Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

When would piping need to be replaced due to an overpressure event where the Yield Stress was exceeded?

bizkitgto

Chemical
Jul 16, 2012
9
In a overpressure situation where design pressure is exceeded, typically stress analysis is done to confirm there was no yielding. In some situations, hydrotesting is done at yield stress.

What if yield stress was exceeded? ASME B31.3 doesn't specify when to replace piping after yield stress is exceeded.

How would you determine if the piping is safe to use after yielding (plastic deformation), and when would you recommend to replace piping?
 
Replies continue below

Recommended for you

Assuming plain old carbon steel piping and assuming the fluid doesn't cause some sort of weird corrosion mechanism related to the steel yielding, I'd say do a visual, if nothing obvious, leave it alone. I'd be more worried about crack like flaws, but again is the piping it "hot" that's probably unlikely to occur, if the piping is cold then I'd definitely look into it.

Keep in mind carbon steel is general "rated" to have at least 10% strain before failure, and in practice is normally in the range of 30-50% when actually tested. 10% strain is huge to see visually, If failing under hoop stress, the pipe would have a very noticeable bulge somewhere.

there are some theories out there that yielding can actually be beneficial, in particular it is suggested it may help stress relieve welds to some degree. Other people dis-agree and suggest it may make existing flaws worse... I'm in camp A, but that's just my opinion.

For context,
what's the different between cold bending a pipe during fabrication and "yield" during operation?
Or forging of seamless pipe versus yield in operation?
Or forging a fitting without heat treatment afterwards?

These processes purposely yield the material, yet we still use these components all the time in piping service.


Also B31.3 is a standard that covers "new construction" only, we often use it to guide is assessing in-service related issues, but that is not what it is intended for. Suggest you dig up ASMBE PCC2 and API 579, one of them probably give a bit more guidance.

Hope this helps,

I guess the official answer if following standards would be "review PCC2 / API 579 or other similar Inspect / Fitness for Service stadands" and follow the recommendations within those documents.
 
More detail required. How and were has the yielding occurred and under what load condition? How do you know?

Codes actually allow yielding in some load cases, e.g thermal stress range or ground substance.
 
More detail required. How and were has the yielding occurred and under what load condition? How do you know?

Codes actually allow yielding in some load cases, e.g thermal stress range or ground substance.

Assume yielding occurred due to overpressure in piping, and it is known and real, measured via pressure transmitter (DCS).
 
Apply a fitness-for-service approach, looking for specific damage mechanisms. And note that merely exceeding yield is not a damage mechanism. We're talking: crack, bulges, dents, gouges, other gross deformations, fatigue damage (somewhat related to cracks), etc. You might need to perform some inspection to ensure that you haven't initiated a crack by your over-pressure event.

Related story - I was called in to perform a fitness-for-service assessment on an air-cooled heat exchanger header box that had suffered from an over-pressure event due to ice formation. There was bulging everywhere - I back-calculated, based on the deformation, that the header box had withstood 10x the design pressure. We did some dye-penetrant testing at a few select locations, performed a hydrostatic leak test (emphasis on the "leak" part - 1.5x design pressure is nothing compared to the 10x design pressure that it withstood), and put it back into service. That was roughly 15 years ago and it's still in service today.

I've even specified deliberately performing an over-pressure event on some equipment that had some dents that really needed to be "popped out".

Now, if the equipment that you are referring to was in creep service, then you may have dramatically shortened its remaining life...
 
I would look at the piping to see if it is deformed and do a wall thickness check to see if its still within limits, check welds for any cracks and probably hydro to the original DP.

As others say, this is a FFS check not a design and construction code which is B31.3. ASME PCC 2 is a better read for this.

Also SMYS has the word Minimun in it. Actual yield is something different.

So if you're say 10% overpressure for yield then you might not have yielded.

And actual wall thickness is often more than the min with tolerance included.

But 50% over and you've probably thinned the pipe somewhere.
 
Also SMYS has the word Minimun in it. Actual yield is something different.

So if you're say 10% overpressure for yield then you might not have yielded.

Can you explain why yielding doesn't happen at SMYS? I was under the impression that SMYS is when plastic deformation starts...

Edit: spelling
 
Last edited:
First, some yielding starts at very low stress.
Annealed steel will show some small permanent deformation well below 0.2% strain.
Secondly most actual material is significantly stronger than the spec minimums.
My world was mostly SS and we had specs with minimums of 25ksi Yield/70ksi UTS and we would ship 45ksi Yield/90ksi UTS.
And third as the material starts to yield it will strain harden slightly and this will cause the load to be shared by adjacent material.
It would take a very sudden load or a very high load to cause gross local distortion.
That said you need to inspect welds, especially at locations where there are changes in size or thickness.
 
Can you explain why yielding doesn't happen at SMYS? I was under the impression that SMYS is when plastic deformation starts...

Edit: spelling
Because the specification is the limit. If a supplier makes a material which when tested is less than SMYS, it gets scrapped. So most vendors will make a steel which exceeds the minimum. The question is by how much. Most times I've ever looked at actual test results they are 5 to 10 % above the minimum. Sometimes a lot more. Go look at the test certificates.
 
And third as the material starts to yield it will strain harden slightly and this will cause the load to be shared by adjacent material.

Thanks for the reply. Can you elaborate on this part more, specially how the "load is shared by adjacent material" (is that a change from before yielding?)?
 

Part and Inventory Search

Sponsor