
is attached. The Local Coordinate System (LCS) of the guide can be viewed through View >
List > Guide and right-click mouse and select the option “Show LCS”.
A guide is input by typing “g” in the Data column or selecting “Guide” from the Data Types
dialog.
The Guide dialog is shown.
Tag
Tag can be 12 characters long. Tags are useful in identifying a support while modeling,
reports.
Friction Coefficient
When a friction coefficient is entered, a nonlinear analysis is performed. In each iteration, the
friction force is calculated which is friction coefficient times the normal force (the vector
sum of local y and local z reaction forces). This friction force is applied in the local x
direction opposing the axial motion of pipe. The solution converges when the displacement
changes by less than 1% between successive iterations.
136
Guide
Stiffness
The default stiffness is rigid which is input by typing “r” or “Rigid” in the Stiffness field. A
non-rigid stiffness may be entered by typing the value of the stiffness in the Stiffness field.
Gap
A clearance between the pipe and the guide, if present, may be entered as a Gap. The gap is
assumed to be symmetric about the guide axis. This gap must be closed before any restraint
forces can develop. If there is no gap, leave this field blank or enter it as 0.0.
Connected to Node
By default the guide is assumed to be connected to a fixed ground point which is not a part of
the piping system. A guide can be connected to another node in the piping system by
entering the node number in the “Connected to Node” field.
Local Coordinate System (LCS)
A guide’s local x-axis is based on the preceding element. If a preceding element is
unavailable, the following element is used to determine the guide’s local x-axis. The local
coordinate system (LCS) may be viewed graphically from the Guide List window using the
menu: View > Show LCS.
Guide forces in global coordinate system are “Print(ed) to file” (in addition to forces in local
coordinate system), accessible from the Print command dialog.
137
Hanger
Variable spring hangers support the dead-weight of piping while allowing vertical and lateral
thermal movement from the installed to the operating condition. CAEPIPE assumes a
spring hanger to always act in the vertical direction.
Also called a “To be designed” hanger, it is input by typing “h (Enter)” in the Data column
(or typing “Han”) or selecting “Hanger” from the Data Types dialog. The Hanger dialog is
shown.
Tag
Tag can be 12 characters long. Tags are useful in identifying a support while modeling,
reports.
138
Hanger
Type
The type (i.e., manufacturer) of the hanger can be selected from the drop-down combo box
“Type.” The following hanger types are currently available:
Hanger Types
ABB-PBS Fee & Mason Nordon
Basic Engineers Flexider (30-60-120) NPS Industries
Bergen-Paterson Flexider (50-100-200) Piping Services
Bergen-Paterson (L) Fronek Piping Tech & Products
BHEL Hyderabad Grinnell Power Piping
BHEL Trichy Hydra Sanwa Tekki(30-60-120)
Borrello Lisega Sanwa Tekki(85-170)
Carpenter & Paterson* Mitsubishi (30-60-120) Sarathi
Comet Mitsubishi (80-160) Spring Supports
Corner & Lada Myricks SSG
Dynax NHK (30-60-120) Gradior
Elcen NHK (80-160)
*CAEPIPE includes catalog from Carpenter Paterson Ltd. based in England.
Number of Hangers
The number of hangers is the number of separate hangers connected in parallel at this node.
The stiffness and load capacity of each hanger are multiplied by the number of hangers to
find the effective stiffness and load capacity of all the hanger supports at this node.
Load Variation
The load variation (in percent) is the maximum variation between the cold and hot loads.
Typical value is 25%.
Hanger Below
Changesthe graphical depiction only.
This should be used to specify whether the Hanger is placed below the pipe. Graphical
symbol changes accordingly. This selection is valid only for Hanger and User Hanger and not
for Constant Support Hanger nor Rod Hanger.
Short Range
Short range hangers are used if the available space is not enough for installing mid-range
hangers. They are considered, however, as a specialty item and generally not used. If a short
range hanger is to be designed, check the Short Range check box.
Connected to Node
node number in the “Connected to node” field. This node must be directly above or below the
hanger node.
139
Hanger
Hanger Design Procedure

1. Calculate Hot Load
Hot load for a variable spring hanger is the actual weight of the pipe (including the
weights of content, lining and insulation) being carried by that hanger. To calculate hot
load, a preliminary sustained load analysis is performed in which all hanger locations are
restrained vertically. If any anchor is to be released (so that the hanger rather than the
nearby equipment takes the sustained load), it is released. The reactions at the hanger
locations from this preliminary sustained load analysis are the hanger hot loads.
2. Calculate Hanger Travel
Vertical restraints (applied in step 1) at hanger locations are removed. Released anchors
(if any) are restored. A preliminary operating load case analysis is performed. If multiple
thermal load cases are specified, only the first thermal load is used for this operating load
case. The hot loads (calculated in step 1) are applied as upward forces at the hanger
locations. Vertical displacements at the hanger locations obtained from this operating
load case analysis are the hanger travels. If limit stops are present, the hot loads are
recalculated with the status of the limit stops at the end of the preliminary operating load
case. Then the hanger travels are recalculated using the recalculated hot loads.
3. Select Hanger
The hanger is selected from the manufacturer’s catalog based on the hot load and hanger
travel. The cold load is calculated as: cold load = hot load + spring rate x hanger travel.
The hanger load variation is calculated as: Load Variation = 100 x Spring rate x travel /
Hot load. The calculated load variation is checked against specified maximum load
variation. The hanger for which both the hot and cold loads are within the hanger’s
allowable working range and the load variation is less than the allowed load variation is
selected. The hanger is selected such that the hot load is closest to the average of the
minimum and maximum loads.
4. Install Hangers
If “Include hanger stiffness” is chosen in the Analysis options: The hanger spring rates
are added to the overall stiffness matrix. The hanger cold loads are used in the sustained
and operating load cases. If “Do not include hanger stiffness” is chosen in the Analysis
options: The hanger spring rates are not added to the overall stiffness matrix. The hanger
hot loads are used in the sustained and operating load cases.
For Lisega hangers, size column will report as Hanger Number, Type and Range instead of
Hanger Number, Range and Type. For example, hanger having a spring rate of 2.1 N/mm,
vertical travel of 30mm with load 440N will be reported as 21D.193 instead of 21D319.
Sustained Displacement during Hanger Selection
An Example Application
Sometimes, rotating equipment vendors (e.g. turbine vendors) require that there be no
weight load imposed on the turbine connections after welding/bolting the pipe to the nozzle
but prior to plant start-up. This can be accomplished only if the pipe is hung by variable
spring hangers like a swing at the nozzle and if the pipe end of such “hung” pipe is almost
140
Hanger
perfectly aligned with the turbine nozzle prior to welding or bolting that pipe end to the
turbine nozzle. In other words, the spring hangers that carry the weight of the “hung” pipe
near the nozzle are to be sized and placed such that the resulting displacements (i.e., 3
translations and 3 rotations) at the pipe end are nearly zero, so that the pipe end need not be
forcibly deformed in order to weld/bolt it to the turbine nozzle. This, in turn, makes sure
that the weight load of the pipe is not imposed on the turbine nozzle prior to plant start-up.
CAEPIPE can be used to perform the above study by carrying out the following steps.
a) During Step 1 of the Hanger Design procedure given above, release all 6 degrees
of freedom of the anchor corresponding to that turbine nozzle, so that piping
weight load will not be transferred to the nozzle during the initial Sustained load
analysis (in which all hangers are pinned, thereby restraining the pipe vertically at
the hanger locations).
b) Review the preliminary Sustained Load Displacements computed during Hanger
Selection via Results > Displacements > Mouse right click > Show sust. disp.
during hanger selection. If the Sustained Load translations and rotations at the
concerned pipe end are NOT nearly zero, try out different hanger locations till
the preliminary Sustained Load displacements at that pipe end are nearly zero.
c) When the Preliminary Sustained Load displacements at the concerned pipe end
are nearly zero, the Support Loads at that Nozzle reported by CAEPIPE for
Sustained Load case would be close to zero.
In results, the sustained displacements during hanger selection when hangers are pinned can
be shown via Results >Displacements>Mouse right click>Show sust. disp. during hanger
selection.
141
Hanger
And here is the Results Window for sustained displacements during hanger selection after
selecting this feature.
142
Hinge Joint
Hinge joint is an expansion joint designed to permit angular rotation in a single plane by use
of a pair of pins that pass through plates attached to the expansion joint ends. Hinge joints
are used in sets of two or three to absorb pipe movement in one or more directions in a
single-plane piping system. A pair of hinge joints, separated by a section of piping, will act
together to absorb lateral deflection. Hinge joints are designed to take the full pressure thrust.
The two sides of the hinge joint shown are joined
by hinge pins which are along the hinge axis shown
in the figure. A hinge is modeled by two nodes,
one on each side of the hinge joint. The two nodes
of the hinge joint are coincident. So, it is a zero
length element, i.e., the “From” and “To” nodes
are coincident. Hence, the DX, DY and DZ fields
in the Layout window should be left blank.
A hinge joint is input by typing “h” in the Type column or selecting “Hinge joint” from the
Element Types dialog.
143
Hinge Joint
The Hinge joint dialog is shown.
Rotational Stiffness
Also called Angular stiffness. Input the stiffness around the rotational (hinge) axis. The
stiffness value may be available from the manufacturer of the hinge joint or from test results.
Otherwise engineering judgment may be used. The stiffness values may be left blank. In that
case a very small value (1 in-lb/rad.) is used internally to avoid dividing by zero.
Rotation Limit
Rotation limit is an upper limit on the rotation of hinge joint in the plus or minus directions.
Rotation limit of 0.0 (zero) means it is unable to rotate (i.e., it is rigid). Rotation limit of
“None” or Blank means that there is no limit on its rotation.
Friction Torque
The hinge joint will rotate only if the external torque exceeds the friction torque. Beyond
that, the rotation is proportional to the rotational stiffness of the hinge joint. The friction
torque value may be available from the manufacturer of the hinge joint or from test results.
Otherwise engineering judgment may be used. If you do not want friction in the hinge joint,
the friction torque value may be left blank.
144
Hinge Joint
When the applied torque is less than friction torque, there is no rotation. When the applied
torque exceeds friction torque, the rotation is calculated as shown above. When rotation limit
is reached, there is no further rotation irrespective of the applied torque. See discussion for
Kbe under Ball joint for additional information.
Weight
This is the total weight of the Hinge joint assembly.Weight is to be input in lbf or kgf and
NOT in mass units. Whenever mass is required for a calculation as in the case of forming
Mass matrix for dynamic analysis, or in calculating inertia force as (mass x acceleration) for
static seismic analysis, CAEPIPE internally computes the mass to be equal to (weight / g-
value).
Axis direction
The hinge axis is specified by the “Axis direction.” See “Direction,” for more information on
specifying a direction using X comp, Y comp and Z comp.
145
Hinge Joint
Example
Assume that we had the model shown below (with an exaggerated deflection) containing 6”
piping with a pair of Hinge joints. Each Hinge joint has the following data: rotational
stiffness of 66 in.-lb./degree, and weight of 35 lb.
The following steps describe the modeling procedure (with auto node incrementing, you do
not have to type in the node numbers below):
 The first node 10 is already defined as an anchor. Press Enter to move to the next
row.
 Type 20 for Node, 3’6” for DX, enter material, section and load names, Guide for
Data. Press Enter to move to the next row.
 Input bend at node 30: Type 30 for Node, press Tab to move to the Type field. Type
“b” and Tab to next column to enter a bend, 2’ for DX, press Enter to move to the
next row.
 Type 40 for Node, enter –1’6” for DY (as the offset from node 30 to node 40), press
Enter to complete the bend and move to next row.
 Type 50 for Node. Type “h” in the Type column. This shows the hinge dialog.
146
Hinge Joint
Enter 66 (in.-lb/deg.) for Rotational stiffness, 35 (lbf) for Weight, 1.0 for Z comp (axis
direction), press Enter or click on OK to close the dialog. This completes the hinge input.
Since there cannot be any offsets (DX, DY, DZ) for the hinge node from the previous node,
the cursor automatically moves to the next row.
 Type 60 for Node, enter –1’6” for DY (as the offset from node 50 to node 60), press
Enter to move to next row.
 Type 70 for Node. Type “h” in the Type column. This shows the hinge dialog. Enter
the hinge data as before and click on OK to move to the next row.
 Input bend at node 80: type 80 for Node, press Tab to move to Type field, type “b”
for Type and Tab to next column to enter a bend. Type –1’6” for DY (offset from
node 70 to bend node 80), press Enter.
 Complete the model through nodes 90 and 100 similar to steps 1 and 2 above.
147
Hinge Joint
See under Expansion Joints for more examples.
148
Hydrotest
Use this load to analyze loading from a hydrostatic test which is performed by filling the
piping system with a pressurized fluid (typically water) to check for leaks, etc., before putting
the system into service.
During hydrotest, all hangers are assumed pinned (i.e., they act as rigid vertical supports).
The hydrotest load is defined by the specific gravity of the test fluid (1.0 for water), test
pressure and whether to include or exclude the insulation weight (because many times the
hydrotest is performed before applying the insulation).
The hydrotest load is input by pressing “h” on an empty row in the Layout window (similar
to pressing “c” for a comment) or on an empty row, selecting “Hydrotest load” from the
Element Types dialog (Ctrl+Shft+T).
The “Hydrotest Load” dialog appears.
After the hydrotest load is input by pressing Enter or clicking on OK, the hydrotest load
appears in the Layout window.
149
Hydrotest
If you need to modify an existing hydrotest load, double click on the row that defines the
hydrotest load to bring up the Hydrotest Load dialog. The hydrotest load is applied to the
rows that follow until changed by another hydrotest load. The hydrotest load can be constant
over the whole model or can be changed in parts of the model.
To analyze the hydrotest load case, the Hydrotest load case must be selected using the
command Loads > Load cases from the Layout window.
The hydrotest load case is analyzed as a sustained load (with no temperature effect
considered) and the resulting stresses are computed using the Sustained Stress (SL) equation
corresponding to the piping code selected for analysis.
150
Jacket End Cap
Use a jacket end cap to rigidly connect all six degrees of freedom of the coincident nodes of
a jacketed pipe (i.e., the node on the core pipe and the corresponding node on the jacket pipe
(“J” node) are tied together so that both nodes have the same displacements and rotations).
A jacket end cap is input at a jacketed pipe node by typing “j” in the Data column or
selecting “Jacket End Cap” from the Data Types dialog.
You will need this to secure the jacket pipe to the core pipe rigidly. Sometimes (depending
on the combination of restraints), you may get a “Stiffness Matrix not positive definite”
error, which may be corrected by inserting a jacket end cap.
151
Jacketed Piping
Jacketed piping is used when the primary state of the pipe contents (fuel, chemicals such as
resins, etc.) needs to be maintained at a specific temperature during transport. An outer
(jacket) pipe surrounds the inner (core) pipe that contains the operating fluid or the chemical.
The jacket provides external heating or cooling as required along the length of the core pipe.
The terminology used here is as follows:
 Jacketed piping refers to the entire assembly, i.e., a core pipe with a jacket on the
outside.
 Jacket pipe refers only to the outside pipe.
 Core pipe refers only to the inside pipe that contains the operating fluid.
In CAEPIPE, jacketed piping need only be modeled once, not twice (as in some other pipe
stress software programs). CAEPIPE models the outer jacket pipe along with the inner core
pipe once on the Layout window. Each row defines a jacketed piping element. The jacket
and the core pipes may have different materials, sections and (P, T) loads.
Jacketed Pipe
A Jacketed pipe is input by typing “JP” under Type or selecting “Jacketed pipe” from the
Element Types dialog. The material, section and load specified in the Jacketed Pipe dialog
apply to the jacket pipe while the ones mentioned on the layout row (next to offsets) apply to
the core pipe.
The Jacketed Pipe dialog is shown.
152
Jacketed Piping
The jacket’s material, section, and load names are input here (1, 6 and 1 as shown).
CAEPIPE retains the properties of a jacket pipe until changed so there is no need to retype
the names of the jacket properties every time you input a jacketed pipe.
The ends of the jacket and core pipes need to be rigidly connected using the “Jacket End
cap” data type (see previous topic). Also, “Spiders” need to be input at locations found in the
physical assembly. You may have to break up the piping into smaller elements to insert
spiders at appropriate locations. See example given later in this section.
In case you are analyzing for wind, it may be more accurate to specify a different load for the
core pipe alone that does not specify the Wind load since the core pipe is not exposed to
wind. Same applies to the core pipe insulation if the core pipe does not have insulation.
Internal nodes
CAEPIPE generates a “J” node for jacket pipes. For example, if you had a jacketed pipe
from node 10 to 20, CAEPIPE generates 10J and a 20J as (internal) jacket nodes (that may
be referenced on the layout screen).
These internally generated nodes may be used to specify data items such as a spider, jacket
end cap, support (hanger, restraint), forces on the jacket. See example later in this topic.
Jacketed Bend
A Jacketed bend consists of a core bend (with a straight portion) surrounded by a jacket
bend (with a straight portion of jacket pipe).
A Jacketed bend is input by typing “JB” in the Type column or by selecting “Jacketed bend”
from the Element types dialog.
153
Jacketed Piping
The Jacketed Bend dialog is shown.
Jacket (properties)
The jacket’s material, section, and load names are input here. The properties of a jacketed
pipe are retained until changed. So, there is no need to retype the names of the jacket
properties every time you input a jacketed pipe.
Core (properties)
Presently these properties are disabled. You need to enter them on the layout row under
Material, Section and Load.
Bend radius
Separate bend radii may be specified for the core and the jacket pipes.
Note that CAEPIPE does not check for interference between the core and the jacket arising
out of differently specified bend radii.
The bend radius for the core pipe is normally the same as that of the jacket pipe, since the
two bends are generally concentric. Use the Render feature in the Graphics window to check
visually for interference, however.
Bend thickness
Separate bend thicknesses may be specified for the core and the jacket bends, if they are not
default jacket and core section thicknesses.
154
Jacketed Piping
Intermediate nodes
You can define additional nodes on the outside jacket of a jacketed bend for locating
supports. You may also use internal nodes generated by CAEPIPE. See Internal nodes
below.
Internal nodes
CAEPIPE generates “C” and “D” nodes for the Jacketed bend on the jacket at the near and
far ends of the bend. The core pipe (bend) has its own “A” and “B” nodes. Example: When
you define a Jacketed bend from node 20 to node 30, 30A, 30B (nodes on core bend), 30C
and 30D (nodes on jacket) are generated. Nodes (30A, 30C) and (30B and 30D) are
coincident only if the core and the jacket pipes have the same bend radii. See figure.
These internal nodes may be used to specify data items such as a spider, jacket end cap,
supports, forces, etc.
Split a Bend/Pipe
A jacketed pipe/bend may be split by using the Split option in the Edit menu in the Layout
window.
Contents Weight
The weight of the contents between the jacket and the core pipes is calculated in the
following manner:
(a) Twice the insulation thickness on the core pipe is added to the outer diameter of the core
pipe. (b) The external area of the core pipe is calculated by using the above diameter (a). (c)
The internal area of the jacket pipe is calculated. (d) The external area of the core pipe (b) is
subtracted from the internal area of the jacket pipe (c) and this result is further used to
compute the weight of contents between the jacket and core pipes.
Jacketed Reducer
See modeling procedure in topic on the Reducer.
Example: Jacketed Pipe/Bend
The figure below shows a Jacketed pipe with a Jacketed bend (at node 20-TIP). Observe the
spider at the far end of the bend (node 20B).
155
Jacketed Piping
The nodes 10J, 20C, 20D, 30J, 20A and 20B are internally generated nodes. You may use
them for specifying data items such as spiders, supports (hanger, restraint), forces, etc.
The pairs of nodes (10, 10J)and (30, 30J) are coincident. The nodes 20A and 20B are
coincident with the nodes 20C and 20D respectively only if the core and the jacket pipes
have the same bend radii.
Note that the core and jacket nodes are not connected even though they are coincident. The
core and jacket pipes have to be supported and connected using supportsand jacket
connections (namely, spiders and jacket end caps). An anchor each at nodes 10 and 10J is
specified. The hanger is at node 30J since it is attached to the jacket.
A jacket connection of the type spider at node 20B acts as an internal guide between the core
pipe and the jacket pipe, that is, it prevents any radial movement but allows sliding, rotating
and bending movement between core and jacket pipes.
156
Jacketed Piping
In case forces transmitted from core pipe to jacket through spiders are required, then spiders
specified on core pipe can be replaced by guides with “connected to” nodes specified as the
corresponding nodes on the jacket pipe.
The end cap at node 30 connects the core and jacket pipes rigidly.
Jacketed Piping Stresses/Ratios
CAEPIPE provides an option for you to display the color-coded stress/ratio contour for
jacketed piping in the graphics window context menu. The default stress contour is for
thecore piping. Upon selecting the command for Jacket stresses as shown below, stress
contour plot for the jacket piping is displayed:
157
Limit Stop
In CAEPIPE, a limit stop is capable of modeling several types of physical supports including
a guide, an anchor, a resting support, a two-way rigid restraint and a rod hanger. Using a
combination of upper and lower gaps (limits), friction coefficient, support stiffness and a
direction for its axis, you can model the above mentioned physical support types.
A limit stop prevents a node from moving beyond a certain distance (called a gap or a limit)
in a certain direction. The node can move freely within the gap. After the gap closes, a limit
stop acts as a rigid or flexible restraint (depending on your input for stiffness) resisting
further movement of the node in the specified direction. If friction is specified, after the limit
is reached, the friction force will oppose movement in the plane normal to the limit stop
direction.
A limit stop is input by typing “l(L)” in the Data column or selecting “Limit Stop” from the
Data Types dialog.
158
Limit Stop
The Limit Stop dialog is shown.
Limits
Also called gaps, these limits, upper and lower, are the gaps present on either side of the
node. The gap in front of the node in the direction of the vector is called the upper limit, and
the gap to the rear of the node is called the lower limit. The gap is measured from the
undeflected position of the node.
Typically, the upper limit is positive and the lower limit negative. In some situations, it is
possible to have a positive lower limit or a negative upper limit, which is the same as
forcefully displacing the node in that direction by the gap specified.
The algebraic value of the upper limit must be greater than the lower limit. For example,
upper limit = –0.125”, lower limit= –0.25”, or upper limit = 0.25”, lower limit = –0.25”.
If a particular limit does not exist (that is, a node can move freely on that side of the node),
then that limit should be left blank (as in the case of a resting support, the upper limit should
be blank assuming Y-vertical and Y comp = 1, as in the figure shown).
If there is no gap at all, then the corresponding limit should be explicitly input as zero. When
zero is entered, the limit stop acts as a one-way restraint in that direction.
Direction
The direction in which the limit stop is oriented must be specified in terms of its global X, Y
and Z components. Or use one of the preset buttons to orient the limit stop axis (see above
image):
1. Vertical: To set the limit stop axis in the(global) vertical direction
2. Axial: To set the limit stop axis along the local-x direction (pipe axis)
3. Shear y: To set the limit stop axis in the local-y directionof pipe
4. Shear z: To set the limit stop axis in the local-z directionof pipe
159
Limit Stop
If you have connected the limit stop node to another node in the same piping model, then
unless the connected node is coincident with the limit stop node, the limit stop direction
must not be input. It is calculated from the locations of the connected node and the limit
stop node, and its direction is oriented from the connected node to the limit stop node.
Friction coefficient
If friction coefficient is specified, a friction force will oppose the movement in the plane
normal to the limit stop direction when the gap is closed. This friction force is displayed in
results under Limit stop support loads.
If you had several limit stops with friction coefficients specified, and you wanted to change
all of those friction coefficients to the same value, use the Change command under the Edit
menu.
Stiffness
The default is set to Rigid stiffness. Other values may be input by estimating the stiffness of
the support. (e.g., for a rod, stiffness = AE/L).
where
𝐴 = Cross-sectional area of rod
𝐸 = Young’s modulus of rod material
𝐿 = Length of rod
Connected to Node
You can connect a limit stop node to another node in the same piping model. During gap
and friction calculations, the relative displacements of the limit stop node are calculated with
respect to the connected node. If you connect the limit stop node to an external fixed point
(Ground point), leave the “Connected to Node” blank. See “Direction” above for
information about how the direction is calculated depending on whether the connected node
is coincident or not with the limit stop node.
160
Limit Stop
Solution procedure
Limit stops require a nonlinear iterative solution. If you specify a friction coefficient, the
following procedure is used for convergence: If the lower or upper limit is reached, the
corresponding reaction force is calculated. The maximum friction force is the product of
friction coefficient and the reaction force. The solution converges when the displacement
varies by less than 1% between successive iterations.
Limit stops are included in dynamic analysis. The status of the limit stops arrived at upon
completion of all iterative calculations for the first operating case (W+P1+T1) is used during
dynamic analysis. If either the lower limit or the upper limit is reached at the end of iterations
for the first operating case, then that limit stop is treated as a rigid two-way restraint in the
direction of the limit stop during dynamic analysis. If both limitsare not reached, then that
limit stop is ignored during dynamic analysis.
Example 1: Vertical 1-way restraint
Assume that you have a vertical 1-way support with the following data: Upper limit = None,
Lower limit = 0, Friction coefficient = 0.3, Direction vector of the limit stop is vertical along
+Y axis.
Model the pipe up to the Limit stop node 30. At node 30, type “l(L)” in the Data column.
The limit stop dialog will be shown.
161
Limit Stop
In the limit stop dialog, press “Vertical” button. The data is automatically entered [0.000 for
Lower limit, None for Upper limit (blank), 1.000 for Y comp (vertical)]. Enter 0.3 for friction
coefficient (this acts in the X-Z plane which is perpendicular to the direction of the limit
stop).
162
Limit Stop
Example 2: Pipe Slide/Shoe Assembly
The assembly is modeled using three limit stops, one in each direction.
 When you start a new model file, node 10 and an Anchor are automatically input,
press Enter to move cursor to next empty row.
 Press Tab in the Node column which puts the node number 20 automatically. Type
5’ for DX, enter material, section and load names, press Enter.
5’ for DX, Tab to Data column and type “l(L)” to open limit stop dialog. Input line
stop with gap (0.5”) along X axis (Notice that this is a one-way restraint; there is only
one stop block along +X). Type Upper limit 0.5”, leave lower limit blank, Direction
as (X comp = 1, Y comp = 0, Z comp = 0), Friction coefficient=0.3, press Enter.
 Create a limit stop in the Y direction. Type 30 for Node, press Tab to move to Type
field, type “L” for Location, choose Limit stop from the Data Types dialog, in the
Limit stop dialog, type Upper limit = 0.5”, Lower limit = 0.0, Direction (X comp =0,
Y comp = 1, Z comp = 0), Friction coefficient=0.3, press Enter.
163
Limit Stop
 Create a limit stop in the Z direction. Type 30 for Node, press Tab to move to Type
field, type “L” for Location, choose Limit stop from the Data Types dialog, in the
Limit stop dialog, type Upper limit = 0.25”, Lower limit = –0.25”, Direction (X
comp =0, Y comp = 0, Z comp = 1), Friction coefficient =0.3, press Enter.
The Layout window is shown below.
164
Limit Stop
The Graphics window is shown below.
Example 3: Limit Stop Connected to Another Node

As shown in the figure below, two 300” long cantilever pipes are connected with an 8”
separation between center lines.
The gap between the bottom of the top pipe and the top of the bottom pipe is
8” – (6.625” + 8.625”) / 2 = 0.375”.
If pipes were free to deflect downward due to deadweight, the top pipe will deflect 1.908”
and the bottom pipe will deflect 1.116” at the free ends. The relative deflection between
them will be 1.908” – 1.116” = 0.792”. This however is not possible because when the
relative deflection exceeds 0.375” the pipes will touch. This situation can be modeled using a
limit stop connecting the free ends of the pipes. In this case the top pipe deflects 1.608”, i.e.,
less than the 1.908” free deflection because it is resisted by the bottom pipe. The bottom
pipe deflects 1.233”, i.e., more than the 1.116” free deflection because additional load is
165
Limit Stop
imposed on it by the top pipe when they touch. The difference between the deflections is
1.608 – 1.233 = 0.375” as expected.
 When you start a new model file, node 10 and an Anchor are automatically input,
press Enter to move cursor to next empty row.
300” for DX, enter material, 6” section and load names, press Enter.
 For the bottom pipe, start with node 30 of Type “From” at DY = –8” and make it
an anchor. On the next row enter node 40 with DX = 300” and 8” section. Press
Enter to go to the next row.
 Enter the limit stop at node 20. Type 20 for Node, press Tab to move to Type field,
type “L” for Location, choose Limit stop from the Data Types dialog, in the Limit
stop dialog, leave the Upper limit blank and input Lower limit = -0.375”. Input the
Connecting Node as 40. Since this limit stop connects two nodes, the direction
should be left blank. The direction is implicitly from node 40 to node 20.
Alternatively the limit stop could be specified at node 40 connected to node 20. The
direction now would be from node 20 to 40. The limits would then be Lower limit =
None and Upper limit = 0.375” since the direction is now reversed compared to the
previous case. Both these cases will give identical results.
The Layout window is shown below.
The Graphics is shown below.

