Situation:
A section of 8" diameter pipe transfering hazardous material was damaged when the overhead portion of the pipe rack was damaged by moving equipment. A long section of the piping was left unsupported and the first ell that turns up to the rack fell (dropped) about 2 feet to the ground.

When I model the existing piping and remove the supports at the bridge and add a new support for the ground, the SUS case fails to the tune of about 200%. However, the pipe is still in service, i.e., no leaks. Temporary supports have been added where the bridge used to be, but CII gives similar overstress values with the added supports due to the original large displacement.

I’ve never checked the OPE case per APP P before now; the results are about 70% (note the EXP stress is low which may explain partly).

Question: Why has the piping not failed catastrophically with a SUS stress ratio of 200%?

Depending on responses, there may be some followup questions.

Thanks for your consideration . . .

Quick follow-up question: how likely is it that the piping has suffered permanent deformation somewhere? None is obvious, but ells could be oval(ed), etc.

Question: Why has the piping not failed catastrophically with a SUS stress ratio of 200%?

Because the pressure boundary was not breached. The allowable stress is based upon a fraction of the material yield point at temperature. "Failing" a Code allowable stress check does not necessarily mean the pipe (per se) "fails". The pipe will have had some local yielding (maybe in several places) but unless three plastic hinges form you may NOT have "catastrophic failure" in the classic sense of a through the section rupture. When the first local plastic deformation happens, there is a redistribution of strain over the system in the immediate vicinity. Linear elastic analyses do not predict the plastic strain distribution. That is not to say that you should not strip off the insulation and thoroughly inspect the pipe. Be sure that the pressure boundary was not "nicked" by the "moving equipment" - no small hole punched through the wall (damage to the insulation will be your guide). Have a look at ASME PCC-2.

You may just end up putting the pipe back in its original position, supported by original supports and putting it back into service. I hope (lessons learned) you put some procedure in place to prevent a similar incident.

2. how likely is it that the piping has suffered permanent deformation somewhere? None is obvious, but ells could be oval(ed), etc.

Probably pretty likely that there is some permanent (plastic) deformation (if you got your sums right). It happens all the time in hot systems before they shake down to elastic response. In and of itself, permanent deformation is no big deal. A single event cannot cause plastic ratcheting.

John

Thanks for quick reply!
I'm not sure I follow the three plastic hinges.
The pipe is uninsulated so is easier to inspect except for where it is in the dirt and where it is 16 feet above grade (can have that inspected tho for nicks, etc.).

As you guessed, the owner wants to put back in its original position while still in service. New bridge is installed. Situation that occured is kind of like a train wreck, just about no way to engineer out.

Biggest issue I see is "what if" pipe fails with system online. Material transported is hazardous.

I just got a copy of ASME PCC-2 and will read - Thanks--David

DavidPE

Given the nature of the service it would be prudent to leak test the piping after it is restored to its normal position (and supports) in accordance with B31.3 Paragraph 345. If you test it in accordance with 345.4 be sure to read 345.3.2.

I am surprised that a piping system in this service is normally in contact with soil. If it runs underground or is normally in contact with the soil it should be a double containment system with a leak monitoring system installed.

There is no such thing as an accident that could not have been prevented. By definition this system would be covered by Process Safety Management (29 CFR 1910.119) which would require an incident investigation with a review of all the procedures that were in place at the time of the accident.

References:

http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=STANDARDS&p_id=9760

http://www.epa.gov/swerust1/ustsystm/leakpipe.htm

Regards, John

John,
The phrasing of your reply sounds as if it is assumed they would shut down the line for repairs. This is what I would prefer they do, but am trying to give them technical ammunition for that position. The owner wants to reposition the piping while it remains online. Otherwise, that section of the plant must be shut down, perhaps, the entire plant.

The piping runs along sleepers for the most part until it is routed up and over a roadway onto a bridge, where it remains in a pipe bridge for many feet. There is no underground piping. Since one of the bridges over a roadway was taken out by a dump truck operator without his bed fully retracted, the first ell that turns up onto the bridge is now supported by the ground (about a 2' drop).

This is definately a covered process, but I was not part of the investigation and only brought on a couple of days ago when the contractor had issues with repositioning the line on the replacement bridge while still in service. They have remained online since the incident (not sure when that was at this point).

So far, I'm not finding any technical reason from a piping stress standpoint for why they cannot reposition the line while in service. Perhaps there is other reference material out there that better addresses this type issue.

thanks again . . . David

Drop of 2 feet is not that much. However I would insist that the line if fully inspected as soon as possible (NDT). It is possible that there are hidden defects. Uninsulated line and therefore low temperature but is it low or high pressure?

When ever repositioning a line I prefer that the pressure is off. Safer.

If the line has permanent deformation and you force it back to its original position you may have high stresses as a result. You might be able to get some idea if you would support the line on load cells in its original position. If the loads would be far out of the calculated loads you may be bending a deformed line.

I am presently reading up on some of the previously referenced materials which has some good info.

The design is P = 150 psig and temp = 150F. Piping section in question is 8" std A53 GrB seamless; painted.

I like the idea of the load cells; I will suggest that as well as the NDT (may have been done but not relayed to me yet). A real unknown is how much the pipe was moved/pulled out of position.

Thanks . . .