Bearing loss-density and motor bearings in general

[MS0000]

Hello,

Motor bearings seems to be almost always 63-series type. Output shaft diameter seems to be dimensioning the bearings as far as I have been making arrangements. When modified lifetimes(Lhmr) of the bearings are considered motor bearings seems to exceeding easily "infinite life" of one million hours.

Q1: I have always chosen 63-series bearings because of their size - I have heard that for the same ID 63 runs cooler than 62. But is there some more scientific explanations? Maybe some rule of thumb for bearing (friction)loss-density when motor bearings are considered?
Q2: Nowadays it is very easy to get over 20year lifetime for bearing when Lhmr is considered. End customers and certifying bodies do accept Lhmr ratings also. I am a bit doubting that Lhmr value is realistic even if I know I have taken many factors into account and I use 50 000 hours as minimum value for Lh10. Have someone else been using some other rule of thumb numbers for Lh10?

Other comments about sizing motor bearing?

Thanks ,

-S

 

[caoimhin1]

With regard to temperature rise, comparing a 63 series with a 62 series with the same diameter, the 62 series should in theory run cooler than a 63 series due to it being a smaller bearing with smaller mass....however, the temperature difference under normal circumstances would be negligible, probably impossible to measure.

Choosing an L10 life is all about how the product will be used. For example, if designing a bearing for a domestic vacuum cleaner or dishwasher motor, an L10 life of 5,000 hours would be more than enough, because the product is only used intermittently. However a motor for a waterworks pump or vacuum pump would need an L10 of 50,000 or 100,000 hours due to it running continuously with reliability a critical factor. So it depends on the motor use. But 50,000 hours isn't a bad choice in general.
Note that bearings almost never reach fatigue life in practice, usually other factors like contamination or grease life kill it off first.

 

[MS0000]

Thanks for your quick post.

In electrical machines there is always heat coming from the rotor shaft to bearing also - might it be the thing why to choose oversize bearings? And in case of inverter driven motors there are insulated bearing housings used - so bearing and also other losses are put into bearing from the shaft - but the heat transfer from bearing to housing is limited because of the insulation on housing. So I think that a small bearing runs hotter because of its limited volume? Does this make sense to others? If yes, my original question is regarding to this effect - is there some rule of thumb value for loss-density of bearings? ... or is this too complicated way of thinking?

About L10:

For applications where I do make my designs is bearing lifetime requirement been always 20yeas equals around 175 000 hours. Bearing manufacturers and even end-customers accept Lhmr instead of Lh10, nowadays. Is electric motor industry just so conservative that Lh10 value is used instead? Lhmr = a1*a23*Lh10

 

[Tmoose]

I believe arbitrary "Size" is way, way down the list of heat generation factors.
A small roller bearing will have a lower speed rating than a larger ball bearing due to heat generation of all the line contacts.

There are many heat generators in ball bearing.

sliding in the contact ellipse, size of the contact ellipse ( adjustable via more open curvature in the races), how the cage/retainer is guided. Even the variations in grease characteristics ( not just quantity ) can make a real difference in heat generation.