Piping Stress Analysis with CAESAR II
Piping Stress Analysis with CAESAR II
2
What is this course about
- The course provides the participants with training on performing piping stress
analysis using CAESAR II software.
- In addition, and to achieve the above properly, the course will cover the basic
background of the piping stress analysis field and the related codes and standards.
3
Introduction to Piping Stress Analysis
Governing Code Requirements contain the allowable values set for each of the
above mentioned parameters.
4
Introduction to Piping Stress Analysis
Why do we perform Piping Stress Analysis?
5
Types of Stresses
Radial (Hoop Stress) Due to Pressure
Wall Thickness Calculation
PD
S
2t
PD
t min
2S
Tension/Compression Due to Bending Tension/Compression Due to Pressure
(Tension)
(Compression) F=PA
M M PD
S i i = SIF
Z Z (Tension) S 6
Z = Section Modulus 4t
D 2 t
Z
4
Categories of Stresses
7
Loads Cause Pipe Loads Cause
to Rupture Fatigue Failure
PIPING CODES
B31.1 Power Piping
B31.2 Fuel Gas Piping
B31.3 Chemical Plant and Petroleum Refinery Piping
B31.4 Liquid Petroleum Refinery Piping
B31.5 Refrigeration Piping
B31.8 Gas Transmission and Distribution Piping Systems
B31.9 Building Services Piping
B31.11 Slurry Transportation Piping Systems
Section III, NB/NC/ND-3600 Nuclear Power Piping
8
Stress due to Sustained Loads
(primary stresses)
PDo 0.75iM A
1.0 S h
4t n Z
Where
• P = internal pressure Do = Pipe outside diameter
• tn = nominal pipe wall thickness Z = section modulus
• i = stress intensification factor Sh = allowable stress (hot)
• MA = resultant moment loading due to sustained loads
9
Stress due to Occasional Loads
(primary stresses)
Where
• Sustained part = same
• MB = resultant moment loading due to occasional loads
10
Stress due to Thermal Expansion
(secondary stresses)
iM C
S A
Z
iM C
OR S A f ( S h S L )
Z
Where
• i = stress intensification factor
• MC = range of resultant moments due to thermal expansion
11
SA = 1.25Sc + 0.25Sh
Allowable Stress Curves
• A106-Gr.C
• A335-P11
• A335-P22
• A335-P91
• A312-TP304
Materials
12
Differences in Material Strength
14
Thermal Expansion Table
B31.1 Appendix B
15
Example
• 4” sch.80, A335-P22
Example Problem
17
Guided Cantilever Example
The layout shown can be idealized as two guided cantilevers each of length L and each absorbing
half of the thermal expansion between the anchors.
Example:
Given: Material = A 312 TP304
Operating temperature, T0 = 8000°F
=8.80 in/100 ft
Outside diameter, D = 10.75 in
Pipe Isometrics
Information
P&ID-Pipe Sizes
Gathering Design Conditions
Class sheets
Vendor Information
Wind/ Seismic Criteria
Math Model
Preparation
Preliminary Stress
Analysis
Check Results NO
Pipe Stress
Nozzle Loads
22
YES
Send Loads To Pipe
Final Stress Analysis Support/Civil Groups
Send Loads To Vendors
Math Model
• Pipe Properties
• geometry
• diameter, schedule
• material
• insulation, fluid
• Stress Intensifications
• Boundary Conditions
• Pipe supports
• Equipment anchors
• Loads
• pressure
• thermal expansion
• wind/ seismic
23
Analyzed vs. Non-analyzed
• Based on Temperature
CS & Low Alloy > 400F (2.7 in/100ft)
SS > 300F (2.6 in/100ft)
• Based on Importance
Sensitive Equipment: (Feedwater, Ext. Steam, Fuel Gas)
Non-standard components: (Relief valves, Expansion joints)
Headers or Equipment with large movements
• Based on Unique Criteria
Cross-country, Yard Piping (solar radiation)
Fiberglass, Nozzles connected to Large Tanks
Underground greater than 150°F
24
CHAPTER 2 – SYSTEM MODELING
25
Introduction to CAESAR II system modeling
Element Types
- Delta Dimensions
- Pipe Element & Pipe Section Parameters
- Elbow Element
- Miter Element
- Rigid Element
- Expansion Joint Element
- Anchors
- Single and Double Acting Restraints
- Friction, Gap & Stiffness
- Connecting Nodes
- Translational & Rotational Restraints
- Modeling of Restraints on Bends
- Bi-Linear Supports
- Rigid Hangers
- Spring Hangers & Supports
- Snubbers
26
- Applied Displacements (consideration in the load cases)
- Restraints with Displacements & Stiffness
- Skewed Restraints
Piping Component and Rigid Elements
Element Dimensions & Node Coordinates
For each node, the differential dimension with respect to the previous node is
defined as delta dimension ( X, Y, Z).
