Ball bearings and friction.

[Neil LickfoldParticipant @neillickfold44316]

John Hain, what an interesting article about bearings.

I have just done a short and not very scientific study on bearings myself. I have 2 standard 61902 bearings with the plastic cage. Both were cleaned with IPA to remove the oil and spun by hand on a mandrel. They both last for about the same time when flicked to see the time of spinning. One is then taken apart, and the inner and outer races are polished with a felt mop and 1um diamond polish compound. The bearing races really do shine after the felt mop. Re assembled and tried to measure the difference in clearance. With a micron dti was unable to show any difference in total bearing clearance. Both seemed to measure only 0.006mm total clearance. The first thing I noticed after cleaning again was the reduction in total noise from the bearing. The polished one was definitely a lot quieter. The standard bearing still spun for the similar time frame of 10seconds, while the polished one consistently was spinning for 15 to 16 seconds.

[Robin GrahamParticipant @robingraham42208]

JohnH/NielL – thanks for your contributions.

John – Swensen's article was very interesting. He has answered my principal question I think, as well as some subsidiary questions about bearing lubrication which were lurking in the back of my mind. I had made some experiments and found that, irrespective of size, my bearings ran for longer after lubrication by light oil (sewing machine oil was the thinnest I had to hand) than they did dry. This appears contrary to accepted wisdom. Further work (by me) is needed!

Neil – thanks for the report of your experiments. While 'unscientific', your results are clearly significant. A 50% increase in spinning time is impressive. I have 1um polish and would like to try this, but I can't see how to take a deep groove bearing apart and reassemble it without damaging something. Can you tell me how you did that?

Robin.

[Neil LickfoldParticipant @neillickfold44316]

The bearing I took apart had a steel cage. The other bearings had a plastic cage. So I recycled the plastic cage from the old bearings and put them into the new ones. To take apart the steel cage, I use the hook from an O ring pick and starting from the inner side, carefully lift up the cage. It just unfolds the tabs and comes out. I do not attempt to recycle the steel cage. Some of the 15X28 bearings have 10 balls instead of 9. With the 10 ball ones, you need to very carefully distort the outer race, and the centre ball can be removed. The 9 ball ones, the centre ball can be easily removed . In my case the 28mm od race was held in a ER40 collet. The Inner race is held on a 15mm mandrel that is a tight thumb press fit. I made the mandrel from AL. I put it in the freezer and the inner race comes off very easy.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Robin Graham on 05/10/2020 23:34:38:

…made some experiments and found that, irrespective of size, my bearings ran for longer after lubrication by light oil (sewing machine oil was the thinnest I had to hand) than they did dry. This appears contrary to accepted wisdom. Further work (by me) is needed!

Neil – thanks for the report of your experiments… A 50% increase in spinning time is impressive. …

Robin.

Funnily enough I was reading about lubricating roller bearings last night whllst looking for something else. Oil plays a different role in plain bearings and in roller bearings.

In a plain bearing, the oil has to form and maintain a hydrodynamic layer between the two metal surfaces. The surfaces 'float' on the oil which is under considerable pressure (up to 30 tons per square inch) and likely to be squeezed out. So plain bearings need plenty of oil, it has to be kept topped up, and forcing it in with a pump may be necessary.

Ball bearings work by rolling rather than floating on a hydrodynamic layer. Although they benefit from lubrication, it's applied much more lightly. My book suggests just enough oil to create a mist as the bearing rotates. Too much oil or grease gets churned, which increases drag and causes frictional heat. (Bad result!)

The amount and type of lubricant needed in different roller bearings varies from none at all to a dollop of grease; I guess it's related to the design and purpose of the bearing. Unfortunately my book doesn't say. I guess optimum lubrication depends on the load, speed and direction of forces applied to the bearing, and what's best for a high-speed light load is wrong for a slow-speed heavy load. This may be why the experiment doesn't match accepted wisdom – it's true in one circumstance, but the result isn't general.

Polishing is a good way of reducing friction though I worry about the way very highly polished gauge blocks stick together!

Dave

[Neil WyattModerator @neilwyatt]

Oil free ceramic bearings?

They are supposed to be freer than steel bearings with oil, other things being equal.

