Rod material / finish for high temp o-ring application...

[quizzical1]

Hi All,

We have an application where we are trying to adapt a dynamic o-ring seal (movement up to .350" range and less than 1"/min rate) for use with a fixed volume of motor oil. The max temp is 350F and the current materials are brass housing, 1/8" dia 303 SS precision ground rod and Viton 90A o-ring. Pressures are up to 3000 psig.

We seem to have no loss of height in the static condition over time, but the unit seems to lose rod height (appr .005" per stroke) which indicates the rod needs to have a better finish (currently electropolished).

The unit will be exposed to water / steam.

My Questions:

Would a better surface treatment (e.g. chrome / nickel plating) provide a mirror finish to eliminate the surface roughness and help reduce the fouling of the surface when exposed to the elements?

Should we consider a different rod material altogether?

Would there be any advantage to increasing o-ring size from an 0XX series to a 1XX series?

Thanks in advance,

Quiz

 

[flexibox]

Wow this is small. Are you experiencing Extrusion of the Oring? At these high pressures and Temperatures, you may see extrusion that may contribute to your dynamic Leakage. Have you inspected the condition of the Oring carefully after you had the leak? Can you report on your findings?

 

[quizzical1]

Hi Flexibox,

No extrusion was evident. We are using 90A duro o-rings and have a diametral guide/rod clearance of .003" (Apple Rubber handbook states .010" max @ 3000 psi) and a 25% wider groove (.125 vs .101 std) to allow for swelling due to the high temp.

What was evident was compression set (flattening on the OD/ID of o-ring), but the unit still operates fine when returned to between 50F - 150F w/o any stroke loss. This leads me to suspect that the o-ring is softening enough at the 350F temp to allow slight leakage past it. As an aside, we also tried Aflas and Kalrez, neither of which worked well at all. Our next test will be to try Viton ETP to see if it has better hi temp operating characteristics.

I know the rod finishes vary a bit (16 uIN - 32 uIN) which is why I wanted to see if there is a std process (e.g. grinding, lapping, chrome plating) we need to do to get the mirror finish (4 - 8 uIN) required for a hydraulic rod seal.

Quiz

 

[MikeHalloran]

I suggest you look at a different seal material entirely; Bal-Seals.

I have found them far superior to elastomer seals for very slow reciprocating service with nonlubricating fluids.



Mike Halloran
Pembroke Pines, FL, USA

 

[EdDanzer]

You need to calculate the expansion of the brass housing, shaft and o-ring to determine if the temperature change is a portion of the problem. The o-ring will grow faster than the shaft and housing increasing the preload and shortening the seal life.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[quizzical1]

Thanks everyone - I really appreciate your collective advice.

Mike - we have tried a spring-loaded Variseal with marginal success. I have contacted Bal Seal for their advice.

Ed - we know the unit performs properly up to about 180F or so w/o leakage. We were instructed from the o-ring mfr to increase the groove width an extra 25% (.125 wide vs .101 std) and this definitely decreased the bleed down decay rate by allowing the o-ring additional room to "breathe" at the elevated temp. But it still loses about .005" height at room temp after every 8 hrs@350F run. The unit performs repeatably w/o any leak down when operated between 35F and 180F.

This leads me to suspect shaft finish. What is the best finish / process to use for sliding rod applications?

 

[MikeHalloran]

Metals are not very satisfactory; we started out using stainless steel, and results improved as we hand-polished it more and more, using Simichrome as the last step. Mirror finish, or damn near.

Best rod material was sapphire. Don't dismiss it out of hand; it can be affordable in reasonable quantities.

Next best rod material, and cheaper than sapphire but not quite as straight, was fire-polished glass. I.e., glass rod was ground and polished, and then its surface was partially melted. This is a high-skill operation that requires developing a partnership with your supplier.



Mike Halloran
Pembroke Pines, FL, USA

 

[btrueblood]

Ah, brings back memories, Mike.

FWIW, the OP stated he was using 303. Changing to a non-sulfurized stainless steel (eg 304) might be a good first step.

 

[EdDanzer]

Have you ever measured the set the o-ring takes if you just do a temperature cycle at no pressure? We have observed many problems with seals that do not fill the groove correctly at high pressure or move significantly when going from low pressure to high pressure and back to low pressure. Wear on the OD of your o-ring could be even more significant as the groove will be rougher than the rod. Making the groove wider may make the OD wear worse at lower temperatures.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[quizzical1]

Thanks Mike - we will give the Simichrome a try.

btrueblood - would switching to 304SS make that big of a difference? How about 316SS (we normally stock 303SS and 316SS)?

Ed - have not tried a non-pressurized unit test yet. We did try 3 units: .085" groove width, .101 (book std) and the 25% wider .125" and noticed the leak down condition got progressively better. The .085 wide o-ring looked like it was molded to be a square o-ring upon disassembly after 40 hrs @ 350F. The .101 wide was heavily flatted on the ID and OD and the .125 wide was slightly flatted.

Upon testing between its normal operation temp range of 20F - 180F the 350F-soaked unit performs fine w/o any noticable loss in stroke (dial indicator used to verify).

What's got me puzzled is we have two sizes: small .125 shaft dia and 006 o-ring / large .250 shaft dia and 108 o-ring. The large unit seems to take a loss (settle) of piston height of .015" and holds while the smaller .125 shaft unit seems to lose .005" height / day (8 hr @ 350F). This leads me to think the shaft finish is the culprit.

Quiz

 

[btrueblood]

316 would probably be fine. Somewhere, somewhen, I have read that the sulfur inclusions in 303 make getting a superfine finish nearly impossible with 303 - the sulfur just keeps leaving pits in the finish, and continued polishing just exposes fresh sulfur inclusions. I generally avoid the stuff for anything with a dynamic seal; also most shops today have plenty of carbide tooling lying around, and so quote machining of 303 (which machines quicker but costs a bit more in raw stock) at about the same price as 304.

 

[quizzical1]

Thanks btrueblood - we noticed the same thing, even on precision ground (+.0000/-.0002) 303SS stock from Carpenter. I wasn't sure if it was the mill run / lot or not, but what you wrote makes plenty of sense.

We'll switch to 316SS anyway. I think that will help in the long run with inhibiting any fouling of the surface finish when the extended rod is exposed to the steam / water.

Quiz