bolts preload in the flange joint 1

[josedipofi]

I have a question regarding bolts preload:
I calculated a flange according to appendix 2 of Asme VIII div.1, and now i have to define the max preload bolts to assemble the joint, i saw several calculations where is showed a pre stress of the bolts of 50% of the allowable stress.(this value is quite common between our vendor)
If i check the flange with that value, the stress of the flange&rigidity factor are too high.
My question is: the preload value to be used during assembly the joint must be checked/verified according to appendix 2 in any case or not? Looks that nobody do that, i mean that after calculating the flange almost no vendor check it against preload of the bolts

 

[TGS4]

You're making the fatal error of confusing design with operation. Use the values presented and provided in Appendix 2 to design your flange, but follow the advice in Appendix S and actually use higher preload in service. Although this would appear to be contradictory, it sort of is, but it's how flanges are designed in Div. 1.

BTW, flanges don't fail due to over-stress. They leak. Be less concerned about the stresses and more concerned about making sure you have enough preload so there's no leakage.

 

[josedipofi]

Thank you for your reply TGS4.
You say that i have to focus more on leakage of the joint than the stress of the flange and in principle i’m agree with you but anyway you have to set a preload value for your bolts or not?
According to appendix S is clear that a higher value of the stress in the bolts is recommended during assembly of the joint (compare with the one using in design the flange) and also according to ASME PCC-1 you find the same, see below what is stated in appendix O of the same code:

(b) The maximum permissible bolt stress (Sbmax) must
be selected by the end-user. This value is intended to
eliminate damage to the bolt or assembly equipment
during assembly and may vary from site to site. It is
typically in the range of 40% to 70% of ambient bolt
yield stress (see section 10).

In the same appendix you will find also some tables where are showed some typical value of bolt stress to apply, previously checked with appropriate calculation (i.e. FEM analysis or Elastic Closed Form Analysis).
Say that my question is: after i designed my flange according to appendix 2, how can i decide which value (preload stress in the bolts) is more suitable for my joint without performing a complicate calculation?
Thank you for your replay

 

[SnTMan]

josedipofi, one fairly common way is to get from the gasket maker a recommended applied gasket stress. Note that this will generally be different from the "y" stress used in Appendix 2. From this a bolt load and in turn a bolt stress can be calculated.

Regards,

Mike

 

[TGS4]

I would second SnTMan's recommendation - ask your gasket manufacturer.

 

[josedipofi]

Sorry gent. but my concern about the flange is sill there!
thank you anyway

 

[TGS4]

Please be more specific about your concerns about the flange. Is it that there will be higher stresses in the flange? Are you concerned about flange rotation? Are you concerned about plasticity in the flange?

 

[josedipofi]

According to ASME PCC-1 you have to check all components of your joint and this imply some not simple calculations, of course you need some data from the gasket maker but this is not enough:

(d) The maximum permissible bolt stress for the
flange (Sfmax) must be determined, based on the particular
flange configuration. This may be found using either
elastic closed-form solutions or elastic–plastic finite element
analysis, as outlined in section O-5. In addition,
when the limits are being calculated, the flange rotation
at that load should also be determined (_fmax). Example
flange limit loads for elastic closed-form solutions and
elastic–plastic finite element solutions are outlined in
Tables O-1 through O-7.
that is what you find in the ASME PCC-1

I’m right when I say that in any case you have to perform some checking of your joint (could be also the gasket maker of course), say FEM analysis instead of different method to prove that the joint is ok?
The discussion is valid for all kind of joint, standard dimension and not but customize flanges are mine main concern
My intention is to understand if there are a simple method that give me enough confident in a calculation of the joints.
I’m also interest which philosophy other companies around the world usually follow
Thank you

 

[TGS4]

If you are actually interested in the "actual" stresses in the flange under any condition, then the simplified calculation in Appendix 2 are not for you. They are design calculations, and not stress analysis calculations.

I have had success in replicating the results of FEA computations using EN-1591. Those calculations are ridiculously complicated, though. I could do either, but I can be quicker with an FEA (personally-speaking only).

 

[SnTMan]

As a practical matter, MOST flanges designed to Apx 2 are able to withstand the bolt loads actually applied to them and give good service. As TGS4 says, analyzing as opposed to designing flanges to APX 2 is hardly worth the effort, especially when using commercial software which "controls" a lot of things for you.

Regards,

Mike