Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment   Page 2 of 2      Prev   1   2
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #16 
Quote:
Originally Posted by Edwin
The outer race is not sliding, but rolling around through a too wide fit. That way there is no heat generated.
(I couldn't find the English word for it, but I hope you understand what I mean)




Similar to a disk rolling within a ring; not too unlike a cycloidal drive.  

Untitled.jpg 

Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #17 
Danny,
yes its a peakvue reading. 1000Hz Fmax/1600L HP filter 1000Hz
Attached is the auto correlation with clear 1X and side band/modulation.

I checked again today, bearing is only warm, slightly above ambient to touch. In live mode, side band frequency did not vary,  but there was the occasional increase in side band amplitude and number.
Interesting (to me anyway) is the "X" button or harmonic family finder doesn't pickup the side bands? GSE2A auto corro.JPG

Ron Brook

Member
Registered:
Posts: 35
Reply with quote  #18 
I have seen this freq sideband many times. It is a bounce mode of the cage. So you can think of it as a loose bearing, but you shouldn't see evidence of it spinning in the housing. I've seen it in new bearings where preload was insufficient.
Edwin

Avatar / Picture

Sr. Member
Registered:
Posts: 131
Reply with quote  #19 
You get me, John!
I was about to show this: (at 0:20) 
 to illustrate.

Off course with the bearing fit it is not a clearance (=movement) of millimeters, but only microns.

__________________
= Cost Effective Asset Management =
Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,403
Reply with quote  #20 
Nok,

To me, it means this.

1 x running speed means there is something shaft related such as imbalance or misalignment.

Harmonics mean there is also looseness.

The sideband spacing indicates the cause of the 1 x running speed so something at 284 cpm is the cause.

The fact that these frequencies all appear in the PV data means that all of these things are sufficient to cause impacting at running speed and 284 cpm.

The epicyclic motion that Edwin suggests would seem to generate rolling friction rather than sliding and might not generate that much heat. It would also be a source of imbalance at 284 cpm that would cause vibration at shaft speed.

What are the twf amplitudes?

There are a few unknowns and theoreticals but I don't think any of them change the answer. It's loose in the housing.

dmcmmc

Avatar / Picture

Member
Registered:
Posts: 69
Reply with quote  #21 
Is this fan belt driven or direct couple driven?
Jim Crowe

Member
Registered:
Posts: 53
Reply with quote  #22 
I have seen similar data on a bearing with both an inner and outer race defect. The autocorrelated waveform has some of the 1X peaks line up exactly and others do not. You also have higher frequency peaks riding on the 1X peaks (just like an inner race defect. If you have a defect on inner and outer races they will both impact at the same time generating a low frequency amplitude modulation. If you know the bearing calculate the period between a roller hitting both races at the same time and it occurring again. Looks like the low frequency modulation is at 4.539 Hz for a period of .22 seconds. The 4th tall peak in your spectrum is not a harmonic of running speed. Put your cursor on it and see if it has exact 1X sidebands.
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #23 
dmcmmc, its direct drive

Danny, to answer your earlier question twf amplitude is 4g's

Jim, the 4th tall peak to the other tall peaks are sidebands of 49.69 Hz, run speed is 49.89Hz,  I probably don't have enough resolution
I did notice that the harmonics of the 1st big peak do sort of drift off from the other tall peaks. 
When I run the fault freq's for 6311 (8ball) bearing over the zoomed correlated waveform I get a sort of match for bpfo & bpfi.
Quote
"If you know the bearing calculate the period between a roller hitting both races at the same time and it occurring again."

Sorry, I'm not mathematically astute enough to do that.
EDIT:
Do I need to multiply bpfo and bpfi until i get a matching number?
like say if
bpfo =2 &
bpfi =3.
6 X bpfo =12  & 4 X bpfi =12 hmm that wont work ???? how would i calculate that?

GSE2ABPFI.JPG  GSE2ABPFO.JPG 

So the question on my lips is it bpfo/bpfi or slipping in housing or on shaft

Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,403
Reply with quote  #24 
I had a client tell me he actually saw a bearing behaving exactly as Edwin described only much slower. I don't have access to his data, though.

Can you get a strobe light on the end of the shaft?
Noknroll

Avatar / Picture

Sr. Member / Supporter
Registered:
Posts: 843
Reply with quote  #25 
still havent got a strobe light [bawl]
stace1g

Sr. Member
Registered:
Posts: 104
Reply with quote  #26 
Struggling to understand how a rotating (or rolling) outer bearing race would modulate 1x vibration?

Are we saying that the outer race rolling around the housing generates its own imbalance force that modulates the 1x? (in this case with impacting of both hence the appearance in peakvue and the harmonics) 

In his book Talor describes a situation with the inner race 'slipping' but this is seen as a discrete frequency lower than run speed that can 'beat' with run speed as opposed to causing any modulation

Gary
Jim Crowe

Member
Registered:
Posts: 53
Reply with quote  #27 
Here is one example I had with heavy modulation at 5.4 Hz in a 2 pole pump motor:

Modulation in the waveform and peaks in spectrum were spaced at 5.4 Hz

The period for the 5.4 Hz modulation is 185 ms

The period for fundamental BPFO was 183.5 Hz for a period of 5.45 ms

The period for fundamental BPFI was 291.4 Hz for a period of 3.43 ms

In 185 ms BPFO will impact 34 times

In 185 ms the inner race will impact 54 times

If we start at zero, with both defects impacting at the same time, every 185 ms both defects will occur again. Therefore the Peakvue data will be amplitude modulated at this low frequency.

The bearing ended up having a significant inner race defect with a slight outer race defect.

Danny Harvey

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,403
Reply with quote  #28 
"Struggling to understand how a rotating (or rolling) outer bearing race would modulate 1x vibration? 

Are we saying that the outer race rolling around the housing generates its own imbalance force that modulates the 1x? (in this case with impacting of both hence the appearance in peakvue and the harmonics)"

Gary,

As I understand it the idea is that because the outer race is rolling in the housing, the shaft center is eccentric by the amount of clearance which increases the radius of the mass imbalance.

I hope that makes sense to you (and me).
stace1g

Sr. Member
Registered:
Posts: 104
Reply with quote  #29 
Understand the concept Danny but for some reason it does not 'feel' right! I quite like Jim's explanation though.

Gary
Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.