Neil

[Ketan SwaliParticipant @ketanswali79440]

Something I have found when making prototypes is that the bearings need to be 'tight' in the sense of keeping all the moving parts coplanar. Any wobble and parasitic vibrations set in which sap the energy surprisingly quickly. I guess there is a trade-off between that and friction. I imagine people making Stirling engines must face the same sort of challenge.

Robin.

In the early days of ARC, we used to sell Stirling engines, and seek advice from Roy Darlington (key member of Stirling Engine Society who sadly passed away in late 2018). His simple advice was to remove all seals/shields and wash the ball raced bearing in paraffin. 'The oil' in paraffin was enough to lubricate the bearing, resulting in the lowest amount of friction.

Most metal shielded bearings have two metal shields. Could you live with one shield, remove the other and wash the bearing grease out?… or remove both shields, wash out the bearing, and clean the bearing when necessary.? Once metal shield/s are removed it is generally difficult to put them back on. They are usually press fitted into a grove on the inner side of the outer ring, or, in some cases, depending on the size and supplier of the bearing, the shield is retained with a snap ring… if you are lucky.

Ceramic ball raced bearings which require no lubrication are available with cage for certain sizes/without cage but with full complement of ceramic balls, where the bearing bore is 3mm and upwards, but I have no idea for how good they would be for your application.

Ketan at ARC

Edited By Ketan Swali on 06/10/2020 15:50:18

[John HaineParticipant @johnhaine32865]

Rex recommends washing out with the metal shields fitted, using several changes of fluid.

Ceramic bearings are used in model gas turbines I think? Not sure about lubrication. I believe they can be used in clocks but can't recall where I've seen it.

Another recommendation from Rex, in another article, is DON'T ATTEMPT TO CLEAN BEARINGS IN AN ULTRASONIC BATH! It basically b*****s them.

[Ketan SwaliParticipant @ketanswali79440]

Posted by John Haine on 06/10/2020 16:00:12:

Rex recommends washing out with the metal shields fitted, using several changes of fluid.

That is okay provided the bearings have metal shields. However, no one really knows how old the grease/oil inside the bearing is, or the type of grease. So, difficult to say how difficult or easy it will be to remove. Worth a try I guess.

Ketan at ARC.

[pgk pgkParticipant @pgkpgk17461]

There used to be a DIY bearing regreasing tool… not recommended since it over fills them.. based on two cones. The bearing is dropped into a conical receptical and a second conical nozzle on a filled syringe pressed against the inner race..pumps grease through the centre and back out via the races. Such a system might be useful for washing grease out though. The conical nozzles provided with dog intranasal vaccines work very well for bearings with inner races up to about 10mm bore. Since those nozzles aren't tolerated well by dogs (we used just to dropper the vax in ) most vets will have plenty of spare throwaways,

Graphite pwoeder is supposed to dave the ability to fill tiny surface defects and reduce friction? If that could be introduced i a tiny quanitiy??

pgk

[VicParticipant @vic]

There was one of these curiosity’s in a film, I think it may have been Iron Man? But in this case it had little to do with bearings or friction but electronics.

**LINK**

You can buy them ready built of course.

**LINK**

[VicParticipant @vic]

Posted by Ketan Swali on 06/10/2020 16:07:33:

Posted by John Haine on 06/10/2020 16:00:12:

Rex recommends washing out with the metal shields fitted, using several changes of fluid.

That is okay provided the bearings have metal shields. However, no one really knows how old the grease/oil inside the bearing is, or the type of grease. So, difficult to say how difficult or easy it will be to remove. Worth a try I guess.

Ketan at ARC.

For several Vacuum engine designs I’ve seen they recommend either washed out plain bearings or remove the shields and wash out. Seemed to work ok for me.

[Ketan SwaliParticipant @ketanswali79440]

Posted by Vic on 06/10/2020 17:58:49:

Posted by Ketan Swali on 06/10/2020 16:07:33:

Posted by John Haine on 06/10/2020 16:00:12:

Rex recommends washing out with the metal shields fitted, using several changes of fluid.

That is okay provided the bearings have metal shields. However, no one really knows how old the grease/oil inside the bearing is, or the type of grease. So, difficult to say how difficult or easy it will be to remove. Worth a try I guess.

Ketan at ARC.

