A quick and basic question: Pipe flange design criteria 1

human909 · Jul 25, 2024

Hi all, Second time this week I've reached out on what is a pretty basic question but it is something outside my normal experience. As I said in the last post feel free to flame away on me, an engineer who is poking his head outside his normal area of competency.

Background: A client approached me with a problem. I'm likely to offer guidance and a suggestion to consult an engineer more experienced in the area, I'm not sure it is worth the risk for me, but it is a good client so I'll offer some friendly guidance on where to seek a solution. Regardless the topic has perked my interest.

The problem:The client has fabricated a bunch of pipework for compressed air with flanges rated for 7bar. The client has now found out that the the customer scope included the option to upgrade the pressure beyond this I believe up to 10bar. The fabricator doesn't want to cut and reweld dozens (possibly hundreds) of flanges and is looking for a solution.

(Under the relevant code AS 2129 it specifies specific flanges for specific pressures/temperatures. These all have predefined dimensions etc.. As I understand it ANSI has similar guidance.)

My questions:
Using some preliminary basic engineering analysis it seems to me that pipe flange design is largely dictated by deflection and sealing requirement rather than yielding or material failure. Is this correct?

One obvious solution that occurs to me is to have half circle flanges to double up the flange. If it is thick enough then deflection and strength criteria seem to be satisfied. Does this make sense?

The bit I don't know enough about is the local code requirement. Eg even if it works from a engineering perspective it still might not comply. Hence I'll probably stay away from the job. It likely isn't worth my time, but it does interest me.

1503-44 · Jul 25, 2024

Yes, misalignment and seal failure is critical, once you get beyond bolting patterns sufficient for axial force and wall thickness due to internal pressure.

It may be possible to solve a deflection problem with a backing ring around the flange, but as I often say, just because you can do something doesn't mean it's a good idea. The underlying flange would still not meet specifications for the increased pressure, as originally intended, and should anything happen in future, regardless of the cause, you will find yourself on the hook for it. Since design of these things is apparently not your speciality, it would put you on difficult footings to explain why you did such a thing. Plus the code apparently requires specific flanges for specific maximum pressures, a problem that is likely not going to go away with modifications. If modifications were indeed even permitted by that code, it should explain the whats and hows in detail. Plus, it's likely to be eventually rejected anyway and it would be your sole burden of proof to justify adequacy. My crystal ball says "Not worth the effort", keeping in mind that the fabricator is trying to put you on the hook for his mistake.

Your theory is correct, failure of many things is not only the result of high stress and loss of sealing is often a critical case. Take comfort in that and walk away.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

human909 · Jul 25, 2024

1503-44 said:
My crystal ball says "Not worth the effort", keeping in mind that the fabricator is trying to put you on the hook for his mistake.
....

Take comfort in that and walk away.

Yes. I agree. It isn't worth the risk, I have plenty of work.

I don't mind stepping outside the norms if there is sound engineering basis for doing so. I'm used to designing bespoke structures and equipment. Particularly for temporary works and/or when I can be on site and observe and monitor things. But signing off on something like this doesn't seem to be worth the fee.

Thanks. [thumbsup]

LittleInch · Jul 25, 2024

I agree with all mr 44 says.

I assume the flanges are the type "D" flanges rated at 7 bar, otherwise he wouldn't have to go and change them.

Flange design is a work of art rather than science, at least for established flanges. It's common knowledge that if you subject ASME (not ANSI) flanges to a design to something like ASME VIII appendix 2 I think, which is how you design "one off" flanges for pressure vessels, a lot of them "fail". But the design is proven over decades so everyone keeps using them.

Flanges get a variety of design issues, from bending / moments in two planes, fatigue, sealing force, bolt stress, shear that simply adding a half backing plate won't solve them.

Especially the bolts.

The other factor is testing. You 7 bar design flange will be good for a 1.5 times (10.5bar) test pressure so will probably actually operate at 10 bar at very close to max allowable or yield. However your 10 bar system will need to be tested at 15 bar. Now your 7 bar flange is at double it's "rated" pressure value. High risk of failure.

The fabricator has a solution - replace ALL the flanges and also identify any other weak bits which are not good for 10 bar. There is no other practical solution that anyone would A) certify (I don't think I would) or B) provide insurance cover for.

I've seen and heard some crazy things in my time, but modifying flanges to exceed their rated Max Working Pressure isn't one I ever seen or even heard of before now. SO you are in unchartered waters which is a strange and lonely place to be when a storm arrives.

Your decision is a good one. IMHO.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

A quick and basic question: Pipe flange design criteria 1

human909

Structural

Recommended for you

1503-44

Petroleum

human909

Structural

LittleInch

Petroleum

Similar threads

Part and Inventory Search

Sponsor