Using Published Yield Stress from MTR in lieu of Code Values 1

[GIBSTJEAN]

Does anyone know of an allowance in B31.3 that lets a Designer use actual Yield Stress values of material taken from the associated MTR rather than the values published in the strength tables?

We have a piping system that had to have extensive stress analysis done with a lot of manipulation of hanger/support types, quantities and locations. The material has arrived and it was ordered as welded instead of Seamless Pipe (0.8 quality factor vs 1.0). Needless to say that drops the strength enough that the stress analysis now fails.

When looking at the actual yield stress for the various heat numbers of the materials, the Strength is well above where we need to be. Roughly 40 ksi vs 30 ksi as published in the strength tables for SA-312 304 ss.

Is there some caveat to the code(s) that will allow us to use known strengths vs minimum acceptable strengths found in the tables?

 

[rneill]

I'm not aware of any such caveat. In fact, I think there is an interpretation that says this practice is not permitted. I don't have the interpretations handy right now so I can't check for you but I'm pretty sure that this isn't allowed.

 

[jte]

For vessels, you'd be subject to Interpretation II-D-01-02, but I don't know of something similar for piping.

Subject: Section II, Part D (2001 Edition); Yield Strength of SA-724, Grade C
Date Issued: July 1, 2002

Question: Is it permissible to use values from material test reports in lieu of the design values shown in Section II, Part D?

Reply: No.

jt

 

[JohnGP]

rneill nailed it.

Subject: ASME B31.3-2002, Para. 302.3.2(d), Bases for Design Stresses
Date Issued: May 29, 2003

Question(1): In accordance with ASME B31.3-2002, for a listed carbon steel material, may the measured tensile strength and measured yield strength as shown in the Material Test Report be used to establish the basic allowable stress values at room temperature per para. 302.3.2, instead of using the basic allowable stress values in Appendix A?

Reply(1): No

The question related to CS materials, but the same response would no doubt be given for SS material.

John

 

[BigInch]

The only thing that mistifies me is why they call that an interpretation. There is no provision I see in the code that comes anywhere close enough to mentioning even the remote possibility of making such a special allowance. How do you interpret something that doesn't even exist. OK, admittedly I'm not well versed in the subject of theology. I mean, I might as well ask for an interpretation on the use of macaronni elbows.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[StevenHPerry]

I think you have your answer, but what was the thought process?

By stress analysis, I assume you mean a flex analysis which will deal with sustained and displacement stress range. E_j affects hoop & sustained stresses. Displacement stress range isn't affected by a lower E_j. Sustained stresses can be reduced without changing the piping.

Depending on the size of system you're dealing with, there could be a cheaper and quicker solution than reordering everything seamless.

I assume the MTR yield strengths are at ambient temperature. Do you have them at elevated temperatures (assuming your process is hot) as well? This would be what is needed if you were going to strike out on a limb and declare a new allowable stress for a specific heat.

For 304ss, the allowables are already taken to 66% or 90% of expected yield for many temperatures. The margin left in there is so slim relative to other materials that perhaps the Committee was counting on most actual material to have yield strengths higher than the minimum (i.e. they had historical data showing this trend, not just wishful thinking). Conjecture only.

- Steve Perry

 

[GIBSTJEAN]

Thanks for the responses. It was probably wishful thinking on my part but I'm not sure it was as outrageous a proposition as BigInch is suggesting. The strength tables must be set up to list the minimum yield and tensile stress that a material can exhibit and still be labelled as that material so for design purposes, in the absense of actual test results, those minimums would be all you could use for your calculations. We rely on the test results of each heat number to classify and use the materials so presumably they can be trusted. A lot of material is being dual speced as well so even if you ordered 304L for the chemical composition and the pipe arrived as 304/304L would it not be acceptable to use the Allowable Stresses for 304?

Either way, not a huge deal yet, that was only the first option.

Steve, I'd like to dig into your train of thought some more since it's not a great option to re-stock a thousand feet of pipe. The design temperature of the system is only 250°F and for SA-312 304ss in B31.3 the allowable stress is 67% of yield right up to 300°F so no change between ambient and design temperature.

The sustained loads are not what's causing the failure either, it is actually Seismic conditions. If the displacement stress range is not affected by the quality factor then this could just be a case of improper input into the flex analysis software. I'm actually reviewing someone else's analysis so I still need to make sure all the inputs into the software are up to snuff.

 

[BigInch]

Steve,

"perhaps the Committee was counting on most actual material to have yield strengths higher than the minimum".

I'm sure you have the feeling in your stomach that there is something about that statement that doesn't "ring" exactly right, no?

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv

 

[StevenHPerry]

Gibs,

Allowable stress due to earthquake is affected by E_j. I don't think it's an error in the inputs.

And a correction to what I wrote yesterday... E_j does affect S_A in an optional formulation of S_A. But it doesn't seem to matter in your case.


BigInch,

I get a lot of gut feelings that don't "'ring' exactly right." But I have developed the impression that the Committee doesn't always have a rock solid justification for everything included in "The Code."

Why does 31.3 give a 33% margin for seismic while 31.1 only gives 20%? If I say 25%, is that any more or less valid?

Again, that's just my impression from my interactions with Committee members.

- Steve Perry

 

[JohnGP]

GIBSTJEAN,
B31.3 would allow you to use the higher basic allowable stresses tabulated in Appendix A if the material was dual certified, but that would only begin to help you in this case if you had designed for 304L.

I can understand the desire to seek out a minimal cost solution to your dilemma - I suppose relocating the piping system to a less seismic prone area is out of the question....
Cheers,
John

 

[BigInch]

"If I say 25%, is that any more or less valid?"
Yes and no. It depends on which code is specified for your project.

I can equally ask, why are there different codes for different types of systems? When you can explain that, you'll also have the answer to your question.

"We have a leadership style that is too directive and doesn't listen sufficiently well. The top of the organisation doesn't listen sufficiently to what the bottom is saying." Tony Hayward CEO BP
"Being GREEN isn't easy." Kermit

http://www.youtube.com/watch?v=hpiIWMWWVco

http://virtualpipeline.spaces.liv