166
Limit Stop
167
Load
Loads on a piping system can be many and varied along its routing. Different piping
segments may experience different pressures and temperatures depending on process
requirements, different loads (snow, wind, etc.) depending on their physical locations and
carry different states of a fluid between different points in a piping system. CAEPIPE offers
a flexible method to input as many loads as required for as many segments or elements as
needed with the least effort.
So, Load allows you to apply a temperature and pressure, specify weight of the operating
fluid and add additional weight (e.g., due to snow load) on each element (if required) or for a
range of elements in the model. Also the wind load can be turned “on” or “off” for each
element (if required) or for a range of elements in the model.
After specifying the requested information here, including a name, use it under the Load
column on the Layout window to associate the load information with an element.
Each load allows up to 10 operating conditions for Temperature and Pressure depending on
the “Number of Thermal Loads” specified under Options > Analysis > Temperature. This
Load is not to be confused with Load cases [which are combinations of load(s)] found under
the Loads menu in the Layout window. Load cases are analysis cases (Sustained case,
Thermal case, Operating case, Static Seismic case, etc.) for which CAEPIPE computes a set
of results.
To define a new load, click on Load in the Header row in the Layout window (or select
Loads under the Misc menu, hotkey: Ctrl+Shift+L). This opens a List window that displays
currently defined loads.
Either you can start typing the load data directly here into the fields or double click on an
empty row to enter data through a dialog.
Depending on the number of thermal loads specified (under Options > Analysis >
Temperature), up to 10 temperature/pressure load sets (T1/P1, T2/P2, T3/P3,…,T10/P10)
can be input for each element or for a range of elements.
168
Load
Up to 10 specified thermal displacements can be entered for Anchor and Nozzle data types.
169
Load
Load Name
Type an alphanumeric name (up to five characters) in this field. The name can be changed
later.
Temperatures
Type up to 10 operating temperatures. The maximum of the 10 temperatures is used to look
up the corresponding allowable stress for the material used in code evaluation.
The other quantities looked up using these temperatures are the thermal expansion
coefficients (alpha) and the temperature-dependent moduli.
Make sure to select number of thermal loads equal to two, three or 10 under Options >
Analysis > Temperature, when you have more than one set of temperature and pressure.
Pressures
Type up to 10 operating pressures that correspond to the 10 operating temperatures above.
The maximum of the 10 pressures is used to calculate the pressure stress term
[𝑃𝐷/4𝑡 𝑜𝑟 𝑃𝑑 2 /(𝐷2 − 𝑑 2 )], specified under Options > Analysis > Pressure.
Specify gauge pressures for Pressures input. Negative (external) pressures may be specified,
too. But, the longitudinal pressure stress term (pD/4t) will still be positive according to the
piping codes. Internal pressure will expand the pipe cross-section radially outward while
external (negative) pressure will contract the pipe cross-section radially inward.
170
Load
Design Pressure and Temperature

CAEPIPE requires that the Design Temperature to be entered should be equal to or greater
than the algebraic maximum of all operatingtemperatures (T1 through T10). Similarly,
Design Pressure to be entered should be equal to or greater than thealgebraic maximum of
all operating pressures (P1 through P10).
Design Temperature so entered will be used to determinethe allowable stress for material,
which is in turn used to compute the Allowable Pressure as per the piping code selected.
Since Allowable Pressure reduces with decreasing allowable stress, to be conservative,the
least allowable stress would then be obtained when the Design Temperatureis no less than
the algebraic maximum of all operating temperatures (T1 through T10).
The Allowable Pressure so computed as per the piping code selected is then compared
against the Design Pressure entered above and reported in the Code Compliance results.
Except as stated above, Design Temperature and Design Pressure are not used anywhere else
during CAEPIPE analysis.
Specific Gravity
Specific gravity is the ratio of the density of a fluid to the density of a reference substance (in
this case, water). Enter the specific gravity of the operating fluid inside the pipe. This input is
used to calculate the weight of the operating fluid, which is added to the weight of the pipe.
Specific gravity is with respect to water.
Additional weight
The value you enter here is taken as weight per unit length of the element and this total
additional weight is added to the weight of the pipe. (Total additional weight = Length of
element x Additional weight per unit length). Additional Weight is to be input in lbf/ft or
kgf/m.
Whenever mass is required for a calculation as in the case of forming Mass matrix for
dynamic analysis, or in calculating inertia force as (mass x acceleration) for static seismic
analysis, CAEPIPE internally computes the mass to be equal to (weight / g-value).
For example, Additional weight could be used to apply the weight of snow on the pipe.
Wind load
Type Y(es) or N(o) to apply or not apply the wind load for this element. When you press
Y(es), the wind load (entered as a separate load under Loads menu) is applied to this element.
For example, this is useful when you have a part of a piping system exposed to wind with the
remaining part inside a building. In such a case, you should define two Loads, all data the
same except that one has Wind load and the other does not. The Load with the wind load is
applied to those elements that are affected by wind.
171
Local Coordinate System
CAEPIPE allows you to create your physical piping system in a mathematically equivalent
3D Cartesian coordinate space with a global origin, which is the point of intersection of three
planes orthogonal to each other, with three axes commonly denoted X, Y and Z (with either
of the latter two vertical).
Y Vertical Z Vertical
Once you begin creating your system from a given point (usually the global origin), you route
your piping system one element at a time until you get to the end of the line(s). An element’s
orientation could be different from another element’s, thereby necessitating an element’s
own “elemental coordinate system,” which is commonly referred to as the Local Coordinate
System (LCS), provided for the purpose of understanding the local forces and moments on
each element. This system can be turned on (graphically) through the View menu > Show
LCS command while you are viewing “Element Forces in Local Coordinates” in the Results
window.
For a straight element (such as a pipe or a beam), the “local x” axis is along the element,
from the “From” node to the “To” node. For a node location such as a guide, the local axes
are based on the previous connected element. If the preceding element does not exist, the
following element is used. The local y-axis and local z-axis are calculated differently
depending on whether the vertical direction is Y or Z and also depending on whether the
element is in the vertical direction.
The local coordinate system may be displayed graphically (for beams and guides in the input
processor and for all the elements in the output processor) by selecting the “Show LCS
(Local coordinate system)” command from the View menu.
In CAEPIPE, the local coordinate system is indicated by the lower case x, y and z letters,
whereas the global coordinate system is indicated by the upper case X, Y and Z letters.
Global vertical axis is Y
Element is not Vertical
The local y-axis of the element lies in the local x - global Y plane (i.e., vertical plane) and
is in the same positive direction as the global Y axis. The local z-axis is the cross product
of the local x-axis and local y-axis.
172
Element is Vertical
The local z-axis of the element is in the global Z direction. The local y-axis is in the
global –X direction.
Global vertical axis is Z
Element is not Vertical
The local z-axis of the element lies in the local x - global Z plane (i.e., vertical plane) and
is in the same positive direction as the global Z-axis. The local y-axis is the cross product
of the local z-axis and local x-axis.
Element is Vertical
The local y-axis of the element is in the global Y direction. The local z-axis is in the
global–X direction.
173
Local Coordinate System for a Bend

For a bend, at the “From” node, the local x axis is along the tangent from the “From” node
to the tangent intersection point. The local y-axis is along the radius and points to the center
of curvature. The local z-axis is the cross product of the local x-axis and local y-axis.
Similarly, at the “To” node, the local x-axis is along the tangent line from the tangent
intersection point to the “To” node. The local y-axis is along the radius and points to the
center of curvature. The local z-axis is the cross product of the local x-axis and local y-axis.
Sign Convention for Element Forces and Moments

The sign conventions for the element forces and moments in the local coordinate system
follow strength of materials conventions, i.e., forces and moments at the “To” node of an
element are positive if they are in the positive local axes directions of the element. On the
other hand, forces and moments at the “From” node of the element are negativeif they are in
the positive local axes directions of the element.
Positive sign conventions for local forces and moments are shown above at the “From” and
“To” nodes of an element. Note that positive directions at the “From” node are reversed
compared to the positive directions at the “To” node.
174
Location
Several times, you may need to input more than one data item at a node, e.g., a hanger and a
Branch SIF to designate the type of tee. At those times, use Location (press “L” in the Type
field or pick Location from the Element Types dialog) to input more than one data item at a
node. For example, you may want to input a lateral restraint at an existing hanger.
Another use for Location is when you want to input a data item at an internally generated
node. Nodes are internally generated by CAEPIPE for bends (A, B nodes, e.g., 20A) and
Jacketed pipes/bends (J, C, D nodes, e.g., 10J).
By design, each row in the Layout window allows only one data item to be entered under the
Data field. Additional data items can be input only through Location (see Examples below).
Ensure that the node you use for Location has already been defined on an earlier row,
ordefined earlier as an intermediate node for a bend or is an internally generated node. In
other words, you cannot use Location on an undefined node.
Example 1: Multiple Limit Stops at a node to modelPipe Slide and Shoe assembly
See Example 2 under the Limit Stoptopicinthis manual.
Example 2: Data at internally generated nodes
Refer to the example given underthe JacketedPiping in this manual. CAEPIPE internally
generates J node on the jacket for a Jacketed pipe and the C and D nodes on the jacket for a
Jacketed Bend. The following 4 data items present in that example are input using the
Location type as shown in the Layout window below.
1. Node 10J is the starting Jacket node which is anchored (node 10 is separately anchored,
since it is the node on the core pipe).
2. There is a hanger at node 30J since the hanger is connected to the Jacket.
3. There is a spacer (spider) at the far end of the bend, node 20B (which is on the core
pipe). Remember that the bend has a jacket on the outside.
4. Node 30 has a jacket end cap
Example 3: Hanger at a Bend Intermediate Node

See Example 6 in the Bend section of this manual to locate a hanger at an intermediate node.
175
Lug EvaluationModule
Lugs (integral attachments) are forged attachments or attachments welded on the pressure-
loaded wall of a straight pipe which transfer piping loadings to the steel framework or
concrete.
Loads on attachments cause local stresses in the pipe wall. Equations to determine these
pipestresses at lug attachments are given in different codes. These local stresses are then
added to the piping system stresses at the attachments. The combined stresses thus obtained
are checked for compliance with the appropriate equations given in those codes.
The Lug Evaluation module implemented in CAEPIPE computes local pipe stresses as per
the following codes for Rectangular and Hollow Circular cross sectional attachments.
1. ASME Section III, Division 1 (2010) – Appendix Y (NC Piping – Class 2)
2. ASME Section III, Division 1 (2010) – Appendix Y (ND Piping – Class 3)
3. EN 13480-3 (2017), Section 11
The details on the implementation of this module are provided in the Section titled “Lug
Evaluation” of the CAEPIPE Code Compliance Manual.
The Lug Evaluation module assumes that the flexibility analysis of that piping system with
CAEPIPE has already been performed, which will have produced a stress report as well as
the forces and moments at the location where the lug is attached to the piping.
Note that this module is separate from a piping stress model file and can be accessed from
File Menu > Open/New command.
176
dialog with input fields you can edit. You will need to enter the data in this dialog. The
different parameters required to be input are explained in detail in the Section titled “Lug
Evaluation” of the Code Compliance manual.
177
Once the required data are input, save the file (Lug evaluation will have a .lug extension).
Now, select File menu > Analyze to calculate stresses and perform code evaluation, which
178
179
Lug Evaluation Module Menus

File Menu
.Analyze.
Analyze command calculates pipestressesat the attachment and compares them to stress
allowable specified by the selected code.
Print.
You can print a Report by using the Print command. You can also preview the report by
clicking the Preview button on the print dialog.
180
Edit Menu
.Edit.
You can edit the data by clicking the Edit command.
181
Options Menu
.Units.
.Font.
182
Sample Problem
Caepipe LUG_XEE03_ASME_RA.lug Page 1
Version 8.10 Aug 10,2018
-----------------------------------------------------------------------------------
Evaluation of Piping at Lug Attachment as per ASME Sec III, Div 1 - NC (2010)
-----------------------------------------------------------------------------------
Input Data:
Lug Evaluation Code: ASME Sec III, Div 1 - NC (2010)
Level: C
Lug Type: Rectangular
Weld Type (Penetration):Full Penetration
Outside Diameter of Run Pipe (OD): 10 (inch)
Nominal Wall Thickness of Run Pipe (Thk): 0.165 (inch)
Half Length of Attachment in Circumferential Direction (L1): 0.25 (inch)
Half Length of Attachment in Longitudinal Direction (L2): 1.5 (inch)
Sustained Stress at Run Pipe (Eq.8): 217 (psi)
Expansion Stress at Run Pipe (Eq.10): 39 (psi)
Sustained + Occasional Stress at Run Pipe (Eq.9): 509 (psi)
Settlement Stress (Eq.10a): 0 (psi)
Thermal + Sustained Stress (Eq.11): 39 (psi)
Allowable Stress at Maximum Metal Temperature (sh): 17100 (psi)
Allowable Stress Range (sa): 42533 (psi)
Yield Stress (sy): 25650 (psi)
-----------------------------------------------------------------------------------
Forces and Moments at Pipe surface: Sustained
-----------------------------------------------------------------------------------
Thrust Load applied to the Attachment (W): 0 (lb)
Circumferential Shear Load applied to the Attachment (Q1): 0 (lb)
Longitudinal Shear Load applied to the Attachment (Q2): 2 (lb)
Torsional Moment applied to the Attachment (MT): 0 (ft-lb)
Circumferential Moment applied to the Attachment (MN): 0 (ft-lb)
Longitudinal Moment applied to the Attachment (ML): 0 (ft-lb)
-----------------------------------------------------------------------------------
Forces and Moments at Pipe surface: Occasional
-----------------------------------------------------------------------------------
-----------------------------------------------------------------------------------
Forces and Moments at Pipe surface: Thermal
-----------------------------------------------------------------------------------
-----------------------------------------------------------------------------------
Forces and Moments at Pipe surface: Settlement
-----------------------------------------------------------------------------------
183
Caepipe LUG_XEE03_ASME_RA.lug Page 2
Version 8.10 Aug 10,2018
-----------------------------------------------------------------------------------
Abs. Max. Forces and Moments occurring simultaneously at Pipe surface
-----------------------------------------------------------------------------------
Thrust Load applied to the Attachment (W**): 0 (lb)
Circumferential Shear Load applied to the Attachment (Q1**): 0 (lb)
Longitudinal Shear Load applied to the Attachment (Q2**): 1343 (lb)
Torsional Moment applied to the Attachment (MT**): 0 (ft-lb)
Circumferential Moment applied to the Attachment (MN**): 0 (ft-lb)
Longitudinal Moment applied to the Attachment (ML**): 0 (ft-lb)
-----------------------------------------------------------------------------------
ASME Sec III, Div 1(2010) Appendix Y (NC - Class 2)
-----------------------------------------------------------------------------------
Evaluation of Piping at Rectangular Attachments
-----------------------------------------------------------------------------------
Section Y-3410: Pipe Stresses at Attachment
-----------------------------------------------------------------------------------
Calculated Allowed Ratio
(psi) (psi)
Sustained Stress (Ssl) [Eq. 8]: (psi) 221 25650 0.009
Sustained + Occasional (Sol) [Eq. 9]: (psi) 3226 46170 0.070
Thermal Exp. Stress (Se) [Eq. 10]: (psi) 43 42533 0.001
Settlement Stress (Sd) [Eq. 10a]: (psi) 0 51300 0.000
Sustained + Thermal Exp. Stress (Ste) [Eq. 11]: (psi) 47 59633 0.001
-----------------------------------------------------------------------------------
Additional check for Full Penetration Weld
-----------------------------------------------------------------------------------
Calculated Allowed Ratio
(psi) (psi)
Stress (SNT**) as per Y-3410 Eq. 5: (psi) 2713 51300 0.053
Shear Stress as per Y-3410 Eq. 6: (psi) 2713 25650 0.106
184
Material
For a project, a piping material engineer usually produces different material lists for different
process and utility systems based on process design conditions, list of components, fluid type
(corrosivity, viscosity), end conditions and temperature, pressure and size ranges.
As a result, a piping system will have its materials list, which will specify the materials a stress
analyst will have to define inside CAEPIPE before analysis. CAEPIPE is so flexible that it
allows each element (such as a pipe, beam, elbow, valve, jacket, bellows, etc.) to have its own
material definition.
Once you define and name a material type, you type in that name on the Layout window
under the column “Matl” to specify the material for an element. A material may be metallic
or FRP. You need to obtain properties (density, Poisson’s ratio, Young’s modulus, mean
coefficient of thermal expansion and allowable stress as a function of temperature) for a new
material not found in the supplied libraries. You will need properties for at least two
temperatures (reference and design) for each material you define. Subsequently, the
temperatures you specify for an element that uses this material must fall within this
temperature range.
The material name you specify on the layout applies to the piping element on that row. For
jacketed piping, you must specify two materials - one for the core pipe (on the Layout
window), and the other for the jacket pipe in its own dialog. The material you specify for a
bend, a jacketed bend, and for a miter bend applies only to that specific element on that
input row.
There are two ways in which you can define materials:
1. By defining a material inside the CAEPIPE model, or
2. By picking a material from an existing material library.
For the sake of convenience, we suggest that you create a separate material library for your
pipe stress project so that you can share it with your team members.
Below, you will see how to create a material inside a model and how to create a new or
modify an existing material library.
Define a Material inside a CAEPIPE model
From the Layout window, click on “Matl” on the header row (or select Materials from the
Misc menu, Ctrl+Shift+M).
185
Material
The List window for materials is shown.
In the Material List window, you can edit inside both panes - the left pane contains Name,
Description, Type of material, Density, Poisson’s ratio (nu), Joint factor etc. and the right
pane contains material properties (usually Modulus of Elasticity E, mean Coefficient of
Linear thermal expansion [Alpha], not instantaneous coefficient nor thermal expansion per
unit length, and the code-specific Allowable stress) as a function of Temperature. CAEPIPE
requires “Weight Density” to be input in lbf/in3 or kgf/m3 and NOT its mass density.
While entering the temperature-dependent material properties, you need not enter
temperatures in an ascending order, although recommended. CAEPIPE will sort the entries
later.
After you are done entering properties for one material, be sure to press Enter when the
cursor is in the left pane, to move it to the next row so you can start entering the next
material. You can insert, copy & paste, delete and edit any material (see under Edit menu).
These panes may change depending on the piping code chosen. For example, for the
Swedish and Norwegian codes, the following window is displayed. This window contains
additional columns for Tensile strength and instead of one Joint factor, it has Longitudinal
and Circumferential joint factors.
186
Material
The European (EN 13480) code has a column for Tensile strength and temperature-
dependent properties have an additional column for fCR (allowable creep stress).
187
Material
To Input a New Material

You can input a new material in three ways:
1. Start typing directly into the fields in the Materials List window.
2. Input through dialogs (shown below for some but not all piping codes).
For the Swedish and Norwegian piping codes, the Material dialog has Longitudinal and
Circumferential joint factors and a Tensile strength field.
188
Material
For the European (EN 13480) code, the Material dialog has a single Joint factor and a
Tensile strength field, while the properties window has fCR (creep stress allowable).
189
Material
3. Pick a material from an existing material library (supplied with CAEPIPE or your own).
Click on the Library button on the toolbar to open the library:
(or select Library command under the File menu):
You will have to open a library file first if it was not previously opened. Select the one of
interest. Note that the libraries with filenames starting with B311 and B313 with year
later than 2010 (e.g., B313-2012.mat) have approx. 200 materials each (from the
respective piping code).
You can select a material from the opened library by double clicking on it or highlighting
it and clicking on OK.
190
Material
Name
Type a Material name, up to five(5) alpha-numeric characters.
Description
Type a description for the material, up to 31 characters.
Type
AL for Aluminum
AS for Austenitic Stainless Steel
CA for Copper alloys annealed
CC for Copper alloys cold worked
CS for Carbon Steel
FR for Fiber Reinforced Plastic piping
FS for Ferritic steel
NA for Nickel alloys 800, 800H, 825
SS for Stainless Steel
TI for Titanium
These material types are used in calculation of the Y factor for allowable pressure at high
temperatures for certain piping codes. Swedish and Norwegian piping codes also use it for
calculating allowable expansion stress range. These codes also need tensile strength.
For Fiber Reinforced Plastic piping, you need to select the material type “FR” to enter FRP
material properties. More information can be found under the section Fiber Reinforced
Plastic Piping in this manual.
191
Material
Density
Density of the material is used to calculate weight load and also mass for dynamic analysis.
CAEPIPE requires “Weight Density” to be input in lbf/in3 or lbf/ft3 or kgf/m3 or gf/cm3
and NOT its mass density. Whenever mass is required for a calculation as in the case of
forming Mass matrix for dynamic analysis, or in calculating inertia force as (mass x
acceleration) for static seismic analysis, CAEPIPE internally computes the mass to be equal
to (weight / g-value).
Nu
The Poisson’s ratio (Nu) defaults to 0.3 if not input.
Joint factor
The joint factor is the longitudinal weld joint factor used to calculate allowable pressure. For
Swedish and Norwegian piping codes, a circumferential joint factor is also input to calculate
longitudinal pressure stress.
Tensile strength / Yield
For Swedish, Norwegian and European (EN13480) piping codes, tensile strength is used in
the calculation of the allowable expansion stress range. For Stoomwezen piping code, tensile
strength is used in the calculation of hot allowable stress.
For B31.3 or B31.12 piping codes, yield is used in the calculation of sustained plus occasional
allowable stress when the temperature is > 4800 C or 8000 F.
Note: Leaving the yield field blank will cause CAEPIPE to issue Assertion failure during
analysis for stress system having the temperature of piping >4800 C or 8000 F.
To create or modify a material library
CAEPIPE offers you flexibility in creating your own material libraries (user-defined
libraries). That way, you do not feel restricted by the offered choices in materials and can
continually keep updating the material libraries with your own materials. To create a library:
From the Main window, select File > New and click on Material Library.
A List window for materials is shown.
192
Material
You must select a piping code first, using the menu command Options > Piping code,
before you start entering properties.
You can, as before, start typing directly into the fields, or enter properties through a dialog.
The only difference is that materials in the library do not have names whereas those in a
model have names.
After you are done entering materials, you must save to a material library file by using the
File > Save command.
Give the file a suitable name. The file will be saved with a .mat extension.
193
Material
Note: It is also possible to Import/Export the material library through a Ascii Material Batch
Library (.mlb) file. See the section titled Import/Export Material Library in Appendix A of
CAEPIPE User’s Manual for more details.
Should you need to change the piping code, then you will need to update all materials’
properties (in this library) according to the new code, or load a new file. Better yet, if you do
not see the library among the supplied files, create a new library for the new code.
Therefore, first make sure that you have selected the correct piping code (under Options >
Analysis > Piping code) in the Layout window. Then, ensure that you use the correct
properties from that code.
CAEPIPE comes with many libraries from several piping codes. When inside a CAEPIPE
model, you can open any library and pick a material from it. Make sure to pick the proper
library (with the proper year), especially between B31.1 and B31.3 libraries because they have
significantly different allowable stresses for the same materials. Also verify the material
properties in these libraries before you use them.
In summary, CAEPIPE is supplied with many built-in material libraries applicable

for different piping codes. Users should ensure that the correct piping code and
the corresponding material library are selected for each CAEPIPE model created
and the built-in material properties have been checked for correctness prior to
their use.
194
Missing Mass Correction
In dynamic analysis using modal superposition, usually an approximate solution is obtained

because only a limited number of modes is considered. (For seismic analysis, typically all
modes up to 33 Hz). The errors in pipe displacements and stresses are usually small because
they are affected relatively little by high modes. The error in support loads may be substantial
because the influence of high modes on support loads can be important. In stiff piping
systems with few low frequency modes, stresses may also be affected significantly.
Using limited number of modes results in some mass of the system being ignored. The
distribution of this “missing mass” is such that the inertia forces associated with it will
usually produce small displacements and stresses. However these forces will often produce
significant support loads, and in stiff systems can produce significant stresses.
A correction can be made by determining the modal contributions to the mass of the system
and obtaining the “missing mass” as the difference between these contributions and the
actual mass.
The inertial force vector for the n th mode is given by

{𝐹𝑛 } = −[𝑀]{𝑢̈ 𝑛 } = 𝜔𝑛2 [𝑀]{∅𝑛 }𝐴𝑛 (1)
Where [𝑀 ] = diagonal mass matrix

{𝑢̈ 𝑛 } = acceleration vector
𝜔𝑛 = circular frequency
{∅𝑛 } = mass normalized eigenvector
𝐴𝑛 = modal displacement for mode n
For X seismic excitation
𝑎 𝑎
𝑆𝑛𝑥 𝑆𝑛𝑥
𝐴𝑛 = {∅𝑛 }𝑇 [𝑀]{𝑟𝑥 } = 𝛤𝑛𝑥 (2)
𝜔𝑛2 𝜔𝑛2
Where {𝑟𝑥 } = displacement vector due to a unit displacement in the X direction

𝑎
𝑆𝑛𝑥 = spectral acceleration for the nth mode for excitation in the X direction
Γnx = mass participation factor in the X direction for mode n
Let 𝑚 = number of modes used in analysis
𝑁 = total number of modes
If it is assumed that the higher modes: m+1 through N are in phase and have a common
𝑎
spectral acceleration𝑆𝑜𝑥 (conservatively taken as the maximum spectral acceleration after the
mth mode), the total inertial force contribution of these higher modes (also known as “Rigid
body force” or “Left out force”) is
𝑁
{𝐹𝑥𝑅 } = 𝑆𝑜𝑥
𝑎 [𝑀] ∑ {𝜙 }
𝑛 𝛤𝑛𝑥 (3)
𝑛=𝑚+1
195
Missing Mass Correction
It can be shown that

𝑁 𝑚 𝑁
{𝑟𝑥 } = ∑{𝜙𝑛 } 𝛤𝑛𝑥 = ∑{𝜙𝑛 } 𝛤𝑛𝑥 + ∑ {𝜙𝑛 } 𝛤𝑛𝑥 (4)

𝑛=1 𝑛=1 𝑛=𝑚+1
Substituting from (4) for the summation in (3)

𝑚
{𝐹𝑥𝑅 } 𝑎 [𝑀] [{𝑟 } (5)

= 𝑆𝑜𝑥 𝑥 − ∑{𝜙𝑛 } 𝛤𝑛𝑥 ]
𝑛=1
Note that there will be missing mass inertia forces in the Y and Z directions, in addition to
the X direction, for X excitation.
The missing mass force vectors for the Y and Z directions are similarly calculated. The
response to each of these three force vectors is calculated and these additional response
vectors are combined with the responses of the first “m” modes.
This feature is currently not available for Time History and Harmonic analyses.
The above described method is based on the technical paper by Powell. See below for
details.
Powell, G.H. “Missing Mass Correction in Modal Analysis of Piping Systems.” Transactions
of the 5th International Conference on Structural Mechanics in Reactor Technology. August
1979: Berlin, Germany.
196
Miter Bend
Miter bends are typically used when space limitations do not allow the use of regular bends
(elbows), or when a miter is more economical to use than a regular bend. Miters are not
fittings. They are fabricated from pipe, to requirements. “The use of miters to make changes
in direction is practically restricted to low-pressure lines, 10-inch and larger if the pressure
drop is unimportant...” (Sherwood 1980).
See figure below for Miter bend parameters.
In this figure, r = mean radius of pipe

S = miter spacing at center line
θ = one-half angle between adjacent miter axes (≤ 22.5°)
Before modeling a miter bend, you should determine whether it is closely or widely spaced.
Closely Spaced Miter
A miter bend is closely spaced when S < r(1 + tan θ).
A closely spaced miter bend is input as a single miter bend element.

