frequent failure of motor thrust bearing 2

[Khader001]

hi ,
the motor has combination of ball bearing and cylindrical roller bearing on the drive end. only cylindrical roller bearing on the non drive end side. The motor running in decouple condition the axial vibration is higher side comparing with horizontal and vertical. The thrust ball bearing failure frequency is every year

Driven machine has sleeve bearing with thrust collar axial float max 0.5mm
flexible coupling with 6.5mm gap between tow hubs

Is this symptom indicating the motor magnetic center not set properly
Motor : 2200HP
RPM :590rpm

 

[electricpete]

edison - that all makes sense to me except "the ball bearing is always loose in the housing to prevent skewing".
It would be much easier to tilt/cock the outer ring in a very loose housing than a tight one, wouldn't you agree?

=====================================
(2B)+(2B)' ?

 

[edison123]

pete - As in two bearing arrangement, the grease cups press the outer races of the both the ball and roller bearings to prevent the skew. Ball bearings don't tilt unlike roller bearings. That's why they are used as locating bearings.

Muthu

http://www.edison.co.in

 

[electricpete]

Thanks edison/kumar. I have zero experience with bearings with that large of a housing clearance and I have no doubt you have seen it a lot. I'd like to explore your comments.

Ball bearings don't tilt

We can tilt the outer ring a limited amount with respect to inner ring before the internal clearnance is used up.
After that further tilting stresses the internal parts.

Assume we are assembling a machine where shaft is straight and true to housing bore...

For bearing with standard fit, the ability of bearing to tilt is limited by the housing. We cannot possibly tilt the bearing enough to stress the bearing within that standard fit.

But for bearing with loose fit, you no longer have that benefit. The bearing can tilt near its full limit of travel without being stopped by the housing. What would cause it to tilt I picture would be an assembly process pushing housing axially relative to bearing and the bearing (maybe slightly off-center at that stage of assembly) sees more friction on one side and tilts. The tilting reduces the clearance. In order to remain cocked it would have to tilt far enough to reduce the clearance to zero. For 6044 bearing D=340mm, B=56mm, it requires less than 1 degree tilt to use up the 0.7mm diametral clearance if my trigonometry is correct. I neglected the rounded bearing corners which would have helped the situation somewhat. Assuming the bearing is capable of tilting 1 degree (under stress), the way to avoid preclude cocking thru design (if cannot be precluded by assembly method) would either be either A - reduce clearance to standard clearance or B - increase clearance even further so that bearing will tilt to its limit without ever contacting the housing.

the grease cups press the outer races of the both the ball and roller bearings to prevent the skew

I agree that if the outer ring face is clamped tightly and squarely on each side of the bearing, there cannot be a skew or tilt. I think maybe the photos show damage over only portion of the bearing outer ring face and portion of the thrust plate face as if there is not a firm square contact between bearing and thrust plate (although I’m still confused about that zone 2). And it suggests to me bearing cocked in the housing. Is that what you see? If not what kind of skew do you see or envision is going on here... inner ring cocked on shaft?

op - can you tell us more specifically what kind of skew you saw?


=====================================
(2B)+(2B)' ?

 

[Khader001]

thanks to all your very useful posts and brain storming

most of your assumptions are matching with mine.
last year I have raised a modification proposal that too enclosed with this post please find the attachment.

As Mr Muthu comments I agree that with cylindrical and ball bearing combinations are working well without any problem. these all motors are driving which has anti friction bearings

the problem which has friction bearing with larger radial clearances, also when the motor is in operation motor and the blower shaft runout was measured blower side 0.15 to 0.20mm motor side 0.15mm the peak raise looks both shaft moving with the same timing

a sectional drawing also attached it may help to get more details

grease: SHC 100 mobile
intervals : 1000 hrs/ 40g

 

[electricpete]

So I think you are envisioning the outer ring moves as the shaft rotates. Either up and down or maybe tilting. But no permanent tilt with respect to housing as if cocked.

I can see why you would be concerned about tigher radial clearance of thrust bearing than radial bearing. Even if thrust bearing does not contact housing radially, it is still clamped so it can still either see radial load or have relative motion at the axial clamping surfaces.

Did you happen to see if the outer ring move as the shaft was manually rotated with thrust plate removed? (that would rule out cocked bearing to my mind).

Can you explain some more the pattern shown in the middle and bottom row of photos? It looks to me as if the damage is preferentially on the upper half of the outer ring and thrust plate. If the only problem is deep groove ball bearing internal clearance tighter than cylindrical beairng internal clearance, we'd expect to see wear 360 around those surfaces, wouldnt' we? (if not cocked bearing maybe it suggests shaft at permanent angle to the housing)

If you implement the proposed solution the spherical bearing is wider than the ball bearing. I gather you will machine the thrust plate or the housing to accommodate that.

Still thinking about the grease. My experience has been bad reliability (low life) with high D*N greased bearings. It seems the motor OEM's routinely go beyond the recommendations of the grease and bearing OEMs. I'll see if I can dig up some bearing manufacturer guidelines on D*N




=====================================
(2B)+(2B)' ?

