ASME BPVC.VIII.1-2023 Man App 2. Flange Connections Calculations

[Gavin_UK]

Hello All,
I am seeking advice on understanding this, I have scanned through the forums and found some similar lines of enquiry, but I still can't understand it this appendix fully. This is based on the 2023 release. I am pretty new to ASME and very new to pressure vessels.
1. Equation 5. (Pg 344.)
Why are we averaging the areas of bolts? The Load, W will be the actual number of bolts x the actual bolt stress. Sure there needs to be a factor of safety in this, but averaging doesn't do this consistently.

2. Just under equation 5, it says Sa shall be not less than tabulated value.... But where is the next calculation to check the Sa value isn't so high it wont damage the bolts?

3. Why isn't Hg = W for operation? Unless the bolts are flexible, comparable to the gasket? Once the bolts are tightened the bolt tension must equal the gasket force.

Thanks for any help.
Gavin

 

[r6155]

Where is the average term?
Regards

 

[Maslov]

1. I assume by averaging you are referring to (Am + Ab)/2. The fact that you are essentially taking Am and Ab, which is Am + extra_area, you end up designing your flange to withstand an additional stress: extra_area/2 * Sa. You are correct that it's not "consistent", but does it have to be?

2. That clause is to prevent damage to the flange by designing it as if the bolts were stressed to Sa, but you won't actually tighten the bolts to that stress or above it. By the definition of allowable stress, you can not go above it (UG-23).

3. There is a pressure*area force on the flange during operation that acts against the bolt force (works alongside the gasket force to try to open the joint).

 

[Gavin_UK]

Hello Both,
Thank you for the replies,
1. Yes, by the average term I meant adding (Am + Ab)/2.
I guess I cant understand why the Bolt load, W, would be related to the minimum area required of the bolts.
I would have thought we should work the out the minimum area, check the proposed design area is larger than this and then use the actual design area.

2. Ah, so am I using equation 5 wrong? Should I be using Wm1 and Wm2 to work out the Sa value, then checking that is greater than the one given in the material tables? If not why say Sa "Should not be less than.." When I actually need to use the Sa value?

3. I agree there is a pressure acting on the bolts, during operation so the stress on the bolts will increase, but unless the bolt stretch significantly the gaskets will still stay at the same compressed height.
So as the bolts are tightened the load on gasket is W, and so the load on the flange from the gasket is W.
But then when internal pressure is applied this gasket force W should stay the same. (In fact there might be small increase in it due to the internal pressure acting on the inner face of the gasket causing it to thicken, pushing more against the flange.)

Any help appreciated.

 

[Maslov]

The wording "should not be less than" seems to be there because normally allowable stresses are maximums you can not exceed. In this case, because we are designing the flange, the strategy is inverted when regarding bolt stresses. Bolts could theoretically be stressed to Sa, so your flange must be able to withstand the resultant stresses when that occurs, thus you use a value of Sa in your formulas that is equal to or higher than allowable. If you wanted, you could apply a safety factor and use a higher Sa, just like you could use a lower allowable when designing shells, heads, etc.

In reality, unless there is some reason, there is no need to overcomplicate things... Just use the tabulated value of Sa, as using a higher value will give you a unnecessarily thicker flange.

As for gaskets, you do bring up a good point. It looks to me like gasket deformation is not the focus of the formulas, they are concerned about loads, so maybe its been simplified to a free body problem with sealing forces. You also have to consider that it's set up for many types of gaskets (elastomer, spiral wound, double jacketed metal, etc.), so you could have a gasket with compression comparable to bolt elongation. This is compounded by creep/relaxation of the gasket over time. Maybe somebody else can chime in.

Regardless, you have to remember these are design formulas, so the numbers and assumptions may not reflect reality, but rather a conservative simplification of it.