Tank Manways

[hopper146]

Say...I have a 24" manway on an oil storage tank, flat face flange, 3/4 inch B7 studs and a neoprene gasket....anyone know where I can find torque specs for this application.?...thanks...

 

[JStephen]

I'n not aware that a torque wrench is normally used for this application- the bolt torque is not especially critical. Probably 50-100 ft-lbs would be typical.

We don't normally see neoprene allowed for oil service, you might check that also.

 

[hopper146]

Ahhhh...Looks like a new mat'l spec, now a garlock 3000 or equiv.

 

[metengr]

Talk to the gasket folks.

 

[arto]

look at ASME PCC-1 "Guidelines for Pressure Boundary Bolted Flange Joint Assembly "

 

[hopper146]

Thanks.

 

[BenThayer]

make sure your gasket is within the scope of the ASME reference.

we have seen ASME reference ranges that one end would either exceed or not meet the requirements for the gasket.

it is pretty bad when a gasket fails and you find out you did not use the minimum required torque or recommended torque sequence.

if you went to a mechanic to replace your head gasket and you asked him what torque he was going to use and he said "we torque it til its tight" would you still let him do the work?

 

[deanc]

http://www.boltscience.com/pages/gasket.htm

 

[JStephen]

A normal 24" API tank manway is fairly thin (much lighter flange than a 150# flange of the same size, 5/16" and up), is flat-faced, and may use either a ring or full-face gasket. It is not an ASME design. Actual pressure will run around 10-20 psi. The bolt pattern is not a normal 150# pattern, and uses more bolts of smaller size. The gaskets are normally cut from rubber-type sheet materials, not the spiral-wound type gaskets used in piping. It is not a head gasket and doesn't need to be treated like one.

Several years ago, I did have a head gasket that failed, but it was the original gasket in the engine, had never been disassembled.

 

[JStephen]

In the boltscience link above, note a sentence down near the end:
"Implicit in this method is that the bolt spacing and the rigidity of the joint flanges are such, that problems will not occur because of deficiencies in these areas"

The problem is that API manway flanges, in the thinner designs are fairly flexible. This isn't normally a problem in sealing up (assuming a pre-determined torque isn't specified) but is a problem in analysis. In particular, unless the flexibility of the flange, the cover plate, the neck, and possibly the tank shell are considered together, there's no good way to estimate the bolt force remaining after the hydrostatic load is applied. You can go at the stiffnesses using the circular-plate and cylindrical shell formulas in Formulas for Stress and Strain, but it quickly becomes a mess.

Unlike 150# flanges, manway flange and cover thicknesses vary according to tank height and manway neck thickness varies according to shell thickness. So if you calculate a tightening torque based on a particular manway, it should only be good for that one manway, and vary somewhat for others.

If you check the API manway cover thicknesses, they seem to be based on about a 20,000-25,000 psi bending stress, treating the cover as a simply supported circular plate. If any significant edge moment due to bolts is superimposed on top of this, you run into some high stresses. API doesn't specify or limit a stress in this application, and the covers seem to function fine, but it is another problem in trying to analyze the manways.

 

[MJCronin]

hooper...

Don't you have more important things to worry about ?

The manway won't leak in a low-pressure tank. The sequence of tightening is not critical, like it would be in high-pressure joint.

Thumb tighten each bolt and then give it a 1/4 turn.

You should have some kind of table or chart for bolt torques for piping flanges.....which are much more important than manways.

If not, you can find one at the Flexatallic website

Or look here:

http://www.eng-tips.com/viewthread.cfm?qid=29658&page=9

MJC

 

[hopper146]

Ahhh yes....I always have more important things to worry about. But....when the client wants a torque value, the client gets a torque value.

 

[JoeTank]

Hopper,
Dude! API doesn't state a torque value, so share that with him. As stated above, these are very thin flanges and sometimes they are not even machined. Somehow, the contractors get them to seal without even a fraction of the dicussions already posted herein. I work with tanks all the time. I've never encountered this question in over thirty years of fielding questions.

Joe Tank

 

[hopper146]

Ahhhh...the story of my life. When inspecting and checking. I've seen everything from 40ft.lbs. to 200#s marked on these tank flanges. Usually hand tightened works just fine. Its the normal pain when the client decides he wants his paperwork for something non-critical such as this.