 

[Khader001]

when the motor shaft rotates manually no movement found on the ball bearing. the movement force coming from the driven machine when the motor operates with rated load and RPM

The thrust bearing wear pattern studied most of the housing and thrust bearing end cover step bearing seating area found worn out around the bearing outer ring.

After installation the thrust bearing alignment was inspected there is no scew found.
one more experiment done while operation in rated load and RPM motor thrust bearing end cover loosen found that motor vibration came down 3.6mm/sec to 1.9mm/sec same time two dial indicators on drive and driven shafts measured 0.15mm in both shaft no difference in runout readings after loosening the thrust bearing end cover its conforming that thrust bearing is oscillating

greasing we under discussing with the mobile to select suitable one

now I collecting all the damage bearings once dismantle bearings and ensure the wear pattern indications

modification till not started looking for more information

 

[BrianE22]

Possibly the bearing seat or the cover is dinged up a bit and assembling them tight is putting an excessive load (bearing not square) on the bearing. Or the pre-load is too high.

 

[electricpete]

I've lost track of what's going on. Info was over several days with comments between (a lot of the clutter is mine) which makes it difficult to focus on all the info provided. Let me post the op's info all together

the motor has combination of ball bearing and cylindrical roller bearing on the drive end. only cylindrical roller bearing on the non drive end side. The motor running in decouple condition the axial vibration is higher side comparing with horizontal and vertical. The thrust ball bearing failure frequency is every year
Driven machine has sleeve bearing with thrust collar axial float max 0.5mm
flexible coupling with 6.5mm gap between tow hubs
Is this symptom indicating the motor magnetic center not set properly
Motor : 2200HP
RPM :590rpm
=====
thanks....
some of the pictures and vibration details are attached pl refer. this motor bearing dismantling is under progress (attachment 14 Jul 16 10:59)
=====
thanks to all your very useful posts and brain storming
most of your assumptions are matching with mine.
last year I have raised a modification proposal that too enclosed with this post please find the attachment. (attachment 17 Jul 16 12:22)
As Mr Muthu comments I agree that with cylindrical and ball bearing combinations are working well without any problem. these all motors are driving which has anti friction bearings
the problem which has friction bearing with larger radial clearances, also when the motor is in operation motor and the blower shaft runout was measured blower side 0.15 to 0.20mm motor side 0.15mm the peak raise looks both shaft moving with the same timing
a sectional drawing also attached it may help to get more details
grease: SHC 100 mobile
intervals : 1000 hrs/ 40g

======
when the motor shaft rotates manually no movement found on the ball bearing. the movement force coming from the driven machine when the motor operates with rated load and RPM
The thrust bearing wear pattern studied most of the housing and thrust bearing end cover step bearing seating area found worn out around the bearing outer ring.
After installation the thrust bearing alignment was inspected there is no scew found.
one more experiment done while operation in rated load and RPM motor thrust bearing end cover loosen found that motor vibration came down 3.6mm/sec to 1.9mm/sec same time two dial indicators on drive and driven shafts measured 0.15mm in both shaft no difference in runout readings after loosening the thrust bearing end cover its conforming that thrust bearing is oscillating
greasing we under discussing with the mobile to select suitable one
now I collecting all the damage bearings once dismantle bearings and ensure the wear pattern indications

modification still not started looking for more information

=====================================
(2B)+(2B)' ?

 

[dArsonval]

Observation:

Photo showing the partially dismantled bearing, and bearing bracket conflicts with the bearing installation arrangement drawing provided by the OP.
The engineering drawing shows the cylindrical roller bearing shouldered on shaft journal first, and held enclosed by a bracket labeled at “A” on my attached photo.
A spacer with dust fin labeled “B” secures the separable NU 1044 inner race and rides with the shaft on the outside of bracket A.
These two components would visibly shroud, hide, or prohibit the NU bearing from even being seen during the dismantling process.
The OP’s picture showing the NU bearing surrounded by a bracket and completely exposed doesn’t “jive” with the engineering drawing.

Are we looking at a scenario where the bearings are not or (were not) in their proper locations to begin with?

John

 

[electricpete]

I interpretted that we are looking at in that photo is the deep groove bearing. Although it also looks like bearing/shaft has been slid towards driven machine since it is sticking out beyond the housing. Not sure exactly how they would do that.

...That's what it looks like to me but I've been wrong before or maybe I misunderstood John's comment. Hopefully op (or John) can clarify.
=====================================
(2B)+(2B)' ?

 

[dArsonval]

I'll plead guilty and admit I've periodically studied the photos on the PDF file, and PowerPoint slideshow more than a couple times this past week.
What I'm seeing is a brass cage cylindrical roller bearing fully exposed, sticking out of the bearing bracket that's in the process of being pulled off
using long rods threaded into the bracket. That bearing we can all see in the photo is a cylindrical roller bearing with a separable inner race.
Behind that bracket (not yet removed during the dismantling phase) is another bearing that according to the drawing supplied by the OP, should be the cylindrical roller bearing.

