Calculating Bolt Torque for Appendix 2 Flanges

[sgtpickles]

I know this topic has been discussed on the forums previously, but in all the instances I looked at this question was never answered:

Is there a reliable method to determine proper bolt torque when designing an appendix 2 flange? I know the rules of thumb are to use a predetermined bolt stress, say 50 ksi for A193 studs/bolts, but I would argue that using these values in all situations may harm the gasket.

Similarily, using the minimum bolt loads calculated from App. 2, Wm1 or Wm2 only provides a means for the lower bound on the total bolt load.

A vendor has recently supplied us with a heat exchanger with a very small PTFE gasket. The minimum bolt load required is calculated to be ~20,000 lbs, yet they specified 130 ft lb torque on each of the twelve 3/4" bolts. This makes a huge total bolt load. This seems to be arbitrarily based on making a bolt stress of 25 ksi. The required bolt area calculates to be about 1/5 of the total bolt area provided. To me this seems like poor design where either the bolt size and/or number should have been reduced or the torque should be reduced.

Is there a proper method to use (not solely based on bolt stress) to determine suitable bolt torque? Is there a target range for the ratio of available bolt area to required area? Five times, as stated above, seems rediculous. Obviously if specific gasket data exists (proper bolt torque of maximum stress before crushing), then one should use that data, but what about in the cases where all you have are the y and m values and need to specify a proper bolt torque?

 

[nickelkid]

You may want to get a copy of the Pressure Vessel Design Maual by Dennis Moss. In it Procedure 2-10, Bolt Torque Required for Sealing Flanges is detailed very well.

 

[sgtpickles]

Thanks for the info RCHandy.

Procedure 2-10 covers the code calculations for the minimum required bolt load and throws on the addition delta F (decreasing compression force in gasket) to Hp. The total tightening force required to seal the joint is the greater of Wm2 or Hp + F. In the case where Wm2 dominates, this procedure is exactly the same as the code. Even in the case where Hp + F dominates, I would still argue that this value is just a minimum required bolting load/torque. I would be inclined to increase the torque up from this value, but not necissarily go up to the point where I base it entirely on bolt stress, if that makes any sense...

 

[arto]

Look in ASME PCC-1 "Guidelines for Pressure Boundary Bolted Flange Joint Assembly"

 

[SnTMan]

...and if possible, contact a gasket manufacturer for the recommended gasket stress.

Regards,

Mike

 

[HeviiGuy]

I've never tried to upload a file to the forum but, I'll give it a shot.

If successful, you'll see that the file contains an Excel spreadsheet that'll help you along. It shows you the intimate relationship of certain important variables. It allows you to calculate torque (for one thing) based on bolt stress, preload etc. It also rightly puts the onus on you to insert your assumed friction factor (K). ;-)

While it's not a full flange stress analysis worksheet, it should set you in the right direction...

Ciao,

HevïGuy
www.heviitech.com

 

[unclesyd]

There is some validity to your assumption about using bolt stress alone to effect the tightness of a bolted pipe joint especially for SW gaskets. Our piping design manual has a similar table to the Dacrco/Garlock table, but has one addition input factor and that is on SW we used different Minimum Gasket compression numbers based on the filler material in the gasket. Like using a graphite filled vs a mica/ceramic filled, where we use a lower number for graphite than that used for mica/ceramic. We gravitate to the higher values, with a max of 55,000 psi bolt stress using B7 or B16, as the temperature and/or pressure increase. Our min flange class for SW gaskets is Class 300 and when in doubt Class 600. The only time we shoot for the max bolt stress is for metallic gaskets which require flanges that can take this bolt loading without cupping.

On sheet gaskets we have a similar situation where we try to limit the bolt stress to 30,000 form the our original spec of 45,000 psi. We found out early on that there were different gasket seating requirements for Teflon vs the Blue Asbestos we were using. On all our low pressure designed flanges we have gone to many small bolts vs a few large ones and again limiting the bolt stress. On all water sytems where we use rubber gaskts we change bolting material and limit the bolt stress to 15,000 psi to 20,000 psi.

http://www.dracomech.com/metalgaskets/torque.html

http://www.garlock.com/ViewSolutionsPage?page=torqueTables

I know this is old hat to most but look at the values for the Class 150 3" and 8" flanges.