i have a 52 " flare line working in range from -170c to 0c. ambient temperature is 20 c. i'm thinking for two type of load cases. Case 1.L1= W+T1+P1 (T1=-170 c) L2= W+T2+P1 (T2=0 c) L3= W+P1 L4= L1-L3 L5= L2-L3 L6= L4-L5 (Algebric) L7= L4,L5,L6 (MAX) OR In this case i am taking 0 c as ambient and 20 c as T2 and finally adding L4 and L5 to get the combined stress range from 170 c to 0 c. Case 2 L1= W+T1+P1 (T1=-170 c) L2= W+T2+P1 (T2=20 c) L3= W+P1 L4= L1-L3 L5= L2-L3 L6= L4+L5 (Algebric) L7= L4,L5,L6 (MAX) pls suggest which load case is technically correct.

I would probably go with a modified Case 1:
L1: W+T1+P1 (OPE) [T1=-170]
L2: W+T2+P1 (OPE) [T2=0]
L3: W+P1 (SUS)
L4: L1-L3 (EXP)
L5: L2-L3 (EXP)
L6: L1-L2 (EXP)
L7: L4,L5,L6 (EXP) - stress only, MAX

My only change is L6. I am uncertain about the signs in your L6 in that algebraic combination of L4 & L5. I would be concerned about differences caused by nonlinear supports.

I would probably not run L7 but refer to the three individual expansion cases instead.

Of course a more detailed review may uncover something missing from this appraoch.

Thanx Dave for ur comments
but i want to discuss some more
when we go with T2=0 c
1. Caesar will not recommend any case like W+T2+P1 for T2=0 c
2. u will not get ant displacements or stresses related to T2.

So the write way when T2=0 c is that we should change the ambient temp =0 c and T2=20 c.
in this case we will get the expansion range from 0c to 20 c.
So the load cases will be
L1= W+T1+P1 (T1=-170 c)
L2= W+T2+P1 (T2=20 c)
L3= W+P1
L4= L1-L3
L5= L2-L3

Need not to make a case L4+L5. Since case L4 already specify the total expansion range.

so when the operating and design temperature are in one side of ambient temp i.e either higher side or lower side we need to make a cumulative case like L4+L5.

But when the operating and design temp are in either side of ambient temp i.e one is at higher side and the other is towards lower side then we have to calculate the cumulative L4+L5 case to get total expansion range.

Please double-check the combinations you made now.
You said ambient == 20 degr. C
OPE range : -170 - 0 degr. C
So OPE case 1 has a DT of 20 - -170 = 190 degr C
OPE case 2 has a DT of 20 - 0 = 20 degr C
In your example above:
L1= W+T1+P1 (T1=-170 c) --> DT = 0 - -170 = 170 degr C
L2= W+T2+P1 (T2=20 c) --> DT = 0 -20 = 20 degr C
Your OPE cases won't be OK this way.

hi corne

ur observation is appreciable
Since Caesar doesn't take temp =0 c so when we have the working temp =0 c then instead of putting 0 we must put 0.001 c to get the effect of 0 c.
we need not to shuffle between ambient temp 20 c and working temp 0 c.

You are correct; when T=0, CAESAR II will not run that thermal load case. The temperature field can hold two values: 1) temperature and 2)the actual thermal strain. You use #2 for materials with no thermal strain data stored in the program. CAESAR II decides what value you are entering by the magnitude of the number. If the number is less than (the default value of)0.005, then it is strain, otherwise it is temperature. You can change this default value in the configuration file. So when T=0, that's strain and a strain of 0 creates no load, therefore no load case.

If I want to run a job at zero temperature, I would probably set T=0.1.

Your point about expansion range - if I have one temperature above ambient and a second temperature below ambient; yes, I guess the total stress range would be the sum of these two excursions from ambient. But I would probably still run the difference between these two operating sets. It is more direct and avoids any question regarding summation technique.