PDA

View Full Version : Ball Bearing Speed Ratings...



Fasttrack
08-11-2014, 08:06 PM
What is the limiting factor when it comes to ball bearing speeds? I have an unusual job that may require two telescoping shafts, each turning roughly 26000 rpm but in opposite directions. Presumably, I should be looking at bearings with 52k rpm rating, correct?

Any recommendations for a high speed, reasonably priced bearing with something close to a 1/2" ID? OD should be as small as possible since it will be installed in another shaft that is rotating. I have some freedom there as far as size goes, but small would be good. I looked at McMaster-Carr for needle bearings and ball bearings. The only thing that was an obvious fit (12mm shaft, 28mm OD, 60k rpm) was an ABEC-7 bearing coming in at ~$200 - a little steep for the application. Looking for something in the $30-40 range.

Thanks!

Willy
08-11-2014, 08:35 PM
What is the limiting factor when it comes to ball bearing speeds? ...
.....Thanks!

Very good question and one that I knew some of the answers to but certainly not all.
Although I don't have a list of suitable bearing candidates for your project I did find a very interesting pdf file that perhaps will give you a better insight into the factors that limit bearing speeds. Hopefully this info will allow you to better make a prospective purchase decision.

http://www.kaydonbearings.com/downloads/catalog300/Kaydon_300_Performance.pdf

Below a list from the pdf of the key factors that limit bearing speeds. I had not considered life requirement but In my opinion this is probably one of the most often overlooked elements in bearing choice.


Bearing diameter
Ratio of bearing diameter to cross-section
Bearing type and internal configuration
Ratio of ball groove radius to ball diameter
Bearing internal fit-up (diametral clearance
or preload)
operating contact angle(s)
Bearing precision (runouts)
Ball separator material and design
Precision of mount (roundness, flatness under load)
Lubrication
Ambient temperature and provision for
Heat dissipation
Seals
Loads
Life requirement

Guido
08-11-2014, 09:43 PM
http://compturbo.com/spotlights/triplex-ceramic

How's about ceramic ball bearings, fitted inside some pretty small turbochargers, spinning at 100K r's and better, almost red hot, very hot and thin lube, mile after mile after mile. Mass produced.

Maybe the above people can help.

--G

jlevie
08-11-2014, 10:18 PM
I suspect that when you get into that speed range that there are no inexpensive bearings that will any sort of a reasonable life.

ikdor
08-12-2014, 03:08 AM
Interesting question and I don't have the answer either. However, I believe that one of the main limiting factors for ball bearings is the strength of the cage and the centrifugal force acting on it. With both rings spinning in opposite directions the cage is standing still so that's not a problem anymore.
You do have huge seal lip speed, so it's going to be hard to keep any contacting seal alive. Better to try a shielded bearing, or better, an open one with oil lubrication.

Igor

EVguru
08-12-2014, 03:22 AM
If you look up the bearings in something like a die grinder, you'll probably see that they're running way over their rated speed. You have to remember that bearings are usually specified at 100,000 hour operating life and you can exceed some of those specifications if you can accept a shorter life.

A cheap die grinder might have a design life of 50 hours. That can equate to many years of hobby 'use'.

macona
08-12-2014, 04:41 AM
Die grinder bearings also run in an air-oil mist which allows them to run at higher speed. That being said the bearing life in die grinders is pretty short compared to the same bearing in a "normal" speed application.

I would probably go with a ceramic ball bearing. You can get them off ebay cheapish. They are used in the mini turbine jet engines which do 70-100k rpm and also turbo pumps which will do up to about 90krpm.

ikdor
08-12-2014, 04:55 AM
I don't know whether you are able to try a bearing or whether it needs to be first time right, but I'd try this one:
The energy efficient bearing E2.6002-2Z/C3 from SKF with a reference speed of 55000 and a limiting speed of 28000, bore = 15mm. They're pretty cheap, about five dollar for instance here:
http://uk.rs-online.com/web/p/ball-bearings/7192036/
Smaller sizes like 6001 and 6000 are rated for even higher speeds.

