Rulon not working...

[tracetrimble]

I'm looking for an FDA-approved bearing material with high abrasion resistance. This application is for the food industry, similar to a packing seal, subjected to low pressure (<5 psi) water-based product, up to 220ºF. I have tested virgin teflon and two grades of FDA reinforced PTFE, including Rulon-641 ($$$!). These materials are wearing too quickly, and surprisingly, the Rulon really didn't perform any better than virgin.

I beieve the PV is within limits, but is also difficult to control because the loader material is 40 duro flat silicone (non-linear modulus, and somewhat difficult to control size tolerances).

I am considering a new FDA-approved Nylatron material. I am aware of the water-absorption characteristics of nylons, but I believe I can live with the swelling in this case, if the wear rate is reduced. The other problem with harder materials like nylon is that the current design uses the PTFE as a gasket as well, so a harder material will need secondary static seals.

For an idea of similar designs, see

http://www.cinchseal.com

and

http://www.mecoseal.com

TIA! Trace

 

[EdDanzer]

Short of building a machine to do P/V testing at given loads, velocities, and wear rates with known materials, all you can do is try stated values. To date our testing of different bearing materials, speeds, loads, and lubrications make most published data just a guide line created for marketing. There is no way to know who published this data, and bad data just keeps being recycled if it is good for marketing. These bogus values have made us to consider building a P/V testing device. But justifying a several thousand dollar machine and spending weeks testing are difficult. One thing for sure having 2 different materials seems to be better than similar material when rubbing together.

 

[MikeHalloran]

Agreed that material mfgr's spec sheet data for PV is always suspect, not because they're dishonest, but because they can't cover all the possibilities.

If you've got an assembly that's already in production, you should be able to assemble a test rig from scrap/ obsolete/ simplified/ returned parts, for nearly zero _visible_ money.

If you have to go through some formal proposal/ review/ approval cycle to build an engineering test rig, you might as well save yourself the aggravation, and quit; any such company is doomed. But that's another discussion.

-Mike-





Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[tracetrimble]

Trying to CYA for most situations is one thing, but stating that your PV (probably the most important spec for a bushing material) is 10 times what it tests out to be is quite another. It's really not worth my time to make an issue of it, but IMO that's a shady business practice, and seems like it would have been outed by now. Live and learn.

I've been testing on a couple production machines so far, but a test stand is in the works. Not much red tape around here for projects like this. I do like the idea of using a manometer to quantify leakage, had not thought of that.

Thanks again for all the feedback.

 

[tracetrimble]

TTT... I'm currently waiting for my test stand to get finished, just trolling for some more input.

Thanks- Trace

 

[EdDanzer]

Have you provided for measuring wear rate, rpm, temperature, torque required, and system monitoring?
When we looked at building a P/V tester the data collection portion would cost the most. If you do not monitor what is happening as it runs the conclusions will be very subjective.
We were considering using a servo motor for the drive so the rpm would be accurate. The most expensive part is measuring the torque. The last concept was to use a load cell on a torque arm.

 

[tracetrimble]

Yes, everything you list except torque will be monitored, but not in an automated manner. I plan to run different setups for a set period of time, and stop and take measurements. While this is not ideal, I don't have budget for automation, and this should get me where I need to be (a working seal). I will also be varying the product temperature & pressure via an immersion heater and head pressure.

 

[tracetrimble]

It would be possible to measure relative torque (as opposed to absolute) to some degree of accuracy using a high end VFD (inverter) that will show current load. This won't work in my case because I'm using a returned 3 hp 60:1 gearmotor for the test stand, which is extremely oversized. I'd be trying to read the difference between something like 1% load and 1.03% load.

 

[btrueblood]

tracetrimble,

I'm a little confused by your post. You say you need a bearing material, and yet it needs to be a seal as well? Usually these two functions are seperated. A bearing is generally going to wear over time, and the wear, and grooves/scratches will simply not seal once the part has run awhile. Use an elastomer seal on the water side of the bearing. The elastomer will wear too, but is more compliant, and squeezes down to take up the slop.

I would suggest using an EPDM packing, O-ring, or u-cup seal (Parker has them in NSF compliant materials), and use whatever you need for a bearing behind this seal. EPDM works to 250 F in water/steam with just about any chemical/salt mix.

PTFE will wear and creep no matter what; you need to keep high preloads on PTFE seals to keep them sealing. Rulon is a glass-fiber or mineral-fiber reinforced PTFE, and once it starts to wear, the fibers will scratch the mating surface and destroy any seal that may have been there. Acetal is no good for water above about 180 F, it just breaks down. Nylon also gets attacked by hot water. Also, don't use FKM materials in hot water or steam (fluorocarbon polymer rubbers, e.g. Viton), the binders used in making the rubber products will break down. Dupont sales engineers have told me that "Viton is not like the other guys' fluorocarbon elastomers", and that Viton brand elastomers will hold up to hot water; I'm waiting for the day when we have to have a seal hold up at 300 F or so, and then will try the duPont product, until then, EPDM is the choice material.

