Flange Finish for Spiral Wound Gasket

[PeterIgg]

Hi
I am looking for assistance in justifying why a flange with a gasket sealing surface finish with approx 80 grooves/inch, roughly 63 micro inch should be acceptable for a spiral wound gasket in a process gas line (LNG plant). I do not know why the valve has this finish, it should be as per ASME B16.5, but if I can avoid re-machining the face - so much the better.

B16.5 does not seem to give any latitude and PCC-1 similarly does not give any assistance. However if someone could give technical reasoning as to why the face could be acceptable, it would be appreciated.

Peter

 

[bimr]

Flange face finish
The ASME B16.5 code requires that the flange face (raised face and flat face) has a specific roughness to ensure that this surface be compatible with the gasket and provide a high quality seal. `A serrated finish, either concentric or spiral, is required with 30 to 55 grooves per inch and a resultant roughness between 125 and 500 micro inches. This allows for various grades of surface finish to be made available by flange manufactures for the gasket contact surface of metal flanges.

http://www.wermac.org/flanges/flanges_stock-finish_smooth-finish.html

Both the pipe flange and valve flange should have the same roughness. The smooth surfaces mate to create a seal and depend on the flatness of the opposing faces to effect a seal.

Spiral wound gaskets require surface finishes in the range 125- to 250-μin.

http://www.pumpsandsystems.com/topics/seals/what-impact-flange-finish-gasket-performance

 

[PeterIgg]

Thanks for taking the time to respond. I thought that that would be the case, just optimistic enough to hope that there was a get out.

 

[racookpe1978]

On a practical and functional basis, the surface of your two flanges clamping the flexitallic (brand name) spiral wound gasket will seal adequately if the two raised face surfaces are parallel and uniform. Not exactly within the specifications understand, but as long as the surfaces have NO SCARS or gouges radially aligned so there is a flow path from inside to outside, the gasket will seal properly.

Your gasket surface (the number of spirals per inch) MUST be "flat enough" to crush the spiral wound gasket edges smoothly and evenly, and still be "rough enough" to engage the tiny edges of the spiral V's of the gasket sealing. That is why the surface should be slightly rough, not perfectly polished to a mirror surface finish. Now, a rubber or paper/cardboard gasket needs a smoother surface.

First thing: Go actually COUNT the grooves. You may find you are actually closer to the specification.
Second: Write the manufacturer, write the flex gasket maker and ASK if the number of grooves actually counted is adequate to seal the joint for your pressure and temperature and service conditions. If so, write an acceptance letter and cite the reference letters.
A general specification is for average conditions for average service factors and average flange surfaces.

 

[rconner]

If you have not already done so it might not hurt to also ask the valve manufacturer for their experience(i.e. why they do what they do).

 

[PeterIgg]

racookpe1978: Yes, that is the practical answer. However the client wants a justification to accept the joint other than it tests ok . But at the end of the day, that is the ultimate acceptance criteria, surely?

rconner: Doing that, but why did they make it that way in the first place??? Not what was asked for!