Hi,

On the exemple at the bottom (CAESAR 7.00.01.1600 Build 141003)
I perform the following load cases:
L1 WW+HP+H HYD
L2 W+T1+P1+H OPE
L3 W+T2+P1+H OPE
L4 W+T3+P1+H OPE
L5 W+T4+P1+H OPE
L6 W+T5+P1+H OPE
L7 W+P1+H SUS
L8 L2-L7 EXP
L9 L3-L7 EXP
L10 L4-L7 EXP
L11 L5-L1 L7 EXP
L12 L6-L7 EXP
L13 L8,L9,L10,L11,L12 EXP

Options:
L2-L7(EXP) Keep Disp/Force/Stress Algebraic
L3-L7(EXP) Keep Disp/Force/Stress Algebraic
L4-L7(EXP) Keep Disp/Force/Stress Algebraic
L5-L7(EXP) Keep Disp/Force/Stress Algebraic
L6-L7(EXP) Keep Disp/Force/Stress Algebraic
L8,L9,L10,L11,L12(EXP) Keep Disp/Force/Stress Max

The allowable of the MAX of EXP cases (L13) is equal to 1.5Sh instead of 1.25SC+0.25Sh as for the others cases (L8 to L12).
Did I miss something or is there a issue?

****
RESULTS:

8(EXP) 10 38.74 192.86 13 38.74 192.86 B31.3
9(EXP) 38.74 192.86 38.74 192.86 B31.3
10(EXP) 38.74 192.86 38.74 192.86 B31.3
11(EXP) 38.74 192.86 38.74 192.86 B31.3
12(EXP) 38.74 192.86 38.74 192.86 B31.3
13(EXP) 38.74 122.90 38.74 122.90 B31.3

***********************************************
Input Echo


PIPE DATA
-----------------------------------------------------------------------------
-----------------------------------------------------------------------------
From 10 To 15 DX= 3,107.000 mm
B31.3 (2012) Cycle Max Switch = ON App.P Sy/St Switch = ON
Sc= 138 MPa Sh1= 82 MPa Sh2= 82 MPa Sh3= 82 MPa Sh4= 82 MPa
Sh5= 82 MPa Sh6= 138 MPa Sh7= 138 MPa Sh8= 138 MPa Sh9= 138 MPa
Sy= 207 MPa Fatigue Curve (1)= 1 Cycles
Dia= 863.600 mm. Wall= 52.370 mm. Cor= 1.0000 mm. Wl Factor= 1.000
GENERAL
T1= 495 C T2= 495 C T3= 495 C T4= 495 C T5= 495 C P1= 86.7000 bar
P2= 69.0000 bar PHyd= 219.0000 bar Mat= (185)A335 P22 E= 212,596 MPa
EH1= 174,931 MPa EH2= 174,931 MPa EH3= 174,931 MPa EH4= 174,931 MPa
EH5= 174,931 MPa EH6= 212,596 MPa EH7= 212,596 MPa EH8= 212,596 MPa
EH9= 212,596 MPa v = .292 Pipe Den=7833.4116211 kg/m3
Insul Thk= 130.000 mm. Insul Den= 135.0000000 kg/m3 Clad Thk= .700 mm.
Clad Den=8000.0000000 kg/m3
BEND at "TO" end
Radius= 1,295.400 mm. (LONG) Bend Angle= 90.000 Angle/Node @1= 45.00 14
Angle/Node @2= .00 13
RESTRAINTS
Node 10 ANC
ALLOWABLE STRESSES
-----------------------------------------------------------------------------
From 15 To 20 DY= 5,740.000 mm
RESTRAINTS
Node 20 X

If would be helpful to have the job. Can you open a Support Request on Smart Support and attach the file please?

Richard Ay,

The issue is not related with the Particular system.

I checked in CAESAR-II 5.3.1, 2013R1 and CAESAR-II 7.00 VERSION

MY OBSERVATIONS:

1. CAESAR-II 5.3.1 the allowables of all EXP case(Alg, Scalar, Max etc.,) is same.

2. From CAESAR-II 2013 R1 version up to LATEST 7.00, the allowable value is reduced for load case EXP(MAX)only as mentioned by Rdml

3.This reduction of the allowables happens for the higher temperature only.

Is this related to f (Stress range reduction factor) for EXP case - COMBINATION METHOD "MAX"??

This really doesn't make sense, so I want to see a job file which illustrates what is said above.

Your MAX combination will display the highest stress in the group but these individual stresses in the group may have differing allowable limits. I do not see the value in using the MAX combination to compare with a single allowable limit.

It appears that, to be conservative, CAESAR II uses the smallest values of Sc & Sh to set the "limit" for this combination so Sc=Sh=82 MPa and 1.25 Sc + 0.25 Sh = 1.5(82).

Dave Diehl,

Thanks for your reply.

Its surprising that in Combination MAX; CAESAR-II uses Sc = Sh;

If CAESAR-II takes (min of Sh) and (Min of Sc) from the group is practical, same as CAESAR-II 5.3.1 does.

In older version CAESAR-II 5.3.1 it doesnt take Sc=Sh. What is the reason for changing this COMBINATION MAX allowable in CAESAR-II 6.0?

on what basis from CAEASR-II version 6.00 considers like this?

I use the MAX combination method for structural response such as restraint loads and pipe position - these have no specific, load case specific, limits. So I do not have experience using this for stress in any version of CAESAR II.

I got many (EXP) cases and I just want which one gaves the maximum stress.
I don't understand the answer of Smartsupport which says: "Customer should use a combination method of Algebraic for determination of expansion stresses. However, having said that, when you have a range between two temperatures (not between a temperature and ambient) then CAESAR II uses the highest Sh as Sc and the lower Sh as the Sh value in the equation.
As Sc =Sh
1.25 Sc + 0.25 Sh = 1.25Sh+0.25Sh = 1.5 Sh.
This is more conservative approach."

1) I indeed use ALG for individual stress. I just want the MAX of theses individual EXP stresses.
2) Where do I speak about RANGE? I want the MAX.
3) It is not my trouble but where the ASME B31.3 says to use Sc = Sh? Why the range is between two positive temperature, I will never install my pipes at 495°C... (SAFETY FIRST, no?)
4) I don't want something conservative but something correct

According to B31.3 Paragraph 302.3.5 the definitions of Sc and Sh are:

Sc = basic allowable stress at minimum metal temperature expected during the displacement cycle under analysis.

Sh = basic allowable stress6 at maximum metal temperature expected during the displacement cycle under analysis.

A "displacement cycle" is a "range". For any two temperatures (including ambient), Sh is the allowable from the hottest of the two temperatures and Sc is the allowable from the coldest of the two temperatures.

The Code doesn't say that Sc = Sh, rather the Code wants the highest and lowest allowable stresses in the range being considered. If you happen to setup a load case where T1 and T2 are the same, then it just happens that Sc = Sh.

Consider the following temperatures:

T1 = 200
T2 = 400
T3 = -50

Then:

EXP1 could be: T1 - ambient, (Sh from T1, Sc from ambient)
EXP2 could be: T2 - ambient, (Sh from T2, Sc from ambient)
EXP3 could be: T3 - ambient, (Sh from ambient, Sc from T3)
EXP4 could be: T2 - T1, (Sh from T2, Sc from T1)
EXP5 could be: T1 - T3, (Sh from T1, Sc from T3)
EXP6 could be: T2 - T3, (Sh from T2, Sc from T3)

This is the intent of the Code. This is not conservative, but rather it is complete.

Thank you Richard,

With your explanation I understand how to solve my trouble:
If I add an (EXP) case with T=amb, I got the result I want.