The Bend Radius (R) is calculated as: R = 0.5 S cot .
197
Miter Bend
Widely Spaced Miter
A miter bend is widely spaced when 𝑆 ≥ 𝑟(1 + 𝑡𝑎𝑛 θ).
A widely spaced miter bend is modeled with as many miter bend elements as there are miter
cuts.
The Bend Radius (R) is calculated as: 𝑅 = 0.5 𝑟 (1 + 𝑐𝑜𝑡 𝜃).
A miter bend is input by typing “m” in the Type column or selecting “Miter bend” from the
Element types dialog.
The Miter bend dialog is shown.
198
Miter Bend
Bend Radius
The bend radius (R) depends on the type of miter (Closely or Widely spaced). It is calculated
as explained previously and input in this field.
Bend Thickness
Input the wall thickness of the miter bend if it is different from that of the adjoining pipe
thicknesses. The Bend Thickness, if specified, applies only to the curved portion(s) of the
equivalent bend(s) of the miter bend.
Bend Material
If the material of the miter bend is different from that of the adjoining pipe, select the Bend
Material from the drop down combo box. The Bend Material, if specified, applies only to the
curved portion(s) of the equivalent bend(s) of the miter bend.
Flexibility Factor
Specify a flexibility factor for the miter bend if different from the piping code’s factor. If you
specify one, CAEPIPE uses it instead of the piping code specified Flexibility Factor. A value
of 2.0, for e.g., will mean that the miter bend is twice as flexible as a pipe of the same length.
Closely spaced
To specify the miter bend as closely spaced, click on the “Closely spaced” radio button.
Widely spaced
To specify the miter bend as widely spaced, click on the “Widely spaced” radio button.
Parameters for 90° Miter Bends

The parameters (dimensions) for 90° miter bends (with 2, 3, 4 miter cuts) in terms of
dimension (A), mean pipe radius (r) and number of miter cuts (N) are shown in the following
table.
Use the table to determine whether the miter bend is Closely spaced or Widely spaced, and if
it is Widely spaced and has 2, 3 or 4 miter cuts, to calculate the dimensions B, C, D, E and R
(equivalent miter bend radius) before modeling it.
A closely spaced miter requires only the miter bend radius (same as dimension A shown in
the figures).
Dimensions for Widely spaced miters only
N S Closely spaced if B C D E R
2 0.828427 A A < 1.707107 r 0.414214 A 0.585786 A - - 1.707107 r
3 0.535898 A A < 2.366025 r 0.267949 A 0.464102 A - - 2.366025 r
4 0.397825 A A < 3.013670 r 0.198912 A 0.367542 A 0.281305 A 0.152241 A 3.013670 r
N = Number of miter cuts
Half angle  = 90° /(2N)
The following pages show the details of how dimensions B, C, D and E were calculated, and
are provided only for your information. The above table is important for your modeling
requirements. For miters with more than 4 cuts, you have to calculate the required
199
Miter Bend
dimensions similar to those shown on the following pages. The next table outlines the
modeling procedure for either miter type.
Miter Modeling Procedure
Determine Miter Type
Closely Spaced Miter Widely Spaced Miter
Any number of cuts 2 cuts 3 cuts 4 cuts
Calculate R, B, C, D
Model miter as a single Calculate R, B and C from above table. and E from above ta-
Closely spaced miter Calculate offsets of nodes using B and C. See ble. Calculate offsets
with Bend radius as Example 2 Widely spaced miter. of nodes using B, C,
dimension A. D and E.
Model this miter as 2 Model this miter as 3 Model this miter as 4
See Example 1, Closely widely spaced miters widely spaced miters widely spaced miters
spaced miter. with R as the bend with R as the bend with R as the bend
radius. radius. radius.
200
Miter Bend
Two Miter Cuts (N=2)
 = 90° / ( 2 x 2 ) = 22.5°
S = 2 A tan =0.828427 A
B = A tan = 0.414214 A
C = 2 A tan cos 2= 0.585786 A
Closely Spaced
The miter is Closely spaced if,
S r ( 1 tan  )
Substituting for S,
2A tan  r ( 1 tan  )
A  1.707107 r
Bend radius, R = 0.5 S cot A
Widely Spaced
R = 0.5 r ( 1 cot  ) = 1.707107 r
Three Miter Cuts (N=3)
 = 90° / ( 2 x 3 ) = 15°
S = 2 A tan =0.535898 A
B = A tan = 0.267949 A
C = 2 A tan cos 2= 0.464102 A
Closely Spaced
S r ( 1 tan  )
Substituting for S,
2A tan  r ( 1 tan  )
A  2.366025 r
Widely Spaced
R = 0.5 r ( 1 cot  ) = 2.366025 r
201
Miter Bend
Four Miter Cuts (N=4)
 = 90° / ( 2 x 4 ) = 11.25°
S = 2 A tan =0.397825 A
B = A tan = 0.198912 A
C = 2 A tan cos 2= 0.367542 A
D = 2 A tan sin 4= 0.281305 A
E = 2 A tan sin 2= 0.152241 A
Closely Spaced
S r ( 1 tan  )
Substituting for S,
2A tan  r ( 1 tan  )
A  3.013670 r
Widely Spaced
R = 0.5 r ( 1 cot  ) = 3.013670 r
Example 1: Closely Spaced Miter

Example Data
Pipe OD = 8.625”, thickness, t = 0.322”

Mean pipe radius, r = (8.625 - 0.322) / 2 = 4.1515”,
Number of miter cuts = 3,
Dimension A = 8”, See 3-cut miter figure.
Look up table (Miter modeling procedure), for an outline of the modeling procedure.
First, determine the type of miter (Closely or Widely spaced) before modeling it.
Look up Summary of Miter parameters, for N = 3. A miter is Closely spaced if A <
2.366025r. This condition is true for r = 4.1515”. Hence, this is a Closely spaced miter.
Steps for Example 1
 Create From node: When you start a new model file, node 10 and an Anchor are
automatically input, press Enter to move cursor to next empty row.
 Construct miter bend: type 20 for Node (simply pressing Tab puts this node number
automatically for you), type “m” under the Type column, type 8” for bend radius,
select Closely spaced, click on Ok. Type 10” for DX, enter material, section and load
names, press Enter.
 Finish the miter bend: type 30 for Node and -10” for DY, press Enter.
202
Miter Bend
Example 2: Widely Spaced Miter

Let us assume the same data as in Example 1 (Closely spaced miter) with only one change,
namely, dimension A (see next figure).
Pipe OD = 8.625”, thickness, t = 0.322”
Mean pipe radius, r = (8.625 - 0.322) / 2 = 4.1515”,
Number of miter cuts = 3,
Dimension A = 12”.
Look up table (for Miter modeling procedure) for an outline of the modeling procedure.
It is essential to determine the type of miter (Closely or Widely spaced) before modeling it.
203
Miter Bend
Determine miter type

Look up table, Summary of Miter parameters, for N = 3. A miter is Closely spaced if A <
2.366025r. This condition is false for r = 4.1515”. Hence, this is a Widely spaced miter. This
miter bend has to be modeled as a series of 3 miters. Next, observe in this table that
dimensions B, C and R (Equivalent miter bend radius) can be calculated for N = 3 cuts.
Calculate required dimensions
With r = 4.1515”,
Equivalent miter bend radius, R = 2.366025 × r = 9.8225”,
Dimension B = 0.267949 × A = 3.2154”,
Dimension C = 0.464102 × A = 5.5692”
See previous figure. After calculating B, C and R, let us now calculate the offsets of nodes 20,
30 and 40 (the 3 nodes correspond with the 3 miter cuts).
Calculate Offsets
Offsets of node 20 from 10: (First miter cut)
DX = B = 3.2154”
DY = 0” (because node 20 is on the horizontal axis).
Offsets of node 30 from 20: (Second miter cut)
DX = C = 5.5692”
DY = -B = -3.2154”
Offsets of node 40 from 30: (Third miter cut)
DX = B = 3.2154”
DY = -C = -5.5692”
Offsets of node 50 from 40:
DX = 0” (because node 50 is on the vertical axis).
DY = -B = -3.2154”
Now, start to build the model in CAEPIPE as shown in Example 1 but with different data
(bend radius = 9.8225”, select Widely Spaced miter, and offsets as shown above).
Steps for Example 2
 Create From node: When you start a new model file, node 10 and an Anchor are
automatically input, press Enter to move cursor to next empty row.
 Construct first miter bend: type 20 for Node (simply pressing Tab puts this node
number automatically for you), type “m” under the Type column, type 9.8225” for
bend radius, select Widely spaced, click on Ok. Type 3.2154” for DX, enter material,
section and load names, press Enter.
 Construct second miter bend: type 30 for Node (simply pressing Tab puts this node
bend radius, select Widely spaced, click on Ok. Type 5.5692” for DX, -3.2154” for
DY, press Enter.
204
Miter Bend
 Construct third miter bend: type 40 for Node (simply pressing Tab puts this node
bend radius, select Widely spaced, click on Ok. Type 3.2154” for DX, -5.5692” for
DY, press Enter.
 Finish the miter bend: type 50 for Node and -3.2154” for DY, press Enter.
Sherwood, David. R., and Dennis J. Whistance. THE “PIPING GUIDE” A Compact
Reference for the Design and Drafting of Industrial Piping Systems. First Edition (revised).
San Francisco: Syentek Books Co., 1980.
205
Node
A Node refers to a connecting point between two elements such as pipes, reducers, valves or
expansion joints. The maximum # of nodes in a model cannot exceed 9,999; while the
maximum node number itself cannot be greater than 99,999.Usually, a node has a numeric
designation. In CAEPIPE, occasionally, you may have a need to reference a node followed
by a letter such as A/B/C/D/J. These are automatically generated internal nodes. A and B
nodes (e.g., 20A, 20B) designate the near and far ends of a Bend node (see section on Bend).
J, C and D (e.g., 10J, 10C, 10D) designate a Jacketed pipe and a Jacketed bend (see section
on Jacketed Piping).
In the Layout window, node numbers are typed under the column titled Node. A node
number may be typed as an integer or an integer followed by one of the letters A/B/C/D/J.
Use “Location” type to specify more than one data item at a node (See section on Location).
The starting node in a piping system is always a node of type “From,” which is usually
anchored.
Nodes not only act as connect points for elements but also act as locations for providing
supports or applying restraints/external forces and moments to the piping system. Each
node has six static degrees of freedom (three dynamic), three translational (in x, y and z
directions) and three rotational (about x, y and z axes). Any or all of them may be restrained
using supports.
Specifying Coordinates
The values typed in the DX, DY and DZ fields on the Layout window are interpreted as
offsets from the previous node. If required to specify absolute coordinates for a node (i.e.,
fix the location of a point in space), the node has to be of type “From” or should have an
asterisk (*) at the end of it (e.g., 20*). In these cases, the numbers entered in the DX, DY and
DZ fields are interpreted as absolute coordinates of the node rather than offsets from the
previous node. If the coordinates for a particular node are duplicated the second set of
values is ignored. An asterisk (*) for a “From” node is ignored too.
You can list all coordinates by selecting Coordinates under Misc menu (or click on the Node
header or right click on any node number). This feature can be helpful for verifying
correctness of the input.
In the Layout window, you can search for a node by using the “Find node” command (under
the View menu, or use Ctrl+F), specify a “Node increment” for automatic node numbering
(under Options menu), and renumber nodes for a range of rows (under Edit menu)
Automatic Renumbering
When you delete a row with a node number in the Layout window, CAEPIPE automatically
renumbers nodes from the top starting with the number you have specified under the main
window > File > Preferences > Automatic Renumbering. Note that this is different from the
user-performed selective renumbering of a range of rows from the Edit menu.
When turned on, deleting a row triggers an automatic renumbering operation inside the
Layout window. So if you do not want such to happen, turn the feature off from the main
window.
206
Nonlinearities
There are several types of nonlinearities in CAEPIPE:

1. Gaps in limit stops and guides.
2. Rotational limits in ball and hinge joints.
3. Rod hangers as one-way restraints.
4. Tie rods with different stiffnesses and gaps in tension and compression.
5. Friction in limit stops, guides, slip joints, hinge joints and ball joints.
6. Buried piping.
An iterative solution is performed when nonlinearities are present. At the start of each
iteration, the overall global stiffness matrix and the load vector are reformulated based on the
solution (displacements) obtained from the previous iteration.
Limit stop
Limit stops are input by specifying direction, the upper and lower limits (i.e., gaps) and
optionally a friction coefficient and a support stiffness that comes into play when either gap
is closed. The upper and lower limits are along the direction of the limit stop and measured
from the undeflected position of the node. Typically the upper limit is positive and the lower
limit negative. The upper limit should be algebraically always greater than or equal to the
lower limit. In some situations it is possible to have a positive lower limit or a negative upper
limit. If a particular limit does not exist (i.e., a node can move unrestrained in that direction),
that limit should be left blank. If a gap does not exist, then the corresponding limit should be
explicitly input as zero.
Solution Procedure
At the end of each iteration, the displacements computed at the limit stop node are resolved
along the limit stop direction. The resolved displacement (in the limit stop direction) is
compared with the upper and lower limits. If the resolved displacement exceeds the upper
limit or is less than the lower limit, the gap is closed; otherwise it is open. The solution is
converged when the displacements computed are within 1% of the corresponding
displacementsobtained at the end of the previous iteration.
If both upper and lower limits (gaps) of a limit stop are open, no stiffness is applied at
thatlimit stop node, and friction does not arise there.
If either one of the two gaps is closed,the support load along the limit stop direction is
calculated as(resolved displacement - gap distance) x user-specified support stiffness. In this
case, if friction is specified at the limit stop, the following iterative procedure is carried out.
1. The resultant displacement (d) in local yz plane is computed at this limit stop node.
2. The support load at the limit stop is calculated as (resolved displacement - gap
distance) x user-specified support stiffness.
3. Using this calculated support load, the friction force (ff) is calculated as ff = friction
coefficient (mu) x support load.
4. Using the friction force (ff) and the resultant in-plane displacement (d), the friction
stiffness (kf) is computed as kf = ff / d.
5. The global stiffness matrix is then updated to include the friction stiffness computed.
207
Nonlinearities
6. Analysis iteration is continued with the updated global stiffness matrix. The solution
is converged when the displacementscomputed are within 1% of the corresponding
displacements obtained at the end of the previous iteration.
7. After the solution has converged and if the gap is closed, the support load and the
friction force at the limit stop are calculated as:
Support load at limit stop = (resolved displacement - gap distance) x user-specified
support stiffness.
y shear = fy = local y displacement x friction stiffness (kf)
z shear = fz = local z displacement x friction stiffness (kf)
Resultant friction force (ff) = sqrt(fy^2 + fz^2)
During hanger design, the hot loads are recalculated with the status of the limit stops at the
end of the preliminary operating load case. Then the hanger travels are recalculated using the
recalculated hot loads.
In dynamic analysis, the status of the limit stops at the end of the first operating load case
(W+P1+T1) is used. If either the upper or lower limit is reached for the first operatingload
case, the limit stop is treated as a rigid two-way restraint in the direction of the limit stop. If
both limitsare not reached, then that limit stop is treated as having no restraint.
Friction
Friction is specified by entering coefficient of friction for limit stop and guide, entering
friction force and/or friction torque for slip joint, entering friction torque for hinge joint and
entering bending and/or torsional friction torque for ball joint.
Friction is modeled using variable equivalent friction stiffnesses (fictitious restraints) in
CAEPIPE as described in the above Solution Procedure. The stiffnesses of these fictitious
restraints are estimated from the results of previous iteration. If friction is included in
dynamic analysis, these equivalent friction stiffnesses computed from the last iteration of the
first operating load case are included in modal and dynamic analyses.
Friction in Limit Stop
If the gap is not closed, there is no normal force and hence no friction. If the gap is closed,
the normal force (limit stop support load) is calculated as explained above. The maximum
friction force is friction coefficient * normal force. The displacement of the limit stop node
is resolved into a plane normal to the limit stop direction (let us call this resolved
displacement: y). Also let ky = equivalent friction stiffness which is assumed to be zero for
the first iteration.
If y is non-zero or y * ky> maximum friction force,
then ky = maximum friction force / y
12
otherwise ky = high stiffness ( 1×10 lb/inch) [This is the case of no sliding]
In the next iteration the equivalent friction stiffness is added to the global stiffness matrix.
The iterations are continued till the displacement y is within 1% of y displacement from the
previous iteration. The friction force is y * ky.
208
Nonlinearities
Friction in Guide
A guide is modeled by adding high stiffnesses perpendicular to the direction of the pipe. The
normal force in the guide is calculated by the vector sum of the local y and z support loads.
Maximum friction force is friction coefficient * normal force. The displacements at the guide
node are resolved in the direction of the guide axis. Let us call this displacement: x. Also let
kx = equivalent friction stiffness which is assumed to be zero for the first iteration.
If x is non-zero or x * kx> maximum friction force,
then kx = maximum friction force / x
12
otherwise kx = high stiffness (1×10 lb/inch) [This is the case of no sliding]
The iterations are continued till the displacement x is within 1% of x displacement from the
previous iteration. The friction force is x * kx.
Friction in Slip Joint
The relative displacements (between From node and To node) for the slip joint are resolved
in the direction of the slip joint. Let us call this relative displacement: x. Also let kx =
equivalent friction stiffness which is assumed to be zero for the first iteration.
If x is non-zero or x * kx> friction force input,
then kx = friction force input / x
12
otherwise kx = high stiffness (1×10 lb/inch) [This is the case of no sliding]
The iterations are continued till the displacement x is within 1% of x displacement from the
previous iteration. The friction force is x * kx.
Similar technique is used for friction torque (using rotations instead of translations).
Friction in Hinge Joint
The relative rotations (between From node and To node) for the hinge joint are resolved in
the direction of the hinge axis. Let us call this relative rotation: x. Also let k x = equivalent
friction stiffness which is assumed to be zero for the first iteration. Maximum friction torque
= friction torque input + hinge stiffness input * x.
If x is non-zero or x * kx> maximum friction torque,
then kx = maximum friction torque / x
otherwise kx = high stiffness [This is the case of no sliding]
The iterations are continued till the relative rotation x is within 1% of x rotation from the
previous iteration. The friction torque is x * kx.
209
Nonlinearities
Friction in Ball Joint

For a ball joint friction in bending (transverse) and torsional (axial) directions is treated
independently. For the bending case, the resultant of the local y and z directions is used.
Otherwise a procedure similar to the one used for hinge joint is used.
Friction in Dynamic Analysis
Friction is optional in dynamic analysis. Friction is mathematically modeled by using
equivalent stiffnesses. If friction is included in dynamic analysis, these equivalent friction
stiffnesses computed from the last iteration of the first operating load case are included in
modal and dynamic analyses.
Misconvergence
During the iterative solution procedure for nonlinearities, a misconvergence is reported in
the following manner:
You have three options:

Continue the iterative procedure for 500 more iterations to see whether the solution
converges, or
Accept the misconvergence (maximum misconvergence is reported, 100% in the above
dialog), or
Exit the analysis processor completely.
An environment variable “CPITER” may be defined to change number of iterations from
the default 500. For example, when “CPITER=1000”, iterations will continue up to 1000
before showing the dialog: Continue, Accept or Exit if there is misconvergence.
The shown misconvergence in the solution is really a quantification of how much off the
results will be, IF you accept it. The maximum misconvergence (100% in the above case) is
shown along with its location where such is happening.
In case of a solution with large misconvergence, you have two options: 1. Change parameters
(mainly gaps and friction values, or removal of unneeded supports) inside the model to
influence the convergence routine, or, 2. Increase the # of iterations. In case neither works,
then use engineering judgment to accept or reject the merit of such a solution.
210
Nozzle
Nozzles are integral attachments of vessels (such as pressure vessels, storage tanks etc.)
which connect with external piping. Nozzles transmit the shell (vessel) flexibility to the
piping system and hence are generally included in piping stress analysis.
Three types of nozzles can be modeled in CAEPIPE, namely (i) a nozzle attached to a
cylindrical vessel relatively far from the ends of the cylinder, (ii) a nozzle attached to a
spherical vessel or torispherical head and (iii) a nozzle attached to a cylindrical shell with a
flat bottom and close to the flat-bottom. CAEPIPE calculates the nozzle stiffnesses (local
flexibility components) according to WRC 297, PD 5500 and API 650 guidelines. See
Annexure II for the procedures according to WRC 297 and API 650.
Note:
Pressure Thrust (End-cap Force) of Pressure P x Inner Area (A) of pipe is not included in
the Support Loads for Nozzles displayed by CAEPIPE at this time. Since CAEPIPE’s
results for numerous problems compare well with the results from other third-party
software, it confirms that the other stress programs are also not including the Pressure
Thrust (End-cap Force) of pipe in the Nozzle Loads at this time. Refer to the Verification
Manual supplied with CAEPIPE for comparison of results with other stress programs.
If you wish to include the effect of Pressure Thrust (End-cap force) due to internal pressure
in your piping on the Nozzle loads, then you will have to compute the same manually (= P x
A) and input it as an external force at the Nozzle Nodes using the Force data type available
with CAEPIPE. Please choose the option “Add to W+P” in the Force data type dialog. By
doing so, the End-cap force (= P x A) will be included in all relevant load cases and
combinations of CAEPIPE. Of course, when the “None” code is selected under Optionss >
Analysis > Code, this End-cap force is included in the only case of “Static”.
Nozzle attached to a cylindrical vessel
211
Nozzle
The coordinate system is as shown in the figure. The six components of the forces and
moments at the nozzle-vessel interface are:
P = Radial load MC = Circumferential moment
VC = Circumferential load MT = Torsional moment
VL = Longitudinal load ML = Longitudinal moment
Of the six components of shell stiffnesses, only three stiffnesses, axial (Kx), circumferential
(Kyy), and longitudinal (Kzz), are computed. The remaining three are assumed to be rigid.
A nozzle is input by typing “n” in the Data column or selecting “Nozzle” from the Data
Types dialog.
The Nozzle dialog is shown. Note that the Displacements button is disabled. It is only
enabled after the nozzle is input (i.e., existing nozzle).
212
Nozzle
Nozzle
OD: Outside diameter of the nozzle.
Thk: Thickness of the nozzle.
Vessel
OD: Outside diameter of the vessel.
Thk: Thickness of the vessel.
L1, L2: Distances from the nozzle to the nearest stiffening ring, tubesheet or the vessel end.
Vessel axis direction
The orientation of the vessel axis in terms of its global X, Y and Z components are entered
here. See example under “Specifying a Direction.”
Nozzle to Spherical / Torispherical Shell
For a nozzle attached to a spherical shell or a torispherical head, check the Spherical Vessel
checkbox and enter the required details.
213
Nozzle
Nozzle attached to Flat-Bottom Tanks

For a nozzle on a flat-bottom tank and close to the flat-bottom, check the Flat-bottom tank
checkbox.
A slightly modified Nozzle dialog is displayed.
214
Nozzle
All the input fields are the same as before except:

L
L is the distance from the flat-bottom to the nozzle centerline.
Reinforcing pad
If the nozzle is reinforced, check this box.
Stiffness Calculation
Here again, only three shell stiffnesses, axialtranslational (i.e., axial to the pipe, the same as
radial to the shell), circumferential bending and longitudinal bending are computed according
to API 650 guidelines. See Annexure II for the procedure. The remaining three shell
stiffnesses are assumed to be rigid.
The flat-bottom tank nozzles are subject to the following limitations (API 650):
Limitations
 Nozzle OD/Vessel OD ratio must be between 0.005 and 0.04.
 L/Nozzle OD ratio must be between 1.0 and 1.5. See graphs in Annexure II, Figures
D-3 through D-14.
Displacements
After a nozzle is input in a stress model, Displacements (translations and/or rotations) in the
global X, Y and Z directions may be specified for that nozzle (for thermal, settlement and
seismic cases). Click on the Displacements button. You will see a dialog similar to one shown
below. Type in specified displacements and press Enter.
215
Nozzle
There are three types of displacements which can be specified:

1. Thermal (up to 10 displacements can be specified, one for each of the thermal loads T1
through T10). Applied only to the Expansion and Operating load cases.
2. Seismic (available for B31.1, ASME Section III Class 2, RCC-M and EN 13480 codes
only). Solved as a separate internal load case and the results so obtained for this case are
added absolutelyto the corresponding results from static seismic and response spectrum
load cases.
3. Settlement (available under ASME B31.1, ASME Section III Class 2, RCC-M and EN
13480 codes only). Applied as a separate load case called Settlement as described below.
Settlement
For certain piping codes (ASME B31.1, ASME Section III Class 2, RCC-M and EN 13480),
a settlement, which is a single non-repeated movement (e.g., due to settlement of
foundation), may be specified. This is applied to the Settlement load case. For those codes
which do not have a separate provision for settlement (like B31.3), specify the settlement as a
thermal displacement (a conservative approach) for one of the temperature load and define
that temperature as equal to reference temperature.
Example 1: Nozzle on a cylindrical vessel
Assume the following data:
Vessel Radius = 900 in.
Vessel Thickness = 1.0 in.
Nozzle OD = 26 in.
Nozzle Thickness = 0.5 in.
L1 = L2 = 1200 in.
6
Elastic modulus for vessel material = 28×10 psi.
216
Nozzle
The first node (10) is already defined as an anchor. To replace the anchor by a nozzle, right
click on the Anchor in the Data column, then select Delete Anchor. Then type “n” in the
Data column to input the nozzle. The Nozzle dialog will be shown. Input the nozzle data in
the dialog. The Layout window looks like the following:
The graphics is shown next.
Nozzle Stiffnesses
The three local shell stiffnesses computed can be viewed using the List command (Ctrl+L in
the Layout window) and selecting Nozzle Stiffnesses. The following window is displayed.
The nozzle/vessel data may be edited here (double click to edit).

Example 2: Nozzle on a spherical vessel
Vessel OD = 1800 in.
Nozzle OD = 26 in.
6
Elastic modulus for vessel material = 28×10 psi.
The first node (10) is already defined as an anchor. To replace the anchor by a nozzle, right
click on the Anchor in the Data column, then select Delete Anchor. Then type “n” in the
217
Nozzle
Data column to input the nozzle. The Nozzle dialog will be shown. Input the nozzle data in
the dialog. The Layout window looks like the following:
Nozzle Stiffnesses
The three local shell stiffnesses computed can be viewed using the List command (Ctrl+L in
the Layout window) and selecting Nozzle Stiffnesses. The following window is displayed.
The nozzle/vessel data may be edited here (double click to edit).

Example 3: Nozzle on a flat-bottom storage tank
Vessel OD = 1800 in.
Nozzle OD = 26 in.
L = 36 in.
218
Nozzle
6
Elastic modulus of the vessel material = 28×10 psi.
No reinforcing pad on the vessel.
Create the piping till the nozzle node. At the nozzle node, enter a Nozzle by typing “n”,
check the Flat-bottom tank checkboxand provide the required data.
The Layout window looks like the following.
The three nozzle stiffnesses computed can be viewed as before by using the List command
(Ctrl+L) and selecting Nozzle Stiffnesses.
LCS (Local Coordinate System) can be displayed for Nozzle element by a few ways:
1. List (Ctrl+L) > Nozzles > View menu > Show LCS, -OR-
2. List (Ctrl+L) > Nozzles > Mouse Right Click on the listed Nozzle > Show LCS, -OR-
3. Results window > Support Loads > Other Support Loads > Nozzles > View menu >
Show LCS, -OR-
4. Results window > Support Loads > Other Support Loads > Nozzles > Mouse Right
Click > Show LCS.
219
Nozzle
220
Nozzle EvaluationModule
One of the qualification requirements for a piping system is to keep the loads imparted by
the piping on equipment nozzles within certain allowable limits. These loads consist of sets
of three forces and three moments, for the various load combinations. There are basically
two types of nozzle load limits: (1) nozzle loads applied to rotating equipment, and (2) nozzle
loads applied to static equipment such as heat exchangers, tanks and vessels.
Nozzle Loads applied to rotating equipment
Rotating equipment consist of equipment with moving parts, such as pumps, compressors,
turbines and fans. The pipe nozzle load limits are developed by the equipment manufacturer
and are intended to prevent malfunction, such as shaft misalignment, or distortion of the
casing that could impede the movement of impellers. These limits are typically based on
actual testing of the equipment, and not on analysis.
Some pump standards have published standard nozzle load limits, but these are only valid
for the particular pumps for which they are published. This is the case for the American
Petroleum Institute’s API-610 and the Hydraulic Institute’s HI 9.6.2.11 standard.
See subsection titled “Pump” from this manual for more details.
Similarly, API 617 provides nozzle load limits for compressors and NEMA SM-23 for
turbines.
See subsections titled “Compressor” and “Turbine” from this manual for further details.
Nozzle Loads applied to static equipment
For Nozzlesconnected to static equipment such as heat exchangers, tanks and vessels, the
pipe load limits are based on stress or strain limits at the nozzle-to-shell intersection, both on
the shell and nozzle sides.At present, CAEPIPE considers pipe load limits based on stress or strain
limits only at the shell side of the nozzle intersection.
Nozzle Evaluation Module
The Nozzle Evaluation module implemented in CAEPIPE computes Allowable Loads and
Local Stresses at Shell for Nozzles connecting to Spherical and Cylindrical Vessels as per the
following codes.
1. Allowable Loads on Nozzle – EN 13445-3 (2009) (hereinafter referred as EN13445).
2. Local Shell Stresses at Nozzle– WRC Bulletin 537 (hereinafter referred as WRC 537).
The Nozzle Evaluation module assumes that the flexibility analysis of that piping system
with CAEPIPE has already been performed, which will have produced a stress report as well
as the forces and moments at the location where the nozzle is attached to the piping.
Note that this module is separate from a piping stress model file and can be accessed from
File Menu > Open/New command.
221
dialog with input fields you can edit. You will need to enter the data in this dialog. The
different parameters required to be input are explained in detail below.
Details on implementation for Calculation of Local Shell Stresses at Nozzles as per WRC
Bulletin 537 are provided in Code Compliance Manual.
222
Code
Selecting the Code as “Allowable Loads on Nozzles – EN 13445-3 (2009)” will allow user to
compute the “Allowable Loads on Nozzles”.
On the other hand, selecting the Code as “Local Shell Stresses at Nozzles – WRC Bulletin
537” will allow user to compute the Local Shell Stresses at Nozzles as per WRC 537 and
perform Stress Compliance as per ASME Section VIII Division 2.
Nozzle to Spherical / Cylindrical Shell
Selecting the option “Nozzle to Spherical / CylindricalShell” will show the dialog boxes as
shown below for the two Codes.
For EN13445, the input data required for the two types are shown below.
223
Similarly, for WRC 537, the data required to be input for the two types are shown below.
Load Case
From the option available, select the Load Case for which the three (3) forces and three (3)
moments are being entered.
Depending upon the selection of “Load Case”, CAEPIPE will compute

Radial/Circumferential, Tangential/Longitudinal, Shear and Combined Stress Intensities as
explained below.
Please note, the display text for “Allowable (All)”inputfield will change
automaticallydepending on the “Load Case” selected.
For example, for “Sustained” Load Case, the display text for “Allowable (All)” input field
will change to “Sustained Allowable (All)”. Similarly for “Sustained + Occasional”, the
display text will change to “Sustained + Occasional Allowable (All)”.
224
Sustained and Sustained + Occasional

