Pillow block bolted vertically causes bearing failure?

[vtwelve]

Greetings to all,
We've had bearing failures on our HVAC air blowers sometimes catastrophically.
The bearings are encased in plummer block,which is bolted onto a vertical support.Can this kind of loading contribute to failure?Should the pillow block be mounted on a horizontal support instead?
Lubrication is fine,alignment is fine[although we found some cases of bad alignment at the pulley end when the inner raceway had a piece broken off]but in other cases the bearing has disintegrated with only the inner race stuck on the shaft.
The bearings have not been impact loaded.
Lubing is done once every 3 months.However the unit manufacturers require the bearings to be lubed bi-weekly for coastal environments at 1000-1500rpm, but in either case the grease was found to be ok after the failures.
Does anyone have a solution,any help would be appreciated.

Many Thanks,
Vexed.

 

[dinjin]

What did your bearing manufacturer have to say?
Why wasn't the grease contaminated by the failed
parts? Where was the grease port? If all of the grease was at the bottom and you only inserted a small amount, it never would get to the top or have a full grease fill i.e. a lack of lubrication.

 

[Fugeguy]

What type of bearing?

Ball bearing, spherical, etc...

 

[vtwelve]

-Ball bearings,the kind with seals at both sides.

Grease port on top of the pillow block.
New grease is apply till little of the grease appears at the seals.In this way we're sure that its not over/under greased.

The grease was indeed contaminated by the failed parts when the failure was detected late,but in a few cases we found out tht some bearings were beginning to fail,and the grease was still ok..

 

[Fugeguy]

Generally, unless the loading is very light I'd stay away from ball bearings in vertical applications.

While it is true the DGBB's can take thrust loads they aren't the best choice.

Can you switch over to PB's with SRB's?

I think the issues will go away as long as the speeds aren't real high.

 

[Tmoose]

Is some design feature provided to make allowance made for axial thermal expansion?
Describe the "beginning to fail" bearings.
Is it the bearing near the belt drive that typically fails?

Are any vibration measurements being done?

Dan T

 

[diskullman]

HVAC applications are very tough on bearings. Your better manufacturers make special bearings for these applications. They are usually a vey loose fit in the housings, to make up for misalignment.

Russell Giuliano

 

[vtwelve]

Thank you all for your answers.

Fan RPM = 1086,27000cfm
Weight = unknown,looks real heavy.

Yes,it is the drive end bearing that fails[we change both bearings then]
Vib measurements are not being done currently,the units are relatively new,only 2 years.Theres no allowance for axial expansion.Theres a lock system,and the set screws are tightened locking the bearings into place.

If we find a play in the shaft or bearing sound on inspection,we assume the bearing is about to fail.
The units had been running normally for a year or 2,the failures are recent behaviour,so we would want to stick to DGBB if we can figure out the root cause.
Some bearings have had part of their inner raceway broken off,so we figure plummers werent bolted exactly opposite ie. shaft was not exactly parallel to the ground,alignment issues can be ruled out,as the belts have no wear patterns on them.
I've attached a photo of a bearing before we removed it[complete failure]

 

[caoimhin1]

A bit clearer now, thanks for the pictures. So its a self lube type pillow block. Assuming the unit is balanced, I would look at lubrication in the first instance. DON'T grease the bearings after assembly. The units are factory charged with the correct amount of grease (usually 35% fill) and do not need an initial greasing. Ask the bearing manufacturer or your distributor for regreasing intervals and quantities...you may only need to fill with a couple of CCs every couple of months. Also check that the grease is compatible with the manufacturers grease which is already in the unit.
If ambient temp is above 100 degC, consider high temperature units. If ambient temp is above 40degC and the distance between the two bearings is greater than 1m, consider making one of the bearings a free end type (maybe machine a slot in the shaft for the grub screw to slide in)

 

[caoimhin1]

Sorry, re-read and saw that you have already asked about regreasing intervals. Vertical mounting makes no difference assuming you are within the load limits of the insert and housing.
What brand of bearing is it? Assume you are using high quality unit such as RHP/SKF/NTN ??

 

[Tmoose]

If the bearing near the pulley sheave is failing I'd review the belt tightening method. It's easy to exert 1000s of lbs of force with a few big belts.

That appears to be a typical pillow block, that uses a bearing with spherical outer race OD. Bearing-to-bearing misalignment (unavoidable with a fabricated support structure) is easily accomodated when the bearing assembly pivots inside the iron housing. All that rust suggests severe contamination, but it >>might<< have happened toward the end of failure cycle as opposed to initiating it. What they lack is capacity for axial expansion. Some HVAC systems seem content to make one (or both!) supports weak in the axial direction, so shaft expansion easily deflects the support. It would seem a vertical shaft would need one bearing to have a stout support to serve as a thrust bearing. Is it really vertical? I think I heard the shaft is parallel to the groundm which I would call a horizontal shaft.

There should be an additional flinger or shield above each bearing to prevent water reaching and ponding on the bearing.

http://new.backoffice.euromoneydigital.com/images/22/25376/09-05_DeBaecke_Improve_fig5.jpg

The shaft appears scored, as if an inner race had been working on the shaft, and the shaft was slid up or down to provide a fresh surface for the next new bearing installation.

 

[vtwelve]

It is a horizontal shaft,but the pillow block is bolted vertically.
Sorry for not being too clear,this other photo - its clearer,panorama.

