Hello all,

I am humbly seeking for your expert advice.

We will be analyzing an existing pipe rack that has settled. Upon checking on site, some parts of the piping had sagged due to settlement on rack supports; while other parts remain the same due to some support are piled so no settlement occurs on those parts. Also, some parts on the piping is not touching the support steel due to the uneven elevation and maybe also due to the see-saw effect of the pipe (imagine a worm wiggling, that is the looks of the piping on the rack). Now the good thing is its still operational. I can assume it is still within code allowables, because after all the sagging and unbalanced supporting, it is still functioning.

It was previously designed by a different company and now the client wants us to fix the pipe rack. The procedure would be pushing the ones that sagged upward to make the pipe rack straight again. They plan on replacing the existing support steel on the ones that settled, so all the TOS on the pipe rack would be the same and the rack piping straight again.

a) Without considering any further settlement after the steel has been replaced and using the current pipe configuration as the initial condition; will the analysis be the same as analyzing settlement on pipes; only that the displacement value for settlement will be upward (positive value) instead of downward (negative value)?

My load case would be:

L1 W+T1+P1+D1+D2 (OPE)
L2 W+T2+P1+D1+D2 (OPE)
L3 W+P1+D1 (SUS)
L4 W+P1+D1+D2 (SUS) *new SUS after pushing the piping upward for alignment
L5 L1-L3 (EXP)
L6 L2-L3 (EXP)
L7 L1-L4 (EXP) *after pushing the piping upward for alignment
L8 L2-L4 (EXP) *after pushing the piping upward for alignment
L9 L1-L2 (EXP)

D1 = measured sag of piping (maybe different in each location)
D2 = amount of sag to be pushed upward in order to align the piping

When looking upon this load case, the D1 and D2 would cancelled out since they will be assumed to have the same value. Please help if my understanding is correct.

b) Since the piping already experienced stress due to the settlement effect, will there be any other special cases to be considered or dynamic analysis concerns since the piping will be pushed upwards upon re-alignment?

Thank you all in advance.

I don't understand what you try to calculate. As you said, after all the sagging and unbalanced supporting, it is still functioning and you assume that piping still complies with Code as allowable. This may be true or not and I understand you are not certifying it is true; important is it was not collapsing.

You plan to establish the original condition. I understand that the original calculation shows is OK vs Code requirements. In case this calculation is not available, you can do it now.

As theory, the settlement effects can be considered as a strain displacement cycle added to the thermal strain 7000 cycles@ EXP allowable(let's say one cycle per day in 20 years). Maybe you will find that the settlement equals 1000 cycles and you evaluate that piping is OK for the next 5887 thermal cycles, one cycle per day in 16 years. It is what you plan to do?

IMO opinion it is more important the procedure to re-establish the original condition and I'm afraid this is rather technical common-sense and experience than calculation.

All you can really assume at this point is that the current configuration out there has not resulted in a stress higher than the Ultimate Tensile Stress (UTS) of the pipe.

We don't know which code you're referring to, or what the intent of the analysis is, so stating whether or not it's code compliant is largely ambiguous. B31.3 does make an allowance for lines in acceptable service being permitted, but my opinion is you invite risk while invoking this clause.

The allowable stress per B31.3 is the lesser of 1/4 UTS or 2/3 Specified Minimum Yield Stress (SMYS). If you are able to see the system "turned off," and there is permanent deflection, then you're above SMYS, and you're definitely not to code (except referenced above), and you might have to look into Fitness for Service (FFS).

For one-time displacements (i.e. permanent settlement), there's a different allowable I believe, and that should be looked at, and you'll need to tailor your results accordingly.