Combined Stresses computed willexclude thefollowing BendingStresses from the Evaluation
of Nozzle to Spherical / Cylindrical Shells respectively.
a. Radial/Circumferential Bending Stresses due to P
b. Radial/Circumferential Bending Stresses due to M1/MC
c. Radial/Circumferential Bending Stresses due to M2/ML
d. Tangential/Longitudinal Bending Stresses due to P
e. Tangential/Longitudinal Bending Stresses due to M1/MC
f. Tangential/Longitudinal Bending Stresses due to M2/ML
Operating
Combined Stresses computed will include all Membrane and Bendingstresses due to P,M1
and M2 / P, MC and ML, Torsional stresses due to MTas well as Shear stressesdue to V1/VC
and V2/VL.
Once the required data are input, save the file (Nozzle evaluation will have a .noz extension).
Now, select File menu > Analyze to calculate loads or stressesas per the code selected, which
225
Loads
For Nozzle to Spherical Vessel, enter the following loadscomputedin piping analysis for the
selected load case.
1. Radial Load (P)
2. Shear Load (V1)
3. Shear Load (V2)
4. Overturning Moment (M1)
5. Overturning Moment (M2) and
6. Torsional Moment (MT)
Similarly, for Nozzle to Cylindrical Vessel, enter the following loads computed in piping
analysis for the selected load case.
1. Radial Load (P)
2. Shear Load (VC)
3. Shear Load (VL)
4. Circumferential Moment (MC)
5. Longitudinal Moment (ML) and
6. Torsional Moment (MT)
Vessel and Nozzle Parameters
For Spherical Vessel, from the Vessel Drawing, read and enter the following parameters.
Vessel Thickness (T), Vessel Mean Radius (Rm), Nozzle Outside Radius (ro), Nozzle
Thickness (t) and Nozzle Mean Radius (rm).
Similarly, for Cylindrical Vessel, from the Vessel Drawing, read and enter the following
parameters.
Vessel Thickness (T), Vessel Mean Radius (Rm) and Nozzle Outside Radius (ro).
Fillet Radius
Fillet Radius is required to compute the Stress Concentration Factors for Tension (K n) and
Bending (Kb) from Figure B-2 WRC Bulletin 537.
Pressure Stress at Shell (Pm)
Pm is the Average Primary Membrane Stress across the cross-section of the vessel away from
Gross Structural Discontinuities such as a Nozzle.
For a Spherical Shell such as Enclosure/Head to a Vessel, Pm due to internalpressure would
be PR/2T,where P = Internal Pressure, R is the Mean Radius of the Head and T is the
Thickness of the Head.
226
For a Cylindrical Shell such as Pressure Vessel/Pre-heater/Tank, to be conservative, Pm due

to Internal Pressure would be the Circumferential Stress = PD/2T, where D is the Mean
Diameter of the Cylinder.
The Stress value thus calculated should be entered in this field.
Bending Stress at Shell (Pb)
Pb is the Primary Membrane Stress proportional to the distance from the Axis of the Vessel
due to External Loads such as Weight, Wind, Earthquake, etc. away from Gross Structural
Discontinuities such as a Nozzle.
For a Spherical Shell such as Enclosure/Head to a Vessel, being a free end, Bending Stress
Pb due to external loads could be 0.0.
For a Cylindrical Shell such as Pressure Vessel/Pre-heater/Tank, Bending Stress Pb due to
external loads could be calculated as M/Z, where M is the Bending Moment on the Shell at
Nozzle location and Z is the Section Modulus of the Shell.
So, Bending Stress Pb could be manually calculated or determinedusing computer programs.
Allowable Stress (All)
Allowable Stress (All) is to be computed and entered depending on the load case selected.
For example, as perClause 5.2.2.4 of ASME Section VIII Division 2 (2017), the Allowable
Stress (All) for both “Sustained” and “Sustained + Occasional” load case is to be entered as
“1.5Sh”, where Sh is the basic allowable stress at maximum metal temperature for Shell.
Similarly,as per Clause 5.5.6 of ASME Section VIII Division 2 (2017), the Allowable Stress
(All) forOperating load case should be entered as “3 (SC + Sh)/2 = 1.5 (SC + Sh)”, where Sc is
the allowable stress at minimum metal temperature for Shell and Sh is defined above.
Stress Concentration Factors (Kn and Kb)
CAEPIPE will automatically compute the values of K n and Kbusing Figure B-2 of WRC 537
when these fields are entered as 0.0.
On the other hand, when thesefields are entered with a value great than or equal to 1.0, then
CAEPIPE will use these values of Kn and Kbwhile computing the stresses as per WRC 537.
Once the required data are input, save the file (Nozzle evaluation will have a .noz extension).
Now, select File menu > Analyze to perform the evaluation, which will be shown right below
the input information.
227
228
229
Nozzle Evaluation Module Menus

File Menu
.Analyze.
Analyze command calculates nozzle allowable loads as per EN 13445-3 (2009) or shell
stressesat the attachment as per WRC Bulletin 537 and compares them to stress allowable
specified by ASME Section VIII Division 2 (2017).
Print.
You can print a Report by using the Print command. You can also preview the report by
clicking the Preview button on the print dialog.
230
231
Edit Menu
.Edit.
You can edit the data by clicking the Edit command.
Options Menu
.Units.
.Font.
232
Pressure Design of Pipe and Pipe fittings
Pressure Design of Pipe and Pipe Fittings according to EN 13480-3 (2017)

Pressure Design of Pipe and Pipe Fittings can be performed using the modules built into
CAEPIPE which are independent of the piping flexibility analysis.
These modules can be launched through Layout frame > Misc > Internal Pressure Design:
EN 13480-3 and Layout frame > Misc > External Pressure Design: EN 13480-3.
Note:
These modules perform Pressure Design of Pipe and Pipe Fittings ONLY using the
equations given in the EN 13480-3 (2017) Code irrespective of the Analysis Code selected
for piping flexibility analysis in CAEPIPE.
In case the pipe stress analysis is performed with an Analysis Code other than EN 13480-3
(2017), the Pressure Design modules will use the material allowable stresses corresponding
to the maximum temperature T1 through T10 entered in the CAEPIPE stress model.
Internal Pressure Design of Pipe and Pipe Fittings
Snap shots shown below present a sample stress model developed to show the Internal
Pressure Design calculations performed by CAEPIPE.
233
234
235
Internal pressure design calculations of pipe and pipe fittings according to EN 13480-3
(2017) are independent of element lengths entered. Hence, these calculations can be
performed from the CAEPIPE model already developed for flexibility analysis. Equations
used for performing Internal Pressure Design as per EN 13480-3 (2017) are provided in
Section titled “Pressure Design of Pipe & Pipe Fittings” in the Code Compliance Manual.
Once the layout of the stress model as shown in the above snap shots is completed, the
internal pressure design is performed through Layout window > Misc > Internal Pressure
Design: EN 13480-3.
When executed, CAEPIPE automatically performs the pressure design calculations for Pipes,
Elbows, Miters, Bends and Reducers for the entire stress model and displays the results as
shown below.
It is observed that the ratios Uf1 and Uf2 are all less than 1.0, confirming that the Internal
Pressure Design requirements of EN 13480-3 (2017) code are met for this stress model.
236
The results shown above can also be printed to the printer or to a file using the option File >
Print.
237
238
External Pressure Design of Pipe and Pipe Fittings

External Pressure Design module will function ONLY when the stress layout is defined with
negative pressure (such as vacuum pressure).
This module first calculates collapse pressure (same as buckling pressure), which is a function
of span length “L” between the stiffeners placed on the piping (shown in figures below).
Since the collapse (buckling) mode of deformation for a pipe element between two adjacent
stiffeners is restrained by these stiffeners, shorter the span length L between the stiffeners,
higher the collapse (buckling) pressure.
The External Pressure Design module assumes that a stiffener is located at each node of the
CAEPIPE model. Hence, ensure that nodes are defined in CAEPIPE model only at
locations where the stiffeners are attached to the piping. Even nodes where flanges or
certain types of supports that restrain the collapse (buckling) mode of deformation should be
included as “stiffener locations”. All other nodes at which the collapse (buckling) mode of
deformation is not restrained (such as resting supports) should not be included in the
CAEPIPE model for external pressure design calculations. In other words, the CAEPIPE
stress model (that was developed for flexibility analysis) needs to be edited before
performing the external pressure design.
Single Pipe
Pipe with bend
Pipe with flange connections
239
Pipe with bend or elbow with ‘L’ measured on extrados
Pipe with mitre with ‘L’ measured on extrados

The procedure given below will help in retaining ONLY those nodes of the CAEPIPE stress
model (originally developed for flexibility analysis) prior to External Pressure Design
calculations.
 Create a copy of the existing CAEPIPE stress model (developed for flexibility
analysis).
 At whichever node the collapse (buckling) mode of deformation is NOT restrained,
navigate to that element node in the layout window and use the option “Combine…”
through Layout window > Edit. This action will remove that node by combining the
two adjacent elements.
 Repeat the above step and remove all other nodes where the collapse (buckling)
mode of deformation is NOT restrained, thereby retaining ONLY the stiffeners,
flanges and supportsthat restrain the collapse (buckling) mode].
 Upon completion, save the model.
Snap shots shown below present a sample model developed to show the External Pressure
Design calculations performed by CAEPIPE. As stated above, a copy of the original stress
model was made and the model has been edited to include only those nodes on pipe where
stiffeners, flanges and supports (that are equivalent to stiffeners from the point of view of
restraining collapse mode of deformation) are attached.
Refer Section titled “External Pressure Design according to SS EN 13480-3” in CAEPIPE
Code compliance manual for details on implementation.
240
241
242
243
Once the layout of the stress model as shown in the above snap shots is completed, the
external pressure design is performed through Layout window > Misc > External Pressure
Design: EN 13480-3.
When executed, CAEPIPE automatically performs the external pressure design calculations
for Pipes, Miters, Elbows, Bends and Reducers for the entire stress model and displays the
results as shown below.
It is observed that the ratio [Pr/(KPc)] is much higher than 1.0 throughout the stress model,
confirming that the collapse (buckling) pressures Pr calculated for all segments of the stress
model are much higher than the corresponding peak negative pressures specified in the
CAEPIPE model. In other words, the potential for any segment of this piping system to
collapse (buckle) is very minimal.
244
The results shown above can also be printed to the printer or to a file using the option File >
Print.
245
Pump
governed by an industry publication — API (American Petroleum Institute) publishes an
API 610 for Horizontal and Vertical inline pumps, ANSI/HI 9.6.2 for Rotodynamic pumps,
and API 617 for compressors, while NEMA (National Electrical Manufacturers Association)
publishes the NEMA SM-23 for turbines. These publications provide guidelines for
evaluating nozzles connected to rotating equipment among other technical information
including the items relevant to piping stress analysis – criteria for piping design and a table of
allowable loads.
Modeling the rotating equipment is straightforward since it is assumed rigid (relative to
connected piping) and modeled only through its end points (connection nozzles).
1. In your model, anchor all of the nozzles (on the equipment) that need to be included
in the analysis.
2. Specify these anchored nodes during the respective equipment definition via Misc.
menu > Pumps/Compressors/Turbines in the Layout window.
CAEPIPE does not require you to model all of the nozzles nor their connected piping. For
example, you may model simply one inlet nozzle of a pump with its piping. Or, you may
model one pump with both nozzles (with no connected piping) and impose external forces
on them (if you have that data). Further, there is no need to connect the two anchors of the
equipment with a rigid massless element like required in some archaic methods. A flange and
an anchor may coexist.
A pump is input by selecting “Pumps” from the Misc menu in the Layout or List window.
CAEPIPE produces API 610 pump compliance report after analysis for 2 types of pumps,
namely Horizontal and Vertical inline. Also, CAEPIPE generates ANSI /HI 9.6.2
compliance report for four types of Rotodynamic pumps: Horizontal or Vertical inline or
Axial split case or Vertical turbine short set pumps. See Section titled “Rotating Equipment
Qualification” from the Code Compliance Manual for related information.
246
Pump
Once you see the Pump List window, double click on an empty row for the Pump dialog and
enter the required information.
247
Pump
Type a short description to identify the pump in Description. You must designate the pump
nozzles as anchors, and the shaft axis must be in the horizontal plane. The nozzle locations
(top, side or end) should be specified for suction and discharge nodes.
See section on specifying a Direction for information on X comp/Y comp/Z comp.
For horizontal pumps, you must enter coordinates for the center of the pump with respect
to global origin. For pumps with two support pedestals, API 610 defines the center by the
“intersection of the pump shaft centerline and a vertical plane passing through the center of
the two support pedestals.” For pumps with four support pedestals, the center is defined by
the “intersection of the pump shaft centerline and a vertical plane passing midway between
the four pedestals.” See Section titled “Rotating Equipment Qualification” of the Code
Compliance Manual for illustrative figures.
You might find it helpful to first model the nozzles as anchors. In some situations, you might
not have the discharge or suction side piping. In that case, here is how you can fix the
location of the other side. Look up the coordinates of the nozzle (anchor) you have already
modeled.
Then, use the coordinates command (menu Misc > Coordinates) to note the (X, Y, Z)
coordinates of this pump nozzle node number. Using its coordinates, you can now arrive at
the required coordinates of the other side’s nozzle and the center of the pump.
Example – API 610 Pump Compliance
248
Pump
249
Pump
250
Pump
In the above image, suction side piping ends at node 380. Discharge piping starts at node
500. In the Pump definition dialog (shown next), you can see that the center of the pump
is just behind the suction nozzle (node 380) coordinates. The reducer between nodes 370
and 380 is a vertically offset eccentric reducer; hence the graphics at the reducer shows a
break.
251
Pump
A simpler example is when you do have piping on both sides of the pump. Consider the
network below consisting of two pipe segments connected by a pump.
(suction side) 10-20-30-…-90-100→PUMP←200-210-…-280-290-300 (discharge side)
The suction side of the pump ends at node 100. The discharge side begins at node 200. Make
nodes 100 and 200 as anchors so that equipment loads can be calculated. A similar method
applies to turbines and compressors too.
A different dialog is shown for vertical inline pumps. Only Description, Suction and
Discharge nodes are required.
252
Pump
API 610 Report

Upon analysis, you will see CAEPIPE produce API 610 reports under “Rotating Equipment
Reports” in Results.
In addition to the input details, for a specific load case, the calculated forces and moments
(on the nozzle and those at the center of the pump), API allowables, ratios (of calculated to
allowable) and status for all of them are reported.
When you see “Failed” entries in this report, you will need to examine the cause of the high
force or moment for that line item. Generally, the high numbers come from the expansion
load but may well come from the weight load. You must reduce these excessive forces and
moments by making the system or intersections more flexible before this pump can become
compliant.
Note: If you have input multiple temperatures, corresponding reports for additional
operating load cases are shown. Use the black right arrow key to see them.
253
Pump
Example – ANSI/HI 9.6.2 Pump Compliance
In the above image, suction side piping ends at node 10. Discharge piping starts at node
30. Pump size is 1.5 x 8 17 with Material Group ASTM A351/A351M – Grade CF8M.
The temperature of the pump is set to 1000 F.
254
Pump
ANSI/HI 9.6.2 Report

Upon analysis, you will see CAEPIPE produce ANSI/HI 9.6.2 reports under “Rotating
Equipment Reports” in Results.
In addition to the input details, for a specific load case, the calculated forces and moments,
ANSI/HI 9.6.2 allowables, ratios (of calculated to allowable) and status for all of them are
reported.
When you see “Failed” entries in this report, you will need to examine the cause of the high
force or moment for that line item. Generally, the high numbers come from the expansion
load but may well come from the weight load. You must reduce these excessive forces and
moments by making the system or intersections more flexible before this pump can become
compliant.
255
Reducer
Use a reducer to join a larger pipe to a small pipe to meet fluid flow requirements. A reducer
is Concentric when the axes at the reducer ends are collinear; Eccentric, when they are not.
Use an eccentric reducer only when necessary to keep the top/bottom of the line level.
In the figures shown below, observe that the two ends of the reducers are of different
diameters. The larger end (at node 20) has the outside diameter and thickness as OD1 and
Thickness 1 (Thk1) with the smaller end (at node 30) having OD2 and Thickness 2 (Thk2).
In case of the eccentric reducer, the eccentricity as shown is between the two axes of the
ends of the reducer. The cone angle, α, is also as shown in the following figure.
An eccentric reducer’s eccentricity is modeled by a change in offsets of the “To” node (node
30 in Figure (b) above). Eccentricity is (ID1–ID2) / 2. See example 2 later in this topic.
A reducer (concentric or eccentric) is input by typing “re” in the Type column or selecting
“Reducer” from the Element Types dialog.
256
Reducer
The Reducer dialog is shown.
OD1, OD2, Thk1, Thk2

These are the cross-sectional properties at the two ends of the reducer. OD stands for
outside diameter and Thk stands for thickness. By default, OD1 and Thk1 contain preceding
section’soutside diameter and thickness, but different values may be typed here. The
Section1 and Section2 buttons can be used to quickly input OD and Thk values from
previously defined sections.
Cone angle
Shown in Figure (a) above, it is used to calculate SIF at the ends of the reducer for certain
piping codes (B31.1, B31.9, ASME Section III Class 2, RCC-M, Swedish and Norwegian).
For these codes, if the cone angle is left blank, the maximum value of the SIF (2.0) is used.
For all other piping codes in CAEPIPE, the cone angle is not used.
SIF Calculation
For B31.1, B31.9, ASME Section III Class 2, EN 13480, RCC-M, Swedish and Norwegian,
as mentioned above, the cone angle (if input) is used to calculate the SIF. If the cone angle is
not input, the maximum value of the SIF (2.0) is used.
For Swedish and Norwegian piping codes, additional input is required which affects the
calculation of SIF.
Knuckles
If the reducer is with knuckles,check this box.
257
Reducer
Delta
If the reducer is without knuckles, specify delta, which is the mismatch (difference in mean
radii across the weld at the smaller end of the reducer). If the reducer is with knuckles, delta
is not used.
For other codes, if the code is not specific about a reducer’s SIF, then a value of 1.0 is used.
Weight, Stiffness and Stress Calculation
The properties such as weight of the reducer, stiffness, contents weight and insulation weight
are based on the average diameter (of OD1 and OD2) and average thickness (of Thk1 and
Thk2).
The stresses at each end, however, are calculated using the actual dimensions at each end.
Example 1: Concentric Reducer
To model a concentric reducer as shown in Figure (a) earlier in the topicwith the data:
8”x4” reducer, OD1=8.625”, Thk1=0.322”, OD2=4.5”, Thk2=0.237”.
Create two sections, 8”/STD and 4”/STD.
 The first node (10) is already defined. Press Enter to move to the next row.
 Complete pipe run till node 20: type 20 for Node, type 1 (ft.) for DX, enter material,
8” section and load names, press Enter.
 Input reducer: Type 30 for Node, press Tab to move to the Type field, type “Re” (to
open the Reducer dialog box, note that by default CAEPIPE displays the preceding
8” section’s properties for OD1 and Thk1).
Press the “Section 2” button to select the section at “To” node.
258
Reducer
Highlight the 4” section and press OK. 4.5” for OD2 and 0.237” for Thk2 will be entered in
the Reducer dialog.
Press OK, type 11” for DX (reducer’s length) and press Enter. Now you are asked if you
want to change section, press Yes. “Select Section” dialog will be shown. Highlight the 4”
section and press Enter. Press Enter again on Layout to move to the next row.
Type 40 for Node, 1 (ft.) for DX, press Enter.
Example 2: Eccentric Reducer

To model an eccentric reducer (as shown in Figure (b) earlier in the topic) with the following
data: 8”x6” reducer, OD1=8.625”, Thk1=0.322”, OD2=6.625”, Thk2=0.28”, eccentricity =
(ID1–ID2) / 2 = 0.958” which is modeled as change in elevation.
Create two sections, 8”/STD and 6”/STD.
259
Reducer
 The first node (10) is already defined. Press Enter to move to the next row.
 Complete pipe run till node 20: type 20 for Node, type 12” for DX, enter material, 8”
section and load names, press Enter.
 Input reducer: Type 30 for Node, press Tab to move to the Type field, type “Re” (to
open the Reducer dialog box, note that the preceding 8” section properties are
already displayed for OD1 and Thk1).
Press the “Section 2” button to select the section at “To” node.
Highlight the 6” section and press OK.

6.625” for OD2 and 0.28” for Thk2 will be entered in the Reducer dialog.
Press OK, type 11” for DX (reducer’s length along X axis), -0.958” for DY (this is the
eccentricity), then press Enter. Now you are asked if you want to change section, press
Yes. “Select Section” dialog will be shown. Highlight the 6” section and press Enter. Press
Enter again on Layout to move to the next row.
 Type 40 for Node, 12” for DX, press Enter.
260
Reducer
Example 3: Jacketed Reducer

A jacketed reducer may be modeled in the following manner: Calculate the averages of OD1
and OD2, and of Thk1 and Thk2 for the two reducers, one for the core pipe and the other
for the jacket pipe. Create two new pipe sections with these averages as OD and Thickness.
Insert two pipes at the location of that jacketed reducer, one for the core pipe and the other
for the jacket pipe, having the corresponding section with average OD and thickness. You
may have to input SIF at the two ends of each reducer using code guidelines for a reducer.
Alternate Method: In a jacketed piping system, e.g., 10-20 is a JPIPE, model the reducer for
the core pipe next as you would normally do in a non-jacketed system, i.e., 20-30 is the core
pipe reducer, followed by a JPIPE between 30-40. Then, connect the jacket nodes 20J and
30J with a jacket reducer (shown below) or a JPIPE (jacket pipe). In other words, on new
rows, connect 20J to 30J with the jacket reducer or a JPIPE with average OD and thickness
as calculated above.
261
Reducer
262
Refine Nodal Mesh
Refinement of Nodal Mesh based on Mass Modeling Frequency

The purpose of this feature is to ensure that there are a sufficient number of mass points for
an accurate dynamic model for the dynamic loading under consideration.
Intermediate mass points along a span are generated based on the free vibration of an
equivalent simply supported beam. Optimum element length is calculated from:
1 𝜋 𝐸. 𝐼. 𝑔
𝐿𝑜𝑝𝑡 = √ . 4√
2 2. 𝑓 𝑤
Lopt = Optimum length required to capture the span dynamic behavior
f = mass modeling frequency
g = acceleration due to gravity
E = modulus of elasticity of pipe material
Although the above equation is valid for any temperature, to generate intermediate nodes, E
is taken at the reference temperature entered in CAEPIPE.
I = moment of inertia of pipe cross section
w = weight per unit length of pipe (including insulation, lining and content)
Intermediate mass points can be automatically generated in CAEPIPE by selecting the radio
button “Dynamic Analysis” through Layout window > Edit > Refine Nodal Mesh. Enter the
Mass modeling Frequency in the dialog box shown and press the button “OK”. See figures
shown below for details.
263
Refine Nodal Mesh
While refining the Nodal Mesh, the new node numbers will be generated by adding the node
increment specified through Layout window > Options > Node increment to get the new
node numbers (without affecting the original node numbers used in the Layout window).
Hence, set the node increment value as required before refining the Nodal Mesh.
Upon refining the Nodal Mesh based on Mass modeling frequency, CAEPIPE will prompt
for renumbering of nodes as shown below.
Press the button “Yes” to renumber the nodes and enter the details required by CAEPIPE in
the dialog box. See snap shot shown above.
264
Refine Nodal Mesh
Example
A sample CAEPIPE model (with graphics and layout details as shown below) was chosen for
verification of implementation. Modal analysis was then performed by defining the cut-off
frequency and number of modes as “110 Hz” and 175 respectively in the CAEPIPE model
through Layout window > Options > Analysis > Dynamics with the node points as defined
by the stress analyst.
Wireframe layout without addition of mass points
Rendered layout without addition of mass points

265
Refine Nodal Mesh
Caepipe dynamic_model_original Page 1

Version 8.00 May 30,2018
-----------------------------------------------------------------------------------
Options
-----------------------------------------------------------------------------------
Piping code = B31.3 (2016)
Do not use liberal allowable stresses
Include axial force in stress calculations
Reference temperature = 40 (F)
Number of thermal cycles = 7000
Number of thermal loads = 1
Thermal = Operating - Sustained
Use modulus at reference temperature
Include hanger stiffness
Include Bourdon effect
Use pressure correction for bends
Pressure stress = PD / 4t
Peak pressure factor = 1.00
Cut off frequency = 110 Hz
Number of modes = 175
Include missing mass correction
Use friction in dynamic analysis
Vertical direction = Y
-----------------------------------------------------------------------------------
# Node Type DX(ft'in") DY(ft'in") DZ(ft'in") Mat Sec Load Data
-----------------------------------------------------------------------------------
1 Title =
2 10 From Anchor
3 20 Bend 11'0" API 54I 54I
4 30 -20'0" API 54I 54I
5 40 Bend -10'9" API 54I 54I
6 50 9'0" API 54I 54I Y restraint
7 60 2'0" API 54O 54O
8 65 10'0" API 54O 54O
9 70 12'0" API 54O 54O Y restraint
10 75 12'0" API 54O 54O
12 85 12'0" API 54O 54O
14 100 11'0" API 54O 54O
15 110 3'0" API 54O 54O Anchor
16 120 3'0" API 54O 54O
17 130 Reducer 3'0" API 54O 54O
18 140 Bend 3'8" API 36I 36I
19 150 Bend 2'11" -4'6" 4'6" API 36I 36I
21 165 7'0" API 36I 36I
23 180 Bend 6'0" API 36I 36I
26 210 12'0" API 36I 36I
28 230 Bend 4'5" API 36I 36I
30 250 12'0" API 36I 36I
32 270 12'0" API 36I 36I
33 280 12'0" API 36I 36I XY restraint
266
Refine Nodal Mesh
-----------------------------------------------------------------------------------
# Node Type DX(ft'in") DY(ft'in") DZ(ft'in") Mat Sec Load Data
-----------------------------------------------------------------------------------
34 290 12'0" API 36I 36I
36 310 12'0" API 36I 36I
39 340 12'0" API 36I 36I
41 360 12'0" API 36I 36I
43 380 12'0" API 36I 36I
45 400 12'0" API 36I 36I
47 420 12'0" API 36I 36I
49 440 12'0" API 36I 36I
51 460 12'0" API 36I 36I
53 480 12'0" API 36I 36I
55 500 12'0" API 36I 36I
56 510 12'0" API 36I 36I Anchor
57 520 12'0" API 36I 36I
59 540 12'0" API 36I 36I
61 560 Bend 4'5" API 36I 36I
62 570 -4'5" API 36I 36I Y restraint
63 580 -12'0" API 36I 36I
65 600 -12'0" API 36I 36I
66 610 -12'0" API 36I 36I XY restraint
67 620 -12'0" API 36I 36I
68 630 -12'0" API 36I 36I Anchor
69 640 -12'0" API 36I 36I
70 650 -12'0" API 36I 36I XY restraint
71 660 -12'0" API 36I 36I
74 690 Bend -9'0" API 36I 36I
76 710 12'0" API 36I 36I Anchor
77 100 From
78 720 Bend 10'9" API 36O 36O
80 740 Bend 9'0" API 36O 36O
81 750 16'9" API 36O 36O
82 760 26'11" API 36O 36O
83 770 Bend 5'8" API 36O 36O
85 790 12'0" API 36O 36O Anchor
-----------------------------------------------------------------------------------
267
Refine Nodal Mesh
-----------------------------------------------------------------------------------
Anchors
-----------------------------------------------------------------------------------
(lb/inch) (in-lb/deg) Releases Anchor
Node KX/kx KY/ky KZ/kz KXX/kxx KYY/kyy KZZ/kzz X Y Z XXYYZZ In Pipe
-----------------------------------------------------------------------------------
10 Rigid Rigid Rigid Rigid Rigid Rigid GCS
-----------------------------------------------------------------------------------
Bends
-----------------------------------------------------------------------------------
Bend Radius Thickness Bend Flex. Int. Angle Int. Angle
Node (inch) (inch) Matl Factor Node (deg) Node (deg)
-----------------------------------------------------------------------------------
20 81 L
40 81 L
140 54 L
150 54 L
180 54 L
230 48 U
560 48 U
690 48 U
720 54 L
740 54 L
770 54 L
-----------------------------------------------------------------------------------
Reducers
-----------------------------------------------------------------------------------
OD1 Thk1 OD2 Thk2 Cone Angle Knuc Delta
From To (inch) (inch) (inch) (inch) (deg) kles (inch)
-----------------------------------------------------------------------------------
120 130 54 0.375 36 0.375
-----------------------------------------------------------------------------------
Restraints
-----------------------------------------------------------------------------------
Node X Y Z
-----------------------------------------------------------------------------------
50 Yes
70 Yes
80 Yes
90 Yes
160 Yes
170 Yes
190 Yes
200 Yes
220 Yes
240 Yes
260 Yes
280 Yes Yes
300 Yes
320 Yes
330 Yes Yes
350 Yes
268
Refine Nodal Mesh
-----------------------------------------------------------------------------------
Restraints
-----------------------------------------------------------------------------------
Node X Y Z
-----------------------------------------------------------------------------------
370 Yes
390 Yes Yes
410 Yes
430 Yes
450 Yes Yes
470 Yes
490 Yes
530 Yes
550 Yes
570 Yes
590 Yes
610 Yes Yes
650 Yes Yes
670 Yes
680 Yes
700 Yes
730 Yes
780 Yes
-----------------------------------------------------------------------------------
Pipe material API: API 5L Grade B
-----------------------------------------------------------------------------------
Density = 0.283 (lb/in3), Nu = 0.300, Joint factor = 1.00, Type = CS
Yield strength = 35000 (psi)
Temp E Alpha Allowable

(F) (psi) (in/in/F) (psi)
-------------------------------------------
-325 31.4E+6 5.00E-6 20000
-200 30.8E+6 5.35E-6 20000
-100 30.2E+6 5.65E-6 20000
70 29.5E+6 6.07E-6 20000
200 28.8E+6 6.38E-6 20000
300 28.3E+6 6.60E-6 20000
400 27.7E+6 6.82E-6 19900
500 27.3E+6 7.02E-6 19000
600 26.7E+6 7.23E-6 17900
650 26.1E+6 7.33E-6 17300
700 25.5E+6 7.44E-6 16700
750 24.8E+6 7.54E-6 13900
800 24.2E+6 7.65E-6 11400
850 23.3E+6 7.75E-6 8700
900 22.4E+6 7.84E-6 5900
950 21.4E+6 7.91E-6 4000
1000 20.4E+6 7.97E-6 2500
1050 19.2E+6 8.05E-6 1600
1100 18.0E+6 8.12E-6 1000
-----------------------------------------------------------------------------------
269
Refine Nodal Mesh
-----------------------------------------------------------------------------------
Pipe Sections
-----------------------------------------------------------------------------------
Nominal O.D. Thk Cor.Al M.Tol Ins.Dens Ins.Th Lin.Dens Lin.Th
Name Dia. Sch (inch) (inch) (inch) (%) (lb/ft3) (inch) (lb/ft3) (inch)
-----------------------------------------------------------------------------------
36I 36" STD 36 0.375 0 0.0
36O 36" STD 36 0.375 0 0.0
54O Non Std 54 0.375 0 0.0
54I Non Std 54 0.375 0 0.0
-----------------------------------------------------------------------------------
Loads
-----------------------------------------------------------------------------------
Static seismic load: X = -0.20, Y = -0.20, Z = -0.20 (g's)

Acceleration load combination = Square Root of Sum of Squares
X spectrum: Malta, NY_b Factor = 1.0000

