PVE-1135
Derek Verstege C.Tech
Laurence Brundrett P. Eng.
1 of 27
Table of Contents ver 4.00 Page 2 of 27
Contents Page
Cover 1
Table of Contents 2
Executive Summary 3
Section - Materials 4
SA-182 F316L 5
SA-193 B7 6
Section - Model Information 7
Model 8
Mesh 9
Error 10
Section - Restraints & Loads 11
Restraints 12
Loads 13
Bolt Preload 14
Reaction Area 15
Reaction Forces 16
Section - Results 17
Displacement 18
Stress - General 19
Stress - Clamp 1 20
Stress - Clamp 2 21
Stress - Ferrule 22
Stress Linearization - Ferrule 23
Bolt Stress 24
Component Cycle Life 25
Drawing - Tri-Clamp 26
Drawing - Tri-Clamp Details 27
Revision(s)
Rev Description Date By
0 Release 24/02/09 DRV
1
2
Executive Summary ver 4.00 Page 3 of 27
Project Information:
Goal:
The Tri-Clamp will be used under ASME B31.3 service. This product cannot be calculated to B31.3 code
rules due to the complexity of its geometry. Instead the rules of ASME VIII-2 are used with B31.3
allowed stresses to determine its acceptability.
Summary Conclusions:
Materials
Material strength properties used in this report are obtained from ASME B31.3, Table A-1. The rules of
ASME VIII-2 are used to determine the stress limits.
Model Information
The model used in the analysis represents 1/2 of the Tri-Clamp due to symmetry. The mesh size has
been refined to reduce the reported error to less than 5%.
Results
Through FEA we found a 0.00068" maximum displacement as acceptable. A 23,600 psi peak stress
results in a 1,000,000 cycle life, all other stresses are below their respective allowables.
Analysis Conclusion:
The Tri-Clamp meets VIII-2 design rules using B31.3 allowed stresses and is acceptable.
Materials ver 4.00 Page 4 of 27
Summary:
Material strength properties used in this report are obtained from ASME B31.3, Table A-1. The rules of
ASME VIII-2 are used to determine the stress limits.
Contents:
Material properties are obtained from ASME B31.3 Table A-1 as this block will be used in B31.3 service.
The rules of VIII-2 have been applied to calculate the maximum allowed stress levels depending upon
location.
Material properties are shown for the SA-182 F316L Clamps & Ferrules, and the SA-193 B7 Bolts. The
stresses of the Ferrule caps are not investigated in this report as they are not used in service.
Please refer to the "Results" section for the reported stress values.
1 Material Stress Limits ver 4.00 ASME VIII-2 Fig 4-130.1 Page 5 of 27
2 Material:
3 SA-182 F316L Material
4 Clamp & Ferrules Application
5 Strength Properties:
6 ASME B31.3, 2008 Source of strength properties
7 500 T [F] temperature
8 14,400 Sm [psi] basic allowable stress at temperature T
9 Sy [psi] yield stress at temperature T (optional)
10 1.0 k [] - stress intensity k factor
11 1.00 E1 [] - weld efficiency factor
12 1.00 E2 [] - casting efficiency factor
13 FEA Properties:
14 ASME VIII-IID, 2007 Source of FEA properties
15 25,900,000 E [psi] - modulus of elasticity (at temperature)
16 0.27 v [] - Poison's ratio
17 Stress Limits:
18 Pm = k*E1*E2*Sm~~general primary membrane stress intensity limit
19 1*1*1*14400 = 14,400
20 Pl = 1.5*k*E1*E2*Sm~~local membrane stress intensity limit
21 1.5*1*1*1*14400 = 21,600
22 Pl+Pb = 1.5*k*E1*E2*Sm~~primary membrane + primary bending stress intensity limit
23 1.5*1*1*1*14400 = 21,600
24 Pl+Pb+Q = Max(3*E1*E2*Sm,2*E1*E2*Sy)~~primary + secondary stress intensity
25 MAX(3*1*1*14400,2*1*1*0) = 43,200
26 Pl+Pb+Q+F = Use fatigue curves~~peak stress intensity limit
27 Comments:
28 (1) Sy material property is not required, more conservative Pl+Pb+Q limits might be computed without it.
