Cold water affects on Hydrostatic Testing of Static 300 Series FKM 90 Vacuum Gland Seals

[johnchrc]

Two questions:

1) Could cold temperature fluid (40F vs. 80F) cause a std o-ring seal to leak after it has already energized and began holding pressure differential?
I know it can cause problems achieving a seal because it becomes hard and loses elasticity. I have seen where we just warmed the fluid with hot tap water and then the seal will energize. I have a different case here where the seal energized even though it was cold but began leaking before it should have. I just wonder if you can contribute this to cold fluid or is something else at play. I have tools my company hydrostatically tests from 10,000 to 15,000 psi with ambient water in Houston climate and hold pressure for 5 minutes with no leaks. Ave Squeeze with pressure according to FEA distortion is 9.7% at 15,000 psi. We have had a cold spell with temps dropping into upper 30F range and have one of two identical static seals leaking with pressure > 10,000 psi. We could stabilize at 10,000 psi. I did not see the test or the tools as they were shipped with waiver stating 10ksi test was OK. They said 75% of the tools could not reach DP above 10ksi.

NOW, We are getting ready to do another round of 15ksi tool tests next week and I'm going to get involved in the assembly and testing. It is still very cold. I want to fill with warm water to test but will have to prepare. I hope this will work. Production want's us to change to tool to a 10 ksi rated tool which makes no sense to me. We already have a 10 ksi tool. Either obsolete or fix the problem. Don't create ECO to change rating. (Sigh)

2) I would prefer to use mineral oil inhibited water or glycol because of the viscosity and anti-corrosion properties but my Manager wants to test "worst case". In service it will not see pure freshwater so I don't agree. What do most hydrostatic testing labs use to check seals in tool for QC purposes? Does anyone know?


- CJ

 

[btrueblood]

1) Yes. The same type of seal as you described, operating under similar temperature change conditions as you described, is what caused the Space Shuttle Challenger disaster.

2) We hydro with city water, it's cheap and we can stand a little rust. But we use a softer o-ring than you also.

 

[Compositepro]

A seal that has been established and is "energized" by pressure can start to leak if there is movement in the seal. Movement can be caused by changes in temperature or pressure.

 

[dgallup]

FKM SHOULD keep working to much lower temperatures, standard grades typically work down to about -20 degrees F. However, the amount of squeeze will be affected by temperature. This will vary with layout and the other materials in the sealing system. I use a spreadsheet that calculates min, max & nominal stretch corrected squeeze at room temperature, min and max service temperature. Since most elastomers have larger thermal coefficients of expansion than the mating components, low temperature compression usually drops down quite a bit. You don't say what the o-ring cross section is or if it's a static or dynamic seal but 9.7% compression sounds very low to me, especially for a static seal.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[johnchrc]

It is a 349 90 duro Viton with PTFE Back-ups. Average squeeze is 15%. Static Seal. Has Xylan 1425 1 mil thick finish on ID land (included in dimension).
Piston Seal
Bore ID 4.965/4.970
Groove OD: 4.613/4.599
Piston OD: 4.960/4.965
.402 wide
R.008 at shoulders and R.016 in corners
ID under groove 4.173
OD of Housing 5.90"

Surface finish is 32 on land and 63 in groove.

At some point we will test this tool again with city water (cold) and then if it leaks will fill with 100F tap water and try. After that, we will disassemble and examine parts.

- CJ

 

[johnchrc]

This does not include material deformation from FEA results which resulted in average of 9.6% squeeze at 15,000 psi. A loss of 5.4% of squeeze. Did not look at thermal contraction/expansion.

I know it holds 10,000 psi and leaks above that. We have not been given chance to analyze this because of other "hot projects". Maybe end of week.

- CJ

 

[btrueblood]

"I know it holds 10,000 psi and leaks above that."

That wasn't clear to me from your 1st post. Temperature may have less to do with the problem than movement of the structure, though there are still potential problems with fluoroelastomers (and -20F may be possible with some, but not all FKM variants, do check with supplier, or do your own test of compression set at various temperatures, it's pretty easy to do). Any sign of extrusion of either backup rings or the o-ring?