Dear

I have some problem about the flange leakage after pre-commissioning.
The rating of flange is 1500#, the size is 20", material is SUS 347 and use ring-joint gasket.
During hydrotest and pre-comming, there was no problem. But during re-start after pre-commissioning, the leakage was occured.

Gasket problem? Or Bolt? I don't have any idea.

Are there any members who have that kind of experience?

Please help me.

Once you have problem at field, it is hard to make the conclusion of failure source because there are so many potential factors such as gasket material, bolting material, excessive bending moment, chloride attack during hydro-test, and other metallurgy problems. My understanding is ; Austenitic stainless steel(321/347) pipe is generally used for high temperature hydrocarbon/hydrogen service in refineries.
You’d better let members get more detail information about your problem, that is, service fluid, temperature, bolt material, and gasket material, max. bending moment. Also, please let us know if you have a leakage problem in only single point or all flanges ?

I would like to point out a possible factor; if a high pressure system fluid is exposed to elevated temperature(extremely high temp), alloy or austenitic stainless bolt materials(A320) may cause flange leakage problem. In this case, Martensite stainless steel bolt(A453-660) material is a solution because its thermal expansion coefficient is much comparable to austenitic steels bolts at high temp. and high tensile load.

Is it resonable to take gasket material young's modulus as the effective gasket modulus of ring type joint in flange leakage module of Caesar-II ?

Oct'91 Coade newsletter page 5 gives practical usage limitation no 3 as follow:

Leakage for flanges with ring type joints can not be predicted.

Does the above mean that flange leakage module of Caesar-II is not applicable for ring type flange joints ?

regards,

sam

The purport of my earlier response was that for ring type gasket, postulation of flange leakage is not very accurate in Caesar-II flange leakage calculation.

In critical services, either spiral wound gaskets or enough care in minimising insertion gap, flange misalignment need to be carefully prevented & criss-cross bolt-tightening sequence in stages (not to the maximum tightening stress at one go) to be followed. Conical spring washer in case of dissimilar bolt/flange material expansion or use of compatible bolt material of similar expansion coefficient as that of flange material is to be followed. In thermal shocks in start-up or shut-down, bolt may be at different temperature from flange. Spring washer will help in this case,too, in maintaining uniform gasket stress.

Maximum allowed preload stress to bolt should be adopted after checking flange design for that condition, as this can avoid the need for any risky hot bolt-tightening.

regards,

sam