Alternatively the length of the element and the direction cosines are entered.
For nodes which start a new disconnected line inside the same model, the global
coordinates are
entered for the first node.
Pipe Elements
The pipe element section properties are entered in the spread sheet as pipe outside
diameter or
nominal diameter , wall thickness, corrosion allowance, mill tolerance and insulation
thickness.
27
All pipe section data are carried forward for the subsequent elements until they are
changed.
Rigid Elements
The weight and length of the element are entered for rigid elements. For standard
flanges and valves which are available in the database, the weight and length can be
input directly by the program.
The stiffness of the rigid element is proportional to the rigidity of the pipe with the
entered diameter and thickness. Therefore, the diameter and thickness for the rigid
element should be considered in the definition of the element.
For the non-zero weight rigid elements, the program adds the weight of fluid as equals
to the fluid weight of an equal length of pipe and adds the insulation weight as 1.75 of
that of equal length of pipe.
For zero length rigid element, it is assumed that they are used for modeling purpose and
no fluid or insulation is added.
28
Modeling of Elbow
29
Modeling of Flanged Elbow
30
Modeling of Elbow with Trunnion
31
Modeling of Reducers
32
Restraint and Displacement
MODELING OF RESTRAINTS
The following are the types of restraints available in CAESAR II input spread sheet:
Anchor: ANC
Translation: X, Y, Z, +X, -X, +Y, -Y,+Z, -Z
Rotational: RX, RY, RZ, +RX, -RX, +RY, -RY,+RZ, -RZ
Guide: GUIDE
Limit Stop: LIM, +LIM, -LIM
Spring Support: XSPR, YSPR, ZSPR
Snubber: XSNB, YSNB, ZSNB, + XSNB, +YSNB, +ZSNB, -XSNB, -YSNB, -ZSNB
Translation Bilinear: X2, Y2, Z2
Rotational Bilinear: RX2, RY2, RZ2
Large Rotation Rod: XROD, YROD, ZROD
For each restraint we define the restraint node and an optional connecting node
(Cnode).
33
Anchors
They represent complete fixation of the point of support against translation &
rotation such as welded or bolted support.
Anchors can also be modeled as 6 restraints provided with stiffness to model the
different flexibility of the anchor in the different directions.
34
Translational Restraints
They can be single acting (directional), i.e. acting in +ve or –ve direction or
double acting, i.e. acting in both +ve and –ve directions.
They are always in the direction of the global axes. The restraints +X
represents the case when the support reaction acts in the +ve X-direction.
For each restraint, we can define three additional input data : coefficient of
friction, gap and stiffness.
The gap is always positive and it is taken to apply in both directions for the
double acting and to apply in single direction for the single acting.
The restraint symbol can be succeeded with the cosine angles for restraints
acting in inclined directions.
35
Rotational Restraints
Guide Restraint
36
Limit Stop
Limit stop is a restraint acting in the axial direction of the pipe. The restraint can be
double acting or single acting in the +ve or –ve directions. A gap can be applied as for
the translational restraint.
Spring Support
The restraints presents a spring with an initial force F, spring stiffness and a maximum
travel (deflection) x, after which the spring becomes rigid support. This is used for an
existing spring support.
For the case of the selection of a new spring support, the spring hanger data sheet is
used.
37
Snubber Restraint
A restraint acting only during the dynamic load cases only and free during
static load cases.
They represent flexible supports with variable stiffness. The stiffness can take
2 values K1 and K2. Fy is the yield load which corresponds with the force at
which the stiffness changes from K1 to K2.
C-Nodes
Applied displacement can be defined for a node as 9 vectors V1, V2, V3, … and each
vector has 6 components X, Y, Z, RX, RY, RZ, corresponding to the 3 translational and
3 rotational degrees of freedom at the node.
It is important to note that if a component is not defined (left blank) that means this
degree of freedom is free and not restrained. Therefore to apply zero displacement
at a node a zero has to be entered.
39
Model of Anchors
40
41
MODEL NONLINEARITIES
CAESAR II handles certain types of nonlinearities in the boundary conditions (model restraints).
These types are:
The solution of the system equations for nonlinear system is an iterative solution. It is important to
bear in mind that in many situations the iterative solution cannot reach a convergent solution and
for nonlinear problems, there might be even more than 1 solution where the software will give
only the first convergent solution.
