Flange Spacers and Sealing

[pogoGo]

Flange spacers are sometimes used when there is a need for flange skillets or when there are insufficient spool lengths. Each spacer has two gaskets surfaces on each end. Since the gaskets are in series, the bolt load transferred to each gasket is the full bolt load whether it is one gasket or two (or ten), so there isn't a need for a higher bolt load.

However, I've seen issues with getting RTJ spacers to seal. All instances I've seen were part of a another supplier's scope so I never found out what the cause was. I'm sure at the end of the day they just increased the bolt load and moved on.

My theory: When it is just two weldnecks and one gasket, the flange rotation caused by the bolt load results in uniform contact pressures on both side of the gasket; both weldnecks essentialy rotate into the gasket. When a spacer comes into the picture, the weldnecks still rotate into the gaskets but the spacer doesn't. If we focus on a single gasket, the side that contacts the weldneck will have a higher contact pressure than the side that contacts the spacer. Perhaps it may even reduce the contact pressure on the spacer side.

The stud lengths also need to be longer than usual when a spacer is used. I'm still thinking if this can have undesirable effects (the spacers I've seen are about 4" thick).

Does anyone have any insight into whether this is an actual phenomenon or am I overthinking this (or thinking incorrectly)?

 

[rtmorgan]

pogoGo,
Not over-thoinking, but here are some of the issues/comments from the dim recesses of my poor memory:
1. The extra stud length is not detrimental, it actually increases the effective length of the stud, adding to the effective elongation of the stud. This helps especially for pressure/thermal cycles.
2. In an RTJ gasket, remember that you are preforming plastic deformation of the gaskets into the groove. This is not instantly achieved, so an additional amount of the seating/sealing stresses are lost with time.
3. Better is to preload to some slightly (+10%??) load, and then follow up with a re-tighten 24 hours or so later. It's not necessary to 'hot bolt'. This helps regain bolt load loss due to all of the bolting/gasket imbedment and relaxation factors.
4. WN flange rotation will reduce bolt loads to a lower equilibrium, gasket seating stresses should be mostly equal on both WN and Spacer at any level of bolt loads.

Not a lot of help, I'm sure, but maybe you can glean some support/thoughts from these.