There's a LOT going on with this failure. And the placement/location of the bearings may explain the frequent failures the OP mentions earlier in the conversation.
It's not uncommon for an expert repair person to follow someones earlier mistake.
Couple this bearing placement question with the PowerPoint slideshow that's suggesting a proposed engineering design change in the whole arrangement makes the situation more complicated.

There's a "To be continued" here we may never get to read about.

Enjoying the forum,

John

 

[Khader001]

Dear All thanks for spending precious time with this issue

I m not sure that motor shaft slides towards driven machine. if there is axial movement on the motor shaft and it creates excessive axial load on the ball bearing then the bearing failure should lead to damage to the balls runway or bearing internal parts. the ball bearing axial face should not get damage because the ball bearing don't have space to move along with the shaft. the motor shaft axil movement limited as the ball bearing axial clearance.

the motor bearings are removed and the damage bearings are in analysis soon I will upload the photos

the intension of the modification is to avoid radial load sharing between ball bearing and cylindrical roller bearing. the spherical bearing selected because it can work well with both loads radial and axial.

what are the complications in spherical roller bearing design kindly write more about this

 

[electricpete]

I would look at the bearing minimum load to prevent skidding. Spherical roller bearing is less tolerant of low load than deep groove ball bearing. And while minimum load ratings to prevent skidding tend to ignore the lubrication type, skidding is more prevalent in greased bearings than oil bearings. You're adding an adapter sleeve in the housing in attempt to help add some radial load I gather, but how much will it share with the cylindrical roller? Especially if the spherical roller has larger internal clearance than the CRB. I'm not sure how we know that the spherical rollers will have any load at all.

Another question either from position as devil's advocate or maybe from a position of ignorance… As you mentioned many bearings operate reliably with deep groove thrust bearing adjacent to cylindrical roller radial bearing (and with large housing clearance at the deep groove). You're blaming the problem here on lower internal clearance for the deep grooves than for the cylindrical rollers, but doesn't that same factor apply to other motors with similar configuration with same bore diameter DGBB and CRB? How is this one different? Maybe others use something other than C3?


=====================================
(2B)+(2B)' ?

 

[electricpete]

I'm not sure that motor shaft slides towards driven machine.

If you're responding to me, I was talking about in the photo. During process of disassembly once you remove that outer cap there is nothing to prevent you from pulling the rotor toward the driven machine (unless maybe the cylindrical roller bearings have lips on them). At any rate John's question remains as to what we are looking at in that photo (and although I don't see what he sees, if it were my motor I'd be intently interested on understanding what he's saying).

=====================================
(2B)+(2B)' ?

 

[dArsonval]

Author of the post has replied, "...the motor bearings are removed and the damage bearings are in analysis soon I will upload the photos."

Look forward to seeing, and reading what you've found.

I'm still confident the bearings were originally installed in the wrong place. Hence, the continued failures.

Staying tuned,

John

 

[Khader001]

Yes if the end cover removed there is nothing to stop rotor sliding towards driven machine

Pl check the attached video I have check the radial clearance of the rotor till 0.05mm rotor move with minimum pressure then rotor stops the movement then the hydraulic pressure was raised then we can see jumping movement from 0.05 to 0.10 and then 0.23mm

Mr John
Can you pl explain your view
As you said the bearing installed in wrong place what do you mean is themagnetic Centre not positioned properly ?

Most of the readings are taken I didn't found any abnormal issues rotor run out 0.01mm
Ball bearing housing concentric to the cylindrical roller bearing housing 0.01mm bearing alignment is good

 

[Khader001]

Ball bearing outer ring pic
Pl comments and the wear pattern

 

[electricpete]

Pl check the attached video I have check the radial clearance of the rotor till 0.05mm rotor move with minimum pressure then rotor stops the movement then the hydraulic pressure was raised then we can see jumping movement from 0.05 to 0.10 and then 0.23mm

The logical interpretation I guess is the first 0.05mm is taking up the internal clearance of the deep groove bearing. After that it takes further force which moves the outer race of the deep groove bearing outer race is pushed (causing sliding against the axial face) until it reaches the point where the cylindrical bearing clearance is taken up. It does tend to support the drawings / dimensions you showed and your "theory of the crime". Is that the way you're looking at it?


=====================================
(2B)+(2B)' ?

 

[electricpete]

Bearing photo observations:
1 - it looks like a track right down the center, which would correspond to zero axial load.
2 - there are many whitish marks on each side of that center load track. Not sure what to make of that. Either frosting from severely degraded lubrication or perhaps denting from many particles in the lubricant. Speculation: if lubricant severely degraded it may be result of overheating but if foreign particles present they would be result of rubbing/movement of the clamping faces against the outer ring
3 - brown stains generally signs of lubricant overheating although doesn't look that sever. Would like to look at all signs of possible overheating together (the whole bearing).

It's tough to tell from this angle whether we're looking at a segment of inner ring or outer ring.
Any other views available?


=====================================
(2B)+(2B)' ?

 

[dArsonval]

Ladies and Gentlemen;

Study the attached photo screen shots. Tell me I'm not going insane.

John