The friction force direction will not only depend on the present system temperature but also on the
history of the displacement and the temperature variation. This explains the fact that for long lines
anchor force will be different for startup and shutdown scenarios for the same system operating
temperature.
42
Operating Conditions
The operating conditions of temperature and pressure are defined for each element. The
input values for the an element are carried forward for the subsequent elements until
they are changed. There can be up to 9 operating conditions.
Inside the temperature field the total thermal expansion (mm/mm) can be entered. The
program takes the value entered as expansion if it is less than Alpha Tolerance as defined
in the setup file.
43
44
CAESAR II input window
45
Input layout
46
First element Input
47
Element summary
48
Elbow screen
49
Rigid element screen
50
Valves data base
51
Valves data base
52
Reducer screen
53
Sif or Tee connections
54
Sif or Tee connections
55
Sif or Tee connections
56
Sif or Tee connections
57
Restraints
58
Restraints
59
Restraint with displacement
60
Nozzles
61
Concentrated Forces
62
Wind input
63
Uniform loads (Seismic)
64
Uniform loads (Seismic)
65
Allowable
66
Allowable
67
Allowable
68
Materials
69
Materials properties
70
Listing window for pipe element summary
71
Edit screen
72
Model menu
73
Execution parameter
74
Include piping files
75
Include piping files
76
Configuration setup
77
Configuration setup
78
Configuration setup
79
Configuration setup
80
Configuration setup
81
Configuration setup
82
CHAPTER 3 – STAIC ANALYSIS OUTPUT
AND SEQUENCE OF CHECK.
83
Load Cases Types and Definitions
Static analysis starts by the definition of the load cases to be considered in the analysis.
The load cases are built by combining the different applied loads and design conditions
in a set of load cases. The loads considered for combination are:
85
Load Combinations
The simple set of load cases, which can be obtained as a recommended for the main
code stress equations, are as follows:
87
For the occasional load cases, we consider the following:
88
The method of combination is algebraic as default. The required method of
combination is defined for each combined load case.
Algebraic: the default combination method for which the displacement and
the force vectors are added algebraically and then the stresses are
computed from the resultant.
89
LOAD CASES INPUT
90
91
92
93
System error check
94
95
The static analysis results report the following:
96
RESULTS REVIEW
97
98
99
100
101
102
Static Analysis Example
103
104
Modified Line Layout & Support
105
106
CHAPTER 4 – SPRING SUPPORTS
AND ADVANCED TOPICS.
107
SPRING SUPPORTS
Spring supports are used to support the weight of the piping system and at the
same time allow movement of pipe such as that due to thermal expansion.
By proper design of the spring support, pipe stresses and forces transmitted to
connected equipment are kept within allowable limits. Spring supports are used
• Even with limited thermal displacements, we may consider using spring supports
in rigid vertical runs to compensate for thermal displacement
equipment. And/or small errors in the fabrication and installation.
We should consider using rigid pipe hangers or supports at the points where the
pipe displacement due to thermal expansion is limited and the piping system is
flexible such that placing rigid support will not introduce large forces or stresses in
the system. 108
where:
There are 2 types of spring supports:
• Variable spring support: the most common type is the variable spring
type. For the variable type, the spring applies a variable reaction
force on the pipe due to pipe displacement.
109
110
111
Variable spring supports can be:
• Spring hangers
• Spring can supports.
The design Codes ASME B31.1 & B31.3 provide requirements for piping supports.
112
The following lists the basic requirements of ASME B31.1:
• Counter weights and hydraulic types can be used for the constant
support. They have to be designed so as to include safety devices
support.
and stops.
113
Many manufacturers provide wide variety of pre-engineered pipe
supports hardware including spring supports. These include:
• Lisega
• Carpenter & Paterson
• Pipe Supports Ltd.
• Piping Technology
• Inoflex
114
Generally there are different spring coil sizes with catalogue number to
indicate the coil size and the support form and details. For each coil size
there are:
• Short Length
• Standard Length
• Long (can be 2 and 3 times the standard)
The selection between the different types will be governed by the amount of
travel, the amount of acceptable load variation and the available space.
For the short spring, the spring rate is high and the amount of travel is
limited. For the longer types, the spring rate is lower and the allowable
travel is larger.
115
C II APPLICATION TO SPRING SUPPORT Refer to the example in
the next slide:
116
117
Hanger Design Control Data
118
119
Special Execution Parameters
120
Hanger Load Cases
121
122
Spring Support additional options
123
EXPANSION JOINT
Expansion Joints are used in piping systems to reduce expansion stresses, reduce piping
reactions on connecting equipment and to isolate mechanical vibration. Basically expansion joints
are used when it is necessary to provide a large amount of flexibility in a small space.
