ASME BPVC Section VIII Division 1 UG-34

[DMay121]

Hey there,

I am currently analyzing a vessel that uses an unstayed flat head that is bolted down with thirty two 1-1/8-8 bolts. The gasket is a full-face typed elastomer gasket that follows sketch (p) of Fig. UG-34. Generally, Appendix 2 is used to carry out calculations for the bolts and the flange or cover, but in the first line of Appendix 2 it states it can only be used when the gasket material is contained completely within the bolt diameter. Seeing that this vessel utilizes a full-face gasket that has bolt holes machined out it, Appendix 2 can't be used.

I know that I can use equation 1 in UG-34 to calculate the thickness of the flat head, but how should I go about calculating bolt stresses and the required thickness of the flat head because of the compression stress of the bolts? Is there a standard I can reference or do I need to use FEA or some textbook equations?

Thanks,
David

 

[mechanical2014]

Hi David,

Try this DS/CEN/TR 1591-5. I have not used this myself yet but know such thing exists.

 

[DMay121]

mechanical2014, I will try to find a copy of this, thanks for pointing me to it.

Also, is my assumption about not being able to use Appendix 2 correct?

 

[SnTMan]

DMay121, full face gaskets are often designed per Taylor Forge Bulletin #45. You are aware this work was done for many years before FEA methods were available, right? Most canned software will do these calculations.

Examination of the relevant UG-34 equation will reveal that the bolting is of no interest in calculating the head required thickness.

You can actually use Appendix 2 methods to establish the required bolting, if taking proper account of gasket geometry, etc.

Regards,

Mike

 

[SnTMan]

DMay121, not having luck uploading this bulletin, will try again later.

Regards,

Mike

 

[DMay121]

SnTMan,

I am aware that that much of pressure vessel design was completed way before the days of FEA.

I am just curious about the effects of overbolting on the flange and cover thickness. The vessel that I am analyzing is a 20 psi vessel that has 32 bolts on a 33.38" diameter. The spacing is definitely more than adequate, but I am trying to understand the effects of what could happen if someone fully torqued all 32 of the 1 inch bolts. I know in appendix 2 they size the flange based on the pressure load and the gasket load, which takes into account the total force that can be applied by the bolts onto the flange.

Also, thank you for trying to upload that bulletin!

David

 

[SnTMan]

DMay121, the difference is that with full-face gaskets, no edge moment due to bolting is considered to act on the flange / head. Refer to UG-34 eqns (1) [Sketch p] and (2) [sketch j] for example. This is why I say the bolting is of no interest to the head thickness.

You will find an old, poor copy of Bullletin #45 attached, maybe you can locate a better one. A discussion and calc form for this is also included in TF Bulletin #502.

Regards,

Mike

 

[DMay121]

When you say no edge moment due to bolting, are you referring to the fact that the gasket compression reaction is at the same location of the bolting reaction, i.e. there is no moment acting on the flange?

Does that being true eliminate the need to consider bolting because there is effectively no bending of the cover or flange occurring?

 

[SnTMan]

Yes, and yes. Again, compare the UG-34 eqns (1) and (2).

This assumes that the gasket is the same width from bolt circle to OD and bolt circle to ID.

See Bulletin #45 for a more general treatment.

Regards,

Mike

 

[DMay121]

Mike,

If the gasket has twice the width inside the bolts than that outside the bolts, then I could definitely see how the equivalent location of gasket reaction would be inside of the bolt holes by a small amount. I found a book that deals with this very topic.

David

 

[XL83NL]

SnTMan, the link you provided did not work.

 

[SnTMan]

One last try...

 

[SnTMan]

XL83NL, DMay121, apologies, I cannot seem to get this file to upload, after several tries. Never had problems before...

Any suggested alternates to Engineering.com?

Regards,

Mike

 

[XL83NL]

Try 4shared, scribd.com or mediafire.

 

[SnTMan]

How bout this?

 

[SnTMan]

Was a bad file name. Had the pound (#) character in it.

Enjoy,

Mike

 

[DMay121]

SnTMan, I have another question for you. In regards to UG-37, area reinforcement calcs, if the required thickness of a head or shell, tr, is less than 1/16", do we use 1/16" for tr? I am trying to understand if UG-16 minimum head/shell thickness requirements govern tr in UG-37 calcs.

Thanks,
David

 

[XL83NL]

That one works SnTMan, thanks

 

[SnTMan]

DMay121, thickness required for pressure (and any other loadings) tr, is exactly that and may be thinner than UG-16 min. thickness, just as thickness of nozzle wall for pressure (and any other loadings), trn, is not subject to either UG-16 or UG-45 mins.

Regards,

Mike

 

[DMay121]

SnTMan,

Is the fact that tr can be less than UG-16 minimum (which makes perfect sense, because the reinforcement area required is dependent on the actual loading) stated anywhere in the code book? Or is that just common sense?

I am asking because I am acting as a third party reviewing calcs that were already completed. It would be nice to point to a specific article that justifies the above statement. Generally, using a higher tr increases the Areq, which means the original calculation could be considered "more conservative" because the area available still met that higher Areq.

Thanks again,
David