29 (2) Refer to VIII-2 Table AD-150.1 for k values
30 (3) The thermal expansion coeficient is only required for studies including thermal stresses
31 (4) Refer to VIII-2 App 4-130 and following for the Pm, Pl, Q and F stress limits
32 (5) Refer to VIII-2 App 4-130 Table 4-120.1 for the correct application of the calculated stress limits
33 (6) Use IID tables 2A and 2B for Sm for VIII-2 studies
34 (7) Use IID tables 1A and 2A for Sm values (S) for VIII-1 studies
35 (8) Use B31.1 Table A for Sm values for B31.1 studies
36 (9) Use B31.3 Table A for Sm values for B31.3 studies
1 Bolt Stress Limits ver 4.00 ASME VIII-2 4-140 Page 6 of 27
2 Material:
3 SA-193 B7 Material
4 Clamp Bolting Application
5 Strength Properties:
6 ASME B31.3, 2008 Source of strength properties
7 500 T [F] temperature
8 25,000 Sm [psi] basic allowable stress at temperature T
9 FEA Properties:
10 ASME VIII-IID, 2007 Source of FEA properties
11 27,400,000 E [psi] - modulus of elasticity (at temperature)
12 0.30 v [] - Poison's ratio
13 Coef [in/in/F]- coefficient of thermal expansion (for thermal stress studies only)
14 Service Stress Limits for Seating + Operating Loads: 4-140
15 Pm = 2*Sm~~membrane stress intensity limit at location without stress concentration
16 2*25000 = 50,000
17 Pm+Pb = 3*Sm~~membrane + bending stress limit at at location without stress concentration
18 3*25000 = 75,000
19 Pl+Pb+Q+F = Use fatigue curves~~peak stress intensity limit
20 Comments:
21 (1) The thermal expansion coeficient is only required for studies including thermal stresses
22 (2) Use IID table3 for Sm for VIII-2 studies
23 (3) Use IID table 3 for Sm values (S) for VIII-1 studies
24 (4) Use B31.1 Table A for Sm values for B31.1 studies
25 (5) Use B31.3 Table A for Sm values for B31.3 studies
Model Information ver 4.00 Page 7 of 27
Summary:
The model used in the analysis represents 1/2 of the Tri-Clamp due to symmetry. The mesh size has
been refined to reduce the reported error to less than 5%.
Contents:
Model
The Tri-Clamp model used in the FEA is shown. A 1/2 model was used due to symmetry, ferrule caps are
included to simulate exit pressure loads.
Mesh
A second order tetrahedral mesh was applied with a global mesh size of 3/32". A "No Penetration"
condition was applied, ensuring that all components act individually as they would in reality.
Refer to the "Material Stress Limits" section for each materials Modulus of Elasticity and Poison's Ratio.
Error Plot
The error plot shown is based on the Cosmos Designer built in error plot estimation function. The error is
less than 5% as per ABSA guidelines, except at discontinuities and contact surfaces where higher error
values are expected. These localized areas of higher error do not reduce the validity of the overall results.
Reference Information:
Including reference components in an FEA to provide appropriate boundary and load conditions.
http://www.pveng.com/documents/content_80.pdf
Mesh Refinement Using the Error Function Results for Areas at Discontinuities.
http://www.pveng.com/documents/content_250.pdf
Mesh Refinement Using the Error Function Results for Areas near Discontinuities.
http://www.pveng.com/documents/content_251.pdf
Error Plots for Bolt Heads and Surface to Surface Contacts Areas.
http://www.pveng.com/documents/content_248.pdf
10
11
12
13
14
Ferrule
15 Bolt
16
17
18
19
Clamp
20
21
22
23
24 Fig-A
25 An exploded view of the "Tri-Clamp" model.
26 The Clamp, Ferrules and Bolts will all be analyzed in this report.
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Summary:
A symmetry restraint is applied to compensate for the use of a 1/2 model and a single fixed point on the
clamp is fixed to prevent rigid body motion. Internal pressure is applied and clamp bolt preloads
simulated. The resulting reaction forces closely match the calculated reaction forces. The model is in
balance and can be used for displacement and stress analysis.
Contents:
Restraints
A symmetry restraint is applied to all faces on the symmetry plane. This compensates for the use of a 1/2
model and provides identical results to that of a full model. A single point on the clamp has been fixed to
prevent rigid body displacement.
Loads
Internal pressure is applied to all internal surface up to the pressure boundary. A clamp bolt preload is
simulated by a bolt connector condition. No other forces exist.
Reaction Area
The reaction area of the Tri-Clamp is shown. The reaction force is equal to the product of the reaction
area multiplied by the pressure.
Reaction Forces
The reported reaction forces closely match the calculated reaction forces in all directions. The model is in
balance and can be used to calculate expected displacements and stresses.
1 Restraints Ver 4.06 Page 12 of 27
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 Fig-A A view of symmetry restraint applied to the section plane of the model. This condition restrains the
25 sectioned faces to the "XY" plane wile permitting movement in the "X" & "Y" directions. This restraint
26 compensates for the use of a half model and provides results identical to a full model.
27
28
29
Point Restraint
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
26
27
28
29
30
31
32
33
34
35
38
39
40
41
42
43
44
45
46
47
48
49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Fig-B An alternative view of Fig-A showing the resulting forces in the bolt under combined loading and
51 operating conditions. These loads are calculated on the Bolt Stress page to determine the acceptability of
52 the bolting.
1 Reaction Area Ver 4.02 Page 15 of 27
2
7 Reaction Area
8
10
11
12
13
14
15
16
17
18
19
20
21
22
Reaction Area
23
24 Fig-A A view showing the reaction area (inside area of ferrules and caps)
25 This area is used to calculate the reaction force in the "Z" direction.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 View showing global reaction forces from analysis "X" = 0 lb, "Y" = 0 lb, "Z" = 10431 lb
25 Calculated Reaction Forces = Analysis Reaction Forces within 0.2%
26 Model is balanced, results are valid.
27 X Axis: reaction forces on the YZ plane caused by loads in the X direction
28 0.00 XArea [in2] - Pressurized area on YZ plane
29 1,500 P [psi] - Pressure
30 0 XForce [lbs] - Added force in the X direction
31 0.0 XReaction [lbs] - Reaction force in X direction reported by FEA program
32 TReactionX [lbs] = XArea*P+XForce ~~ Theoretical X reaction force 0*1500+0 = 0
33
Summary:
Through FEA we found a 0.00068" maximum displacement as acceptable. A 23,600 psi peak stress
results in a 1,000,000 cycle life, all other stresses are below their respective allowables.
