I am currently evaluating an existing 20" 304 ss steam line that appears to have yielded. A vertical loop is supported at the top center by a dual type F can setup with a pre-insulated slide support(the pipe is not guided). Upon a recent shutdown the pipe drifted approximately 12" laterally to one side of the support creating unequal spring loading. Other info that I have been given leads me to believe that support friction elswhere in the piping system may have contributed to the problem. I was told that the crew at the facility was unable to fully center the pipe on the support due to pipe stress. This is a test facility that has been operating for about 5 years at approximately 40 cycles per year with a max operating temp of 750 F.

Is it possible for a poorly supported system to induce yielding or is it likely that the problem lies elsewhere? The stress analysis of the system predicts a max stress ratio of 80% of the b31.1 allowable.

Q1: "Is it possible for a poorly supported system to induce yielding"
A1: Yes, of course...

Q2:"...is it likely that the problem lies elsewhere?"
A2: Anything is possible.

"The stress analysis of the system predicts a max stress ratio of 80% of the b31.1 allowable." What loads are you talking about in creating this thermal displacement, Sustained loads, Occasional loads or what?


The line must have been installed centered on the support but something caused it to move laterally, perhaps thermal strain, perhaps an occasional load...

1st order field survey is what your model is based upon, match the real world layout, support/retstraint condition?


2cd Evaluate the cause of the lateral shift... what amount of force either uniform load or single point load would cause the lateral shift assuming that it does not shift due to suatained or thermal loads?

Justa couple of ideas off the top you get the idea I hope.....

The 80% stress ratio stated above is in the expansion case. An as-built stress model was built prior to the support problem described above.

What has your experience been with polyethylene slide plates? Will accumulation of dirt/grit increase the sliding friction substantially?

Thanks

"What has your experience been with polyethylene slide plates? Will accumulation of dirt/grit increase the sliding friction substantially? "


ABSOLUTLEY!!!! I have witnessed said phenomena in real life!

But, is it not an expected behaviour of any slide bearing ?

Slide design should protect the bearings from dirt/grit and temperature with thermal barrier for PTFE slide.

I have used F type variable spring will base on slides made with large ball bearings in applications near boiler feed pumps. But, here too, ingress of dirt/grit can damage the system.

regards,

sam

Any experience with thermal bowing? The elbow located on the bottom of the expansion loop is within 20 ft or so of a spray desuperheater. I believe there may be a large temperature gradient between the outer and inner walls of the 20" elbow.

Thermal bowing is not due to the temperature gradient through the pipe wall, but the temperature difference between the top and bottom wall. I would probably use the average metal temperature of the pipe for each bottom and top surfaces.

Your analsysis indicates a displacement stress range at 8o% of the allowable at a max.... The system has probably shaken out and is now self sprung.

I recomend leaving well enough alone..... Ona system shutdown I would recomend taking the trapeezed spring assembly installing travel stops, take it off the line, reinstall it correctly with guide stops and pulling the stops out.


You may want to also do some UT examination of the elbows in this area. The Desuperheater may have cause some thermal stresses or thinning due to erosion. Does the desuperheater have adequate recomended straight length downstream of it before the elbow?

80% of allowable puts the stress at about 22 kpsi. This is still considerably below material yield so why would local yielding hence self springing be a factor here?

There is approximately 20 ft between the desuperheater and the elbow. 11.5 ft is recommended.

Due to the pipe location one spring can is more heavily loaded than the other so I don't think I will be able to re-position the pipe and can loads properly with travel stops installed on the cans. I would prefer not to fight the spring force when re-positioning but I don't see how. Am I missing something?

There is a flanged joint at the bottom of the expansion loop riser that I plan to disconnect. From there I thought I could re-center the pipe on the dual spring support and also be able to determine if any localized yielding occurred. Does this make sense?

You as well as other users of the code need to read all the notes that appear in Appendix A The note applicable to ASTM A 312 TP 304 that I refer to is note 4…
“(4) *Stress values printed in italics exceed two-thirds of the
expected yield strength at temperature. Stress values in
boldface are equal to 90% of expected yield strength at
temperature. See paras. 302.3.2(d)(3) and (e).
“

The Sh value at 750F foir this material is 90% of yield at temperature so we can take 15.6 x 1.1 = 17.16 KSI is yield at temperature below your 22 KSI calculated stress therefore self spring has occurred!

The fact that you are desuperheating makes me wonder about SCC your layout appears to be adequate for the straight run requirements but if you have any chlorides present in your incoming water feed the desuperheating will concentrate them.

Being within the code allowable I assume self springing is acceptable. Is this correct or would you recommend being more conservative than the code when designing 304 ss systems in order to prevent self spring? Or would you say if properly supported self spring would not pose problems such as described above?

Yes self springing is not an issue per se and is quite common at high temperatures...

having said that I recomend you try to leave the system in its current range.... that is do not yank the pipe or spring it around when its cold....

Since the pipe network appears to be self sprung I am considering two options for resolving the spring support problem:

1. block spring cans, shift spring support so it is centered under new pipe position, and install pipe guide.
2. heat pipe(anticipating the pipe will re-center itself), install pipe guide, and adjust springs in hot condition.

I think both scenarios present potential problems. Any thoughts?

option 1 is preferred it will cause the least disruption to the thermal range already established and should provide for a more acceptable support configuration.

Remember the more you change things the more you cause the system to settle into a new range, so minimize your changes.

On a field survey of an existing SH steam ring header I discovered a spring can had never had its stop removed.... the system had been running for over 30 years at that point... rather than reset the range by removing the stop I had the stop tack welded into place. Sometimes more harm can be done when "fixing" things then simply leaving well enough alone.