Y spectrum: Malta, NY_b Factor = 0.0430
Z spectrum: Malta, NY_b Factor = 1.0000
Mode sum = SRSS Direction sum = SRSS
-----------------------------------------------------------------------------------
Wind Load 1
-----------------------------------------------------------------------------------
Shape factor = 0.60
Wind direction: X comp = 0.000, Y comp = 0.000, Z comp = 1.000
Elevation Pressure
(feet) (psf)
0 15
15 15
30 15
45 15
60 15
-----------------------------------------------------------------------------------
Pipe Loads
-----------------------------------------------------------------------------------
Load T1 P1 T2 P2 T3 P3 Specific Add.Wgt Wind
Name (F) (psi) (F) (psi) (F) (psi) gravity (lb/ft) Load
-----------------------------------------------------------------------------------
36O 100 125 Y
36I 100 125
54O 100 125 Y
54I 100 125
-----------------------------------------------------------------------------------
From the Modal Analysis results shown below, it was noted that the CAEPIPE was able to
extract 92 modes with highest frequency being 95.967 Hz.
270
Refine Nodal Mesh
-----------------------------------------------------------------------------------
Frequency Period Participation factors Modal mass / Total mass
Mode (Hz) (sec) X Y Z X Y Z
-----------------------------------------------------------------------------------
1 1.875 0.5335 -5.7615 -0.0004 0.5255 0.0979 0.0000 0.0008
2 3.288 0.3041 -6.7721 -0.0000 0.0003 0.1353 0.0000 0.0000
3 5.971 0.1675 -4.1919 0.0000 -0.0001 0.0518 0.0000 0.0000
4 6.147 0.1627 0.0909 -0.0000 -0.1021 0.0000 0.0000 0.0000
5 7.800 0.1282 -0.4240 -0.0000 4.5032 0.0005 0.0000 0.0598
6 8.102 0.1234 0.0029 0.0001 0.3321 0.0000 0.0000 0.0003
7 9.933 0.1007 0.0861 -0.0000 -4.2655 0.0000 0.0000 0.0537
8 10.146 0.0986 -0.6501 0.0002 1.6688 0.0012 0.0000 0.0082
9 10.241 0.0976 -2.1689 -0.0000 0.0048 0.0139 0.0000 0.0000
10 11.279 0.0887 3.2684 -0.0002 -7.9718 0.0315 0.0000 0.1874
11 11.628 0.0860 6.7199 -0.0002 5.8034 0.1332 0.0000 0.0993
12 13.116 0.0762 -0.0013 1.4576 -3.8221 0.0000 0.0063 0.0431
13 13.276 0.0753 2.6364 0.0000 -0.0001 0.0205 0.0000 0.0000
14 13.931 0.0718 3.3527 0.0030 -3.2879 0.0332 0.0000 0.0319
15 16.027 0.0624 -3.6866 0.0000 -0.3181 0.0401 0.0000 0.0003
16 16.038 0.0624 -2.9281 -0.0001 -0.7810 0.0253 0.0000 0.0018
17 17.802 0.0562 -1.8832 0.0179 2.3004 0.0105 0.0000 0.0156
18 21.780 0.0459 0.0000 2.7930 3.3879 0.0000 0.0230 0.0339
19 22.412 0.0446 -0.0006 -3.2159 -1.3274 0.0000 0.0305 0.0052
20 22.683 0.0441 -0.3117 -0.0005 0.0747 0.0003 0.0000 0.0000
21 24.003 0.0417 -0.5955 -0.0000 0.1696 0.0010 0.0000 0.0001
22 24.910 0.0401 2.7763 -0.0000 -0.0070 0.0227 0.0000 0.0000
23 26.421 0.0378 -0.2314 0.0001 0.1348 0.0002 0.0000 0.0001
24 27.698 0.0361 -1.3316 0.0000 -0.0002 0.0052 0.0000 0.0000
25 29.713 0.0337 -1.2585 -0.0000 -0.0049 0.0047 0.0000 0.0000
26 30.079 0.0332 -0.3437 0.0000 -0.1955 0.0003 0.0000 0.0001
27 30.146 0.0332 0.0215 -0.0001 0.3494 0.0000 0.0000 0.0004
28 31.946 0.0313 -0.6062 0.0012 -1.6290 0.0011 0.0000 0.0078
29 33.670 0.0297 0.9928 -0.0040 2.8297 0.0029 0.0000 0.0236
30 36.072 0.0277 0.0014 1.4997 -2.1277 0.0000 0.0066 0.0134
31 37.812 0.0264 -0.4989 -0.0074 2.1572 0.0007 0.0000 0.0137
32 39.097 0.0256 1.2371 -0.0000 -0.0086 0.0045 0.0000 0.0000
33 41.172 0.0243 -0.0001 1.2001 -6.2487 0.0000 0.0042 0.1152
34 41.250 0.0242 -0.3414 -0.0000 1.6185 0.0003 0.0000 0.0077
35 42.488 0.0235 0.7544 -0.0001 -0.0047 0.0017 0.0000 0.0000
36 43.804 0.0228 -0.4830 0.0108 1.2389 0.0007 0.0000 0.0045
37 46.065 0.0217 -0.0011 0.0686 -0.0001 0.0000 0.0000 0.0000
38 46.766 0.0214 -0.0098 -0.5622 -0.0023 0.0000 0.0009 0.0000
39 47.504 0.0211 -0.2337 -0.0029 -0.2475 0.0002 0.0000 0.0002
40 49.063 0.0204 0.0015 0.3511 0.0007 0.0000 0.0004 0.0000
41 49.295 0.0203 -0.0000 -1.0018 -0.0000 0.0000 0.0030 0.0000
42 49.985 0.0200 1.3732 -0.0000 -1.3809 0.0056 0.0000 0.0056
43 50.052 0.0200 -0.1938 0.0000 -1.9635 0.0001 0.0000 0.0114
44 52.178 0.0192 -0.9811 -0.0281 -0.6345 0.0028 0.0000 0.0012
45 52.335 0.0191 0.0000 -0.8918 -0.0000 0.0000 0.0023 0.0000
46 53.313 0.0188 -0.0715 -1.0283 -0.1169 0.0000 0.0031 0.0000
47 53.496 0.0187 -1.0974 -0.0751 -1.8548 0.0036 0.0000 0.0101
48 54.481 0.0184 0.0407 -1.8917 0.0845 0.0000 0.0106 0.0000
49 56.408 0.0177 0.4159 -1.3502 0.0358 0.0005 0.0054 0.0000
50 56.457 0.0177 2.6140 0.2176 -0.1253 0.0202 0.0001 0.0000
51 56.920 0.0176 1.8036 0.0012 0.0177 0.0096 0.0000 0.0000
52 57.929 0.0173 -0.2764 -0.0033 -0.0311 0.0002 0.0000 0.0000
53 58.372 0.0171 -1.1204 0.0366 0.3623 0.0037 0.0000 0.0004
271
Refine Nodal Mesh
-----------------------------------------------------------------------------------
-----------------------------------------------------------------------------------
54 58.736 0.0170 0.0013 -0.4068 0.0005 0.0000 0.0005 0.0000
55 62.338 0.0160 -0.0004 1.7473 2.3502 0.0000 0.0090 0.0163
56 63.166 0.0158 1.4680 0.0042 1.3132 0.0064 0.0000 0.0051
57 63.180 0.0158 1.1531 0.0001 -1.1840 0.0039 0.0000 0.0041
58 63.512 0.0157 0.0000 -0.1612 0.0000 0.0000 0.0001 0.0000
59 64.003 0.0156 -0.0020 -0.5988 0.0007 0.0000 0.0011 0.0000
60 64.339 0.0155 -0.0000 -1.1254 0.0000 0.0000 0.0037 0.0000
61 65.107 0.0154 -1.0551 -0.0210 1.1888 0.0033 0.0000 0.0042
62 65.192 0.0153 0.0016 0.3043 0.5961 0.0000 0.0003 0.0010
63 65.893 0.0152 -0.0078 -0.0572 0.0105 0.0000 0.0000 0.0000
64 69.245 0.0144 -0.0005 -0.9214 0.0006 0.0000 0.0025 0.0000
65 70.044 0.0143 -2.3140 -0.0000 -1.0324 0.0158 0.0000 0.0031
66 70.911 0.0141 -3.2315 -0.0000 -1.9392 0.0308 0.0000 0.0111
67 73.355 0.0136 -0.9182 0.2454 0.2391 0.0025 0.0002 0.0002
68 73.445 0.0136 -0.0085 1.5999 0.0018 0.0000 0.0075 0.0000
69 74.015 0.0135 0.0021 1.9142 0.0703 0.0000 0.0108 0.0000
70 74.156 0.0135 -0.0000 2.9214 -0.0000 0.0000 0.0252 0.0000
71 74.905 0.0134 0.0017 -3.8949 0.0017 0.0000 0.0447 0.0000
72 76.249 0.0131 -0.1108 0.0489 0.0287 0.0000 0.0000 0.0000
73 76.553 0.0131 0.0001 1.4453 0.0002 0.0000 0.0062 0.0000
74 76.599 0.0131 0.0000 3.2442 0.0000 0.0000 0.0310 0.0000
75 78.151 0.0128 -0.0001 -5.7802 -0.0001 0.0000 0.0985 0.0000
76 78.399 0.0128 0.1428 -1.8259 0.2005 0.0001 0.0098 0.0001
77 79.572 0.0126 -0.0000 -3.1388 -0.0000 0.0000 0.0291 0.0000
78 80.352 0.0124 -0.0611 -0.0303 0.0321 0.0000 0.0000 0.0000
79 80.590 0.0124 -0.4718 0.0055 -0.0030 0.0007 0.0000 0.0000
80 81.366 0.0123 0.0014 -1.7246 1.7171 0.0000 0.0088 0.0087
81 81.537 0.0123 -0.6787 0.0000 2.6001 0.0014 0.0000 0.0199
82 83.538 0.0120 2.6049 -0.0000 0.2580 0.0200 0.0000 0.0002
83 83.555 0.0120 0.0186 0.0064 -0.0224 0.0000 0.0000 0.0000
84 85.143 0.0117 -0.9456 0.0093 -0.0136 0.0026 0.0000 0.0000
85 86.260 0.0116 -0.0033 -0.7082 1.6260 0.0000 0.0015 0.0078
86 86.553 0.0116 0.1187 -0.0708 1.3196 0.0000 0.0000 0.0051
87 87.389 0.0114 -0.0038 -0.2207 -0.7108 0.0000 0.0001 0.0015
88 87.678 0.0114 1.8995 -0.0000 -1.1843 0.0106 0.0000 0.0041
89 90.219 0.0111 -0.3088 -0.2777 -0.2039 0.0003 0.0002 0.0001
90 93.121 0.0107 -0.5735 -0.1326 0.4279 0.0010 0.0001 0.0005
91 95.224 0.0105 0.0000 4.5908 0.5409 0.0000 0.0622 0.0009
92 95.967 0.0104 0.7252 -0.0000 -0.6981 0.0016 0.0000 0.0014
-------------------------------
Total 0.7887 0.4496 0.8526
----------------------------------------------------------------------------------
Now, the original model shown above was refined through the feature Layout window >
Edit > Refine Nodal Mesh. The Mass modeling frequency was set to 110 Hz for refining the
mesh for “Dynamic Analysis” as shown below.
The resulting refined model with additional mass points added by CAEPIPE is shown in the
snap shots below.
272
Refine Nodal Mesh
273
Refine Nodal Mesh
274
Refine Nodal Mesh
Modal analysis was then performed using CAEPIPE for the refined model (with additional
mass points automatically added) for mass modeling frequency 110 Hz and found that the
CAEPIPE was able to extract all the modes below the cut-off frequency 110 Hz specified in
the analysis options. Please see the modal analysis results obtained from CAEPIPE shown
below.
Caepipe dynamic_model_refined Page 1

-----------------------------------------------------------------------------------
-----------------------------------------------------------------------------------
1 1.755 0.5698 -5.7696 -0.0003 0.5155 0.0982 0.0000 0.0008
2 2.855 0.3502 6.7662 0.0000 -0.0002 0.1350 0.0000 0.0000
3 5.521 0.1811 -4.1344 0.0000 -0.0001 0.0504 0.0000 0.0000
4 6.151 0.1626 0.0878 -0.0000 -0.1088 0.0000 0.0000 0.0000
5 7.722 0.1295 0.4309 0.0000 -4.5502 0.0005 0.0000 0.0611
6 8.119 0.1232 0.0315 0.0001 0.3840 0.0000 0.0000 0.0004
7 9.824 0.1018 0.0839 -0.0000 -4.3165 0.0000 0.0000 0.0550
8 10.063 0.0994 2.3014 0.0000 0.0054 0.0156 0.0000 0.0000
9 10.134 0.0987 -0.6185 0.0003 2.3174 0.0011 0.0000 0.0158
10 11.111 0.0900 2.0288 0.0003 -8.3004 0.0121 0.0000 0.2032
11 11.613 0.0861 -7.1492 0.0006 -4.4135 0.1508 0.0000 0.0575
12 12.635 0.0791 -0.0006 1.5513 -3.8503 0.0000 0.0071 0.0437
13 13.233 0.0756 2.6877 0.0000 -0.0007 0.0213 0.0000 0.0000
14 13.640 0.0733 2.9690 0.0034 -4.0182 0.0260 0.0000 0.0476
15 15.725 0.0636 3.7891 -0.0000 0.3498 0.0423 0.0000 0.0004
16 15.805 0.0633 3.1354 0.0024 0.9356 0.0290 0.0000 0.0026
17 16.409 0.0609 2.2658 -0.0169 -2.5046 0.0151 0.0000 0.0185
18 17.349 0.0576 -0.0000 -2.7191 -3.1311 0.0000 0.0218 0.0289
19 20.661 0.0484 0.0007 3.2479 1.4079 0.0000 0.0311 0.0058
20 22.737 0.0440 0.2479 0.0001 -0.0877 0.0002 0.0000 0.0000
21 23.733 0.0421 0.7055 -0.0000 -0.0963 0.0015 0.0000 0.0000
22 23.782 0.0420 -2.9878 0.0000 0.0053 0.0263 0.0000 0.0000
23 26.507 0.0377 0.1027 0.0005 -0.1910 0.0000 0.0000 0.0001
24 27.187 0.0368 -0.6336 0.0000 -0.0045 0.0012 0.0000 0.0000
25 29.021 0.0345 -1.1191 0.0000 0.0002 0.0037 0.0000 0.0000
26 29.061 0.0344 0.3323 -0.0000 0.1679 0.0003 0.0000 0.0001
27 30.289 0.0330 -0.4261 0.0036 -1.0079 0.0005 0.0000 0.0030
28 31.215 0.0320 1.0457 -0.0087 2.3123 0.0032 0.0000 0.0158
29 33.583 0.0298 0.4478 -0.0064 1.5750 0.0006 0.0000 0.0073
30 36.362 0.0275 -1.1690 0.0000 -0.0052 0.0040 0.0000 0.0000
31 37.505 0.0267 -0.0033 1.4151 -2.1430 0.0000 0.0059 0.0135
32 37.885 0.0264 0.4183 0.0121 -2.4989 0.0005 0.0000 0.0184
33 40.194 0.0249 0.1951 -0.0000 -1.5005 0.0001 0.0000 0.0066
34 40.968 0.0244 0.0001 -1.1979 6.4128 0.0000 0.0042 0.1213
35 43.038 0.0232 0.3511 -0.0019 -1.2983 0.0004 0.0000 0.0050
36 43.372 0.0231 -0.4795 0.0000 -0.0105 0.0007 0.0000 0.0000
37 46.870 0.0213 0.0021 -0.0008 0.0010 0.0000 0.0000 0.0000
38 47.541 0.0210 -0.0144 -0.7462 -0.0080 0.0000 0.0016 0.0000
39 48.377 0.0207 -0.2963 -0.0021 -0.3487 0.0003 0.0000 0.0004
40 50.010 0.0200 -0.4830 0.0001 -1.8955 0.0007 0.0000 0.0106
41 50.132 0.0199 -0.0043 -0.3850 -0.0040 0.0000 0.0004 0.0000
42 50.285 0.0199 0.0000 1.2526 0.0000 0.0000 0.0046 0.0000
43 51.219 0.0195 1.2828 0.0000 -1.6462 0.0049 0.0000 0.0080
44 51.778 0.0193 1.4654 0.0565 1.8643 0.0063 0.0000 0.0103
45 53.561 0.0187 0.0000 -1.1367 -0.0000 0.0000 0.0038 0.0000
46 53.672 0.0186 -0.1396 0.0329 0.4238 0.0001 0.0000 0.0005
47 54.731 0.0183 0.0027 -1.2178 0.0180 0.0000 0.0044 0.0000
48 56.221 0.0178 -0.0024 -2.0037 0.0429 0.0000 0.0118 0.0000
49 57.051 0.0175 -1.6501 -0.0050 -0.0543 0.0080 0.0000 0.0000
50 57.365 0.0174 -0.2648 -0.0208 0.4221 0.0002 0.0000 0.0005
51 58.161 0.0172 -0.0620 -1.7053 0.0076 0.0000 0.0086 0.0000
52 58.480 0.0171 -2.8218 0.0672 0.2197 0.0235 0.0000 0.0001
53 59.590 0.0168 0.0011 -1.6797 -2.3900 0.0000 0.0083 0.0168
275
Refine Nodal Mesh
Caepipe dynamic_model_refined Page 2
-----------------------------------------------------------------------------------
-----------------------------------------------------------------------------------
54 59.697 0.0168 1.3631 -0.0003 -0.0002 0.0055 0.0000 0.0000
55 60.969 0.0164 -0.0001 1.4746 0.5326 0.0000 0.0064 0.0008
56 61.166 0.0163 -0.0003 0.3240 -0.0009 0.0000 0.0003 0.0000
57 63.791 0.0157 1.1014 -0.0167 1.6815 0.0036 0.0000 0.0083
58 65.476 0.0153 0.0000 -0.0239 -0.0000 0.0000 0.0000 0.0000
59 66.097 0.0151 -2.0302 -0.0513 0.5933 0.0122 0.0000 0.0010
60 66.996 0.0149 0.6629 0.0000 -1.1896 0.0013 0.0000 0.0042
61 67.413 0.0148 0.0051 -0.4949 -0.0031 0.0000 0.0007 0.0000
62 67.702 0.0148 0.0000 -1.4164 -0.0000 0.0000 0.0059 0.0000
63 69.973 0.0143 -0.0065 -0.2969 0.0052 0.0000 0.0003 0.0000
64 70.117 0.0143 0.0002 -2.2385 0.7487 0.0000 0.0148 0.0017
65 70.175 0.0143 -2.4152 -0.0000 -0.9261 0.0172 0.0000 0.0025
66 74.234 0.0135 3.4185 0.0003 1.5639 0.0345 0.0000 0.0072
67 74.560 0.0134 0.0042 1.0701 -0.0017 0.0000 0.0034 0.0000
68 74.843 0.0134 -1.3329 0.8236 0.1534 0.0052 0.0020 0.0001
69 76.927 0.0130 -0.4107 -1.5744 0.2386 0.0005 0.0073 0.0002
70 78.576 0.0127 -0.1023 0.0017 -0.0013 0.0000 0.0000 0.0000
71 80.082 0.0125 0.0006 -0.5746 1.6438 0.0000 0.0010 0.0080
72 80.485 0.0124 0.0002 -2.1498 0.0002 0.0000 0.0136 0.0000
73 81.701 0.0122 -0.2466 -0.0537 0.0757 0.0002 0.0000 0.0000
74 81.749 0.0122 -0.0002 3.4973 0.0000 0.0000 0.0361 0.0000
75 82.787 0.0121 -0.0004 4.3124 -0.0023 0.0000 0.0549 0.0000
76 84.316 0.0119 1.6969 0.0000 -3.1197 0.0085 0.0000 0.0287
77 85.065 0.0118 -0.0026 -0.4712 0.8256 0.0000 0.0007 0.0020
78 85.440 0.0117 0.0000 3.7852 -0.0000 0.0000 0.0423 0.0000
79 85.507 0.0117 -0.0001 1.4996 0.0009 0.0000 0.0066 0.0000
80 87.100 0.0115 -0.0680 0.0853 -1.2077 0.0000 0.0000 0.0043
81 87.358 0.0114 -0.0000 3.5186 -0.0000 0.0000 0.0365 0.0000
82 87.548 0.0114 1.1884 -0.0035 -0.0409 0.0042 0.0000 0.0000
83 87.820 0.0114 0.1975 -0.0174 0.6688 0.0001 0.0000 0.0013
84 88.217 0.0113 -0.0001 6.8293 -0.0003 0.0000 0.1376 0.0000
85 88.587 0.0113 0.0003 -0.6062 -1.1192 0.0000 0.0011 0.0037
86 89.553 0.0112 2.9233 -0.0000 0.1706 0.0252 0.0000 0.0001
87 91.995 0.0109 -0.2639 -0.2144 -0.3238 0.0002 0.0001 0.0003
88 92.948 0.0108 0.1126 0.0508 0.0909 0.0000 0.0000 0.0000
89 93.055 0.0107 1.1698 -0.0041 0.6341 0.0040 0.0000 0.0012
90 102.283 0.0098 0.0000 -0.0302 -0.0000 0.0000 0.0000 0.0000
91 103.208 0.0097 0.0131 5.0475 0.3381 0.0000 0.0751 0.0003
92 103.322 0.0097 -0.6291 -0.0242 0.3139 0.0012 0.0000 0.0003
93 109.288 0.0092 -0.7946 -0.1550 0.6055 0.0019 0.0001 0.0011
94 110.299 0.0091 1.2405 -0.0000 0.3050 0.0045 0.0000 0.0003
-------------------------------
Total 0.8148 0.5606 0.8574
276
Relief Valve Load Analysis
General
During an overpressure event, the discharge of a PRV imposes a load, referred to as a
reaction force, on the collective installation. The flowrate and associated reaction force
increase from nominally zero to some value, remain relatively constant at that value for the
duration of the release, and then decrease to zero again, i.e., when the relief valve opens, the
discharge fluid creates a jet force that acts on the piping system. This force increases from
zero to its full value over a time frame similar to the opening time of the valve. The relief
valve remains open until sufficient fluid is vented to relieve the overpressure situation. As the
valve closes, the reduction in flow reduces the jet force to zero.
Simplified Analysis Approach
American Petroleum Institute’s API 520, Part II (1994), provides a basis for calculation of
the reaction force in the event of a vapor or a two-phase release directly to the atmosphere.
There is no discussion in this section of API 520, Part II, about the reaction force developed
during a liquid release. Furthermore, no guidance is presented with respect to applying these
results or determining if an installation is acceptable; instead, the burden is placed on the
designer to ensure that the installation is appropriately designed. While this may be
reasonable for the design of new facilities, evaluating the adequacy of existing facilities
becomes much more complicated.
The formula (section 2.4.1.1) in US Customary units from API 520, Part II (1994), for vapor
relief devices discharging to the atmosphere, is shown below:
W kT
F  ( AP )
366 (k  1) M
where,
F = Reaction force at the point of discharge to the atmosphere,(lbf.)
k = Ratio of specific heats (CP/CV) at the outlet conditions
W = Flow rate of any gas or vapor, pound mass (lbm.)/hr
CP = Specific heat at constant pressure
CV = Specific heat at constant volume
T = Temperature at the outlet, °R
M = Molecular weight of the process fluid
A = Area of the outlet at the point of discharge, in2
P = Static pressure within the outlet at the point of discharge, psig
Using the reaction force computed from the above formula along with the following PRV
parameters, namely
 Valve Opening Time,
 Valve Closing Time and
 Relief duration (all obtained from the PRV manufacturer),
One can generate a PRV load profile and apply it in CAEPIPE to perform a simplified
analysis.
277
Detailed Analysis Approach