Ben T.

 

[tracetrimble]

Thanks for the reply. Take a look at this:

http://www.cinchseal.com/solids.htm

I'm simply adapting this design to my needs.

 

[btrueblood]

tracetrimble,

Ok, I looked at the cinchseal website. I'm assuming the rulon parts are the rotor cups. I stand by my earlier post, the rulon is going to wear, and you will need a substanial preload to be maintained on the stainless endplates to maintain any sealing. You might try putting belleville washers under the bolt heads on the ss plates to "take up" the slop that will develop over time, also, look at adding more bolts, and possibly an outer clamp ring (stiffening ring) if you think the wear is not occuring evenly. This assumes that the "spacer blocks" allow the plates to float axially a little bit. If not, I'd replace the spacer blocks with either a softer material to allow the endplates to be compressed inwards a little, or get rid of them and put a spring or similar in their place.

I'd still put an O-ring groove on the plastic rings, to provide a good (bubble tight) seal where they mate to the stainless plates.

Good luck!

Ben T

 

[EdDanzer]

Btrueblood.
The reason to do a P/V test for seal material is that in order to seal two materials must contact and operate as a zero clearance bearing in a small area.

 

[tracetrimble]

Here's a screenshot of a concept drawing xsecn view.

http://img.photobucket.com/albums/v280/tracetrimble/untitled.gif

The rotor cups are stainless, and the Rulon is in sheet form. When the housing is clinched down by the mount bolts, the silicone elastomer loads the rotor cups. The amount of squeeze is determined by the design gap (currently only about .030"). The Rulon on the right side also acts as a gasket between the housing and endwall.

 

[tracetrimble]

Also, Viton is a very pricey material due to EPA regs on its mfg process, but it does perform as advertised. I used to work for a hydraulic cylinder mfr, and it was the only solution for many applications.

 

[tracetrimble]

I should also mention that this is in a food plant, and is broken down for sanitation daily. I wouldn't think creep should be too much of a concern?

Is there any way to edit posts on this board? I'm not familiar with this BBS platform. Thanks y'all!

 

[btrueblood]

I don't see where this device acts as a bearing at all, altough I'm just going by what I can see on the website drawings. Looks like it's just an axially loaded seal. There doesn't seem to be any radial load bearing path. If that is the case, an elastomeric seal would do the same job cheaper and better. It may wear more quickly, or less quickly, and yes, a test is the best way to find out what will work best. But: if you want to SEAL something, use an elastomer. Rulon is a lousy seal material. It will creep rapidly if not otherwise constrained.

And, yes, Viton works wonderfully as a high-temperature fuel/oil seal. It <may> NOT work for hot water or steam, I know that other fluorocarbon elastomers don't. EPDM DOES work for hot water, and is used (by me) all the time for rotary shaft seals, it is the best thing going for that application IMO. It would make a lousy bearing, though, and thus I'd put a bearing outside the water to hold the shaft in place. Elastomers work well for sealing, because they have compliance, and tend to extrude themselves to fill voids. Hard plastics don't do this, thus they tend to leak.

 

[EdDanzer]

Btrueblood,
A thrust bearing is a face contact bearing. This proposed seal is similar to face seals used in construction equipment final drives and track rollers. These seals consist of a rubber load ring and a metal sealing ring with lapped faces contacting. The faces receive some lubrication from the oil to provide long life. If grease is used the seal life is poor. I will agree the flexibility of elastomer materials will improve sealing. The problem we see is as pressure and or speed go up seal life goes down. I high pressure hydraulic seals the P/V of the materials appears to have a significant factor on seal life.

 

[tracetrimble]

Thanks for all the great feedback. This site is awesome.

True it is not a "bearing", but the face seal contact is under thrust load from the elastomer as described above. PV is the only way to quantifiably compare materials under consideration.

We currently use a loaded silicone seal in this application, but the problem is that the seals wear grooves in the shaft after years in service, and then do not work as well. The shafts are extremely expensive to replace. The attraction of this design is that there is no sliding contact with the shaft itself, and all wear items are inexpensive and easily replaced. Ideally, this design will retrofit onto worn shafts as well.

Thanks again -

 

[Gibit]

You may want to also consult bearingisolators.com and refer to the Air mizer PS.

 

[ApC2Kp]

tracetrimble,
A couple of ideas for improving the wear on the shaft:
1) Ceramic coating of the seal area on the shaft. Plasma arc applied coatings are done on turbines / severe service rotating equipment. Pricey, but economics should consider costs of shutdowns and lost production.
2) Some time ago, research had suggested seals installed in a skewed (small angle to shaft) groove resulted in longer life and better lube distribution over the seal. It would seem to be true with larger area to spread wear, instead of line contact of conventional groove that is perpendicular to the rotating shaft. It would probably require a CNC machine to do the skewed groove geometry.