Flange Torque

[sshultz460]

We are setting up a flange torquing guideline and I am having alot of trouble deciding on what to use. Basically my problem is too much information. (how odd)

I have guidelines from my bolt manufacture to go to 52ksi bolt stress, and then you look at ASME PCC-1 Bolted flange joint assemblies and it gives you a diffrent number based again on bolt size. Even Hytorc says to torque to bolt size, and everyones numbers are diffrent.

Then I have an flange specific recomended torques (6"-300 series, 8"-300 series etx) from my gasket manufacturer (none of these torques will exceed 60 ksi bolt stress, and usually they are fairly close to the 52ksi stress numbers, some closer than others).

I work for a pipeline company so thermal stress isn't much of a concern.

All of these gaskets and flanges are to B16-20 (we use rasied face flanges and spiral wound gaskets)

My main question is this: Is it better engineering practice to have a flange specific program, or to torque a 14" 900 flange to the same value as a 16" 600 flange (just one example)

Per bolt stress (52ksi) you would take both of them to 1676 ft-lbs.(1 1/2-8 b7 bolt)

Per Gasket manufacturer you would take a 14" 900 series to 1164 ft-lbs and the 16" 600 series to 1600 ft lbs

Any help would be greatly appreciated

 

[unclesyd]

The gasket manufacturer is giving you the actual torque needed on each flange to seat a particular gasket. This value is based on the calculations for a particular the type gasket and flange and has nothing to do with developing a specific bolt stress. A commonly used bolt stress of 50,000 psi is used a guide for preloading bolts on flanged connections. In most cases as your example points out this will seat most gaskets but it will be too high for the softer gasket materials. This is where the ASME flange calculations or a manufacturers recommendation comes into play.
We used a value of 50,000 psi bolt stress for all CS process flanges with the exception of Class 150. On large SS flanges there was always a flange calculation based on the flange type, gasket material , and bolting material with the corresponding torque value to seat the gasket.

If your flanges are predominately CS Class 300 and above using spiral wound gaskets I see no problem setting a torque value based 52,000 psi bolt stress for each size bolt/stud used. Make sure you clearly state the conditions or basis for your table.

When there is any doubt or concern always run a flange calculation.

 

[tothepoint]

Class 600 and higher flanges, with B7 bolting, will have more than enough load available to seat spiral wound gaskets, so bolt torque should be established such that there is sufficient elastic strain of the fasteners. Typically, 50% yield of a B7 bolt (52 ksi) is an ideal assembly bolt stress in this situation.

 

[byrdj]

I have a related question.

I am not familar with "Class XXX" or ASME code B-16.

Does this code or class of flange have a mechanical limit for the compression of the gasket?.

The flanges I see on turbines have a gasket groove, so when pulled the flange is pulled metal to metal, the gasket compression is per gasket manufatures recomendation. The Bolting is then tighten farther to provide the proper clamping force.

However, I have experanced a lot of water piping tongue and groove flanges, with spiral wound installed, that will compress the gasket per the bolt loading. Thus loading the gasket proper is insufucent to hold the flange together, and torquing the bolts to hold the flange, will over crush the gasket. (proably wrong type gasket for flnage type)

Sorry to side track sshultz460's inquiry, but this has been a question to me for years now.

Thanks

 

[CoryPad]

byrdj,

I can't answer your questions regarding ASME code, but I can say that you described a well-designed system, i.e., one that allows the correct gasket compression when the flanges are in solid contact. Thus, the bolted joint is rigid, so the fasteners can be preloaded to a high level which should avoid problems with preload loss, fatigue, etc.

Regards,

Cory

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[sshultz460]

My big thing is that if you do a bolt size specific program, your clamping force for each series flange varies greatly.

1 1/2 bolt is used on 150 300 600 and 900 (and greater)

Each series flange has a diffrent bolt radius number of bolts and that makes thier clamping force on the gasket diffrent from series to series.

ASME B16-20 says a min of 30kpsi of total clamp force on all spiral wound gaskets.

Should I just math out 70-90% of max compression for every flange I have?

 

[TimtheToolMan]

Go to Boltup.com to get calculated torque values for both standard flanges and special flanges.

 

[1969grad]

The critical factor is the face stress on the gasket. For spiral wound gaskets the min for seating is about 10,000 psi and the max to avoid crushing is about 25,000 psi.

The problem is that the same torque can result in different bolt tensions depending on the lubricant. Old torque tables like found in Crane were based on oil and graphite. Lubricants available now have lower friction values which can result in higher bolt stresses for the same torque. In other words, you need to specify different torque values for different lubricants.

BTW, I know of two major gas transmission companies that prohibit the moly lubricant. Don't know why unless they are concerned with contractors over torqueing and crushing the gasket because the moly lube has the lowest friction coef.

Another problem you will have is the published friction factors for the different lubricants vary. That coupled with the fact that the associated equation to be used with the friction coef is an approximation.

Some people have suggested that torqueing bolts has an accuracy of at least 20%.

Hence, the increase in popularity for tensioning which takes the lubricant friction factor out of the equation. I should add that tensioning is normally used for size 20" and greater flange sizes in the pipeline industry.

The problem is that for some Class 600 flanges the number and size of bolts are capable of producing a gasket face stress that crushes the gasket. It is not an across the board problem so you will have to look at each size.

So I would suggest that you start with an allowable gasket face stress and work backwards to the bolt load and then the torque value based on the type of lubricant.

Final thought is that higher torque values do not mean a better seal.