When to Use Appendix 2? 1

[PipingEquipment]

I was told here that any flange over 24" has to be calculated using appendix 2. I just accepted this as a rule to the code, but now I'm wondering where this 'rule' is located.

In UG-44 it states that the following ASME standards are acceptable and in these standards (UG-44(h)) it lists B16.47-Large Diameter Steel Flanges NPS 26-60...

So my question is, when is it required that Appendix 2 is used instead of one of these standards?

 

[TGS4]

You use Appendix 2 when your flange ISN'T covered by one of the standards - such as a specialty flange, or one that isn't compliant with one of the specs. For what it's worth, use B16.5 or B16.47 whenever possible, up to the rated pressure/temperature in those standards.

 

[PipingEquipment]

Thanks for confirming what I thought, man this misinterpretation of that 'rule' has cost the company a lot of money over the years before I was here, I'm glad I caught that!

 

[michigander]

Keep in mind that 22" flanges are not recognized by B16.5. Use of a 22" flange requires an Appendix 2 design confirmation.

 

[rmw]

To say what has already been said differently, Appendix 2 is for non 'store bought' flanges.

rmw

 

[Gorpomon]

What do you mean "calculate a flange" Are you referring to bolt stress or flange stress?

It's my understanding that even with B16.5 or 16.47 flanges, if you want to calculate minimum bolt load, you would use Appendix 2. Can someone please verify this?

 

[TGS4]

Appendix 2 is never used to calculate actual bolt loads to use in assembly. Appendix S contains guidance on this aspect.

The bolt loads in Appendix 2 are for designing the flange ONLY.

 

[EngAddict]

I think you will find if you have a moment at the flange face you will also need to check it is compliant even if it is a stock flange.

 

[TGS4]

EngAddict - Check it - to what criteria? Also, ASME B16.5 and B16.47 say the following:

2.5.1 General. Use of flanged joints at either high or low temperature shall take into consideration the risk of joint leakage due to forces and moments developed in the connected piing or equipment. Provisions in paras. 2.5.2 and 2.5.3 are included as advisory with the aim of lessening the risks

So, unless you have high temperature (which 2.5.2 describes as temperatures in the creep range) or low temperature (which 2.5.3 describes as below the ductile-to-brittle transition temperature), the Standards seem to imply that there is a margin inherent for external loads. There are papers out there (don't make me spoon-feed them to you, please) that indicate that for some standard flanges, external loads, as measured by the equivalent pressure method, could be as high as the design pressure itself, again.

Whatever you do, do NOT take Appendix 2 as a measure of "risk of leakage". I'll say it again, Appendix 2 is for designing flanges, not ensuring that they won't leak. Do not use the bolt loads calculated in Appendix 2 in actual operation. See Appendix S for guidance on this matter.

 

[EngAddict]

I am aware of the equivalent pressure method. Like you said their is some margin for external loads but of course there is a limit. I was not just referring to leaking of the flange rather stress in the flange caused by external moments. Most of the time the loads are applied at the shell intersection, not the flange face but not always. We have found it prudent to check design in most of these cases.

 

[TGS4]

Don't get me started on the Appendix 2 stress calculations. Suffice it to say they are horrendous and do NOT represent any stress that resembles reality. Have you ever done an FEA on a flange, and found that the stresses in Appendix 2 don't match anything you calculate? I have...

And, how do flanges "fail"? They leak. Have you ever seen an over-stressed flange? I haven't. Would you know what one looks like? Probably excessive deflection. But you want to know how it failed - it leaked!

For a good primer on this specific topic, check out this paper

http://www.asmedl.org/getabs/servle...543000127000001&idtype=cvips&gifs=Yes&ref=no.

 

[Mandeep1]

I came across a peculiar thing in ASME Sec. VIII Div. 1 App. 2 calculations,

Take a 300lb 24in std B16.5 flange with standard B15.20 spiral wound gasket. The required flange thickness is about 4in something for these design conditions.

Then use a 600lb 24in flange, the required flange thickness jumps to 8.3in. The longitudinal hub stress becomes twice of the 300lb flange.

Next, analyze a 900lb 24in flange, the flange required thickness is reduced back to 4.282 in.

The design conditions are 200F and 350 psi, 0.0625in internal corrosion. The material is SA234WPB.

The culprit is the 'f' factor. This shows that a 600lb flange is worst than a 300 lb flange in this particular case per App. 2.

What we think is that some factors are going off the edge when computing the 'f' factor.

Has anybody seen this before?


best regards,
Mandeep Singh

 

[TGS4]

That's exactly what I'm talking about - the entire "stress" evaluation in Appendix 2 is wonky.

BTW - note that the pressure/temperature ratings in B16.5 have NO consideration for corrosion allowance. I know - it's scary, isn't it.