Lifting a pipe for quick repair of support

[NovaStark]

Hi all,

I currently have a steam piping in-service which appears to have a pipe support resting on top of a guide support and has bent one of the guides.

Now the pipe carries steam at around 750 psi at 700 degF.

Likely what happened is that during start up and shutdown there was undesirable thermal growth perpendicular to the guide. So as the pipe moved downwards, the pipe support came into contact with the guide plate. I uploaded a simple sketch below. The small rectangle at the bottom of the trunnion is the guide plate which is bent.

Regardless of the system being in service or not, is there any way to judge if lifting the pipe about 8-12 inches (so the guide plate can be removed and replaced) will adversely affect the remainder of piping or even cause excessive localized stresses near to the lifting point ? Or will modeling of the piping be required with an imposed displacement be required to see how other pipe supports will behave ?

 

[KevinNZ]

8-12in is a long way. A few mm would be OK. For your lift you need get a pipe stress engineer to look at it.

 

[racookpe1978]

True a pipe stress analysis for the cold condition would tell absolutely. Doubt - that is a long way to force pipe to move for that low a temperature at design conditions.

For a quick check look for any pipe hanger in the vertical run above the kinked support plate restraints. That maximum movement of those supports would tell the max movement in the cold-to-hot change.

 

[NovaStark]

8-12in is a long way. A few mm would be OK. For your lift you need get a pipe stress engineer to look at it.

Well 8 to 12 might be a stretch to give a welder some comfort however most likely they'd try to lift it like you said a few mm and cut out the plate and weld back using SMAW.

True a pipe stress analysis for the cold condition would tell absolutely. Doubt - that is a long way to force pipe to move for that low a temperature at design conditions.

That's why I believe that during some transient condition, some excess (temporary) high temperature was experienced.

For a quick check look for any pipe hanger in the vertical run above the kinked support plate restraints. That maximum movement of those supports would tell the max movement in the cold-to-hot change.

I will look into this and see.

But in general, if the pipe is to be lifted I would say at most 3-5 mm above the guide support would be sufficient correct ?

 

[mk3223]

IMO, for the high P & T steam line, it's better to shut down the system for repair. But, first of all, you want to figure out if it was caused by the pipe stress issue or the operation issue. So, you can fix the problem correctly.

 

[racookpe1978]

I don't want to comment on possible fixes (how high to lift the pipe to clear the interference) without a photograph (or 2 or 3) of the problem.

 

[DekDee]

Couple of suggestions
Don't lift the pipe at all if it is going to be a major drama.
Put a temporary support next to the trunnion, arc gouge the weld off (without damaging the trunnion) and replace with same size plate
or
Put a temporary support next to the trunnion, cut through the weld (horizontally) and replace with slightly thicker guide plate.
Cheers,
Shane

 

[lilliput1]

If you just replace the plate chances are this will happen again. Use spider type guide and locate it away from the elbow. Use pipe expansion chart and pipe flexibility chart to determine minimum required distance of guide to elbow. For example at 700F pipe expansion is about 5.64" per 100 ft of pipe. If longest leg of the elbow is 40 ft and the pipe is 6" size, expansion of the longest leg is 2.26", minimum distance og guide to elbow is 21 ft. This provides flexibility at the elbow, Support at near the elbow should be spring type support.

 

[dhengr]

NovaStark:
You haven’t been at all clear about how that support has failed, how it works and is detailed, or what movement caused the failure, magnitude, direction, etc. I would suspect that the movement was horiz. along the length of the lower horiz. pipe leg, it was not in/out of the page in your sketch. It would appear that the pipe system can tolerate that movement, it did not fail. You also need to delineate the details of the pipe system, lengths, sizes, support types and locations on both legs, etc. etc. You must make some judgement about the strength, flexibility, stiffness, stresses in the pipes under that movement, etc. The idea that you might lift that pipe elbow 8 to 12" to make a repair seems way beyond any reasonable exaggeration to make a point, it suggests you better really study the situation at hand before you act.

In its crudest form, to suggest detail intent, I would replace that vert. leg and base plate support at the elbow, with a WF shape 3' +/- long under the horiz. pipe centerline, and parallel to it. It would have a pipe saddle on top to support the pipe, and it would have a guided sliding mechanism down at the floor/footing to allow the pipe longitudinal expansion, without particularly imparting significant secondary forces/stresses into the pipe system. Undoubtedly, there are off-the-shelf items in the piping field for this purpose.

 

[NovaStark]

You haven’t been at all clear about how that support has failed, how it works and is detailed, or what movement caused the failure, magnitude, direction, etc. I would suspect that the movement was horiz. along the length of the lower horiz. pipe leg, it was not in/out of the page in your sketch. It would appear that the pipe system can tolerate that movement, it did not fail.

I have uploaded some photos of the pipe support for a better visual guide. From the photos it looks to be that the pipe grew vertically and then out of the page then back down. Which I suspect only occured during a combination of transient conditions. I could be wrong in my preliminary thoughts here. But yes the pipe did not fail which I am guessing is because there was enough flexibility in the system. There is another trunnion support (just a rest support) about 8 feet downstream of the elbow and two control valves as well. At other plants of the same design, this support rests fine in the guide.

All other supports nearby appear to be ok (no near spring supports it seems)

Link for photos (upload to engineering.com wasn't seeming to work)

You also need to delineate the details of the pipe system, lengths, sizes, support types and locations on both legs, etc. etc. You must make some judgement about the strength, flexibility, stiffness, stresses in the pipes under that movement, etc.

I believe there may be sufficient flexibility in the system to just reset the support in a manner that DekDee suggested but only when the system is shut down.

The idea that you might lift that pipe elbow 8 to 12" to make a repair seems way beyond any reasonable exaggeration to make a point, it suggests you better really study the situation at hand before you act.

I agree but that is a suggestion that I was told by a higher up who does not really understand the mechanics of a piping system with respect to stiffness or how it affects other supports, etc. Which is why I wanted to know if it is a common practice to just lift the pipe, cut out the plate and then weld a new one i.e. basically reset the piping or if the common practice would be to model the entire piping and see if the supports are inadequate and so on (this means that I need to convince management that it is going to be reoccurring problem.) If I am going down that route, I will in fact check temperatures and pressures that the line is seeing or if condensate is in the line and so on.

[and yes I have higher ups who will question why I would want to spend time looking at these things rather not immediately knowing the answer, lifting the pipe and so on]

 

[lilliput1]

You need a spring type support or hanger at the elbow so as to allow the pipe to expand or contract without overstressing the support or pipe. Guides direct the direction of expansion of the pipe. Guides should be located at certain distance from the elbow so that the pipe is flexible enough to bend without overstressing the pipe. Steam is very dangerous. 700F is very high and would reduce the allowable stress in piping. Schedule 40 steel pipe service temperature limit is about 1000F.

 

[NovaStark]

Well one assumption that I have made is the plant designer did the actual stress analysis correctly.

This situation seems to be anomaly when compared to other plants with the exact same piping arrangement. So a redesign may not be absolutely necessary.

 

[lilliput1]

Visualize what will happen if the pipe expands between anchor points and think how the support at the elbow will function. Maybe on the other installation the vertical leg is not anchored in the riser but further at a horizontal; portion far from the vertical drop.

 

[NovaStark]

I will have to double check at the other sites and verify. I will also spend time verifying that all supports on this line are intact. I have gotten some recent history showing that a section of the line was upgraded with additional steam traps. So this line may not be equivalent to the other sites.