For several Vacuum engine designs I’ve seen they recommend either washed out plain bearings or remove the shields and wash out. Seemed to work ok for me.

Yes Vic, I agree. John Haine was responding to my comment, which I made at 15:48, to which I responded.

If you read the post I made at 15:48 first, perhaps the rest, including what John said will make sense.

Ketan at ARC.

[brickyParticipant @bricky]

As Michael suggested knife edges on agates was used by Riefler on his regulators .I built one and a 30lb pendulem is driven by a 1.5lb weight.

Frank

[VicParticipant @vic]

Ah, ok Ketan, I didn’t read it all.

[Robin GrahamParticipant @robingraham42208]

Thanks for further comments which have been read and are in the process of digestion. All the bearings in my bearings drawer (marked 'bearings' confusingly – it doesn't usually work as literally as that) are of the double metal shielded type with (AFAIK) steel cages. I'll unpick one and investigate further. I can certainly do away with at least one shield and wash out as Ketan suggested.

Following Neil's suggestion of ceramic bearings I had a look around and came across this video on YouTube

The guy is concerned mainly with frictional losses in bicycle wheel bearings, but it's in the same area. His conclusion seems to be that ceramic bearings aren't all they're cracked up to be, for bikes at least.

I think the main thing I have taken from this discussion is that there isn't an easy answer and I need to experiment to see what works best for my application. Which I can now do with a better sense of direction than I had before raising the issue – thanks.

The videos showing 'Swinging Sticks(R)' linked to in Vic's reply are interesting from both mechanical and psychological perspectives. The tabletop devices there run very slowly compared to my version – they are described as 'creating the illusion of perpetual motion'. Well, maybe. I guess that illusion arises because the mind focuses on the motion of the main beam and can't immediately grasp the way energy is being transferred to-and- fro between the two pendulums.

I first made one of these devices as an illustration in the course of of a discussion about the solution of differential equations which I had with an unusually able A-level student. That was a while ago- I don't know where it took him, but it left me with with an abiding interest in the chaotic evolution of systems governed by simple laws . Which might 'explain' many things. But I risk verging on the political, so shall cease and desist.

Robin

 

Edited By Robin Graham on 07/10/2020 10:45:36

Edited By Robin Graham on 07/10/2020 10:47:23

[CirclipParticipant @circlip]

"There used to be a DIY bearing regreasing tool"

John "Bogstandard" described one he made on "Madmodder" a few years ago.

Regards Ian.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Robin Graham on 02/10/2020 00:25:02:

Some time ago I made a double, or chaotic, pendulum for my own satisfaction. My wife showed a video of the thing to a friend of hers, and now I'm instructed to make another one.

I didn't worry myself too much about the bearings first time round- the device ran for about 30 seconds from an initial shove, which was long enough to display chaotic behaviour. But I'd like to make the MKII run longer if I can.

[…]

.

There’s an App …

**LINK**

https://apps.apple.com/gb/app/double-pendulum/id1377507794

MichaelG.

[Neil LickfoldParticipant @neillickfold44316]

A machined phenolic cage has less drag over the brass/steel/plastic cages. With ceramic balls a full compliment has very little drag, despite the balls rubbing and bouncing off each other. What we have noticed in model engine bearings, is that the ceramic ball requires a different radius to be effective compared to the steel ball setup. The work I did with some bike bearings was that there is a definite gain to a ceramic ball in a bike wheel hub as long as you can keep the dirt out. We lubricated it with a pfpe oil and only 0.1cc of oil was used in each bearing. The oil is to keep the race surfaces from rusting. These were set up in October 2019 for a indoor track bike, and are still doing great. I am fortunate to have a selection of ceramic balls in some nominal sizes, and also in the plus and minus range in about 2.5 micron diameter changes. They measure the balls to 5 decimal places of millimetre. If they are too loose, they do not roll correctly and then skid. Too tight is just friction. I make the full compliment with a small groove that just allows a ball to be pushed into place. In general I use 52% of the ball diameter as the radius for the groove. Sometimes I use 51% if I want less end float on a particular bearing. This is if I am making my own races though. Most commercial races are in the 52% to 54% of ball radius , depending on the design etc for deep groove bearings.

[Author]

Posts