ASME B31.1 - Buried Piping Allowables

[mfoote]

I'm wondering if anyone can provide an explaination for something I have recently noticed with B31.1, Appendix VII on buried piping.

In the example provided in the appendix, the author is using A106 Gr B piping, which has a yeild stress of 35 ksi. In the appendix the calculated allowable (using eqn 15 in the appendix) is 37.5 ksi - i.e. beyond yeilding.

I've heard explainations about permitting secondary stresses in underground (i.e. fully restrained) piping systems to exceed yeilding before, but this was related to European piping design standards. I was always under the impression that the B31 codes would never permit secondary stresses to exceed yielding (on a global sense).

 

[StevenHPerry]

I was always under the impression that the B31 codes would never permit secondary stresses to exceed yielding (on a global sense).

Personal understandings to follow. I don't have any connection to the B31 committees!

Not true! S_A can exceed the minimum S_y.

All Appendix VII Eq 7 seems to do is adapt base code eq 1b (liberal allowable).

Eq 1b for cold, continuously supported A106B at atmospheric pressure, will give you (2.5 * 17.1) = 42.75ksi.

S_C considers longitudinal pressure stress so this just flips the S_L term from base eq 1b from the allowable stress side to the stress value side to give you App VII eq 15.

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[StevenHPerry]

... and perhaps more importantly:

S_A is an allowable stress range. The problem then isn't that S_y might be exceeded (local yielding is acceptable) but that 2*S_y should never be exceeded.

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[mfoote]

Steve,

Thanks for your replies. Sorry, where does the 2*S_y limit come from? I've never come across this before.

I've always understood that local yielding is acceptable. I figured that was more related to the SIF's at branches, elbows, etc and generally not explicitly predicted by the code calculations. What I mean by this, (perhaps I'm mistaken), is that for a branch suppose I calculate the stresses to be within the allowable stress S_A. However, in reality when the branch is loaded, parts of the branch fitting are locally yeilded and the stress comes back down in that part of the fitting and everything is happy - i.e. no leaks, cracks, etc.

On the flip side, what I'm talking about is a calculated stress in the straight pipe (i.e. no SIF's) section that exceeds yeilding. Therefore, in this case, the steel would globally exceed the yield. Hence, in the first case, S_E<S_A<S_y, in the second case, S_E > S_y. In this case, and because these are displacement stresses, the pipe would yield - likely leading to a thicker wall throughout and the stresses would then come back down a bit.

Sorry - I hope I'm not confusing matters.. primarily I'm concerned about the issue of displacement stresses exceeding the yield point and what is happening to the metal after it has yeilded while in operation.

Cheers!

 

[StevenHPerry]

Imagine straight line between two anchors ereceted strain free. One anchor stays still, one anchor moves significantly enough to cause a bending stress (argument's sake, +9 arbitrary stress units) in excess of yield (5 stress units). Hold this new configuration and the tension stress on the "outside" of the bend will reduce to the yield stress (+5 stress units).

Now unpin the anchor. The pipe will return only part of the way to the original installed condition. It is now stress free (0 stress units). It has self sprung.

Pin the anchor again and bring it back to the installed location and you have a stress in the pipe again (-4 stress units). The range of stress didn't change, but the magnitude from neutral did. Each subsequent cycle will take us back to +5 and then -4 from neutral but no more yielding occurs.

Now if the stress range was 11 stress units versus the 5 stress unit yield strength, yielding would occur at the peak and valley of every cycle. This isn't a 7000 cycle piping fatigue failure, it's a beer can pop tab failure after 5 cycles.

The above is my current understanding. I'm sure I've got a few details incorrect but I think the concept is correct. Hopefully someone smarter than me can chime in.

- Steve Perry

http://www.linkedin.com/in/stevenhperry

This post is designed to provide accurate and authoritative information in regard to the subject matter covered. It is offered with the understanding that the author is not engaged in rendering engineering or other professional service. If you need help, get help, and PAY FOR IT.

 

[a234f56]

You will find relevant useful data in the attached pdf file.