The main factor which influences the decision to use expansion joints is limited space and large
diameter low pressure piping.
There are two general styles of expansion joints; Bellow type, designed in accordance with
Appendix X of the ASME B31.3 Code and slip-joint type designed according to Article 304.7.4 of
ASME B31.3 Code. Slip type joint can absorb displacement in the axial and torsional direction,
while bellow type expansion joint can absorb displacement in axial, lateral and angular direction.
Bellows type expansion joint is the more commonly used type of expansion joints and is used in
many different types of arrangements as single, universal, hinged or gimbal joint.
Expansion joints are constructed out of sheet metal (mostly stainless steel, but can be any other
metallic material) through rolling in cylindrical shape with corrugations (also called convolutions)
and welding. The sheet is made of 1 or several layers (plies). The expansion joints element can
vary in terms of the material, the number of convolutions, the number of plies, the pressure
capacity, the stiffness and the allowable movement. 124
For most part, these details are taken care of by the expansion joint manufacturer. A typical
expansion joint piping design proceeds as follows:
- Estimation of the required amount of displacement in the different directions is made. This is
usually done by assuming that the expansion joint has zero stiffness and the free movement of
the piping is calculated.
- Based on the design temperature and pressure a standard expansion joint is selected from a
manufacturer catalogue.
- The properties of the selected expansion joint are inserted in the piping model and the
response of the system as stresses, forces and displacements is checked.
- For each bellows there is a limit to the cumulative axial, lateral and bending displacement that
can be absorbed. This limit is also related to the allowable number of stress cycles according to
the fatigue life of the element. The main factor in the design of the bellow element is the fatigue
life under the applied number of stress cycles.
- In case the applied cumulative displacement is above the bellows limit, increasing the number
of convolutions can be the solution.
125
BELLOW ELEMENT DISPLACEMNET
126
END PRESSURE FORCE
Bending stiffnesses from EJMA (and from most expansion joint manufacturers) that are to be
used in a finite length expansion joint model should be multiplied by (4) before being used in any
piping program. Bending stiffnesses from EJMA (and from most expansion joint manufacturers)
that are to be used in a ZERO length expansion joint model should be used without modification.
For completely flexible element stiffness is entered as 1 and for completely rigid element stiffness
is entered as 1 E+12
Expansion joint can be defined with zero length to simulate hinged connections.
Four stiffnesses are required to define the stiffness of the expansion joint:
Axial Stiffness
Transverse (lateral) Stiffness 129
Bending Stiffness
Torsional Stiffness
The transverse and the bending stiffnesses are directly related when a finite length joint is
defined. In this case the bending stiffness should be left blank and the transverse stiffness
entered.
When the expansion joint has a zero length, none of the expansion joint stiffnesses are related. In
this case the user must be sure that a value is entered into all four of the stiffness fields.
F = ka x = Ka X
Where:
N = number of convolutions
130
ka = N Ka
x =X/N
The lateral stiffness can be estimated approximately from the axial stiffness if not
given explicitly by the supplier:
Kt = 1.5 Ka (D / L )2
The approximate relationship between the bending stiffness (as used in the piping
analysis software) and the axial stiffness:
131
MODELING OF EXPANSION JOINT
Expansion joint modeler can be used to model the complete joint with all elements
(including the rigids) from different suppliers catalogues. The expansion joint element is
specified by specifying the delta dimension and the bellow element stiffnesses. The tie-rods
are modeled by use of rigid or pipe elements.
132
Expansion Joint Selection Example:
Line: 20” Std wall thickness
133
Solution:
Input Data
134
135
Expansion stress for the present configuration: 136
Expansion joint is to be inserted in the run from node 130 to node 160. This will
eliminate the need for pressure force anchors since the expansion joint will be tied
rodded and will take only lateral displacement.
137
Sizing of the Expansion Joint:
Release the anchor at node 160 and find the displacement at this node for the total
thermal expansion range.
Select the expansion joint which correspond to the required displ and number of cycles from the
standard catalogue.
For Flextronics catalogue, the maximum lateral displ for a single bellow element is 29.1 mm.
Select a universal tied expansion joint. The joint is made up of 2 bellow elements 4 corrugations
each. From the catalogue page 35, diameter DN 500, pressure 10.5 barg, and 4 corrugations:
The total length of the exp joint = 200 + 241 + 1000 + 241 + 200 = 1882 mm
Restraint between 141 and 144 to represent the tie rods as: Z, RX, RY
140
141
142
C II has a built in catalogue of Flextronics and other suppliers. The selected exp jt including
the type, number of corrugations, … is modeled.
