ASME BPV VIII-1/VIII-2 - External Loads on Standard Flanges - An Update 7

[TGS4]

Last year, Interpretation BPV VIII-1-16-85 was issued that stated

Question: When superimposed static or dynamic reactions, as required by UG-22, are specified for the design of a pressure vessel, is it permitted to use an ASME standard product as defined in UG-11(c) produced in accordance with an ASME Standard referenced in UG-44 without performing supporting calculations that indicate the component is acceptable for all design conditions?
Reply: No

This caused much consternation in the pressure vessel and piping community because there was no "official" method to to perform such calculations. All that was available was the "Equivalent Pressure Method" whereby the internal pressure plus the equivalent pressure was set less-than-or-equal-to the rating pressure. That approach is excessively conservative.

In 2013, however, PVP2013-97814 was written to address this issue. But many users complained that their jurisdictions did not/would not accept such a published (peer reviewed) paper by a reputable author, because it did not have the approval of the ASME Boiler and Pressure Vessel Code Committee. So, with the help of other Code Committee members, I undertook to get the method described in PVP2013-97814 published as a Code Case.

Today, Code Case 2901 has been Administratively Board Approved. Look for it to be officially published in the next Code Case Supplement. In the meantime, please find an unofficial copy in the link. Please use this Code Case and spread word of it to all of your colleagues.

 

[SnTMan]

Thanks, TGS4, good to know.

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[XL83NL]

Hi TGS4, thanks for the heads up. Good news. I remember seeing that paper being presented @ PVP 13 Paris. Ive referenced the paper a few times here on eng-tips, but few people were familiar with it. I hope this will push it's acceptance one step further!

 

[tigny]

Hi dear Sirs,

is the Code Case 2901 available on the ASME website the official Code Case ? (we are waiting for our "official" paper version of the 2017 supplement 3 and 4).

Long live Warren Brown.

best regards,

 

[TGS4]

Indeed the official electronic version is available for download at the link that you provided. The paper version will be issued in due time.

 

[DK44]

Is it mandatory by code to check standard flange rating considering equivalent pressure instead of design pressure.

 

[Trestala]

I see that this Code Case refers explicitly to Welding Neck Flanges, how about other type of flanges (slip-on, threaded).

Also, could this concept be also applied "officially" for ASME VIII-1, Appendix 2 WN flanges? (e.g. small diameter towers with girth flanges with wind/seismic loads). Some commercial software already consider additional equivalent pressure due to wind or seismic for the design of girth flanges as per Appendix 2.

 

[TGS4]

Is it mandatory by code to check standard flange rating considering equivalent pressure instead of design pressure.

The interpretation that I quoted at the top indicated that performing a check is required. There is no requirement of what form that check may take.

Also, could this concept be also applied "officially" for ASME VIII-1, Appendix 2 WN flanges? (e.g. small diameter towers with girth flanges with wind/seismic loads). Some commercial software already consider additional equivalent pressure due to wind or seismic for the design of girth flanges as per Appendix 2.

No. If you want to incorporate external loads to Appendix 2 flanges, apply U-2(g) and refer to Division 2, Paragraph 4.16.

I see that this Code Case refers explicitly to Welding Neck Flanges, how about other type of flanges (slip-on, threaded).

The research backing up this Code Case was specific to WN flanges. When there is appropriate research on the other flange designs, we will likely update.

 

[Trestala]

No. If you want to incorporate external loads to Appendix 2 flanges, apply U-2(g) and refer to Division 2, Paragraph 4.16.

Noted and thanks. I believe Compress does the Appendix 2 Method with modified "P" value from our previous project. But I would go with your advice to use U-2(g) + VIII-2, Para. 4.16.

 

[Soave]

Thanks TGS4

 

[OnG_Engr]

TGS4 – Thanks for refreshing this topic.

All,

I have a Customer who requires API 660 be used to evaluate nozzles since last year. This has caused some problems with CC2901. A solution I worked out with my Customer is to not use end load calculations for flange pressure rating. Instead, the Customer is required to use a hydrostatic system test or other method to confirm our equipment’s flanges seal with Customer piping. The question I’ve posed to TGS4 is what the desired pressure should be to proof test the system.
I may have stumbled on a possible answer. Compress outputs a required flange pressure rating when one enables it to consider external loads. I don’t know how the Codeware people get this number but it looks like a starting place.

Thoughts? Tom Barsh?

 

[AWDMIKE]

As another point of interest, I recently reviewed supplier calculations (they used Compress) whereby I noticed the nozzles had projections which were about 12" longer than what we had specified.

After some digging and reverse engineering, I realized what Compress did. In the event where you apply loads at the nozzle to shell junction (forces and moments), Compress will calculate the moment at the face of the flange by translating the forces and moments but decrease the moments by a value which is equal to a force times the distance from the nozzle to shell junction to the flange face. I'm not sure why it would decrease it; it's possible that the moment would be higher at the face of the flange. So the farther the flange face from the junction (in other words, the longer the projection) the lower the moment Compress calculated would be on the flange face. Having a 22" long nozzle wasn't what we wanted, so we got ourselves out of trouble by doing something else, but it was interesting to figure out what Compress was doing in that situation.

 

[Trestala]

AWDMIKE, there is an option in Compress where you can specify if the loads you input are acting on the shell-nozzle junction or at the flange face.

Typical pipe flexibility studies would model a section of the shell where the nozzle is located, and the ends of that shell would have fixed boundary conditions. The source of the force and moment comes from the weight and expansion of the attached piping + wind/seismic, so the moment would be smaller at the flange face compared to the nozzle-shell junction, i.e. the piping is the one exerting forces while the pressure vessel is fixed.

In any case, you can specify in Compress if the loads are acting on the shell-nozzle junction or at the flange face. If you specify at the flange face, it would then increase the moment to translate it to the shell-nozzle junction.

 

[AWDMIKE]

Thanks Raymond. I did notice within Compress that you could specify the forces and moments at either the junction or the flange face. My comment was to the situation where you specify the values at the junction and how the new values are calculated by Compress at the flange face.