The raceways have a larger osculation reducing slip losses, non contacting shields, low friction grease and a low friction cage design. All of this is positive in your application. End of sales pitch :-)

Igor

EVguru
08-12-2014, 05:20 AM
Die grinder bearings also run in an air-oil mist which allows them to run at higher speed.

Not in electric grinders!


I would probably go with a ceramic ball bearing. You can get them off ebay cheapish. They are used in the mini turbine jet engines which do 70-100k rpm and also turbo pumps which will do up to about 90krpm.

Having spoken to some model gas turbine builders, they rate the bearings at about 40 hours of full load running. That is quite a lot of flying between rebuilds though.

Fasttrack
08-12-2014, 09:38 AM
Thanks guys. Lots of good suggestions. A 50-100 hour life would be more than sufficient at this point. The product in question is an early prototype and proof-of-concept. I think ceramic is the way to go and I would like to find one that would be used in a turbocharger for the "next generation", since we will eventually need longer lifetimes.

Igor, that looks like an interesting product to build the prototype around.

I have contacted SKF to get some more information about speed ratings and to see what they have to say about this particular problem.

A.K. Boomer
08-12-2014, 09:47 AM
http://compturbo.com/spotlights/triplex-ceramic

How's about ceramic ball bearings, fitted inside some pretty small turbochargers, spinning at 100K r's and better, almost red hot, very hot and thin lube, mile after mile after mile. Mass produced.

Maybe the above people can help.

--G


that's where my head went to too... that's some crazy RPM's and anything sealed or even shielded will most likely overheat and also pop it's cover off, I think @ 50,000 + you can forget about grease...

Fasttrack
08-12-2014, 09:51 AM
that's where my head went to too... that's some crazy RPM's and anything sealed or even shielded will most likely overheat and also pop it's cover off, I think @ 50,000 + you can forget about grease...

No grease, for sure. Since I don't require long lifetimes for the prototype, I have no problem removing seals and washing out grease, if a "standard" bearing will get me 100 hours or so at 50k rpm without blowing up. I used to do this for high mileage vehicles back in high school.

ikdor
08-12-2014, 10:01 AM
It won't be easy to find the turbo ones for sale as they are only produced for the turbo manufacturers. It is a cartridge with two raceways, special cages and it combines with the housing to make a fluid dynamic film damper.
Your requirements are not that dramatic, so there seems no need to go that exotic.

Igor

Willy
08-12-2014, 10:05 AM
....................... if a "standard" bearing will get me 100 hours or so at 50k rpm without blowing up. I used to do this for high mileage vehicles back in high school.

I used to get what seemed at least like 50k rpm out of my vehicles in high school as well. But yeah not for long.:o

A.K. Boomer
08-12-2014, 10:33 AM
No grease, for sure. Since I don't require long lifetimes for the prototype, I have no problem removing seals and washing out grease, if a "standard" bearing will get me 100 hours or so at 50k rpm without blowing up. I used to do this for high mileage vehicles back in high school.


Iv done quite a bit of that too, some bearings feel great yet are running dry and the owner is on a budget so carefully remove seals and clean and re-pack, but you have to be careful, you have to leave all kinds of extra room, you can't pack them to the hilt or they will pop their seal covers off, found that out with a mitsu --- luckily it had timing belt change intervals of only 50,000 miles...

I generally leave about half space air,

@ 50,000 rpm's I don't know what id do, light oil - but your going to almost immediately toast your seal lips i would think, then there goes your oil...

J Tiers
08-17-2014, 11:00 AM
bearings fail by fatigue (or overheating).

They are rated for X number rotations at a given load, and for lesser loads the total number of rotations goes up. At higher speeds you get to the total number faster. it's really the total number of flexings of spots on the races, which get loaded by each ball as it goes by. After "X" number, the spots spall off, and failure has started.

