Rolling Element Bearing Identification 4

[BrianBT]

A 2MW shaft alternator was overhauled and new ball bearings fitted. The DE bearing was identified as an SKF 6232 MC3. This bearing was also the thrust.
After one hours run time this bearing failed.
After disassembly although the bearing was stamped as an SKY the design was unknown to them and was certaily spurious. However the design was such that it would have almost certainly cost more than a standard bearing to produce.

Instead of plain balls the elements were oval shaped with a 'button' on the long axis which located in a complimentary depression in the cage thus allowing the elements to rotate in one axis only and therefore could not accept the thrust when the alternator was running

Anybody have any ideas about identifying this bearing as the bearing distributor who supplied the bearing only reacts with a 1000yard stare when we try to quiz him

 

[diamondjim]

http://www.americanbearingworks.com/pages/1/

6232-M/C3 bearings
American Bearing Works

615 Blue Oak Court
El Dorado Hills, Ca 95762
Orders or Info:

916 933-2450
Toll Free 800 463-2450
Fax: 916 933-2459
-----------------------------------

http://www.thrige-electric.com/pdf/lakc.pdf

Could the bearing have been installed backwards.
I assume you mean it is a thrust bearing on
the commutator side?
Does it have barrel type rollers?
I had trouble understanding what the bearing
looks like.

 

[BrianBT]

The shape of the elements is not that you see in spherical bearings. They are very distinct being oval ie. rounded on the ends. A button is also formed in these ends which directly locates in a complimentary point on the cage.

Certainly the bearing is in no way a 6232 bearing. Due to the design it can take no axial loading (thrust) as it is transmitted onto the cage. This we believe was the cause of failure.

Thanks for the replies so far

 

[diamondjim]

Can you post a picture of a drawing
of the cross section of the bearing?
Of what value is the bearing part
number 6232 mc3 that you referenced?

 

[diskullman]

a 6232 is a radial ball bearing 160mm x 290 x 48. The M stands for Machined Bronze retainer. The Brand name most likely is SKF.It is not designed to be a thrust bearing, although larger ball bearings can take a limited amount of thrust.
If your roller is an oval shape, then it is not a ball bearing, but a spherical roller bearing. What you are describing almost sounds like the old New Departure Hyatt barrell roller bearing.
If you could post a picture to a web address and post the address here, I'm sure someone will be able to tell you what you have, but in any event, a 6232 is quite common and should be readily available anywhere. I assume you are the victim of a mislabled box, or a mis-identified bearing from the surplus market.

Russell Giuliano
Unique Technologies Associates - Cobra Solid Lubricant

http://www.uniquetechnologies.com

 

[BrianBT]

Many thanks for all the replies and my apologies for the delay in sorting out these photos. I have to say that 'New Departure Hyatt' was unknown to me.

Please find below addresses to the photos possibly the most telling of which is the second to last

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) - damaged drive end bearing.JPG

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) - Drive end brg brass cage broken.JPG

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) - damaged Drive end brg mounted on rotor shaft.JPG

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) drive end brg - after cutting open.JPG

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) drive end brg - bearing balls found with welded on pivot cups!.JPG

http://www.marineengineering.org.uk/Repairs/Shaft

Alt (S) drive end brg - close up view of broken brass cage.JPG

 

[diamondjim]

Brian, I had to use the following url

http://www.marineengineering.org.uk/

then selected "shaft"
and then given 4 choices and 1 gear
box pinion drive showed a bearing.
Is that the one you are talking about?

 

[diskullman]

I get a "File not found" message

Russell Giuliano
Unique Technologies Associates - Cobra Solid Lubricant

http://www.uniquetechnologies.com

 

[BrianBT]

Gentlemen

Very sorry, the URL's contained spaces therefore would require cut and pasteing into your browser. Please try the following

http://marineengineering.org.uk/Repairs/1.JPG

http://marineengineering.org.uk/Repairs/2.JPG

http://marineengineering.org.uk/Repairs/3.JPG

http://marineengineering.org.uk/Repairs/4.JPG

http://marineengineering.org.uk/Repairs/5.JPG

http://marineengineering.org.uk/Repairs/6.JPG

 

[diamondjim]

Those were amazing pictures of a type
of ball bearing that I have never seen
before. I assume the cages were two
piece rings riveted together to hold
the balls in place. Really seems like
an inefficient design as the balls always
track on the same surfaces whereas in
other styles the bearings with cages the
balls can rotate. How fast do the bearings
turn? I cannot tell if it is a four point
contact ball bearing with a gothic style
arch in the raceways or not. If the contact
angles were 30 degrees, the bearings could
take thrust as well as radial loads. But
I assume that this bearing was strictly
a radial design and a two point contact
bearing because of the cage design. The
bluish color of the balls I think indicates
being exposed to very high heat.
Thanks for sharing such an unusual design.

 

[unclesyd]

I have seen many bronze bearings with two piece riveted steel cages. As the photographs apparently show high wear, could have the retainers “machined” down the balls to the current shape?

Scanned a couple of older books 30's and 40's no luck on the exact shape.

You might give these people a call as they have seen almost everything.

http://www.bmcbearing.com/services.htm

 

[BrianBT]

With respect to the machining of the elements by the cage. Due to the run time being one hour and the uniform shape of the elements I would imagine this is unlikely.

