Rubber gasket failure 2

[rcooper]

We have had a rubber gasket fail as shown in the photographs. A supplier is suggesting the gasket was misaligned when assembled and bolts were hit through the gasket to form the new holes (hence his photograph). I believe this is impossible. Even the most incompetent fitter would not and could not do this.

The gasket is in a full face to half face joint.

There appears to be a crease in the inside diameter of the gasket. I believe this caused a leak to occur. The resultant flow of water, going from 8.5 bar (g) to 0 bar (g) cavitated and caused the damage to the gasket within 100 hours of operation. The holes in the gasket are where the shoulder of the half face flange is, so will be where there is a high pressure drop in the leaking fluid.

The gasket is PN16 200mm nb.

Does anyone have any experience of this and what the mechanism of failure is likely to be?

Many thanks

 

[rcooper]

Further photograph

 

[rconner]

Forgive my ignorance but what is a "half face" joint (the last photograph surely looks strange!)

 

[rcooper]

rconner - it's probably a colloquialism.

BS EN 1092 refers to them as raised face flanges. That is what I need to start referring to them as.

I should have refered to the full face flange as a flat face flange.

 

[btrueblood]

Agree with rconner, your second photo shows what looks like a gasket that is 2 pipe sizes too large for the flange. The i.d. of the gasket should be within a few mm of the pipe or flange opening i.d., and it looks like the bolt holes don't line up with the holes in the gasket?

 

[gne77]

You're pretty brave to state what an incompetent fitter wouldn't do! I think the possibilities are infinite! I'd have to agree with your supplier. If I had to take a wild guess on the events that led up to failure... fitter needs a flange friday afternoon, finds one that fits an IPS pipe flange probably of some other size, makes it work, beer time.

 

[pch1]

I've seen failure of gaskets as indicated in the photographs, BUT it has been caused by fitters assembling tight fitting spools together and sliding the faces of the flanges together to align the bolt holes, instead of pushing the spools apart to correctly fit the flange faces.

I have seen those exact same crease marks on NCAF gaskets on large diameter, low pressure steam piping systems (Usually the exhaust end of a steam turbine). Initially all is well but over a very short time the gasket joint fails.

 

[pch1]

Why didn't the person in charge have the correct tools for the job?

The gasket doesn't seem to cover much of the raised face area at all, and if you are going to bolt up a flat face to a raised face flange then you need an adaptor as the flat face flange could be of more brittle material than the raised face and possibly crack when tightened up. Could the flat face flange be thin enough to have bowed as it was tightened up to the raised face, thereby helping the gasket to failure?

 

[rcooper]

Thanks all for the replies.

The gasket is the right size for the flange. The second photo shows the supplier demonstrating his hypothesis that the fitters incorrectly fitted the flange by pushing bolts through the rubber rather than through the bolt holes. When the correct bolt holes are used, the flange and gasket look normal. I don't know what the actual fitting the supplier is demonstrating on, it looks like a sharp reducer. It is probably all they had lying around in their workshop.

pch1, a star for you. There is a slight pinch/fold in the gasket near where all the damage happened. I suspected this was the origin of the problems, with the flow of water causing the further damage. It is good to hear somebody else has seen this damage and knows the cause for it in his circumstances.

It is worth noting, this is the bottom flange on a borehole pump rising main, between the pump and the main. It is therefore not possible to pressure test this joint following assembly and the gasket may have been leaking from day 1.

 

[rcooper]

And another star to pch1 for suggesting the flat face may have bowed. I am not in work tomorrow, but will check this in Friday.

I agree with everyone about not fixing a flat face to a raised face. Should not be done. I know about the problems with the flat face cracking, but hadn't thought about the bowing.

It's one of the joys of my job, the project managers don't involve the engineers until something goes wrong, then want to know why their projects are not working as they should have.

 

[rconner]

While I am not going to underestimate the ingenuity of "fitters", I am not sure I understand completely the particular shoddy assembly theory apparently proposed by the "supplier" with the second photograph [I guess to at least explain away the (disassembled after the blowout?) appearances of the first photograph?]. I believe I see some rubber duress radiating inward from the proper gasket bolthole (as if by the shank of a bolt as the gasket tried to blow or extrude out in tightening?) at the lower left of the first photograph, whereas it appears the "new hole" (flap?) the supplier apparently alleges was created by the installer somehow driving a (smaller than normal?) bolt through the flap looks suspiciously too close to the proper hole to be a smaller bolt driven through an apaprently properly spaced adjacent flange hole. How is that possible?

I aAm just curious also of how did the leak failure/blow-out?failure actually manifested itself? [Is it possible the gasket for some reason just "blew out" (moved radially outward as if moved outward by pressure from between normal confinement at least in the half flange annular area) -- I ask this due to the rather large what appear to be stress marks radiating inward from adjacent gasket bolt holes, in fan or seashell shaped fashion, as if by great force or movement against the bolt shank at the two gasket bolt holes in the first photograph)?]