Section 2.4.2 of API 520, Part II (1994) also states the following.
“Pressure relief devices that relieve under steady-state flow conditions into a closed system
usually do not create large forces and bending moments on the exhaust system. Only at
points of sudden expansion will there be any significant reaction forces to be calculated.
Closed discharge systems, however, do not lend themselves to simplified analytical
techniques. A complex time history analysis of the piping system may be required to obtain
the true values of the reaction forces and associated moments.”
Such complex time history analysis of the piping system can be carried out as follows.
 Perform a fluid transient analysis on the piping system using a fluid dynamics
software tool such as, “PipeNet”, “RELAP”, "ROLAST", etc.
 Apply the resulting output obtained (forces as a function of frequency) at the bend
node after the relief valve in a pipe stress analysis software (CAEPIPE).
 Compute forces, moments and stresses in the piping system due to this loading.
As one can see, this method is detailed, time consuming and expensive and hence, not
covered here.
Example 1:
Step 1:
By assuming the following data, one can apply the relief valve loading in CAEPIPE. Please
see the model below for details.
1. Reaction force (F) computed using the formula above = 6854 lb.
2. Relief Valve Opening time = 8 ms (milliseconds)
3. Relief Valve Closing time = 8 ms
4. Relief duration = 1 s
6. Pressure = 475 psig
7. Temperature = 51°F
The steps followed in generating the model are given below.
278
279
280
Step 2:
After creating your piping model (with node 75 being the center node of the discharge bend
where the PRV reaction force will be applied),
Select “Relief valve loading” from CAEPIPE Layout window > Misc and enter the data in
the dialog box as shown in the figure below.
281
Step 3:
After entering the data as shown in the dialog above, press the button “OK”. Using the
above input values for Relief Valve Loading, CAEPIPE internally generates a time-history
loading function, which is then applied on a single degree-freedom spring-mass system with
each intermediate frequency (between 0.0 Hz and the maximum frequency) to generate the
“Force Spectrum Load” shown below.
282
Step 4:
Apply the Force Spectrum Load thus generated at the bend center node 75 after the relief
valve in downward direction (-FY by specifying negative Scale Factor) as shown below.
Step 5:
Check “Force Spectrum” for analysis through Layout window > Load cases. Click on OK.
283
Step 6:
Save and Analyze the model. After analysis, CAEPIPE displays Occasional stresses which
include the effects of the PRV load.
Step 7:
Another load case called “Force Spectrum” will be available for which you can study
displacements, support loads, support load summary (for sizing supports), etc.
284
285
Restraint
This support type is a convenient way to specify a translational two-way rigid restraint in the
global X, Y and Z directions.
A restraint is input by typing “re” in the Data column or selecting “Restraint” from the Data
Types dialog. Alternately, simply typing “X” or “Y” or “Z” in the data type field inputs a
restraint in the respective direction and moves the cursor to the next row.
The Restraint dialog is shown.
Use the check boxes to apply the restraint in a particular direction (both ways). Click on the
vertical button for a rigid vertical restraint. All three directions may be checked too.
12
Rigid restraint has a stiffness of 1×10 (lb/in.).
286
Rigid Element
Use this element to model any “stiff” (relative to pipe) inline component.
12 12
The stiffnesses of 1 x 10 (lb/inch) in translational directions (axial and shear), and 1 x 10
(inch-lb./rad.) in rotational directions (bending and torsional) are used.
A rigid element is input by typing “ri” in the Type column or selecting “Rigid element” from
the Element Types dialog.
The Rigid element dialog is shown.
Weight
The required input is weight. It is applied as a distributed load along the length of the rigid
element. To this empty weight, the Additional weightspecified under Load column is added
(to include weight of snow etc.).
Weight is to be input in lbf or kgf and NOT in mass units. Whenever mass is required for a
calculation as in the case of forming Mass matrix for dynamic analysis, or in calculating
For Sustained load case analysis, weights of content, insulation and lining (as calculated using
insulation thickness and its density as well as lining thickness and density) are added
internally in CAEPIPE when the option “Add Content, Insulation and Lining weights
(CIL)” is turned ON.
For the Empty Weight load case analysis, only weights of insulation and lining will be
included, while the weight of content will be excludedwhen the option “Add Content,
Insulation and Lining weights (CIL)” is turned ON.
Material density does not affect the weight of the rigid element.
287
Rigid Element
Thermal expansion of the rigid element is calculated using the coefficient of thermal
expansion from the Material column and temperatures from the Load column. Wind load is
calculated using the section properties inclusive of insulation thickness.
Rigid, weightless Elements
These may be needed when you want to account for some hard-to-model element’s thermal
growth, or to connect the center line of a large pipe to its outside surface, again to account
for its (radial) thermal growth/contraction (which impacts the branch line), or to model a
rigid, massless link between two points on the stress model.
To model any of these, input a rigid element, type zero for weight and ensure that the
corresponding Load (specified on the Layout window under the Load column) used for this
element does not have any Additional weight specified.
288
Rod Hanger
A rod hanger is a rigid one-way vertical support. The rod hanger node is rigidly supported
against downward movement but able to move freely in the upward direction. That is, the
rod hanger is rigid in tension (downward movement) and has no stiffness in compression
(upward movement). CAEPIPE considers a rod hanger to always act in the vertical direction.
A rod hanger is input by typing “ro” in the Data column or selecting “Rod Hanger” from the
Data Types dialog.
By default, one rod hanger without a connected node is input. The number of hangers and
the node to which it may be connected to may be specified in the Rod Hanger dialog.
Number of Hangers
The number of hangers is the number of separate rod hangers connected in parallel at this
node.
Connected to Node
By default the rod hanger is connected to a fixed ground point which is not a part of the
piping system. A rod hanger can be connected to another node in the piping system by
entering the other node number in the “Connected to” nodefield. This node must be directly
above the rod hanger node.
A rod hanger in CAEPIPE functions as a vertical limit stop, that is, it functions as a
nonlinear one-way restraint. It is rigid in -Y direction and fully flexible in +Y direction (in a
Y-vertical system). The rod hanger offers no resistance in +Y direction.
Rod hanger results are included in the hanger report, which reports results for the first
operating case (W+P1+T1). In the hanger report, a rod hanger’s spring rate may be shown
either as Rigid or zero, the latter potentially confusing to the user.
289
Rod Hanger
It simply means that there is possibly liftoff at the rod hanger location for the first operating
case. You can confirm this by studying vertical displacement (Y or Z) at the rod hanger for
the first operating case (which will be 0 or positive). If this vertical displacement is zero, it
means the rod hanger is in tension and its sprig rate is shown as Rigid; on the other hand, if
this vertical displacement is positive, then the rod hanger is in compression and its spring
rate is shown as zero. You can find reports for other operating load cases under Support
Loads > Other Supports > Rod Hangers.
Liftoff (i.e., zero spring rate and a positive operating condition displacement) indicates that
the rod hanger may not be needed and hence could be removed. You will need to study the
effect on the system at other supports after removing the rod hanger.
In dynamic analysis, the status of the rod hanger from the first operating case (W+P1+T1) is
used, i.e., if the rod hanger is in tension in the first operating case, a rigid vertical two-way
restraint is used in dynamic analysis. If the rod hanger is in compression in the first operating
case (possible liftoff), no vertical restraint is used at that location in dynamic analysis.
290
Section
A Section denotes the cross-sectional properties of a pipe used to build a piping model. You
may define as many sections as needed. To define each section, you will need properties such
as outside diameter (or Nominal Dia.), thickness of pipe, corrosion allowance, insulation,
inside lining, and a name that is used under the Section column on the Layout window while
building your model.
Click on “Sect” on the Header row or select Sections from the Misc menu.
CAEPIPE presents a List window that lists all defined sections in the model (none defined
yet in the image below). Double click on an empty row to define a new one.
The section dialog is shown.
291
Section
Section name
Type an alpha-numeric name (up to 5 characters long) in this field. Example: If you have
three 8” sections with different schedules, you could name them thus: 8-STD, 8-80 and 8-
80S.
Nominal Dia, Schedule
Four databases of pipe sizes are built-into CAEPIPE — ANSI (American National
Standards Institute, default), DIN (Deutsche Industrie Norm), JIS (Japanese Industrial
Standard) and ISO (International Organization for Standardization).
When you click on the drop-down combo box for pipe sizes, CAEPIPE shows the list of
pipe sizes that pertains to the selected database (ANSI, DIN, JIS or ISO). ANSI pipe sizes
range from 1/8” to 48”, DIN from 15 to 1600, JIS from 8A to 1500A, and ISO from 15 to
1000. Select the required nominal pipe size and schedule (wall thickness). To change to a
different database (JIS, DIN or ISO), click on the appropriate radio button. On selection,
CAEPIPE populates the correct OD and Thickness.
For pipe sizes you do not see on the list, each database allows you a nonstandard definition
(“Non std” in the pipe sizes list) too. In other words, you are not restricted only to the
choices available in the databases. You may define any size and thickness as needed. You will
need to enter the Outside diameter and Thickness of such a nonstandard pipe, in addition to
the other parameters.
Corrosion Allowance
The corrosion allowance reduces the wall thickness of the pipe and is used to calculate the
allowable pressure for the pipe section. Additionally, for some piping codes (B31.3, B31.4,
B31.5, B31.8, B31.12, B31.1 (1967), CODETI, Canadian Z183 and Z184), corrosion
allowance is used for reducing the section modulus and cross-sectional area only for
calculating sustained and occasional stresses.
Mill Tolerance
The mill tolerance (in percent) is also used to reduce the wall thickness of the pipe while
calculating allowable pressure. For example, if the mill tolerance is input as 12.5 (%), the pipe
thickness (while calculating allowable pressure) is = 0.875 x nominal thickness.
Reduced thickness = (1 − Mill tolerance/100) × nominal thickness − Corrosion allowance
If defined while modeling, corrosion allowance, mill tolerance, insulation and lining densities
are automatically carried forward while defining a new section.
292
Section
Insulation
Type the pipe insulation density and thickness here. Click on the Insulation button for the
insulation library, or enter your own.
Highlight the desired insulation material and press Enter. The insulation density is entered on
the section property dialog. CAEPIPE uses insulation thickness and density to calculate the
insulation weight which is added to the weight of the pipe. Insulation thickness is also used
to calculate the projected area exposed to wind load(s).
Lining
Lining is used to prevent internal corrosion that might occur during transportation of a gas
or a liquid. CAEPIPE has the ability to model these protective coatings inside the pipe.
Lining is different from insulation. Insulation is around and outside of the pipe. Lining is on
the inside of the pipe. Both have respective thicknesses and densities, which are used to
calculate the respective weight which is then added to the weight of the pipe. See previous
figure.
While calculating the weight of the liquid/gas inside the pipe, CAEPIPE accounts for lining
thickness by reducing the pipe’s internal diameter by twice the lining thickness.
293
Section
Note:
CAEPIPE requires “Weight Density” to be input in lbf/in3 or kgf/m3 and NOT its
“Mass Density” for insulation and lining.
Whenever mass is required for a calculation as in the case of forming Mass matrix for
dynamic analysis, or in calculating inertia force as (mass x acceleration) for static
seismic analysis, CAEPIPE internally computes the mass for each item to be equal to
(weight / g-value).
Using values input for Insulation Thickness, Insulation Density, Lining Thickness
and Lining Density for each pipe section in astress model, CAEPIPE will compute
their weight and include the same in theanalysis. In addition, CAEPIPE considers
Insulationand/or Lining as integral parts of the piping, and that there is no relative
motion between insulation and piping and lining and piping.Accordingly, CAEPIPE
does not account for any friction between pipe and insulation and pipe and lining.
294
Skewed Restraint
A Skewed Restraint is a two-way support that resists translation along or rotation about any
specified direction at a node. You have to use either a manufacturer-supplied stiffness or
calculate it for the support you want to model.
Use this restraint to model sway braces, sway struts and similar supports. You can also use
this to model vertical/horizontal supports, though it is used more commonly to resist lateral
forces.
The figure below shows an application.
A skewed restraint is input by typing “sk” in the Data column or selecting “Skewed restraint”
from the Data Types dialog.
295
Skewed Restraint
The Skewed Restraint dialog is shown.
Type
Translational: Use this type to restrain translation along the specified direction.
Rotational: Use this type to restrain rotation about the specified direction.
Stiffness
Type in the translational or rotational stiffness of the support. As an illustration, assume that
you had a rod (in tension only) which you were modeling as a skewed restraint. You can
calculate the stiffness (required to be input) in the following manner: Assume a 2.5 in. dia.
rod 2 feet long, modulus of elasticity of rod material = 30×106 psi.
𝜋
The translational (axial) stiffness is 𝐴𝐸/L= (2.5)2 × 30 × 106 /24=6,135,925 lb./in.
4
𝐸 𝐽
The rotational stiffness is 𝐺𝐽/L = 2(1+𝜈) × 𝐿 = 1,843,727 in. -lb./rad., where 𝐺 is the
shear
modulus, 𝜈 is the Poisson’s ration and 𝐽is polar moment of inertia.
Direction
If you have no “connected to node,” the direction in which the skewed restraint is oriented
must be specified in terms of its global X, Y and Z components. See topic on specifying a
Direction.
If the skewed restraint node is connected to an externally fixed point (ground), then for the
Direction vector components (X comp, Y comp, Z comp), you can specify the offsets (DX,
DY, DZ) from the skewed restraint node to the fixed point.
Or use one of the preset buttons to orient the skewed restraint axis:
1. Axial: To set the axis along the local-x direction (pipe axis)
2. Shear y: To set the axis in the local-y direction
3. Shear z: To set the axis in the local-z direction
296
Skewed Restraint
If you have connected the skewed restraint node to another node, then the direction must
not be input. It is calculated from the locations of the skewed restraint node and the
connected node, and it is oriented from the skewed restraint node towards the connected
node. In order for CAEPIPE to calculate the direction, the skewed restraint node and the
connected node must not be coincident.
Connected to node
If the skewed restraint node is connected to an externally fixed point (ground), leave the
“Connected to node” blank. You may connect a skewed restraint node to another node that
is not coincident with the skewed restraint node. Note that during skewed restraint force
calculations, the relative displacement of the skewed restraint node is calculated with respect
to the connected node.
Example: Modeling a Sway Brace
Assume that we need to model two sway braces in the same arrangement as shown in the
figure at the beginning of this section. The translational stiffness of the sway braces is given
as 894 lb./in. As can be surmised from the figure, the orientation of the sway braces (sway
struts in the figure) is at 45° from the Y- and Z-axes. We shall model the support on the
right hand side first followed by the support on the left hand side.
The following steps describe the modeling procedure:
 Create node (on pipeline) where support is required. In this case, the node is 50.
Position highlight on this row.
 First support (right): Type “sk” in the Data column to open the skewed restraint
dialog box.
Ensure that Type is set to Translational; if not, click on the Translational radio button.
Type 894 for Stiffness, type 1 for Y comp and –1 for Z comp, press Enter.
297
Skewed Restraint
 Second support (left): type 50 for Node on an empty row, press Tab to move to next
field, press “l(L)” for Location. This will open the Data types dialog.
Select Skewed restraint by clicking on it to open the skewed restraint dialog. Enter the
skewed restraint dialog similar to the first skewed restraint except in this case type 1 for Z
comp, press Enter.
298
Skewed Restraint
299
Slip Joint
A slip joint allows for axial (through telescopic action) and torsional movement between
adjacent pipes (due to thermal expansion or contraction). The joint itself can be fixed using
an anchor if so designed. Slip joints are susceptible to lateral buckling due to internal
pressure, and may become less effective when subjected to small bending loads. Proper
guiding to prevent buckling and keeping the two telescopic parts concentric are therefore
necessary.
Since the primary purpose of a slip joint is to absorb axial growth, the joint is ideal for
placing it towards the end of long pipe runs, while its growth is directed axially by the use of
one or more guides.
A Slip Joint is input by typing “s” under the Type column or by selecting “Slip joint” from
the Element types dialog.
The Slip joint dialog is shown.
A slip joint manufacturer should be able to provide you the required data for a slip joint.
A slip joint will have axial deflection or rotation only when the external forces exceed the
friction force or friction torque respectively. If the pressure thrust area is input, CAEPIPE
imposes a thrust load of: Pressure x Thrust area on both nodes of the slip joint. The weight
is the empty weight of the joint. The contents, insulation and additional weight are added to
the empty weight. A slip joint is considered to be rigid in lateral directions in CAEPIPE.
Weight of the slip joint is input in lbf or kgf and NOT its mass. Whenever mass is required
for a calculation as in the case of forming Mass matrix for dynamic analysis, or in calculating
300
Slip Joint
Example:
Assume that we want to model a (telescoping action) slip joint that allows only axial
movement with no torsion. The outer sleeve of the joint is anchored to hold it in place while
the other end is free to translate axially. The axial (friction) force has to exceed 1,100 lb. (you
will need to get this datum from a manufacturer’s catalog) to make the slip joint move and
the slip joint cannot rotate about the axial direction. So, the data would look similar to that
shown next between nodes 30 and 40.
And, the modeling on the layout screen would look thus:
301
Slip Joint
See the topic on Nonlinearities for related information.
302
Snubber
A snubber provides only translational restraint in a specified direction for seismic and
dynamic cases only. In other words, a snubber engages only during movements caused by a
dynamic load. It does not restrain against static loads such as weight and thermal.
A snubber is input by typing “sn” in the Data column or selecting “Snubber” from the Data
Types dialog.
The Snubber dialog is shown.
The stiffness defaults to Rigid, however a stiffness may be input for flexible snubbers. See
section on specifying a Direction for information on X comp, Y comp, Z comp. A snubber
can be made active in any direction by using this combination of Direction Cosines (X comp,
Y comp, and Z comp).
Since the snubber is considered to be attached to an externally fixed point, for the Direction
vector components (X comp, Y comp, Z comp), it is easier to specify the offsets (DX, DY,
DZ) from the snubber node to the fixed point.
303
Spider
Use a spider (also called a spacer) to connect the coincident nodes of a jacketed pipe (i.e., the
node on the core pipe and the corresponding node on the jacket pipe). The spider acts as an
internal guide. At the spider location, the local x-axis is calculated along the pipe direction.
The spider connects the local y and z translations for the core and jacket nodes. It prevents
any radial movement but allows sliding, rotating and bending movement between core and
jacket pipes. No gap is allowed between the core pipe and the spider. See section on Jacketed
pipe for related information.
A spider is input at a jacketed pipe node by typing “sp” in the Data column or selecting
“Spider” from the Data Types dialog.
304
Static Seismic Load
Starting Version 10.30, CAEPIPE has the built-in feature to calculate g-load values using the
procedure given in ASCE/SEI 7-16 as described below.
.Static Seismic Load – ASCE/SEI 7-16.
Guidelines from ASCE/SEI 7-16 “Minimum Design Loads for Buildings and Other
Structures” are explained below.
Structure Occupancy Category (Risk Category):
Table 1.5-1 of ASCE/SEI 7-16 provides the Risk Category of Buildings and Other
Structures for Flood, Wind, Snow, Earthquake, and Ice Loads Structure. Based upon your
project specification, select the Structure Occupancy Category from the options available.
Site Class:
Based on the site soil properties, the site shall be classified as Site Class A, B, C, D, E & F.
Refer Chapter 20 of ASCE/SEI 7-16 for details on Site Class. Depending on your project
specification, select the Site Class from the options provided.
Note: Para. 11.4.3 of ASCE/SEI 7-16 states that “Where the soil properties are not known
in sufficient detail to determine the site class, Site Class D shall be used unless the authority
having jurisdiction or geotechnical data determines Site Class E or F soils are present at the
site”.
305
Static Seismic Load
Mapped MCE Spectral Acceleration at Short Period S(S):

The USGS maintains a website http://earthquake.usgs.gov/designmaps with which the site
latitude and longitude, as well as the value of S(S) can be retrieved.
Users can also retrieve the above said values from https://hazards.atcouncil.org.
For example, for Centralia, WA, the value of S(S) is retrieved as 1.125 with Structure
Occupancy Category III and Site Class D from the link https://hazards.atcouncil.org as
shown below.
Component Height in Structure (z)

Component Height in Structure (z) is the point of attachment of components with respect to the
base. As per para. 13.3.1 of ASCE/SEI 7-16, for components at or below the base, z shall be taken as
0. In addition, the value of z/h need not exceed 1.0.
For example, for a Piping attached to a Boiler Nozzle located at an Elevation 150’0” with a
Ground Level of 25’0”, z can be computed as 125’ (= 150’ – 25’).
Structure Height (h)
Structure height (h) is the average roof height of the structure with respect to the base. From
the above example, the average roof height of structure (h) with respect to the base can be
calculated as 125’ (= 150’ – 25’).
Component Amplification Factor, a(p)
Table 13.6-1 Seismic Coefficients for Mechanical and Electrical Components of ASCE/SEI
7-16 provides values of Component Amplification Factor “a(p)” for various components. As
per para. 13.3.1 of ASCE/SEI 7-16, the component amplification factor is from 1.00 to 2.50.
For example, Piping in accordance with ASME B31, including in-line components with
joints made by welding or brazing, the value of Component Amplification Factor “a(p)” is
listed as 2.50.
306
Static Seismic Load
Component Response Modification Factor, R(p)

Table 13.6-1 Seismic Coefficients for Mechanical and Electrical Components of ASCE/SEI
7-16 provides values of Component Response Modification Factor “R(p)” for various
components. As per para. 13.3.1 of ASCE/SEI 7-16, component response modification
factor should be from 1.00 to 12.00
For example, Piping in accordance with ASME B31, including in-line components with
joints made by welding or brazing, the value of Component Response Modification Factor
“R(p)” is listed as 12.00.
Importance Factor, I(p)
Para. 13.1.1 of ASCE/SEI 7-16 provides a guideline to arrive at component Importance
Factor “I(p)”. The value of “I(p)” can be between 1.0 and 1.5.
Allowable Stress Design Factor, ADS(a)
Enter Allowable Stress Design Factor, ADS(a) as specified in the project specification. If this
data is not available, then it can be specified as 1.0.
Para. 13.1.8 of ASCE/SEI 7-16 states that the earthquake loads determined in accordance
with Section 13.3.1 of ASCE/SEI 7-16 shall be multiplied by an Allowable Stress Design
Factor ADS(a) of 0.7.
Example:
For a power plant required to operate in an emergency, located in Centralia, WA, USA, with
Occupancy Category as III, Site Class as D (with Stiff soil) and Importance Factor 1.5, use
ASCE/SEI 7-16 and compute the design earthquake load coefficient for piping required to
operate the plant.
1. From the map above with Occupancy Category as III and Site Class as D, the Mapped
MCE Spectral Acceleration at Short Period ‘S(S)” = 1.215.
2. Component Height in Structure (z) = 150’0” – 25’0” = 125’0”
3. Structure Height (h) = 150’0” – 25’0” = 125’0”
4. Component Amplification Factor, a(p) = 2.50 (as per Table 13.6-1 of ASCE/SEI 7-16, for
Piping in accordance with ASME B31, including in-line components with joints made by welding or
brazing)
5. Component Response Modification Factor, R(p) = 12.00 (as per Table 13.6-1 of
ASCE/SEI 7-16, for Piping in accordance with ASME B31, including in-line components with
joints made by welding or brazing)
6. Importance Factor, I(p) = 1.5 (given)
With the data provided above,
a. Site Coefficient at Short Period, F(a) = 1.014 (as per Table 11.4-1 for S(S) = 1.215 is
and Site Class D)
307
Static Seismic Load
b. As per equation 11.4-1 of ASCE/SEI 7-16, Maximum MCE Spectral Acceleration at

Short Period S(MS) = F(a)*S(S) = 1.014 * 1.215 = 1.232 g
c. As per equation 11.4-3 of ASCE/SEI 7-16, Design Spectral Acceleration at Short
Period S(DS) = (2/3)*S(MS) = (2/3)*1.232 = 0.821 g
d. As per equation 13.3-1 of ASCE/SEI 7-16,
Horizontal Seismic g-load value = H(g) = 0.4*S(DS)*[a(p)/R(p)]*I(p)*[1 + 2(z/h)]
= 0.4*0.821*(2.5/12.0)*1.50*[1+(2*(125/125))] = 0.308 g
e. As per equation 13.3-2 of ASCE/SEI 7-16,
Maximum Horizontal Seismic g-load value H(g.max) = 1.6*S(DS)*I(p)
= 1.6*0.821*1.5 = 1.97 g
f. As per equation 13.3-3 of ASCE/SEI 7-16,
Minimum Horizontal Seismic g-load value H(g.min) = 0.30*S(DS)*I(p)
= 0.30*0.821*1.5 = 0.369 g
g. As per Para 13.3.1 of ASCE/SEI 7-16,
Vertical Seismic g-load V(g) = 0.20*S(DS)
= 0.20*0.821 = 0.164 g
As per Para. 13.3.1 of ASCE/SEI 7-16, H(g) >= H(g.min) and H(g) <= H(g.max)
Hence,
Horizontal Seismic g-load value H(g) = 0.369 g.
Vertical Seismic g-load value V(g) = 0.164 g
From the above, as per Para. 13.1.7 of ASCE/SEI 7 – 16,
308
Static Seismic Load
Allowable Stress Design Horizontal Seismic g-load value = H(g)*ADS(a)

= 0.369 * 0.70 = 0.258 g
Allowable Stress Design Vertical Seismic g-load value = V(g)*ADS(a)
= 0.164 * 0.70 = 0.115 g
CAEPIPE Output
309
Static Seismic Load
.Static Seismic Load – ANSI A58.1-1988.

The procedure from ANSI A58.1-1988, “Minimum Design Loads for Buildings and Other
Structures,” is used in the below example to calculate the g-load values for input into
CAEPIPE.
Piping is assumed to be equivalent to equipment, thus giving a force coefficient, Cp as 0.3.
1. First, based on the map below (Contiguous 48 States+Alaska/Hawaii/Puerto Rico),
identify the seismic zone and its corresponding coefficient (Z).
Seismic Zone Coefficient Z

4 1.0000
3 0.7500
2 0.3750
1 0.1875
0 0.1250
Determine the Importance Factor, I as follows:
I = 1.5, For piping required in an emergency or piping with contents representing a
significant hazard to human life
I = 1.0, For other piping
Calculate the acceleration (g’s) as 0.3 × Z × I
Example:
For a power plant required to operate in an emergency, located in Anchorage, Alaska,
calculate the design earthquake load coefficient for piping required to operate the plant.
1. From the map above, the seismic zone is 4; from the table above, the corresponding
310
Static Seismic Load
Coefficient Z is 1.0.
2. For piping required to operate in an emergency, the Importance Factor I = 1.5.
3. Calculate horizontal acceleration as 0.3 × 1.0 × 1.5 = 0.45g, Vertical g-load (g) = 0.75 *
horizontal acceleration = 0.34g.
The above accelerations need to be applied in the desired horizontal and vertical directions
as long as the piping system is predominantly routed at grade level.
311
Supports
CAEPIPE provides many support types, as listed under the Data types menu (See section on
Data Types in the CAEPIPE User’s Manual).
You can use one or more such support types at one nodal location to model almost all types
of real-world pipe support hardware, thereby incorporating accurate mathematical
representation of those supports in the CAEPIPE stress model. For example, you could use
two lateral limit stops with unequal gaps to simulate a pipe shoe (see example under Limit
Stop).
To input more than one support at a node, use the “Location” data type.
Anchor
An anchor can be modeled as a flexible or rigid support which by default restrains the three
translations and three rotations either in the global or local directions at the applied node (six
degrees of freedom). Use this to model all anchor blocks, and nodes where piping connects
to equipment (pumps, compressors, turbines, etc.). See section on Anchor for further
details.
Restraint
A restraint is a two-way rigid support which restrains the translations (negative and positive
directions) along the specified global directions. You can apply a restraint in all the three
directions at the same time. See section on Restraint for further details.
Skewed Restraint
This is a flexible two-way support that can be oriented in any direction. Use this support to
resist either translational movement along or rotational movement about the specified
direction. Use this to model rigid or flexible sway struts and sway braces. See section on
Skewed Restraint for further details.
Hanger, User hanger, Rod hanger, Constant support
These should be used as vertical supports only. Use a Hanger when you want to design (i.e.,
select from a built-in catalog) a variable spring hanger(s) for your piping system (there are
30+ hanger catalogs built-in to CAEPIPE for your convenience). Use a User Hanger when
you want to analyze piping system with existing variable spring hangers. Use a Rod Hanger
for a rod hanger assembly. Use a Constant Support to design a constant support or a
constant force hanger. See corresponding sections for further details.
312
Supports
“Bottomed-out” Springs
To analyze this situation, use a variable spring hanger and a limit stop at this node. Type in
the maximum allowable hanger travel for one of the limits of the limit stop. Once the hanger
traverses the maximum distance allowed, the limit stop becomes active.
Guide
A Guide is a rigid or flexible restraint which resists lateral pipe movements (in directions
perpendicular to the axis of the pipe). You can specify an annular gap, if required, inside the
guide. A friction coefficient is optional. Use a Guide to model U-straps, U-Bolts, pipe guide
assemblies, pipe slides and similar supports. See section on Guide for further details.
Limit stop
A limit stop, a nonlinear restraint, can be oriented in any direction with a gap specified on
both sides of the pipe. A limit stop allows free movement for the distance of the gap and
then acts as a rigid or flexible restraint.
A Line Stop is a support that restricts axial movement of pipe. This support can be modeled
using a limit stop with its direction oriented along the pipe’s axis. Use this support to model
pipe slide assemblies, pipe skirts and similar arrangements. See topic on Limit Stop for an
example.
A limit stop can be used to model 1-way supports for pipe racks where vertical downward
movement is restrained while upward movement is not. See example for pipe rack modeling
in the Beam topic.
Support Tag
All the Supports described above and Nozzle can have Tags (Support Tags). Each Tag can
be up to 14 characters long. Tags are useful in identifying supports while modeling, reviewing
of reports and in field erection. Tag Name entered in this field is shown in all reports.
Refer to the respective CAEPIPE support term in this manual for further details.
313
Supports
314
Tees
Physical tees, even though integral components, are not modeled as such in CAEPIPE.
Instead, they are modeled as three pipes coming together at a common node.
In this figure, you can see that three pipe elements (20-30, 30-40, and 30-60) come together
at node 30 to form a tee.
315
Tees
Modeling
The method of modeling is as simple as its representation.
Step 1:
You could model the pipe run first from 20 to 30 (1st element), then from 30 to 40 (2nd
element), and finally, from 30 to 60 (3rd element). See example layout window below.
Alternately, you could model from 20 – 30 – 60, and then 30 – 40. Or from 60 – 30 – 20,
and then 30 – 40. Modeling order is immaterial to analysis (but sometimes complicates the
merge files process. See Merge under Layout window > Merge in the User’s Manual).
Step 2:
There is one final step remaining. You need to designate the type of tee connection it is. In
this example, the tee is a “Welding Tee”. So, on any row that contains node 30 (i.e., row #4
or row #8), type “Br” (Branch SIF) and select “Welding Tee” from the drop down list. Since
row #4 already has a hanger specified, you can use row #8 to specify the Branch SIF.
The shown Branch SIF list is piping code-dependent. In other words, the list of tee types
shown comes from the selected piping code that dictates how an SIF for each of the listed
tee types is calculated. See Section titled “Piping Code Compliance” from the Code
Compliance Manual for information on Branch SIFs for different piping codes.
316
Tees
Based on a more rigorous analysis, if you have another more accurate SIF value for a joint
you want to insert (instead of the code’s), then skip Step (2) above and use User-SIF data
type at the same node to input your own SIF.
For 45° laterals, a piping code committee member opines as follows:

Use the SIF for an unreinforced fabricated tee and evaluate the branch stress using the
section modulus of the branch. The connection footprint on the run pipe itself has a section
modulus greater than the branch section modulus and this will compensate for the
unreinforced fabricated tee SIF which should be lower than the unreinforced fabricated
lateral SIF. Ignore the fact that a reduced outlet branch connection requires an “effective
section modulus” in accordance with ASME B31.1 (2014) Para. 104.8.4(C) or ASME B31.3
(2014) Para. 319.4.4(c) and just calculate the intensified stress as the unreinforced fabricated
tee SIF x the branch moment / branch section modulus (whether you use the SIF times the
resultant moment approach of B31.1 or the in-plane and out-plane SIFs times the in-plane
and out-plane moments, respectively, approach of B31.3, use the same philosophy of
ignoring the “effective modulus.”
Pay particular attention to the fabrication of the lateral making sure that the Code required
cover fillet dimensioned t(c) in ASME B31.1 (2014) Fig. 127.4.8(D)(a) or ASME B31.3
(2014) Fig. 328.5.4D(1) meets the required size. If the cover fillet is larger than t(c), that
improves the lateral SIF. The figures and dimensions shown are a bit unfortunate, especially
for a lateral because a strict reading of the Code would seem to require a bigger weld on the
obtuse side of the branch and a smaller weld on the acute side of the branch when just the
opposite is true. Having a constant cover fillet weld leg length all around the branch would
improve the requirement and the committee has been working on that. Personally, I think
the larger the run and branch pipes are, the larger the cover fillet should be. Ask for or
calculate whether area replacement requirements ASME B31.1 (2014) Para. 104.3.1(D) or
ASME B31.3 (2014) Para. 304.3.3 are met noticing that the required reinforcement for a
lateral is greater by the factor (2 - sin alpha).
317
Tie Rod
Tie rod is a nonlinear element with different stiffnesses and gaps in tension and compression,
used to model tie rods in bellows, chains, etc. The force versus displacement relationship for
a tie rod is shown below. Around expansion joints, a tie rod continuously restrains the full
pressure thrust while allowing only lateral deflection, bending and torsional rotation.
When the tie rod is in tension, and the displacement is greater than the tension gap, tension
stiffness is used. If the displacement is less than the tension gap, zero stiffness is used.
Similarly when the tie rod is in compression, and the displacement is greater than the
compression gap, compression stiffness is used. If the displacement is less than the
compression gap, zero stiffness is used. A Tie rod is input by typing “t” in the Type column
or selecting “Tie rod” from the Element Types dialog.
The Tie Rod dialog is shown.
318
Tie Rod
A tie rod can be made “Tension only” by setting the compression stiffness to zero. Similarly
it can be made “Compression only” by setting the tension stiffness to zero. Both Tension
and Compression stiffnesses cannot be zero. If there is no tension or compression gap, leave
it blank or specify it as zero. See the earlier “Expansion Joints” topic for examples.
319
Turbine
governed by an industry publication — API (American Petroleum Institute) publishes an
API 610 for pumps and an API 617 for compressors while NEMA (National Electrical
Manufacturers Association) publishes the NEMA SM-23 for turbines. These publications
provide guidelines for evaluating nozzles connected to equipment among other technical
information including the items relevant to piping stress analysis – criteria for piping design
and a table of allowable loads.
Modeling the equipment is straightforward since it is assumed rigid (relative to connected
piping) and modeled only through its end points (connection nozzles).
1. In your model, anchor all of the nozzles (on the equipment) that need to be included in
the analysis.
2. Specify these anchored nodes during the respective equipment definition via Misc. menu
> Pumps/Compressors/Turbines in the Layout window.
CAEPIPE does not require you to model all of the nozzles nor their connected piping. For
example, you may model simply one inlet nozzle of a turbine with its piping. Or, you may
model one turbine with all its nozzles (with no connected piping) and impose external forces
on them (if you have that data). Further, there is no need to connect the anchors of the
equipment with a rigid massless element like required in some archaic methods. A flange and
an anchor may coexist.
A turbine (like a pump or a compressor) is input by selecting “Turbines” from the
Miscellaneous (Misc) menu in the Layout window. CAEPIPE, upon analysis, produces a
NEMA SM-23 turbine compliance report. See Section titled “Rotating Equipment
Qualification” in the Code Compliance Manual for related information on NEMA SM-23,
for Turbines.
Once you see the Turbine List window, double click on an empty row for the Turbine dialog
and enter the required information.
320
Turbine
A short description to identify the turbine may be entered for Description. The nozzle nodes
must be anchors and the shaft axis must be in the horizontal plane. Some of the nozzle
nodes may be left blank if they are not on the turbine (e.g., extraction nodes).
See under Pumps for related modeling tips and the topic on specifying a Direction for
information on how to specify X comp/Y comp/ Z comp for Shaft axis.
321
Turbine
NEMA SM-23 Report
See Section titled “Rotating Equipment Qualification” of the Code Compliance Manual for
more information on how to interpret the NEMA SM-23 report.
322
User Hanger
Use the “User Hanger” type for analyzing piping systems with existing variable spring
hangers, which are different from spring hangers that need “to be designed” (for which you
use the “Hanger” data type).A user hanger is input by typing “u” in the Data column and
pressing Enter or selecting “User Hanger” from the Data Types dialog.
The User Hanger dialog is shown.
Spring Rate
The spring rate is required. For a constant support user hanger, input the spring rate as zero.
Number of Hangers
Type in the number of separate hangers connected in parallel at this node. The stiffness and
load of each hanger are multiplied by the number of hangers to find the effective stiffness
and load of the hanger support at this node.
Hanger Load and Load type
Input the hanger load, if known. Otherwise, leave it blank and CAEPIPE will calculate the
load.
The hanger load may be specified as hot or cold using the Load type radio buttons.
323
User Hanger
When Cold Load and Spring Rate are input for User Hanger
Snap shots shown below are from a CAEPIPE model with two (2) User Hangers defined at
Nodes 20B and 115B. In this model, Spring rate and Cold load are input for each User
Hanger as given below.
User Hangers with Spring Rate and Cold Load UserHanger_Spring_ColdLoad
120 Z X
115
0.669 125
40
38
0.197
110
105
100
35
30
25
20
10
15
Mar 1,2017
User Hanger at Node 20B User Hanger at Node 115B
324
User Hanger
Analysis Options
Details of Layout
325
User Hanger
From the analysis results, Hanger Report as well as Displacements and Support Loads for
Sustained & Operating load cases are presented below.
326
User Hanger
Given below are the Steps performed by CAEPIPE to arrive at the results shown above for
User Hangers when “Cold load” and “Spring rate” are input.
Step 1: Compute Hot Load from Preliminary Sustained Load Analysis
Performs a preliminary sustained load analysis and computes hot loads for the two User
Hangers by replacing them with Vertical Restraints.
To verify this step, remove the User Hangers at nodes 20B and 115B and instead add
Vertical Rigid Restraints at those two nodes and perform a sustained load analysis. From the
Analysis results, you will observe that the “Loads on Restraints” at nodes 20B and 115B for
Sustained load case will be identical to the hot loads reported above in the Hanger Report for
the model with User Hangers.
327
User Hanger
Step 2: Compute Vertical Travel from Preliminary Operating Load Analysis

