Variation in wear of SST ball on Sapphire (non standard jewel bearing)

[KENAT]

I have a small motorized microscope stage design I'm working on.

In it we have a couple of different types of hardened stainless steel (440C or 440F) balls/ball ends on sapphire bearing.

One type in horizontal plane is a simple linear plain bearing where the ball end slides across the sapphire pad. An initial proof of concept (PoC) had a little trouble where one sapphire pad seemed to wear the matching ball end, but we believe this was due to some roughness on the sapphire - and the action of the ball wearing makes logical sense.

The other type is in vertical plain where ball tips of ball screws push against sapphire pad. These balls rotate when the screws are turning, they also slide when the stage is moving in the other axis or potentially could be turning and sliding at the same time if moving in both axis. On the PoC this worked fairly well over thousands of cycles. Some of the running was done dry and some was done with a little grease.

We've now built a full prototype of the stage and are having trouble with the vertical bearings. The balls/ball ends are chewing up the sapphires after just a small number of cycles - maybe a few hundred. This is with just the sliding motion - no simultaneous rotation element. These bearings are running dry as we can't really seal the stage and our customer base aren't good with preventive maintenance so we don't really want to have to use lubricant.

Initially when using custom ball ends we noticed they were rougher than we'd specified, they were replaced with off the shelf balls of smaller radii but we still see the problem.

My research has taught me that Sapphire is anisotropic in terms of hardness, but even in the 'soft' direction it should be twice as hard as the SST on Knoop scale. Surface finish may also be a contributor but we call out a 2 micro inch finish on the Sapphires and are buying from a company that makes jewel bearings as well as optical windows.

We did do a crude test of scratching sapphires from the original PoC and the new prototype with a carbide scribe. The new sapphires scratched fairly easily and cleanly while the old ones just kind of grazed slightly & required more force to do so.

Any thoughts on what might be going on and/or how or what we need to specify to correct it?

Thanks,

(I'm posting here as the bearing forum didn't look particularly active.)

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[btrueblood]

Thre are a lot of ways to make artificial sapphire, not all sapphires are equal. The cheapest method is, I believe, to sinter powdered alumina, but it can yield a softer/weaker product if not fully crystallized.

These guys pull sapphire from partial? melts, and can give you fully crystalline material, there may be others:

http://americas.kyocera.com/kicc/pdf/Kyocera Sapphire.pdf

They are spendy, though.

 

[btrueblood]

Hm, something else you said - you changed to a smaller diameter ball? That could cause the contact stresses to increase, which might be causing the sapphire to spall. Any sign of deformation/denting on the steel ball?

 

[KENAT]

We're buying the sapphire from a nominally reputable company that supplies them both for optical windows and Jewel bearing application. While I can't confirm on their website I'm pretty sure they're made from single crystal boules however, given what we're seeing we have some doubts. When I get a chance I want to follow up with them - a colleague has talked to them but I wonder if anything was lost in translation as it were.

On the balls, the original PoC used relatively small balls but the herzian stress was OK and it had OK life as far as we took it.

On the prototype originally we had larger effective dia ball ends (to further reduce herzian contact stress for some reliability concerns during shipping etc.) and they caused a wear problem.

We switched back to the same size balls as on the PoC, switched to fresh sapphires, and still saw the problem.

In both cases the balls showed little wear even under magnification.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MikeHalloran]

It appears that most of the accumulated knowledge about steel/sapphire bearings has come from steel ball ends spinning on flat jewels, and cylindrical steel journals riding in doughnut holes.

Translation of a steel ball end on a planar jewel would appear to be sparsely charted territory.

How do you feel about having the steel ball end rotate in a cup jewel, and having the planar backside of the jewel translate on a steel surface? That's the least worst thing I can think of right now.

... but you should be talking to application engineers at your sapphire supplier or their competitors, e.g. ITI, Bird Precision, Swiss Jewel, etc. They must have done something like your application already. If you are lucky, you will find someone who has failed enough to map out the portion of the solution space that's likely to bring success.






Mike Halloran
Pembroke Pines, FL, USA

 

[KENAT]

One of the companies you mention Mike is who we got our sapphires from, I'm a bit concerned by the lack of intelligent input from the guy my colleague has been talking to but like I say something may have been lost in translation. Sadly I've been told this morning to focus on something else so haven't got back to them myself.

We thought about some variation of the ball in sliding cup idea a while back, seems to have some issues but maybe we'll think about it again.

Thanks Mike & btrue.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​