Contents:
Displacement Plots
The displacement plot is as expected and acceptable. The maximum reported displacement is 0.00068".
Stress Plots
A peak stress of 23,600 psi is seen at the bolt head to clamp contact surface, this stress is used to
calculate a cycle life. All other areas are below their respective allowable stress, no further analysis is
required.
Cycle Life
A peak stress of 23,600 psi results in a estimated 1,000,000 cycles.
1 Displacement Ver 4.06 Page 18 of 27
2
5
Rotation of clamp
6
10
13
14 Radial Growth
15
16
17
18
19
20
21
22
23
24 Fig-A A view of the displacement plot with superimposed original geometry. Results are magnified 500X.
25 Displacement of the ferrules is radially outwards and axial due to internal pressure.
26 The clamps rotate about the center due to the "pulling" force of the ferrules.
27
28
29
30
31
32
33
34
Bolt head displacement
35
36
37
38
39
40
41
42
43
44
45
46
47
Joint separation
48
49
10
11
12
13
14
15
16
17
18
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Fig-B An alternative view of Fig-A; a 1/4 section has been removed to show stress distribution.
51 A peak stress of 23,600 psi is located beneath the head of the bolt. This stress will be used in a cycle life
52 analysis. Each component will be analyzed individually on the following pages.
1 Stress Ver 4.06 Page 20 of 27
2
Stress location beneath
3
head of bolt
4
10
11
12
13 Stress at ID caused by
14 outwards bending of clamp
15
16
17
18
19
20
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Fig-B An "ISO Clipped" view of Fig-A, scale capped at SA-182 F316L general membrane allowable of
51 14,400 psi. The only stresses in excess of 14,400 psi are at local areas; these areas are allowed 21,600 psi
52 as per VIII-2 rules and will be investigated on the following page.
1 Stress Ver 4.06 Page 21 of 27
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 Fig-A An "ISO Clipped" view of Fig-A, scale capped at SA-182 F316L local membrane allowable of 21,600
25 psi. No stresses in excess of 21,600 psi exist, the clamp results are acceptable.
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
1 Stress Ver 4.06 Page 22 of 27
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Fig-B An "ISO Clipped" view of Fig-A, scale capped at SA-182 F316L general membrane allowable of
51 14,400 psi. Stresses in the hub region in excess of 14,400 psi will be further analyzed through a stress
52 linearization study on the following page.
1 Stress Linearization ver 2.34 Page 23 of 27
2 x
3
Lineariztion Line
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
30
31 15,000
32
33
Sn
10,000
34 St
35 Sh
5,000
Stress
36
Tnt
37
Tnh
38 0
39
Tth
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
40 Tresca
-5,000
41 Pm
42 Pm+Pb
43 -10,000 Peak
44
45
-15,000
46
47 Distance i to j
48
34
Bolt Stress
35
36
41500
37
38
41000
39 40500
Stress (psi)
40 40000
41 Bending
39500
42 Membrane
43
39000
44 38500
45 38000
46
37500
47
1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3
48
Location
49
50
Cycle Life ver 4.04 Page 25 of 27
ASME VIII-2 2004 Ed. 2006 Add, Appx. 5 For series 3XX high alloy steels, nickel-chromium-iron alloy,
nickel-iron-chromium alloy, and nickel-copper alloy for temperatures
Fig 5-110.2.1 graph - Select graph not exceeding 800F and Sa > 28.2ksi
23,600 Str [psi] - Enter stress value
27,400,000 ET [psi] - Modulus of elasticity at operating temp
1,000,000
100,000
Stress
10,000
1,000
1.0E+00
1.0E+01
1.0E+02
1.0E+03
1.0E+04
1.0E+05
1.0E+06
Cycles
A .50 1.0
.263
D D
CL
R.063
.5
.2
3.25
C
B B R.99 C
.153 R.858
R.685
.115
40
CL
A SECTION A-A DETAIL C
SCALE 2.5 : 1
.281 THRU
.75
(2) PLCS
B
C B
Tri-Clamp
SURFACE ROUGHNESS 32 in CUSTOMER
PVEng
THIRD ANGLE PROJECTION PROJECT
2 TYP PVE-1135
DRAWN BY DATE SIZE DRAWING NO REVISION
DRV 25/02/2009
MATERIAL:
As Noted
FINISH:
CHECKED BY
LB
APPROVED
DATE
25/02/2009
DATE
B PVE-1135
SCALE WEIGHT
0
LB 25/02/2009 1:1 1.966 lbs SHEET 2 OF 2
8 7 6 5 4 3 2 1