Has anyone noted or done a study regarding the vibration signature of c2, c3 and c4 clearance bearings?

Does the bigger clearance automatically translate into harmonics of running speed in the spectrum?

When running at the designed temperature, in a perfect world etc. all clearances should end up in the perfect match made in the bearing laboratory...(heaven). In all other normal cases it will rattle. So if you fit a C4 where C2 is optimal, it will rattle more. If you fit C2 where C4 is required it will get stuck. So is that why all modern electrical motors have C3, because they are designed to run darned hot? Only my simplified view of bearing match making and may need some more input. Olov

Oli,

From what I've seen, I would agree with you.

I have a 3585 rpm fan that is going through bearings about every 6 weeks. They are using an SKF 1217A bearing with a c4 clearance and it shows many harmonics of running speed in the spectrum but little impacting in the waveform.

The bearing manufacturers reps have made just about every recommendation under the sun. Changed from grease lube to oil and back, c3 and c4 clearance (not sure about c2), different bearings. I'm beginning to think that neither the manufacturers reps or the mechanics that are doing the installation know what they are doing.

They don't seem to be interested in my opinion, so I'll just keep calling for replacements every six weeks until they either run out of bearings or mechanics.

What bearing or both? What do they look like when getting them out? Is it running hot or more likely to match the thinking cold? Questions, questions no answers. Have you checked alignment, maybe to good so there are no bearing load? That may be just as bad, best is when things are Swedish "lagom", eg. the bucket of bear is finished when passed by the last guy around/under the table. Olov

Oli,

Both bearings have had problems. Right now the coupling side is fine and the fan side is the one with the harmonics of running speed. 3 x is the largest harmonic and axial is highest amps. This is an overhung fan. I've only seen one set and they were both worn with a single groove completely around the outer race.

I seriously doubt that the alignment is too good but even a blind squirrel finds an acorn now and then.

Danny, are these ball bearings or roller bearings? How is the fan coupled? Would that be a Dodge Paraflex (rubber donut) coupling by chance? Was the worn outer race groove offset towards one side?

I see ball bearings on smaller overhung fans, but the larger ones all have double row cylindrical bearings... they tend to have a lot of "rattle" if set up too loose, but will run that way for a long time, unless of course they used antiseize between the shaft and the tapered sleeve.... then the sleeve eventually spins on the shaft, shaft diameter is reduced, sleeve gets loose in the bearing, bearing clearance opens up, fan gets out of balance, vibration ratchets up, bearing starts to beat itself up, etc, etc, etc.

Rusty,

It is a Rex Omega I think, but the same principal. Orange two piece element.

Now I remember you saying that these can cause axial loading if not set properly and that may be the case here. I'll suggest that they check it for whatever that's worth.

This is a double row ball bearing in a pillow block housing. I don't remember about the axial position of the wear. I just got a quick glance at one of the bearings.

Danny, chances are the coupling they are using is not rated for 3600 rpm. The higher the speed, the smaller the diameter (and the lower the torque rating). If the coupling is rated 1800 rpm, then at 3600 rpm the centrifugal force is "elongating" the element radially, which shortens it axially, thus placing a constant axial pull on the shafts, and thus the bearings. The harmonics you are seeing may actually be a rub in the bearing, generated by high axial loading.

Had this happen on a 3600 rpm fan. The first time, the fan and motor bearings failed at the same time. They neglected to replace the donut coupling with the Thomas coupling I spec'ed (and which they actually purchased), and a few months later the fan shaft actually slid in the bearings (Yes, they used antiseize on the tapered sleeves!), pulling the wheel into the housing and cracking the blades (a $20K inconel wheel).

Would never have believed this if I hadn't seen it for myself.

quote:

I've only seen one set and they were both worn with a single groove completely around the outer race.

You said you don't remember exactly, but let's revisit this. You only saw one track for a double row ball bearing similar to attached? In other words, you only saw a pattern created by one row, not the other row?

If so, that seems like it would fit Rusty's scenario of axial loading if the bearing is like the 1217 bearing sketch attached. Imagine that you apply an axial load on the inner race in a direction toward the left. Then only the left-most row of balls is pushed into contact with the spherical-section outer race, and the right most row of balls does not see any axial load.

I thought the original question was a good one and I was eager to hear the answers. I would suspect you are more likely to have harmonics for a loose bearing, particularly if the loading is low, but I don't have any data to back it up and have never seen any studies.

This message has been edited. Last edited by: electricpete,

DannysBearing.pdf (9 Kb, 44 downloads)

Rusty,

I really think that may be the case here too. I'll get on it as well as they will allow.

Pete,

I'm really just not sure but I'm more and more inclined to think Rusty is right.

Thanks guys.

Danny,

I have noticed vibration problems, espeically outer race defects most everytime I have monitored a component using the ball bearing self aligning bearings such as the 1217 bearings mentioned.

With balls contacting a spherical outer race the tangent pt of contact is "very" small. Consider putting a large crane on soil without wide pads beneath, it will sink. My theory is the same for these bearings.

To me these bearings only last if used in "very" light load, high speed applications. They are successful at decreasing temperatures in bearings due to the small amount of contact area.

Changing to 22217 bearings would provide spherical roller bearings still rated for the rpm you are running.

The Id and Od is the same but the width is slightly wider which would mean changing bearing housings as well.

my opinion only.

Mike

22217.pdf (10 Kb, 21 downloads)

Mike,

They changed from those to 1217k's and I think are planning on changing back.

With all the variations on bearings at that size, I's sure they will eventually get some that works.

Thanks,

Danny,

Your original question was something that I was always wondering about. This probably could be answered by running a controlled test in a lab.

But back to your situation...
Self aligning bearings ( both ball or spherical elements), as it is known, do not take a lot of axial load. To be more precise, there should be a certain ratio between static radial and axial loads, something like 10:1. So you may want to verify that there is sufficient radial static load.

david

quote:

I have noticed vibration problems, espeically outer race defects most everytime I have monitored a component using the ball bearing self aligning bearings such as the 1217 bearings mentioned

Yep, couldn't agree more. They are a nuisance.

quote:

Changing to 22217 bearings would provide spherical roller bearings still rated for the rpm you are running

We've successfully done this sort of thing in some applications.
The only thing you will have to watch is that there is sufficient radial load to keep the rollers rolling: if they start to skid, you will be in a bit of strife. If Rusty's (very perceptive) diagnosis is right, then spherical rollers might cause more grief due to the potentially high axial-to-radial load ratio.

We had some issues with bearings on a 3600 rpm fan where I used to work. SKF recommended Carb bearings. You may want to look into this in your case.

Danny,

In your application and you're changing them every 6 weeks or so, has this been something thats been occurring since new or only recently?

Which size of Omega are you using? Are your running it off a VFD by chance?

Spud,

It never used to be a problem as far as I know. I used to do this plant but just restarted after several years of no pdm due to budget problems, so I really don't know much about the history. They are not the most responsive client I have either.

I'm not sure what size, but I would say at least an 11. No VFD.

Where in Alberta?

quote:

Reply

Danny,

Is the motor driving this equipment a sleeve or ball bearing?

Gary

Gary,

It is a sleeve bearing motor but I am assured that the magnetic center is set correctly.

Where do you work in Big Lick?

Danny,

Field Service Technician, Lloyd Electric Co. (EASA Member Motor Shop)

This situation sounds familar, maybe in a foundry, or maybe not.

Gary