Bolt Loads per Appendix 2 2

[gd1215]

Why is it that when I calculate bolt loads for seating and operating conditions, my seating bolt load comes out significantly higher than my operating condition? Is this the load I would use to specify installation torque? Seems awfully high. For instance, a 30" R.F. manway cover with rubber gasket gives a bolt load of ~98,000 lb. (100 psi internal pressure) for my operating condition but ~380,000 lb. for my seating condition. Refer to Section 2-5(e). Seating load seems to account for actual bolt areas whereas operating load only accounts for required bolt areas. Totally confused!

 

[twjag]

The purpose is to protect against "over bolting".

The following is taken form Taylor Forge Bulletin 502 Modern Flange Design 1964 (Fifth) Edition, Section III Basic Design Considerations, Precautions Against Excess Bolting:

"Where excess bolting cannot be avoided, the designer should make some provision against this known overload potential in the bolting. The ASME Code requires that at least 50% of this potential overload be recognized, and built into the design moment."

Note that the portion (Am+Ab)/2, in Formula 4 of App 2-5(e) is the same as Am + (Ab-Am)/2, where the first term is required area & the second term accounts for 50% of the excess bolting area. As noted, W need only be checked "at atmospheric temperature and before application of internal pressure".

 

[JamesBarlow]

If your seating load is less than your operating load the joint would open and you would loose your seal. The extreme difference between the two numbers is one of those ASME things. If 10% over your operating pressure is good enough, then %200 must be better.

 

[gd1215]

Thanks for the reply. The required torque for the bolts based on 380,000 lb. would then be ~226 ft-lb. Seems reasonable for 1-1/4" bolts but our customer kind of gasped at it. They use the "tighten it until it stops leaking method" and I am sure that is not 226 ft-lb! But I assume the high torque requirement is so that the vessel can meet the actual pressure rating for a 150# vessel (275 psi @ 100 deg. F).

 

[butelja]

The seating load is REQUIRED to make the gasket conform to the irregularities in the flange faces. If only the torque corresponding to the operating load is applied, then the gasket may not seal, as it hasn't been deformed enough to compensate for flange imperfections. Even at 226 ft*lbf, your bolt stress is only on the order of 10 KSI.

Did you say RUBBER gasket? For rubber gaskets, the minimum design seating stress is either 0 psi or 200 psi depending upon the durometer. Double check the value of Y that you used in your calculations. It seams that you used a much higer value.

 

[gd1215]

I used 200 psi for "y". However, I think my confusion is which load to use to calculate the bolt torque. In Appendix 2, one calculates a required bolt load in 2.5(c) and then a bolt load for flange design in 2.5(e). The bolt load in 2.5(e) is significantly higher than in 2.5(c). Now there are two bolt loads - which one do I use for the torque equation? I used 2.5(e) which is the load used for flange design. That is why I have such a high torque value. Did I use the wrong load??

 

[twjag]

As previously described, "W" from 2.5(e) is a protection against over bolting. Min required load/bolt at intial bolt tightening = AmSa/N, with N=number of bolts. Max bolt stress should not exceed Sa, based upon area Ab.

 

[twjag]

For soft gaskets in raised face design, I recommend you also check for potential crushing of the gasket.
Per origninal TF Bulletin 502, Nmin=AbSa/2y(pi)G.
We have found the above to provide good field results.
If N < Nmin, consider increasing N, or using stiffer gasket material with larger y gasket seating load factor.