Moments on spherical ends/ellipsoidal heads

[nick262b]

Hi all,


Can anyone point me in the right direction for assessing moment loads on a dished/ellipsoidal end, such as over rotation of the dished end flange due to excessive bolt load? (in this case for a floating head heat exchanger)

Can you use ASME VIII flange design calcs and if so how does one accommodate the quite different hub/neck arrangement from that illustrated in the code, i.e. with the hub and neck parallel to the applied bolt load. Can I do some simplistic moment calculations to estimate the load at the faux hub.

(See attached pic sorry for my lack of word skills!)

Thank you for your time.

Regards,

Nick

 

[SnTMan]

nick262b, I am making the assumption that your are working to Sec VIII, Div 1. If so you should designing your floating head cover to Apx 1-6, and I further assume, Figure (d).

The moment is calculated as given in the nomenclature. No calculation for flange rigidity is given (or apparently required), but a conservative method is to calculate as a loose ring flange per Apx 2, 2-14.

Note that in Apx 1-6 the continuity between head and flange is unaccounted for. Codecalc has a method to account for this continuity and other software may have as well. (I am not familiar with the details of this method as I do not bother with it.)

Regards,

Mike

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

 

[nick262b]

Hi Mike,

Thank you for your reply.

I have just had a look through apx 1-6 and yes you are right in that we have as per fig (d). In this case we are dealing with an existing piece of equipment and wanted to calculate the maximum bolt load we could apply to the joint. In reality it is likely that we could apply significant bolt load without worrying about this calculation as the joint is substantial, it was more out of interest that I wondered how one would asses for over rotation due to excessive bolt load in this instance (or if over rotation is a potential issue to consider). As you have pointed out the calculations in Div 1 do not account for the continuity between the flange ring and dished head but I will investigate if our Finglow software can perform the calculation you have described and how this is done.

Again just out of interest:

Why would you not normally bother with using a continuity method, do you see it as overkill or do you see it as unnecessary as per the code requirements?

Have you ever in your experience come across a bolted head that has been damaged through over-tensioning?

Once again thanks for your time Mike its very much appreciated.

Regards,

Nick

 

[SnTMan]

nick262b, my primary reason is I have no familiarity with the theory, as I have never seen any material on it. In other words I don't know what the software is doing. Another reason is FH flange thickness is often governed by cross-over area, a calculation giving a thinner flange (if that is what the software is doing) is of no benefit.

And no, I haven't, but my field experience is not extensive.

Good luck with the research,

Mike

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

 

[r6155]

Forget the software, use hand calculations.
Regards
r6155

 

[nick262b]

I'll re post if I find anything of use, I think some hand calcs will be best, who knows maybe I'll get bored and build a model!

Thanks again guys.

Nick