It appears that the bearing is two point contact although to be honest I am not entirely sure

The only advantage I could imagine for this style would be the accurate positioning and support of the bearing elements under high speeds although restricting their movement to one axis I would imagine makes them very susceptible to smearing. Could it be that this design prevents or at least reduces the acceleration of the bearing as it enters the load zone?

30's & 40's- crumbs!

 

[diamondjim]

Brian,
You are right about the cage design
preventing the acceleration in the
escape from the load zone. I think
a standard cage with balls would
be just as effective and allow more
grease or oil in the bearing.
The cage design looks like overkill.
I could not get any feel on how large
the ball diameters are. I think the
bearing could be made with a filler
plug to load the balls in the id of
the inner or od of the outer. The
filler plug would be in the fixed ring
and out of the load zone. A standard
centerline cage could be used.

 

[diskullman]

Rather than speculate as to what this bearing is,I think it would be safe to replace it the bearing the manufacturer originally called for. I have never seen any bearing made like this, and my assumption would be that this is the effect of wear and excessive thrust loading.
It may have been an angular contact bearing (7232), installed in the wrong direction. I cannot tell from the pictures, because I cannot determine if the races are equal thickness on both sides.

Russell Giuliano
Unique Technologies Associates - Cobra Solid Lubricant

http://www.uniquetechnologies.com

 

[diamondjim]

It is quite obvious that the bearing
shown is not an SKF 6232-M/C3 bearing.
Was the SKF bearing the one that the
manufacturer specified originally?
Do you know for a fact that you have
high axial loads? I am surprised that
it was not a roller or tapered roller
design if you had thrust loads as well
as radial loads.
Do you have a picture of the original
bearing? Why was it replaced?

 

[diskullman]

Going back to your first post, WHO identified the original bearing as a 6232? Was it the distributor that sold it, or the manufacturer of the equipment. If it was the distributor, he may have mis-identified a 7232 angular contact radial-thrust bearing, and provided a 6232 radial bearing. I believe the bearing in the picture IS a 6232, with severe wear to the balls, and distortion of the cage, caused by the thrust load applied. I beleive those balls were originally spherical, instead of "oval shaped with a button on each end"

Russell Giuliano
Unique Technologies Associates - Cobra Solid Lubricant

http://www.uniquetechnologies.com

 

[diamondjim]

Were there only 6 balls in this bearing?
The cage design looks like it was designed
for a conrad style bearing. After the
rings were displaced and balls loaded, the
balls were then forced around the inner race
and then the cage segments were riveted together.
In the first and second pictures, are we
looking at a metal seal or is this the
cage? If the balls wore that much as shown,
then the cage would have been the carrier of
the radial loads acting on the lands of the
races. I bet this may have been someones
attempt to rework an existing design as I find
it hard to believe that a bearing company would
make such a design.
Please keep us informed about your progress.

 

[ischgl99]

That looks like a 6234 that had excessive axial loads. I suspect the bearing was running with a very high axial load and the balls were turned down or extruded by the inner and outer race to the shape they are now in. The axial load will cause the inner and outer race to do this on each side and if the heat is high enough as indicated by the dark blue of the balls, then it can easily deform the balls into this shape. The damage to the cages also points in this direction. The cage is cracked and bulges at each opening, so as the rollers were deformed from the axial load, they pushed outward.

It definitely is not a spherical roller bearing. The cage would have square openings instead of round.

Do you still have the old bearing that the distributor tried to match? If so, I would bet it is an anglular contact bearing as previously mentioned.

Mike

 

[ischgl99]

typo, meant to say 6232

 

[BrianBT]

Many thanks for the comprehensive replies. I will try to give all the information I have. Please understand that my involvement with this was after the initial bearing failure of the replacement alternator. I had limited advisory control over events that followed.

The 2 MW air cooled alternator was retro - fitted as a replacement for a water cooled unit that was irreperably damaged after rotor pole support failure.

The new unit was in service for 4 months when the DE bearing failed. This was original fit SKF 6232 and was mounted with spring preloading allowin a small degree of axial movement. The NDE was a 6300 series bearing and wa fixed. I could not agree more that thus design is unusual and the use of a 7000 series bearing would have been more appropriate.

However, the manufacturer indicate that several of these units are in general use and such failure is unusual.

Investigation revealed a serious misalignement probably from initial istallation as much as 11mm at the NDE from centreline- severe fretting damaged was evident on the coupling.

The shaft was damaged iwo bearing landing. A repair of sleeving with compatible material was carried out.

At this point the spurious bearing discussed was fitted. I understand that this was offered against the requirement for a 6232 MC3 although I was not privy to this.

This bearing run for 1 hour on test progressively upto 1MW electrical load without obvious fault

After a period the set was again restarted and the load quickly increased to 1.5mw and failure occurred.

Unfortunately I did not see the bearing before fitting so I cannot comment if distortion has occured to the cage. I can confirm that the bearing was without seals. The cage was made from brass in two halves and riveted together

However, the suggestion that the elements may have been distorted due to a combination of loading in excess of capacity and heat I believe to be a very valid one.
I believe that if viewed before damage then the bearing would have the general appearace of a 6232 deep groove bearing

Gentlemen, I think unless something surprising comes to light from SKF ( they have taken receipt of the bearing as I understand it was stamped as to be one of theirs) then I will post the information otherwise we can draw a line under this one.

Many thanks

Diamondjim- sorry, of course the balls accelerate out of the load zone not into it. Showing my ignorance