ASME B31.3 Code Compliant Flange Bolting

[ColonelSanders83]

Hello All,

I have been charged with ensuring B31.3 code compliance for a packaged process plant being built at my company. All the piping comes under the same criteria, but I have chosen the highest pressure and temperature line for an example.

6" sch 40 pipe ,A105 600# flanges
Fluid: Gentron 245fa
Design temperature: 500 F
Design Pressure: 880 psi

In specifying gaskets we chose Flexitallic style CG gaskets. Bolting will be as follows
A193-B7
B31.3 Ultimate per table A-2 = 125 ksi
B31.3 Yield per table A-2 = 105 ksi
B31.3 Allowable stress per table A-2 = 25 ksi (within my temp range)

The problem comes about from the Flexitallic installation procedure which calls out a target stress of 40 to 50 ksi. A little bit of research showed these values came from ASME PCC-1 (table 1) which is referenced for use in flange bolting by ASME B16.5 (Par 2.2) which in turn is called out by B31.3 for use in specifying compliant flanges.
A quick check using the BPVC Section VII div. 1 app. 2 rules for flange design showed that for the gasket in question a bolt stress of 11.8 ksi should be adequate to maintain joint integrity for both operating and gasket seating loads.

I am planning to specify bolt torques leading to about a 20 ksi per bolt stress for these joints. My question is as follows.

What is normally done in industry when specifying code bolting? Use of the vendor information would lead to a code violation on bolt allowable tensile stress. Ignoring the vendor recommendations and using my own calculation would lead to code compliance but I am concerned with the joint integrity since I didn't follow vendor recommended practice on installation. Any advice from experienced hands would be greatly appreciated.

 

[TGS4]

Follow PCC-1 and Flexitallic's recommendations.

ASME Section VIII, Division 1, Appendix 2 is for the DESIGN of the non-standard flanges. Refer to Appendix S that indicates that much higher actual assembly bolt stresses may be required.

 

[ColonelSanders83]

Thank you for the response. My only question now is why do the ASME codes specify limits on bolt stresses in tension that are knowingly violated in every installation? This doesn't seem to make a lot of sense, why not specify the allowable as 2/3 of fastener yield instead? 2/3 of yield is usually the upper limit of sustainable faster torque anyway.

 

[TGS4]

The "allowable" bolt stresses are for the design of flanges ONLY. They have no relation to the actual bolt stresses that you will use in actuality.

 

[thermmech]

Great point. You use those to size the flange, i.e. to calculate the developed bolting load that will cause the bending moment.
This is where the Taylor-Forge method becomes funny, and where you realize that some of the factors in the moment load equations are actually fudge factors.

 

[TWP]

thermmech, would you pls explain your last sentence. thanks in advance. TP