I am doing pipe stress of transporting liquid ethylene. The pipe is 12" Sch10 stainless steel 304L. The design temperature is -115 degree C. For such a low temperature pipe system, is there any difference in stress anlysis with the normal "hot pipe"? Any special attention?

Does any one have experience on it? Your help is highly appreciated. Thanks a lot.

In a strict sense your -115 C( -117 F) ethylene pipeline is cold, but NOT cryogenic cold.

To perform a Caesar II stress analysis, you would need take the dry-out temperature (assume T2) into account (a dry-out operation is essential so that the piping be moisture-free prior to a cold flow.)

Accordingly the flex temperature could become T2-(-115)+20 C, and the expansion stress induced thereof should be compared with B31.3 Code stress range allowable.

This is one condition of CAESAR II analysis one usually adds to cold lines, sufficing also for the liquid ethylene applications.

But for real cryogenic services, pipe stress analysis can get a whole lot more involved--e.g., as performed for liquefied natural gas (LNG) pipelines. Take the flange stress due to a thermal shock effect as a subject, it can captivate a career (like mine).

In the early 1980s there was a proposed B31.10 Code dedicating to Cryogenic Piping. Unfortunately, the proposal fell through.

FOLLOW-UP QUESTION:

If the pipe system were really cryogenic, what special considerations do we newbies have to note when modeling and anylizing these cases?

Thanks

Thanks Darren. But then what is cryogenic cold(-115 C is -175 F), in your view?

Is the T2 ambient temp? We use nitrogen to clean all the moisture. Then precool the whole 2 km pipeline.

What are the load cases after introducing T2? The following load cases ok or not?

L1: W+T1+P1(ope)
L2: W+T2+P1(ope)
L3: W+P1(Sus)
L4: L1-L3 (Exp)
L5: L1-L2 (Exp)

Thank you very much.

To River,
The cleaning (purging) of pipeline with nitrogen sounds normal. Precooling the line with nitrogen also seems reasonable - it is something to more alert about, since I think the boiling point of liquid nitrogen 1 atmosphere would be -320 deg F.
That would possibly be a (partial?) load case, and most people would agree to be a cryogenic low temperature.

To Richard,

No. The nitrogen is only for purging service. The precooling is by circling liquid ethylene.

Given T1>70F, T2<<70F, and Ambience at 70F, on August 18, 2004 I would build the load cases as follows:

1. W+P1+T1, or +H+D1
2. W+P1+T2, or +H+D2
3. W+P1, or +H
4. L1-L3
5. L2-L3
6. L4-L5, Set ALGEBRAIC
7. L4+L5+L6, set MAXIMUM

Case 7 determines B31.3 stress range compliance.

Tomorrow is another day to regret, and to err again. Good luck and good bye!