That's not quite out the window for 50k rpm, but there are other considerations.

Too much lube will get "stirred" to high temp, which even at fairly low temps starts to lower the temper and hardness, of the bearings, and you then get a rapid collapse as they get pounded and heated to a high temp, with mis-shapen lumps found when you look at the balls in the failed bearings.

The bearing manufacturers have recommendations, and have bearings which stand very high rpm, but you DO pay for that special type. Do a thorough search.

darryl
08-17-2014, 04:20 PM
Has anybody mentioned magneto bearings- that might do what you want. Just a thought.

J Tiers
08-17-2014, 05:17 PM
You probably do not want any standard grease-lubed bearing at 50k rpm.... including magneto types. I believe there ARE greased bearings for that speed, and above. It's not nearly the top end speed of ball bearings. But it won't be just standard bearings, most likely. And not standard grease, not packed in the usual way.

I've seen failed bearings from that sort of speed, lubed with too much oil. The balls were sorta square when inspected, as well as black and cooked-looking.

The bearing manufacturers know all about it, it's old stuff for them. Check their info, call and ask them, etc.

Just a little oil, in a fine spray, is often recommended. Can be from a slinger, just slightly wet with oil, slinging off tiny mist drops, as shown in the old New Departure manual, no need for some type of spray nozzle.

What you want to avoid is any chance of getting a big glob at one time... just one regular drop-sized glob can start a failure.

ikdor
08-18-2014, 05:16 AM
Well Jerry, don't assume too much about the knowledge of the bearing manufacturers, or their means to locate this knowledge inside the company :-)
There is some info on bearing speed here:
http://www.skf.com/group/products/bearings-units-housings/ball-bearings/principles/speeds/reference-speed/index.html
But what you can take away from it, is that the reference speed is based on a low load of 0.05C0 (this is very low so you need well balanced assemblies!) and viscosity of the oil or grease.

In general there is no straightforward model for what happens exactly when the rings rotate in opposite directions. The cage spins relatively slowly as it runs at 5-10% of the 52.000rpm, so the cage forces are limited. The ball speed inside the cage pocket is still very high though, so this will probably be the limiting thermal factor.
With grease you're creating some issues with the outer ring rotating as the grease might not wish to get back to the inner ring. If possible this application would work better when oil lubricated, as it's easier to get the oil everywhere and it aids the cooling of the bearing.
Do make sure the bearing itself can get rid of the developed heat, so try to not make the bore out of stainless :-)

If you do wish to go with a greased bearing, then maybe a hybrid like this would be fair choice: 6001-2RSLTN9/HC5C3WT, ref=75000rpm lim=50000rpm. It's not a precision bearing but probably still $100. Or perhaps the 6201 version as it has better load rating with slightly lower speed.
For oil lubrication you could just take out the seals.

Good luck,
Igor

J Tiers
08-18-2014, 08:14 AM
or their means to locate this knowledge inside the company :-)


That *is* often an issue.

However, there always are folks who DO know, better knowledge now than there used to be, really. With all the consolidation, there are few if any minor marginal companies left. The ones still around have high power engineering available, but perhaps not for the peasants visiting the website.

Any serious customer for specialty bearings will have "their people" talk to "the bearing co. people", and get it sorted out. Before that happens, the customer will be vetted as serious, annual quantities will be known, etc, so the efforts do not go down a rat hole. The question is whether the application here justifies getting the bearing co. engineering department involved.

If not, then a bearing rated for max rpm well in excess of the 52K rpm as a max, will be the best bet, with the best possible estimates of the loading under worst case conditions used to determine the life estimate. As you mention, unbalance may be worse than actual outside loads applied.

Essentially it is an *engineering problem*, in and of itself. I have seen otherwise smart companies fail miserably at this task..... when they tried to do it all themselves and pick bearings and lube systems without help.