Vertical Restraints added in Step 1 above at User Hanger locations are removed. The hot
loads (calculated in Step 1) are applied as upward forces at the hanger locations. Vertical
displacements at the hanger locations obtained from the operating load case analysis are the
hanger travels.
To verify this step, remove the Vertical Restraints at Nodes 20B and 115B from the model
developed in Step 1 above. Apply the hot loads as upward forces at Nodes 20B and 115B by
choosing “Add to W+P” in CAEPIPE’s Force dialog and perform the analysis. Vertical
displacement results at the User Hanger locations for the Operating load case will be
identical to the Vertical Travel values reported above in the Hanger Report for the model
with User Hangers.
Step 3: Perform Detailed Analysis
Performs once again the Sustained and Operating load case analyses by including the “Spring
Rate” and “Cold load” input into CAEPIPE model at User Hanger locations.
To verify this step, to the model developed under Step 2 above add a skewed restraint in
vertical direction at each User Hanger location with its stiffness equal to the “Spring Rate”
for that User Hanger. In addition, add to the sustained load case (i.e., choose “Add to W+P”
in CAEPIPE’s Force dialog) an upward force at each skewed restraint node that is equal to
the Cold load for the corresponding User Hanger. The resulting Step 3 model is then
analyzed. Displacements for the Sustained and Operating load cases obtained from this Step
3 model will be identical to the displacements obtained for Sustained and Operating load
cases reported above for the model with User Hangers.
Now, from the Step 3 model results, we observe the following at the two skewed restraints.
Sustained load case
Support load at Node 20B [A] = Spring Rate x Sustained displacement at Node 20B
= 800 lb/in x 0.248”
=198.4 lb (comparing well with CAEPIPE result of 198 lb)
Support load at Node 115B [B] = 340 lb/in x 0.166”
= 56.44 lb (comparing well with CAEPIPE result of 56 lb)
Operating load case
Support load at Node 20B [C] = Spring Rate x Operating displacement at Node 20B
= 800 lb/in x 0.575”
= 460 lb (comparing well with CAEPIPE result of 460 lb)
Support load at Node 115B [D] = 340 lb/in x 1.030”
Using the above Support load and Cold load, CAEPIPE is computing and reporting the
Loads on User Hangers as follows.
328
User Hanger
When the option “Include hanger stiffness” is turned ON

Sustained load case
Load on Hanger at 20B = -Cold load + Spring Rate x Sustained displacement at Node 20B
= -Cold Load + A
= -3787 + 198.4 = -3588.6 lb (matches with Hanger loads results)
Operating load case
Load on Hanger at 20B = -Cold load + Spring Rate x Operating displacement at Node 20B
= -Cold Load + C
= -3787 + 460 = -3327 lb (matches with Hanger loads results)
When the option “Include hanger stiffness” is turned OFF
Sustained load case
Load on Hanger @ 20B = -Hot load = -3327 lb
Operating load case
Load on Hanger @ 20B = -Hot load = -3327 lb
329
User Hanger
When Hot Load and Spring Rate are input for User Hanger
Snap shots shown below are from a CAEPIPE model with two (2) User Hangers defined at
Nodes 20B and 115B. In this model, Spring rate and Hot load are input for each User
Hanger as given below.
User Hangers with Spring Rate and Hot Load UserHanger_HotLoad
Y
Z X
120
115
0.669 125
40
0.197 110
20
15
Mar 3,2017
User Hanger at Node 20B User Hanger at Node 115B
330
User Hanger
Analysis Options
Details of Layout
331
User Hanger
From the analysis results, Hanger Report as well as Displacements and Support Loads for
Sustained & Operating load cases are presented below.
332
User Hanger
Given below are the Steps performed by CAEPIPE to arrive at the results shown above for
User Hangers when “Hot load” and “Spring rate” are input.
Step 1: Compute Vertical Travel from Preliminary Operating Load Analysis
Performs a preliminary operating load analysis by including the hot loads input as upward
forces at the hanger locations. Vertical displacements at the hanger locations obtained from
the operating load case analysis are the hanger travels.
To verify this step, remove the User Hangers at Nodes 20B and 115B and instead apply the
respective hot loads as upward forces at those two nodes by choosing “Add to W+P” in
CAEPIPE’s Force dialog and perform the analysis. Vertical displacement results at the User
Hanger locations for the Operating load case will be identical to the Vertical Travel values
reported above in the Hanger Report for the model with User Hangers.
333
User Hanger
Step 2: Compute Cold Load

The Cold load is then computed using the Hot load entered and Vertical Travel obtained
from Step 1 above as stated below.
Cold load = Hot load + Spring Rate x Vertical Travel
For example, at Node 20B,
Cold load = 3327 + 800 x 0.576 = 3787.8 lb.
Similarly, at Node 115B,
Cold load = 1473 + 340 x 1.030 = 1823.2 lb.
Step 3: Perform Detailed Analysis
Performs once again Sustained and Operating load case analyses by including the “Spring
Rate” and “Cold load” (obtained in Step 2 above) into CAEPIPE model at User Hanger
locations.
To verify this step, to the model developed under Step 1 above add a skewed restraint in
vertical direction at each User Hanger location with its stiffness equal to the “Spring Rate”
for that User Hanger. In addition, add to the sustained load case (i.e., choose “Add to W+P”
in CAEPIPE’s Force dialog) an upward force at each skewed restraint node that is equal to
the Cold load computed in Step 2 above for the corresponding User Hanger. The resulting
Step 3 model is then analyzed. Displacements for the Sustained and Operating load cases
obtained from this Step 3 model will be identical to the displacements obtained for Sustained
and Operating load cases reported above for the model with User Hangers.
Now, from the Step 3 model results, we observe the following at the two skewed restraints.
Sustained load case
Support load at Node 20B [A] = Spring Rate x Sustained displacement at Node 20B
= 800 lb/in x 0.248”
=198.4 lb (comparing well with CAEPIPE result of 198 lb)
Support load at Node 115B [B] = 340 lb/in x 0.166”
Operating load case
Support load at Node 20B [C] = Spring Rate x Operating displacement at Node 20B
= 800 lb/in x 0.576”
= 460.8 lb (matching with CAEPIPE result of 461 lb)
Support load at Node 115B [D] = 340 lb/in x 1.030”
= 350.2 lb (matching with CAEPIPE result of 350 lb)
Using the above Support load values and Cold load, CAEPIPE is computing and reporting
the Loads on User Hangers as follows.
334
User Hanger
When the option “Include hanger stiffness” is turned ON

Sustained load case
Load on Hanger at 20B = -Cold load + Spring Rate x Sustained displacement at Node 20B
= -Cold Load + A
= -3787.8 + 198.4 = -3589.4 lb (matches with Hanger loads results)
Operating load case
Load on Hanger at 20B = -Cold load + Spring Rate x Operating displacement at Node 20B
= -Cold Load + C
= -3787.8 + 460.8 = -3327 lb (matches with Hanger loads results)
When the option “Include hanger stiffness” is turned OFF
Sustained load case
Load on Hanger at 20B = -Hot load = -3327 lb
Operating load case
Load on Hanger at 20B = -Hot load = -3327 lb
Connected to Node
node number in the “Connected to” node field. This node must be directly above or below the
hanger node.
335
User SIF
User SIF (Stress Intensification Factor) may be used to specify SIF at a node where there is
normally no SIF internally calculated (i.e., at a non-bend or non-tee node) or to override any
internally calculated SIF at the node.
Use this for any component that needs an SIF value such as non-right angle tees,
nonstandard tees or branch connections, flanges, etc., for which the chosen piping code does
not specify a SIF, or you want to override the code’s SIF. For example, in case of a bend or a
tee, CAEPIPE calculates the SIF according to the selected piping code. To override the
calculated SIF, specify a User SIF. Note that a User SIF is applied to all elements that come together at
this node.
A User SIF is input by typing “user s” in the Data column or selecting “User SIF” from the
Data types dialog.
Depending on the piping code selected, either a single value User SIF (B31.1)or in-plane,
out-of-plane and Axial values of User SIF (B31.3) or in-plane, out-of-plane, Axial and
Torsion values of User SIF (B31J) may be input. The corresponding dialog will be shown.
B31.1 code
B31.3 code B31J Turned ON
336
Valve
Use this element to model any type of valve. A valve is relatively more rigid than a pipe.
CAEPIPE uses the data input to calculate the rigidity.
A Valve is input by typing “v” in the Type column or selecting “Valve” from the Element
Types dialog.
The Valve dialog is shown.
Weight
The weight is the empty weight (without contents, insulation, etc.). CAEPIPE applies this
weight as a uniformly distributed load along the length of the valve. Additional weight, if
specified, is treated as a concentrated weight offset from the center of the valve.
CAEPIPE requires “Weight” to be input in lbf or kgf and NOT its “Mass”. Whenever mass
is required for a calculation as in the case of forming Mass matrix for dynamic analysis, or in
calculating inertia force as (mass x acceleration) for static seismic analysis, CAEPIPE
internally computes the mass to be equal to (weight / g-value).
Length
If the valve length is input, the DX, DY, DZ in Layout is adjusted to match the valve length,
(assuming that the local x-axis of valve is in the same direction as the local x-axis of the
337
Valve
preceeding element). If the valve length is left blank, the valve length is calculated from DX,
DY, DZ input in Layout.
Thickness X
The thickness multiplier (Thickness X) is used for stiffness calculation (i.e., the thickness of
the pipe section is multiplied by Thickness multiplier by increasing only the OD of the valve
and not changing its ID in the calculation of the valve stiffness). Typical value for Thickness
multiplier is 3 which is the default value if left blank.
Insulation weight X
The insulation weight multiplier (Insulation weight X) is used if the valve has additional
insulation compared to adjacent pipe (i.e., weight of insulation calculated from the insulation
thickness of the pipe section is multiplied by Insulation weight X multiplier). Typical value
for insulation weight multiplier is 1.75 which is the default value if left blank.
Additional weight
The additional weight is a concentrated weight which may be specified at an offset from the
center of the valve, such as for a valve operator. As stated above, CAEPIPE requires
“Additional Weight” to be input in lbf or kgf and NOT its “Mass”.
Valve Library
Cast Iron, Steel and Alloy valve (Flanged and Butt Welding ends) libraries are provided. The
Type of Valve, Connection Type and Rating are indicative in the filenames listed in the
libraries. Valve weights are included for different categories of valves. If necessary, you may
create your own user-definable valve library. A new valve library can be created from menu
File > New > Valve Library in the main opening CAEPIPE window.
338
Valve
The valve library may be accessed by clicking on the Library button of the Valve dialog.
Navigate to the folder called “Valve_Library” or similar under your CAEPIPE program files
folder.
Selecting one of these valve types will display a list of valves for you to select from.
339
Valve
Select one from the displayed list. The weights of the valves (in the Extended Library) are
provided. If a valve is flanged, the mating flanges at the two ends of the valve must be
separately input using the Flange data type and their corresponding weight. Please confirm
the data with your valve manufacturer’s catalog, and input the correct weight.
340
Valve
Calculation of Moment of Inertia

The inside diameter of the valve is calculated from the section O.D. (outside dia.) and the
section Thickness.
I.D. = Section O.D. − 2 ×Section Thickness
Then the new O.D. and Thickness for the valve are calculated as:
New Thickness = Section Thickness × Thickness X multiplier
New O.D. = I.D. (inside dia.) + 2 × New Thickness
The moment of inertia for the valve is now based on the New O.D. and New Thickness. For
a thin-walled pipe, Thickness X and Inertia multipliers are approximately the same. The
weight of the contents of the valve is based on the I.D. as calculated above.
The weight of the insulation is the weight calculated from the section O.D. and insulation
thickness and density (from section properties) multiplied by the insulation weight multiplier.
Angle/Relief Valve
Angle and relief valves which have the outlet pipe at some angle (typically 90°) from the inlet
pipe may be modeled by two valves one after the other and at that angle. The total weight of
the actual valve must be divided between these two modeled valves.
341
Weld
Use this data type to input a Weld at a node. Type “w” in the Data column or select “Weld”
from the Data types dialog.
The Weld dialog is shown. Four types of welds are available: Butt, Fillet, Concave Fillet, and
a Tapered transition. The type of the weld should be selected from the “Type” drop-down
combo box.
Butt weld and Tapered transition require the input of weld mismatch. Mismatch is the
difference in the mean radii across the weld.
The SIF for a weld is calculated according to the selected piping code (see Section titled
“Piping Code Compliance” from the Code Compliance Manual for details) and is
incorporated in the stress calculations. If you have an unlisted weld type, you could specify
the SIF for it using the User SIF data type.
Any SIF value specified using the “User SIF” Data item will always overwrite any other SIF
value calculated/determined at that node using any other method(s).
342
Wind
.Wind Load – ANSI A58.1 - 1982.

CAEPIPE calculates Design Wind Force F as follows to calculate wind load acting on each
pipe element.
F = q x Shape factor x Af (lbs)
where,
Af is the area of the piping plus insulation projected on a plane normal to the wind direction
q = Dynamic pressure due to wind (lbs/ft 2) = 0. 0.00256V2 x I2 x (Kz Gz Cf)
V = Basic Wind Speed, V (mph), from Basic Wind Speed Map, for your region.
Shape factor = 0.6 for Circular cross-section (to be input into CAEPIPE for Wind)
“Section 6” of ANSI A58.1 - 1982, Minimum Design Loads for Buildings and Other
Structures can be referred to arrive at the values for [I 2 x (Kz Gz Cf)].
Importance Factor, I2, is determined according to structure category and location:
Value of I2 for: Other Areas Within 100 mi
of Ocean line
Chemical plants, refineries, industrial facilities, and 1.00 1.11
power plants not required inan emergency
Power Plants required in an emergency 1.15 1.24
Determine combined Velocity Coefficient, Gust Factor, and Force Coefficient, K z Gz Cf)
according to height of piping;
Height of Piping KzGzCf
At or below 50’ above ground 1.23
Above 50’ at or below 100’ above ground 1.44
Above 100’, at or below 200’ above ground 1.68
Note:
If Wind Speed as a function of Elevation is input, the factor SQRT[(I2 x (Kz Gz Cf))] has to
be multipliedwith the Actual Wind Velocity as a function of elevation to arrive at the Wind
Speed that is to be input at each elevation manually.
For example, if the Actual Wind Speed is 60 mph at an elevation of 50’, I 2 as 1.00 and
Kz.Gz.Cf as 1.23, SQRT[1.00 x 1.23] = 1.109. So, V = 60 x 1.109 for up to 50’ elevation =
66.54 mph. This should be input in the CAEPIPE Wind dialog.
If Pressure as a function of Elevation is input in CAEPIPE, the factor [I 2 x (Kz Gz Cf)] and
the Shape Factor have to be incorporated in the Pressure input at each elevation manually.
This is because CAEPIPE currently uses the Shape Factor input into Wind dialog only for
Velocity vs Elevation option and not for Pressure vs Elevation option.
343
Wind
.Wind Load – ASCE/SEI 7-16.

The determination of wind loads for the structural design of buildings is a complex subject that
many building codes simplify by presenting tables of net wind pressures versus height above
grade. Wind loads on a building in any particular locality depend on many factors, including
recorded wind speeds in the area, the terrain around the building, and the shape and height of the
building. It is now common for wind tunnel model tests to be conducted for tall buildings to
determine wind loads - which may be more severe than the minimum code requirements.
American Society of Civil Engineers Standard (ASCE) SEI 7-16 contains detailed information
and formulas for computing wind loads on buildings in various geographic locations. The
procedure described in this code has been updated in CAEPIPE starting Version 10.30 and can
be accessed through this option Layout Window > Misc > ASCE/SEI 7-16.
By defining the basic wind parameters in the dialog shown, CAEPIPE will compute the Design
Wind Force internally as per ASCE/SEI 7-16 and apply the same to the piping system. For
details on implementation, refer Appendix “ASCE/SEI 7-16” from Code Compliance Manual.
Structure Occupancy Category
344
Wind
Structure Occupancy Category (Risk category) can be I, II, III or IV as provided in the Table
below.
Basic Wind Speed (V)

Depending upon the Risk Category, Basic Wind Speed (V) can be determined from the Figures
26.5.-1A, 26.5-1B and 26.5-1C provided in Chapter 26 of ASCE/SEI 7-16.
For example, Basic Wind Speed (V) at Caledonia, MS (near Aberdeen, MS) with Risk Category
III is 114 mph (as per Figure 26.5-1B of ASCE/SEI 7-16).
Alternatively, one can use the link https://hazards.atcouncil.org/#/ to determine the Basic Wind
Speed (V) as per ASCE/SEI 7-16 by entering the location as shown below.
Wind Directionality Factor (Kd)
345
Wind
This Wind Directionality Factor (Kd) accommodates the cross-sectional shape of the structure.
The wind directionality factor (Kd) can be determined from Table 26.6-1 of ASCE/SEI 7-16.
This table is provided below for quick reference.
For example, Chimneys, Tanks, and Similar Round Structures, Kd = 0.95.

Exposure Category
Exposure Category can be B, C or D.
For each wind direction considered, an exposure category that adequately reflects the
characteristics of ground surface irregularities shall be determined for the site at which the
building or structure is to be constructed. For a site located in the transition zone between
categories, the category resulting in the largest wind forces shall apply. Account shall be taken of
variations in ground surface roughness that arises from natural topography and vegetation as well
as from constructed features.
The exposure category for an individual structure shall be based upon the site conditions that will
exist at the time when all adjacent structures on the site have been constructed, provided that
their construction is expected to begin within one year of the start of construction for the
structure for which the exposure category is determined.
For further details, refer para. 26.7.3 of ASCE/SEI 7-16 on Exposure Categories.
Topographical Parameters
Topography or large vertical displacements of the ground surface can have a significant effect on
the wind speed profile. The wind flow in a realistic environment is not merely over a single
ground feature such as hills, ridges, escarpment, but as well over undulating and mountainous
terrain. It is important to understand that the flow over one hill will affect that around the next.
The effects of undulating and mountainous terrain are almost similar to those of a very rough
surface. Ridges and escarpments are mainly two dimensional land feature, and hills are mainly
three dimensional.
346
Wind
Hills differ from ridges in that the wind can diverge over sides in addition to speeding up over
crests. The speed-up effects of a hill are thus generally less than that those of a ridge of the
identical slope. In general, the wind increases its speed when it moves up the windward slope of
a hill or a ridge. The maximum increase in wind speed is usually experienced at or near the crest.
Parameters such as Hill Type, Height of Hill or Escarpment (H), Crest Distance (Lh), Height
above ground level (z) and Distance from Crest to Site (x) are required by ASCE/SEI 7-16 for
computing the Topographical Factor (Kzt) required in computing the Velocity Pressure (q z).
Refer para. 26.8.2 of ASCE/SEI 7-16 for more details on Topographical Parameters.
Type of Surface
Type of Surface can be “Moderately Smooth”, “Rough” or “Very Rough”. This parameter is
used to compute the force coefficient (Cf) required in computing the design wind force (F).
For further details, refer para. 29.5 from ASCE/SEI 7-16.
Gust-effect Factor (G)
As per para 26.9.1, Gust-effect factor for a rigid building and other structures is permitted to be
taken as 0.85.
Upon defining the above parameters, the user can apply Wind Load by selecting the wind code
as “ASCE/SEI 7-16” through Layout Window > Loads > Wind1/Wind2/Wind3/Wind4.
347
Wind
.Wind Load – EN 1991-1-4 (2010).

Eurocode 1 EN 1991-1-4 Action on structures (wind load) and EN 1991-1-4 (2010) contains
detailed information and formulas for computing wind loads on buildings in various geographic
locations. The procedure described in this code has been included in CAEPIPE starting Version
10.30 and can be accessed through this option Layout Window > Misc >Wind Load – EN 1991-
1-4 (2010).
By defining the wind parameters in the dialog shown, CAEPIPE will compute the Design Wind
Force internally as per EN 1991-1-4 (2010) and apply the same to the piping system. For details
on implementation, refer to Appendix “EN 1991-1-4 (2010)” from Code Compliance Manual.
Fundamental value of the basic wind velocity can be obtained from National Annex for EN
1991-1-4.
Terrain Category
Based on the location of the structure, Terrain Category classification can be obtained from the
Annex A and Table A.1 of EN 1991-1-4.
348
Wind
349
Wind
Directional and Seasonal Factors (Cdir & Cseason)

In order to calculate the Basic Wind Velocity as per Expression (4.1) of EN 1991-1-4 (2010), we
need to determine the directional and seasonal factors (Cdir &Cseason). National Annex for EN
1991-1-4 simplifies this calculation as the suggested values of these factors are equal to 1.0.
Terrain Orography
The recommended value of the Terrain Orography factor is 1.0.
Where orography (e.g. hills, cliffs etc.) increases wind velocities by more than 5% the effects
should be taken into account using the orography factor Co(z).
The procedure to be used for determining Co(z) may be given in the National Annex. The
recommended procedure is given in A.3 of EN 1991-1-4 (2010).
350
Wind
The effects of orography may be neglected when the average slope of the upwind terrain is
less than 3°. The upwind terrain may be considered up to a distance of 10 times the height of
the isolated orographic feature.
Turbulence Factor (Kt)
Turbulence factor (Kt) may be given in the National Annex. The recommended value for Kt
is 1.0.
Roughness Length and Minimum Height
Roughness Length (Zo) and the Minimum Height (Zmin) are provided for each Terrain
Category in Table 4.1 of EN 1991-1-4 (2010).
Upon defining the above parameters, the user can apply Wind Load by selecting the wind code
as “EN 1991-1-4 (2010)” through Layout Window > Loads > Wind 1/Wind 2/Wind 3/Wind 4.
351
This page is blank
Annexure I
Dynamic Susceptibility Method
The “Dynamic Susceptibility” Method for Piping Vibration –A Screening Tool for
Potentially Large Alternating Stresses
Dr. R. T. Hartlen, Plant Equipment Dynamics, Ontario, CANADA
Summary
The enhanced output of the “Modal Analysis” load case in CAEPIPE shows modal
frequencies and mode shapes AND now two new outputs called “dynamic stresses” and
“dynamic susceptibility”. The dynamicstresses are the dynamic bending stresses associated
with vibration in a natural mode. That is to say, the modal analysis result has been generalized to
include the alternating bending stresses associated with the vibration in a natural mode. The
dynamic susceptibility for any mode is the ratio of the maximum alternating bending stress to
the maximum vibration velocity. This “susceptibility ratio” provides an indicator of the
susceptibility of the system to large dynamic stresses. Also, the associated animated mode
shapes include color-spot-markers identifying the respective locations of maximum vibration
velocity and maximum dynamic bending stress. The susceptibility ratio and the graphics
feature provide incisive insights into the reasons for high susceptibility and how to make
improvements. This new feature is illustrated by application to the CAEPIPE “Sample
problem” stress model.
1. Dynamic Susceptibility: New Analytical Tool Available for Vibration of Piping
When addressing vibration issues, the piping designer does not have the specific
requirements, nor the analytical tools and technical references typically available for other
plant equipment such as rotating machinery. Typically, piping vibration problems only
become apparent at the time of commissioning and early operation, after a fatigue failure or
degradation of pipe supports. Discovery of a problem is then followed by an ad hoc effort to
assess, diagnose and correct as required. The “Dynamic Susceptibility” analysis, now
included in CAEPIPE, provides a new analytical tool to assist the piping designer at any
stage, from preliminary layout to resolution of field problems.
CAEPIPE’s Dynamic Susceptibility feature utilizes the “Stress per Velocity” method, an
incisive analytical tool for “screening” the vibration modes of a system. It readily identifies
which modes, if excited, could potentially cause large dynamic stresses. Furthermore, it
reveals which features of the system layout and support are responsible for the susceptibility
to large dynamic stresses. At the design stage, the method allows the designer to quickly
identify and correct features that could lead to large dynamic stresses at frequencies likely to
be excited. Where problems are encountered in the field, the method provides quick and
incisive support to efforts of observation, measurement, assessment, diagnosis and
correction.
The technical foundation of this method lies in an underlying fundamental relationship
between the kinetic energy of vibratory motion, and the corresponding potential energy
stored in elastic stresses. That is to say, the kinetic energy at zero displacement and maximum
system velocity must equal the stored elastic energy at zero velocity and maximum
displacement. This implies a fundamental relationship between vibration velocity and
dynamic bending stresses, which is the foundation of the stress per velocity approach for
“susceptibility screening” of vibration modes.
The key analytical step is to determine, mode by mode, the ratio of maximum dynamic stress
to maximum vibration velocity. This ratio will lie in a lower “baseline range” for
354
uncomplicated systems such as classical uniform-beam configurations. For more complex

systems, the stress / velocity ratio will increase due to typical complications such as three-
dimensional layout, discrete heavy masses, changes of cross-section and susceptible branch
connections. System modes with large stress-velocity ratios are the potentially susceptible modes.
The Stress / Velocity method, implemented in CAEPIPE as the Dynamic Susceptibility
feature, automatically and quickly finds these modes and quantifies the susceptibility. Evaluation of the
results, including special-purpose color animation, helps to identify which details of layout
and support are responsible for the large stresses.
This technical note is to present and explain the “dynamic susceptibility” outputs now
included in the modal analysis load case, and to illustrate by application to the standard
CAEPIPE “Sample problem” system.
2. Underlying Fundamental Basis of the Method
2.1 Kinetic Energy and Potential Energy; Vibration Velocity and Dynamic
Stresses
The underlying theoretical basis for the Stress / Velocity method is a deceptively
straightforward but universally-applicable relationship between kinetic energy and potential
(elastic) energy for vibrating systems. Stated simply, for vibration at a system natural frequency, the
kinetic energy at maximum velocity and zero displacement must then be stored as
elastic (strain) energy at maximum displacement and zero velocity. Since the strain
energy and kinetic energy are respectively proportional to the squares of stress and velocity, it
follows that dynamic stress, , will be proportional to vibration velocity, v. For idealized
straight-beam systems, consisting of thin-walled pipe and with no contents, insulation or
concentrated mass, the ratio  / v is dependent primarily upon material properties, (density  and
modulus E) ,and is remarkably independent of system-specific dimensions, natural-mode number and
vibration frequency. For real continuous systems of course, the kinetic and potential energies are
distributed over the structure in accordance with the respective modes shapes. However,
integrated over the structure, the underlying energy- equality holds true. Provided the spatial
distributions are sufficiently similar, i.e. harmonic functions, the RMS or maximum stress will
still be directly related to the RMS or maximum vibration velocity.
2.2 The “Screening” Approach
As stated above, for idealized pure beam systems the stress-velocity ratio will depend
primarily upon material properties.
For real systems, the spatial patterns of the mode shapes will depart from the idealized
harmonic functions, and the stress-velocity ratios accordingly increase above the theoretical minimum or
baseline value. System details causing the ratios to increase would include the three-
dimensional layout, large unsupported masses, high-density contents in thin-walled pipe,
susceptible branch connections, changes of cross section, etc. The more “unfavorable "the system
layout and details are, the larger the  / v ratios for some modes will be.
Thus, the general susceptibility of a system to large dynamic stresses can be assessed by
determining the extent to which the  / v ratios for any mode exceed the baseline range.
Furthermore, by determining which particular modes have the high ratios, and whether these
modes are known or likely to be excited, the at-risk vibration frequencies and mode shapes
are identified for further assessment and attention. This is the basis of the Stress / Velocity
355
method of analysis and its implementation as the “dynamic susceptibility” feature in