It is important to note that the stiffness of the bellow element in the lateral direction is
much higher
than the stiffness of the whole joint. 143
The following are the screen of CII modeled Expansion Joint :
144
145
Equipment Code Compliance
Equipment Code Compliance
Where:
F= resultant force on the individual nozzle
M=resultant moment on the individual nozzle
De=nominal pipe size of the connection
Equipment Code Compliance
Fx<= 50Dc
Fy<= 125Dc
Fz<= 100Dc
Mx<= 250Dc
My<= 125Dc
Mz<= 125Dc
Fc+Mc/2<=125Dc
b) The resultant applied force (FRSA, FRDA) and the resultant applied
moment (MRSA, MRDA) acting on each pump nozzle flange shall
satisfy the equations below:
[FRSA / (1,5 × FRST4)] + [MRSA / (1,5 × MRST4)] ≤ 2
[FRDA / (1,5 × FRDT4)] + [MRDA / (1,5 × MRDT4)] ≤ 2
Equipment Code Compliance
• The forces and moments can be larger than 2 times the table 4
values provided that the principal stress in the nozzle does not
exceed 41 MPa. The principle stress is calculated based on a
pipe with same diameter as the nozzle and schedule 40.
P = / 2 + (2 / 4 + 2 ) 0.5 < 41 MPa
=F/A+M/z& =F/A+T/2z
Equipment Code Compliance
Equipment Code Compliance
Equipment Code Compliance
Equipment Code Compliance
Equipment Code Compliance
• The input for C II include the data of the suction and discharge
nozzles:
• Nozzle node number.
• Nozzle orientation ( top, side or end).
• Nozzle diameter.
• Nozzle distance from the base point.
• Nozzle loads, which can be obtained from the output file if
nozzle.
• node number is specified.
Equipment Code Compliance
• Discharge Lines
Equipment Code Compliance
• Suction Lines
Equipment Code Compliance
177
Feedwater Suction Consideration
178
Feedwater Discharge Design Considerations
179
Feedwater Discharge Design Considerations
180
Blowdown System Design Considerations
181
Blowdown
182
Small Bore Critical Design Considerations
183
Condensate System Design Considerations
184
Condensate Discharge
185
Fiberglass Piping Design Considerations
186
Underground Piping Design Considerations
187
Piping Connected to Large Tanks Design Considerations
188
Large Tanks
189
What is a water hammer or steam hammer
190
These photos are an example of the powerful forces generated by steam hammer. In this
case, the steam hammer occurred in a power plant, on a high temperature steam header, and
caused many pipe failures.
As a result the pipe moved off it’s supports and was displaced by as much as 1 m. The power
plant was out of service for 3 months in order to repair the damages.
191
How is water Hammer caused
PA PA P2A P1A
F=0 F=P1A-P2A
Water hammer most commonly occur when a valve is closed quickly and suddenly stops the flow of
liquid in a pipeline. This causes the velocity to decrease (ex: V1=10 V2=0)
These pressure waves generate impact forces on opposing elbows that act in the axial direction of pipe
that cause the pipe to sway with large pipe deflections in a violent manner that damage the pipe
supports and equipment F= ΔP. A . Ls/C.t
Valve Closure
A Pipe flow Area, Ls Pipe length, t Valve closure time flow
192
time
t
How is water hammer caused
Bumper
Support
194
Sources of water hammer
195
Guidelines for Analyzed systems
196
Guidelines for Analyzed systems
197
Guidelines for Analyzed systems
198
Guidelines for Analyzed systems
199
Guidelines for Analyzed systems
200
Guidelines for Analyzed systems
Consider support
enhancements such as
bumpers for pipe that is
very flexible where thermal
and dynamic movements
can be very significant (i.e.
at loops).
201
Guidelines for Analyzed systems
202
Guidelines for Analyzed systems
203
Guidelines for Analyzed systems
204
Guidelines for Analyzed systems
205
Guidelines for Non analyzed systems
• Condensate system,
• Spray water system,
• Fuel oil,
• Low temperature extraction, etc.
206
Guidelines for Non analyzed systems
Review the pipe support scheme and ensure that there are:
• Use the minimum load criteria for the first lateral restraint and
ensure that there is adequate stiffness such as adding a brace
etc. if this does not add to the cost significantly
207
Guidelines for Non analyzed systems
Review the pipe support scheme and ensure that there are:
• For axial restraints use the minimum load criteria and adequate
stiffness such as brace on a post etc.
209