 

[JJPellin]

A few thoughts occur to me. We have several hundred fin-fan exchangers that use ball bearings. In general, ball bearings can run very well in a fan application if they are properly applied and maintained. We re-grease most of our fans every 3 months. This is less than the bearing manufacturer recommends. But we have seen improved reliability by greasing less often. You can over-grease a bearing and cause it to overheat and fail.

I cannot tell from the pictures if these bearings have a spherical seat or not. The least expensive bearings of this type have the outer race mounted solid in the housing with no capacity to align. The more expensive bearings have spherical seats as described above. This is very much preferred. The solid mounted bearings are going to be very sensitive to how the mechanic aligns them.

For axial growth, I would certainly recommend a fixed bearing on the drive end and a floating or expansion bearing on the non-drive end. Any axial expansion of the shaft can result in very high loads on these bearings.

Does the grease fitting actually lubricate the bearing? I have seen bearings of this type where the grease fitting only lubricated the spherical seat and did not allow any grease directly into the bearing. You said the bearing was sealed. This suggests it is lubed-for-life and cannot be re-greased. caoimhin1 makes an excellent point about grease compatibility. Our bearings are provided pre-greased with polyurea thickener. If we were to add grease with lithium based thickener to these bearings, the thickeners would be incompatible. Incompatible thickeners tend to react and loose their stiffness. Basically, they turn to liquid and run to the bottom of the housing.

As noted previously, loads from the belts can be very high. I would suggest review of your belt design and procedures for alignment and tension of the belts. Move the belts in as close to the bearing as possible and set the tension as low as you can without experiencing slip. Consider converting to a cog belt if you are using V-belts.

I don’t see any immediate problem mounting the bearing sideways like this. But, I have never seen this done before. All of our fans use pillow block bearings on horizontal shafts and flanged bearings on vertical shafts. All of our bearings are mounted in the more conventional orientation. I would think your failures are most likely one of these:

• Misalignment between the two bearings and use of a bearing that cannot accommodate misalignment. Recommend changing to spherically seated bearing.
• Axial misalignment between the bearings caused by the used of two fixed bearing with no accommodation for axial shaft growth. Recommend changing to one fixed and one floating bearing.
• Over-lubrication or incompatible lubrication. Recommend you verify correct grease to add for compatibility with pre-charge grease. Only grease every three months. Only add the required amount of grease. Do not grease until it is full and grease is seen coming out of seals.
• Overload from the belt drive. Verify belt alignment and tension. Recommend changing to cog belt rather than V-belts.



Johnny Pellin

 

[Tmoose]

looks like the fan handles very humid or chemical laden air. That's too bad.
The fan appears to be suffering from corrosion along with everything else. Is this atmosphere the same as the mfr intended the fan to run? If so, I think their choice of materials and finishing is poor.

That structure looks relatively flexible in the axial shaft direction. I'd give a push on the structure or shaft with the bearing loose to see if they move relative to each other 1/8 inch or so with 50-100 lb force +/-. If it does, the requirement for an expansion bearing is reduced.

 

[diskullman]

Take a look at Browning's AH Air Handler bearings

http://www.emerson-ept.com/eptroot/public/prod/bearings/mountedball/brwnAirHandling.asp

Russell Giuliano

 

[vtwelve]

Sorry - Was out of loop for a v.long time.
If its not too late I want to thank each and everyone of you for your inputs - duly copied and applied!- Learnt a lot too..
The bearings have been ok?' since mounting 2 months back but we monitor it - just in case.
Just so that I do not leave any questions unanswered...

--The blower handls cool air around 15-20°C and the blower shaft can be pushed to move axially when the bearing is loose.
--Running in an ambient T = 20°C, can the shaft undergo any appreciable expansion??
--I attach a manual of SKF the bearings YET 210[yes,we read manuals!] To save you time pg 46 might be of interest.I'm inclined to think it might be that the grease cannot retain its lubricating properties below 40°C for long.
--The belts are always checked by rule of thumb if we push against the middle it should go in by an inch, then its not too tight or loose,just right.
Having said that,we still havent pinpointed the actual fault that caused the failures.But its running now so I guess thats all that matters!
Will keep you posted.

 

[Tmoose]

re: shaft expansion - if a steel shaft is 2 feet long between bearings, it does not take many degrees temp change to change the length a few thousandths. 20F ~ 0.003". The force required to prohibit that expansion is enormous. If the support structure is stout, the axial bearing load can approach enormous too. (Whether by design or by accident Non stout structures are common in HVAC equipment, and are required to use non expansion bearings.) Sources of heating and cooling exist everywhere. Bearing heat generation, and even sun coming in a window.

 

[vtwelve]

I understand...Lack of allowance for axial expansion coupled with loss of lubricating properties of the grease at low temp. has definitely contributed to the failures.

Quote from SKF Manual :
'when operating temperatures are below 40°C for longer periods,the grease life is shortened because oil separation is reduced'

I had been informed by the vendor that DGBB's could take a certain amount of axial load but that seems not to be the case here....

 

[Tmoose]

ball bearings have an axial load rating. I would work to understand the actual axial load, and the axial stiffness of the structure supporting the pair of bearings on each shaft. Unless there is a common steel base plate shared by both bearings the axial stiffness of a few angles and channels will not be high. That support flexibility >>can<< work (intentionally, or by blind luck) to harmlessly accommodate the shaft axial expansion/contraction without excessive axial loads.

The blower creates a certain amount of axial load. It would be interesting to see how that compares to the bearing rating.

What did the OEM have say about this?

How did the bearing disintegration compare to standard failure analysis guides (although disintegration can hide more than it reveals.).