CAEPIPE.
2.3 Relation to Velocity-based Vibration Acceptance Criteria
There are various general and application-specific acceptance criteria based upon vibration
velocity as the quantity of record. Some, in order to cover the worst case scenarios, are
overly conservative for many systems. Others are presented as being applicable only to the
first mode of simple beams, leading to the misconception that the stress / velocity
relationship does not apply at all to higher modes. In any case, there are real and perceived
limitations on the use of screening acceptance criteria based upon a single value of vibration
velocity.
The dynamic susceptibility method turns this apparent limitation into a useful
analytical tool! Specifically, large stress / velocity ratios, well above the baseline values, are
recognized as a “warning flag.” Large values indicate that some feature(s) of the system make
it particularly susceptible to large dynamic stresses in specific modes.
3 What the Dynamic Susceptibility Method Does
3.1 General Approach
The Dynamic Susceptibility method is essentially a post processor to fully exploit the modal
analysis results of the system. Mode shape tables of dynamic bending stress and vibration
velocity are searched for their respective maxima. Dividing the maximum stress by the
maximum velocity yield the “/v ratio” for each mode. That ratio is the basis for assessing
the susceptibility to large dynamic stresses. Larger values indicate higher susceptibility
associated with specific details of the system.
3.2 Specific Implementation in CAEPIPE
The Stress / Velocity method has been implemented as additional analysis and output of the
CAEPIPE modal analysis. The modal analysis load case now includes additional outputs and
features as follows:
Dynamic Stresses This output provides the “mode shapes” of dynamic bending
stresses, tabulated along with the conventional mode shape of
vibration magnitude.
Dynamic Susceptibility This output is a table of s/v ratios, in psi / ips, mode by
mode, in rank order of decreasing magnitude. In addition to
modal frequencies and s/v ratios, the table also includes the
node locations of the maxima of vibration amplitude and
bending stresses.
With the dynamic susceptibility output selected, the animated
graphic display of the vibration mode shape includes the
added feature of color spot markers showing the locations of
maximum vibration and maximum dynamic bending stress.
These outputs will assist the designer through a more-complete understanding of the
system’s dynamic characteristics. They provide incisive quantified insights into how specific
details of components, layout and support could contribute to large dynamic stresses, and
into how to make improvements.
356
4 What the Dynamic Susceptibility Method Does Not Do Directly

The Stress / Velocity method of assessment, and its implementation in CAEPIPE as
dynamic susceptibility, is based entirely upon the system’s dynamic characteristics per se. Thus the
vibration velocities and dynamic stresses employed in the analysis, although directly related
to each other, are of arbitrary magnitude. There is no computation of the response to a
prescribed forcing function, and no attempt to calculate actual dynamic stresses. Thus the
dynamic susceptibility results do not factor directly into a pass-fail code compliance
consideration. Rather, they assist the designer to assess and reduce susceptibility to large
dynamic stresses if necessary, in order to meet whatever requirements have been specified.
An example follows next.
5 Illustrative Example of the Dynamic Susceptibility Analysis
The “dynamic susceptibility” feature of CAEPIPE will be illustrated here by application to
the standard CAEPIPE Example system. The modal analysis was performed for frequencies
up to 200 Hz, resulting in a reporting-out for 12 modes.
The frequencies range from mode 1 at 14.5 Hz to mode 12 at 192 Hz. In two instances, very
similar horizontal and vertical modes appear in pairs, i.e. modes 3 & 4 and 7 & 8.
The relevant features of this system can be readily identified and understood, by reference to
the dynamic-susceptibility table and the animated graphic display of mode shape. Results will
be considered here in the order of decreasing susceptibility.
357
5.1 Axial movement of long pipe run (large added mass in motion)
From the dynamic susceptibility table, the top of the list is mode 2 at 20.8 Hz, having a
dynamic susceptibility of 651 psi / ips.
From the animated graphic display, note that the maximum dynamic bending stresses are at
the anchored point, node 50. Note also that the dominant motion is a “Z” motion of the
straight run between nodes 20 and 40 (i.e., in effect an axial motion of that run as a rigid
body). The designer’s interpretation here is that the vertical rise from node 50 to node 40 is
effectively a cantilevered beam with an effective large added mass at the tip; that feature of
layout accounts for the high susceptibility.
5.2 Effects associated with the valve (local rigidity to bending, and added mass)
The next-highest values of susceptibility are for the two pairs of modes, modes 7 & 8 at 129
and 133 Hz, and modes 3 & 4 at 27.7 and 31.2 Hz. As will be shown here, these are
associated with effects of the valve,
358
The susceptibility for modes 7 & 8, respectively 594 and 589 psi / ips, is attributable to the
rigidity of the valve element within an otherwise flexible pipe run. This can be seen from a
close look at the animated graphic. Notice that these relatively high frequency modes feature
a reversal of bending curvature along the run between nodes 30 and 80. Notice also that
there is a stronger localized curvature on approach to the valve body. The designer’s
interpretation here is that, since there cannot be any curvature of the rigid valve itself, there
must be a more concentrated curvature of the adjacent pipe.
The dynamic susceptibility of modes 3 & 4, respectively 522 and 526 psi / ips, is associated
with the more straightforward “concentrated mass” effect of the valve.
From the animated graphic, these modes feature a large amplitude vibration at the valve. The
kinetic energy of this added mass must be stored as strain energy in the flexing (i.e., spring)
element, resulting in elevated dynamic stresses.
5.3 Beam modes with “moderate” added mass effects of adjacent spans
Modes 1,5 and 6, with frequencies of 14.5, 47.4 and 52.4 Hz, show progressively decreasing
“intermediate to low” values of susceptibility at respectively 458, 384 and 339 psi / ips.
359
Reference to the animated graphics shows that these modes involve predominantly transverse
vibration (as contrasted with the prominent axial movement of mode 2) and involve little
participation at the valve (which accounted for the elevated susceptibility of modes7 & 8 and 3
& 4). Notice that these modes, 1, 5, and 6, involve varying degrees of the influence of
effective added mass of adjacent spans, and of length of the cantilevered span contributing
most to stiffness.
5.4 Modes approaching the “simple-beam baseline” behavior
Modes 10, 11 and 12 have significantly higher frequencies, 164 to 192 Hz, and
correspondingly short wavelengths. Consequently, the vibration pattern tends to be
transverse beam vibration “within the span,” with little or no effect from connected spans or
the valve. For these modes, the susceptibility ratios range from 256 to 272 psi / ips. These
values are approaching the baseline values for uncomplicated mode shapes of the pipe
section and pipe contents of this system.
360
NOTE: Mode 9, at 138 Hz, is clearly an exception, with a susceptibility of only 104 psi / ips,
well below the baseline level. From the animated display, it can be seen that this is not really a
“bending” mode; rather, the spring effect for this mode is an axial stretching of the run
between nodes 80 and 30. Consequently, the bending stresses are low, as reflected in the
abnormal susceptibility ratio. In effect, this mode lies outside the intended application of the
dynamic susceptibility approach. Notice however, that the low susceptibility ratio has in
effect “flagged” this mode as “not a bending mode”; that in itself provides the designer
additional insight into system characteristics and behavior.
5.5 Summary Comment
As per paragraphs 5.1 to 5.4, the dynamic susceptibility method has incisively identified the
key features of the Sample model, with respect to potentially large dynamic stresses. This of
course is a relatively simple system. An experienced designer, with some appreciation of
dynamics, might view the results as obvious. However, the method will do the same job,
automatically and directly, on any larger or more complex system for which nothing is
obvious!
6 Summary of “Dynamic Susceptibility” Analytical Capability
The stress / velocity method, implemented in CAEPIPE as the “Dynamic Susceptibility”
feature, provides quantified insights into the stress versus vibration characteristics of the
system layout per se.
In particular, the dynamic susceptibility table identifies specific modes that are susceptible to
large dynamic stresses for a given level of vibration. The larger the stress / velocity ratio, the
361
stronger the indication that some particular feature of layout, mass distribution, supports,
stress raisers, etc., is causing susceptibility to large dynamic stresses.
The animated mode-shape display identifies, by the color-spot-markers, the locations of the
respective maxima in dynamic stress and vibration velocity. Review of these animated plots
will reveal the offending pattern of motion, and provide immediate insight into what features
of the system are responsible for the large dynamic stresses.
Finally, the “dynamic stresses” table provides the distribution of dynamic stresses around the
system, i.e., in effect, the mode shape of dynamic stresses to go along with the conventional
mode shape of vibration. This information allows identification of other parts of the system,
if any, with dynamic stresses comparable to the identified maximum.
7 Suggested Applications and Associated Benefits
7.1 At the Design Stage
At the design stage, the dynamic susceptibility feature allows the designer to quickly
determine whether the system may be susceptible to very large dynamic stresses. This could
be a broad look at all frequencies, or could be focused on particular frequencies where excitation is likely
to occur. On identifying high susceptibility, the designer can then make changes to improve
the design. It is important to note that this method is based upon the dynamic-stress versus
vibration-velocity characteristics of the system per se. There is no need to specify a forcing
function and perform a response calculation and stress / fatigue analysis. However, where such
analysis is a requirement, the dynamic susceptibility module can assist the designer to achieve a
system layout that will meet the requirements and criteria.
7.2 Commissioning, Acceptance Testing
The dynamic susceptibility feature can also contribute to planning acceptance testing and
associated measurements where these are undertaken whether by formal requirement or by
choice. Locations for measurement of vibration or dynamic strain can be selected based
upon knowing the locations of the maxima and the distribution of vibration and dynamic
stress. Reference to the dynamic susceptibility results can help assure that the modes of most
potential concern are well covered by the minimum set of practically-achievable measurements. Furthermore,
mode-specific acceptance criteria can be readily established to avoid the restrictions of generally
over-conservative guideline type criteria, while providing assurance that any highly-susceptible situations are
identified and addressed.
7.3 Troubleshooting and Correction
As mentioned earlier, when vibration and/or fatigue problems are recognized at start up or
early operation, there is typically an ad hoc program of observation, measurement,
assessment, diagnosis and correction. It is not uncommon for there to be some uncertainty
about what to measure and what is acceptable. The dynamic susceptibility module can
contribute very effectively in these situations.
Normally, the overall symptoms, approximate frequency and pattern of vibration are known
to some extent from observation and/or a few measurements. After modeling the system,
and obtaining the dynamic susceptibility results, the subsequent steps can be highly focused on
specific frequencies and locations, the optimum measurements, and system-specific acceptance
criteria.
362
Equally or more importantly, the proposed solution options can be modeled and evaluated to make
sure they will achieve the required improvement.
7.4 General
The dynamic susceptibility module does not apply directly to meeting code or other formal stress analysis
requirements. However, it is an incisive analytical tool to help the designer understand the
stress / vibration relationship, assess the situation and to decide how to modify the design if
necessary. It can be used for design, planning acceptance tests, and troubleshooting and
correction.
8 Information for Reference
The Stress /Velocity method for screening piping system modes was developed and brought
to the attention of SST Systems by Dr. R.T Hartlen of Plant Equipment Dynamics Inc.
The background material provided here is intended to provide only a concise summary of
the underlying fundamentals, the universality for idealized systems, and the expected detail-
dependent variations for real systems. The stress / velocity method, although not yet widely
known and applied, is fundamentally theoretically sound. However, complete theoretical
rigor is beyond the scope of this note.
For users who may wish to independently examine and validate the underlying theoretical
fundamentals, a few key references are provided. References 1, 2 and 3 deal with
fundamentals. References 5 and 6 deal with application to piping. The CEA research projects
reported in References 3 and 4 were initiated and guided by Dr. Hartlen.
References
1. F.V. Hunt, Stress and Strain Limits on the Attainable Velocity in Mechanical Systems,
JASA, 32(9) 1123-1128, 1960
2. E.E. Ungar, Maximum Stresses in Beams and Plates Vibrating at Resonance, ASME
Journal of Engineering for Industry, v84, n1, pp149-155, 1962
3. R. Elmaraghy et al, Correlation of Vibratory Stress, Velocity and Sound, Canadian
Electrical Association Project, G197, Feb 1982
4. J.D. Tulk, Correlation Between Dynamic Stress and Vibration Velocity in Complex
Piping Systems, Canadian Electrical Association Project G521, March 1988
5. Michael P. Norton, Acoustically Induced Structural Vibration and Fatigue - A
Review,
6. Third International Congress on Air-and Structure-borne Sound and Vibration, June
1994, Montreal, Canada
363
This page is blank
Annexure II
Nozzle Stiffness Calculations
Nozzle Stiffness Calculations (WRC 297)
Six stiffnesses are shown below at the nozzle-vessel interface — three are calculated and the
other three are assumed rigid.
The coordinate system is as shown in the figure. The six components of the forces and
moments at the nozzle-vessel interface are:
P = Radial load MC = Circumferential moment
VC = Circumferential load MT = Torsional moment
VL = Longitudinal load ML = Longitudinal moment
Of the six components of stiffnesses, only three stiffnesses, axial (Kx), circumferential (Kyy),
and longitudinal (Kzz), are calculated. The remaining three are assumed to be rigid.
Several graphs are given at the end of this annexure. The stiffness coefficients are obtained
by interpolating logarithmically from these graphs.
The first two, Figures D-1 and D-2, are used to calculate nozzle stiffness coefficients for
Nozzles on cylindrical vessels. Figure D-1 is used to calculate the axial stiffness coefficient
and Figure D-2 is used to calculate circumferential and longitudinal stiffness coefficients.
366
Nomenclature
𝐷 = mean diameter of vessel
𝑑 = outside diameter of nozzle
𝑇 = thickness of vessel
𝑡 = thickness of nozzle
λ = (𝑑/D)√𝐷/T
Λ = 𝐿/√𝐷𝑇
𝐿 = unsupported length of cylinder
2
= 8𝐿1 𝐿2 /(√𝐿1 + √𝐿2 )
𝐿1 = distance from nozzle center line to vessel end
𝐿2 = distance from nozzle center line to vessel end
𝐸 = modulus of elasticity of vessel material
Axial Stiffness(𝑲𝒙 )
4.95𝐸𝑇2
𝐾𝑥 = 𝛼 × (1)
𝐷√𝛬
where
𝛼 = stiffness coefficient read from Figure D-1
Circumferential Stiffness(𝑲𝒚𝒚 )
𝐾𝑦𝑦 = 𝛽 × 𝐸𝑇3 (2)

where
𝛽 =1stiffness coefficient read from Figure D-2.
The bottom three curves in Figure D-2, marked Circumferential moment MC are used to
find𝛽.
Longitudinal Stiffness(𝑲𝒛𝒛 )
𝐾𝑧𝑧 = 𝛾 × 𝐸𝑇3 (3)

where
γ = stiffness coefficient read from Figure D-2.
The top two curves in Figure D-2, marked Longitudinal moment ML are used to findγ.
367
Calculation of Nozzle stiffnesses for Nozzles on Flat-bottom tanks

This procedure is similar to the previous one.
As before, only three stiffnesses are calculated as the other three are assumed to be rigid. The
ones that are calculated are axial(𝐾𝑥 ), circumferential(𝐾𝑦𝑦 ), and longitudinal (𝐾𝑧𝑧 ).
For Nozzles on flat-bottom tanks, twelve graphs are given at the end of this annexure,
Figures D-3 through D-14. Six are for “with reinforcing pad (on vessel)” with the other six
for no reinforcing pad on the vessel. The stiffness coefficients are obtained using the
appropriate graph.
368
Nomenclature
𝑅 = Mean radius of vessel
𝑡 = thickness of vessel
2𝑎 = outside diameter of nozzle
Axial Stiffness(𝑲𝒙 )
𝐾𝑥 = 𝐾𝑅 × 𝐸 × (2𝑎) (4)
where
𝐾𝑅 = axial stiffness coefficient.
Circumferential Stiffness(𝑲𝒚𝒚 )
𝐾𝑦𝑦 = 𝐾𝐶 × 𝐸 × (2𝑎)3 (5)

where
𝐾𝐶 = circumferential stiffness coefficient.
Longitudinal Stiffness(𝑲𝒛𝒛 )
𝐾𝑧𝑧 = 𝐾𝐿 × 𝐸 × (2𝑎)3 (6)
where
𝐾𝐿 =longitudinal stiffness coefficient.
The graphs for stiffness coefficients follow:
369
Figure D-1:Stiffness coefficient for axial load on nozzle
370
Figure D-2:Stiffness coefficients for moment loads on nozzle
371
Figure D-3:Stiffness coefficient for axial load (with reinforcing pad)(L/2a = 1.0)
Figure D-4:Stiffness coefficient for circumferential moment (with reinforcing

pad)(L/2a=1.0)
372
Figure D-5: Stiffness coefficient for longitudinal moment (with reinforcing pad)(L/2a = 1.0)
Figure D-6:Stiffness coefficient for axial load (with reinforcing pad)(L/2a = 1.5)
373
Figure D-7:Stiffness coefficient for circumferential moment (with reinforcing pad)

(L/2a = 1.5)
Figure D-8:Stiffness coefficient for longitudinal moment (with reinforcing pad)(L/2a=1.5)
374
Figure D-9:Stiffness coefficient for axial load (no reinforcing pad)(L/2a = 1.0)
Figure D-10:Stiffness coefficient for circumferential moment (no reinforcing pad)(L/2a=

1.0)
375
Figure D-11:Stiffness coefficient for longitudinal moment (no reinforcing pad)(L/2a = 1.0)
Figure D-12:Stiffness coefficient for axial load (no reinforcing pad)(L/2a = 1.5)
376
Figure D-13:Stiffness coefficient for circumferential moment (no reinforcing pad)

(L/2a = 1.5)
Figure D-14:Stiffness coefficient for longitudinal moment (no reinforcing pad)(L/2a=1.5)
377
Index
378
reducing, 35
A supported by a hanger, 37
flexibility factor, 27
absolute sum, 80, 81
acceleration, 195, 310, 311 intermediate nodes, 28
spectral, 195 long radius, 26, 27
material, 27
vector, 77, 81, 82, 195
radius, 27
additional weight, 19, 168, 171, 287, 337, 338
short radius, 26, 27
anchor, 3
displacements, 4 SIF, 28, 35
releases for hanger selection, 4 tangent intersection point, 26
thickness, 27
settlement, 5
bottomed-out springs, 313
stiffness, 3
angle branch connection
valve, 341 nonstandard, 336
branch line, 48, 131
animate
branch SIF, 47
mode shape, 83
buried piping, 50
ANSI pipe sizes, 27, 292
API, 68, 246 general procedure, 50
API 610 report, 246, 253, 255 ground level, 53
API 617, 246 nomenclature, 54
API 617 report, 68
API 650, 211, 215 C
axes
elastic element, 85 circumferential
local beam, 13 joint factor, 186, 188, 192
local coordinate system, 172 stiffness, 212, 366, 367, 368, 369
axial force, 109 closely spaced modes, 81
axial stiffness, 210 cold load, 140
cold spring, 65
comment, 67
B compressor, 68
ball joint, 9 concentrated mass, 71
friction, 207 cone angle, 256, 257
connected node, 160, 289
rotation limit, 10
hanger, 139
stiffness, 9
beam, 12 skewed restraint, 296
additional weight, 19 constant support, 72
user defined hanger, 323
AISC library, 16
beta angle, 13 coordinates, 131, 206
end releases, 13 core pipe, 151, 155, 175, 304
load, 18 core properties, 154
local coordinate system, 20 corrosion allowance, 291
cut pipe, 65
material, 13
orientation, 20
section, 14 D
bellows, 24
stiffness, 24 data types, 73
tie rods, 318 density, 192
bend, 26 design
examples, 28 hanger, 138
180° bend, 31 pressure, 170
45° bend, 30 temperature, 170
90° bend, 29 DIN pipe sizes, 27, 292
base supported, 39 direction, 74
flanged, 32 displacement vector, 77, 80
379
Index
displacements force, 124

anchor, 4 force spectrum, 125
nozzle, 212 convert time function, 128
dynamic analysis, 77 load, 129
closely spaced modes, 81 read from a text file, 128
effective modal mass, 80 frequencies
friction, 208, 210 closely spaced, 81
modal analysis, 77 frequency
modal equations, 78 circular, 79, 195
orthogonality, 78 natural, 355
participation factors, 79 friction, 208
response spectrum, 80 ball joint, 210
support motion, 79 guide, 209
time history, 81 hinge joint, 209
dynamic susceptibility, 83 in dynamic analysis, 208, 210
environment variable, 83 limit stop, 208
method, 354 slip joint, 209
friction coefficient, 136, 160, 207, 313
E friction force, 136, 158, 160, 208, 300
friction torque, 9, 144, 208, 209, 300
effective modal mass, 80 From node, 131
eigenvalues, 77
eigenvector, 77
normalized, 195
G
elastic element, 85 gasket diameter, 106
stiffness, 85 generic support, 133
elbow, 26 global
element forces and moments coordinate system, 172
in local coordinate system, 174 origin, 53
element stresses guide
FRP, 101 friction, 209
environment variable friction coefficient, 136
CPITER, 210 gap, 137
dynamic susceptibility, 83 local coordinate system, 137
HARTLEN, 83 stiffness, 137
exit, 210
expansion joints, 24, 86
bellows, 86, 88, 91
H
tied bellows, 86, 88, 91 hanger, 138
below, 139
F catalog, 140, 312
cold load, 139
fiber reinforced plastic piping (FRP), 101, 102 connected node, 139
flexibility factor, 102 constant support, 72
moduli, 102 design procedure, 140
SIF, 102 hot load, 140
siffness matrix, 102 load variation, 140
flange, 105 manufacturer, 140
allowable pressure, 106 number of, 139
equivalent pressure, 106 report, 289
flange and bolt stresses, 112 rod, 289
gasket diameter, 106 short range, 139
library, 108 stiffness, 139
rating, 106 to be designed, 138
flange and bolt stresses, 112 travel, 140, 208, 313
flange module, 112 type, 139
flexibility factor user, 323
bend, 27 harmonic analysis, 82
FRP, 102 hinge joint, 143
miter bend, 199 angular stiffness, 144
380
Index
axis direction, 145 static seismic-ASCE/SEI 7-16, 305

example, 146 temperature, 170
friction, 209 wind, 171
friction torque, 144 load cases
rotational limit, 144 cold spring, 65
rotational stiffness, 144 force spectrum, 129
Weight, 145 hydrotest, 150
hydrotest loads
load, 149 acceleration, 311
load case, 150 force spectrum, 126
local coordinate system, 172
I element forces and moments, 174
for a beam, 20
input for a bend, 174
new material, 188 for a guide, 137
insulation, 293 for a pipe, 172
density, 293 local forces, 174
thickness, 293 location, 175
intermediate nodes, 28 logarithmic interpolation, 80, 366
internal nodes, 155 long radius, 26, 27
longitudinal
joint factor, 188, 192
J pressure stress, 192
jacket end cap, 151 longitudinal stiffness, 212, 366, 367, 369
jacket pipe, 185
jacket properties, 154 M
jacketed bend, 153
internal nodes, 155 material
jacketed pipe, 152 create library, 17, 192
internal nodes, 153 define, 185
jacketed piping density, 192
jacketed bend, 153 description, 191
jacketed reducer, 155 input, 188
JIS pipe sizes, 27, 292 name, 191
joint factor, 192 selectr from library, 190
circumferential, 192 type, 191
longitudinal, 188, 192 missing mass correction, 195
miter bend
closely spaced, 199
K flexibility factor, 199
knuckles modeling procedure, 200
reducer, 257 parameters, 199
widely spaced, 198, 199
modal analysis, 77
L modal displacements, 78
lateral tee, 49
left out force, 195 N
limit stop, 158
friction, 208 node, 206
nonlinearities, 207 internal, 155
solution procedure, 207 list coordinates, 206
lining, 293 specifying coordinates, 206
load nominal diameter, 292
additional weight, 171 nonlinearities, 207
Design pressure, 171 ball joint, 210
Design temperature, 171 friction, 208
pressure, 170 guide, 209
snow, 171 hinge joint, 209
specific gravity, 171 limit stop, 208
static seismic-ANSI A58.1-1988, 310 misconvergence, 210
381
Index
number of iterations, 210 reducing tee, 48

slip joint, 209 releases for hanger selection, 3
types, 207 relief valve, 341
nonstandard report
branch connection, 336 flange, 110
pipe, 292 response spectrum, 80, 216
nozzle, 211 restraint, 286
axial stiffness, 212, 366, 367, 369 stiffness, 286
circumferential stiffness, 212, 366, 367, 369 rigid body force, 195
coordinate system, 212, 366 rigid element, 287
displacements, 215 stiffness, 287
example weight, 287
on a cylindrical vessel, 216, 217 rod hanger, 289
on a flat-bottom storage tank, 218 stiffness, 289
longitudinal stiffness, 212, 366, 367, 369 rotation limit
on a cylindrical vessel, 211 ball joint, 10
reinforcing pad, 215 hinge joint, 144
settlement, 216
stiffness calculations, 366
stiffness coefficients, 366
S
Nozzle Evaluation, 221 section, 291
NRC Guide 1.92, 81 ANSI pipe sizes, 292
number of modes, 195 corrosion allowance, 292
number of thermal loads, 19, 168 DIN pipe sizes, 292
insulation, 293
O JIS pipe sizes, 292
lining, 293
occasional load, 130 mill tolerance, 292
name, 292
P nominal diameter, 292
nonstandard, 292
pad thickness, 47 schedule, 292
participation factor, 79, 195 section modulus, 292
period, 80 settlement
pipe skirt, 313 anchor, 5
pipe slide assembly, 313 nozzle, 216
pipe slide/shoe assembly, 163 shear
Poisson’s ratio, 192 area, 15
pressure deflection, 15
gauge, 170 shear deflection, 15
negative, 170 short radius, 26, 27
pump, 246 SIF
branch, 47
FRP, 102
R reducer, 257
reducer, 256 user, 336
concentric, 256 weld, 342
example, 258 sign conventions, 174
cone angle, 256, 257 skewed restraint, 295
delta, 258 connected node, 297
eccentric, 256 direction, 296
example, 259 example
example sway brace, 297
concentric, 258 rotational, 296
eccentric, 259 stiffness, 296
jacketed, 261 translational, 296
knuckles, 257 type, 296
SIF, 257 slip joint, 300
stresses, 258 friction, 209
weight, stiffness and stress calculation, 258 friction force, 300
382
Index
friction torque, 300 Tee, 315

pressure thrust area, 300 tensile strength, 192
snow load, 19, 168, 171 thermal loads, 4, 168, 216
snubber, 303 tie rod, 318
stiffness, 303 stiffness, 318
specific gravity, 171 time history, 81
spider, 304 turbine, 320
spring rate, 72, 140, 289, 323
SRSS, 80
stiffness
U
anchor, 3 u-bolt, 313
ball joint, 10 user hanger, 323
bellows, 24 constant support, 323
elastic element, 85 load, 323
guide, 137 spring rate, 323
hanger, 139 stiffness, 323
hinge joint, 144 user SIF, 336
limit stop, 160 u-strap, 313
nozzle, 211, 366
reducer, 258
restraint, 286 V
rigid element, 287 valve, 337
rod hanger, 289 additional weight, 338
snubber, 303 angle, 341
tie rod, 318 calculation of moment of inertia, 341
user hanger, 323 insulation weight X, 338
valve, 338 length, 337
stiffness matrix, 77, 81, 82, 140, 207, 208 library, 339
FRP, 102 relief, 341
stresses stiffness, 338
FRP, 103 thickness X, 338
occasional, 292 weight, 337
reducer, 258 velocity vector, 81
sustained, 292
support motion, 79
supports W
anchor, 312 weight multiplier, 341
generic, 133 weld, 342
guide, 313 wind, 343
hangers, 312 wind load, 171
limit stop, 313 WRC 297, 211
restraint, 312
skewed restraint, 312
sway brace, 295, 297, 312 Y
sway struts, 295, 312
Y factor, 191
T Z
tangent intersection point, 26, 174
zero length element, 9, 143
tee
SIF, 47
383

CAEPIPE Technical Reference Manual

Uploaded by

Copyright:

Available Formats

CAEPIPE Technical Reference Manual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CAEPIPE Technical Reference Manual

Uploaded by

Copyright:

Available Formats

™

For Info and Support on CAEPIPE, contact:

SST Systems, Inc.

SECTION .............................................................................................................................................. 291

Example 2: Rigid Vertical Support with Foundation Settlement

The Ball joint dialog is shown.

Kbe = (applied moment/rotation) …….. (2)

The Beam dialog is shown.

A dialog for inputting beam material is shown.

A dialog to input beam sections is shown.

To create or modify a Beam section library

A List window for Beam section is shown.

A list of beam loads is shown.

Global vertical axis is Z

Example 1: Pipe Rack using Beams

The “Generate” dialog is shown below:

The “Limit Stop” dialog is shown below:

The graphics is shown below:

Example 2: Base Supported Bend using a Beam

The Bellows dialog is shown.

Example 1: 90° Bend

Example 2: 45° Bend and Pipe routing along a smooth curve

Example 3: 180° Bend

Example 4: Flanged Bend

The graphics is shown below:

The rendered graphics is shown below:

Example 5: Reducing Bend

The graphics is shown below:

The rendered graphics is shown below:

Example 6: Bend Supported by a Hanger

Click on Hanger, the hanger dialog is shown.

The rendered graphics is shown next.

Example 7: Base Supported Bend

Vertical Restraint Support

Double click on Restraint. This will open the Restraint dialog.

The graphics is shown below:

The rendered graphics is shown below:

Example 8: Circular Piping

Ring Radius = Bend Radius (in) 96 Bend Angle 22.5

R' = Absolute Coordinates Offset Distances

Layout using Absolute Coordinates

Layout using Offset Distances

The Branch SIF dialog is shown.

The Tee is as shown at node 30 in the graphics window below.

Two Soil types

Cohesionless soil: Downward load = D × (Soil density × bottom depth × Nq × dq

Dialog for cohesive soil:

Soil S2 is associated with section 12B:

k  Ck N h wD = 80 x 7.518 x 0.075 x 12.75 = 575.127 psi

The Cut pipe dialog is shown.

The Concentrated Mass dialog is shown.

The Constant Support dialog is shown.

or press Ctrl+Shift+D to open the Data Types dialog.

𝐿 = √(X2 − 𝑋1)2 + (Y2 − 𝑌1)2 + (Z2 − 𝑍1)2 = 5.385′

cos2α + cos2β + cos2γ = 0.5570922 + 0.7427822 + 0.3713922 = 1.0

Premultiplying (4) by {an }T , {an }T [K]{am } = ω2m {an }T [M]{am } (6)

which is the orthogonality condition for eigenvectors.

It may be shown that

[𝛷]𝑇 [𝑀][𝛷]{𝐴̈𝑛 } + [𝛷]𝑇 [𝐾][𝛷]{𝐴𝑛 } = [𝛷]𝑇 {𝐹(𝑡)} (15)

Substituting from equations (12) and (13) in equation (15),