Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
Nagesh

Member
Registered:
Posts: 7
Reply with quote  #1 
May be this got asked somewhere earlier im not sure but I was wondering when you guys would be replacing an anti friction bearing.

Because sometimes I see bearing peaks (and their harmonics) in the envelop (SKF instrument) or peakvue (CSI) but not in velocity. And there wouldn't be much sound from the equipment either.

Will there always be bearing peaks (may be small) in the envelop or peakvue albeit small? I for myself tend to replace the bearing after I see the peaks in velocity. What is the best practice here?

Regards,
Nagesh
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,397
Reply with quote  #2 
There is a notion that you need to look at bearing frequencies in velocity.  I don't agree. In my view always look at the spectra in velocity but ALSO in acceleration.  The higher order peaks will be better viewed in acceleration. the lower end better in velocity. Do both.  higher peaks in accel tend to be early, the higher the low end velocity peaks are an indicator of greater severity.  And always look at the waterfall plot for last 8, 10 or 12 surveys.  Gives you an instant view of any deterioration.

In my experience Peakvue or Enveloping gives very very early indications except when a cracked inner race. rgds
OLi

Sr. Member
Registered:
Posts: 1,913
Reply with quote  #3 
IMHO either the damage get older and by that as it contains lower frequency components and get higher in velociy. Or the damage spread and you can find more places damaged and by that a increase in higher frequency content. I think both is as bad and indicate the failure getting closer. When various things require you to run a little to long it is mostly found that bearings run quite some time longer than expected and sometimes that is lucky.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #4 
It is generally accepted that there are four stages of bearing failure.  One excellent resource that discusses both sound and vibration is at http://www.engr.uconn.edu/~aly/Courses/ME343/Reading_1_Sound%20and%20Vibration%20in%20Rolling%20Bearings.pdf.  Many other resources site that 1% to 5% of life remains at stage 3.  IMHO, you ideally want to "catch" the bearing somewhere between stage 2 and stage 3.  There are two reasons for this approach, one being that a bearing expert can generally tell at that stage of failure the root cause of the failure.  At the latter of stage 3 and definitely stage 4, often the damage is so severe that one can't determine the root cause of failure.  Two, generally catching it at the early part of stage 3 minimizes consequential damage.

One thing that can sometimes be experienced, especially with inner race and outer race failures, is a stage where the vibration tends to reduce in amplitude.  That is when the sharp edges of pits and spalls start to smooth over.  Once that has occurred consider shutting down the machine for maintenance.
RRS_Dave

Sr. Member / Supporter
Registered:
Posts: 216
Reply with quote  #5 
Quote:
Originally Posted by John from PA

One thing that can sometimes be experienced, especially with inner race and outer race failures, is a stage where the vibration tends to reduce in amplitude.  That is when the sharp edges of pits and spalls start to smooth over.  Once that has occurred consider shutting down the machine for maintenance.


Anyone who uses Odyssey and Spike Energy knows this is an indicator that can't be ignored. When the spike drops suddenly, it's time to commence to commence[biggrin]
John from PA

Sr. Member
Registered:
Posts: 963
Reply with quote  #6 
Quote:
Originally Posted by RRS_Dave
When the spike drops suddenly, it's time to commence to commence[biggrin]


Priceless...
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,397
Reply with quote  #7 
Quote:
Originally Posted by OLi
IMHO either the damage get older and by that as it contains lower frequency components and get higher in velociy. Or the damage spread and you can find more places damaged and by that a increase in higher frequency content. I think both is as bad and indicate the failure getting closer. When various things require you to run a little to long it is mostly found that bearings run quite some time longer than expected and sometimes that is lucky.


Couuldn't agree more Oli. rgds

ps, coollecting a 58 dryer cylinder (sorry 56) dry end (inc about 124 felt rolls) today.  In up to 29 deg C ambient. Lose a few Kilos then (I hope)
OLi

Sr. Member
Registered:
Posts: 1,913
Reply with quote  #8 
You need to refill the water or you get sick, 2-3 liters per 5-6 hours is really needed.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
Vibe-Rater

Avatar / Picture

Sr. Member / Moderator / Supporter
Registered:
Posts: 1,397
Reply with quote  #9 
Definitely Oli. They have electrolyte drink reservoirs and electrolyte icy poles!  The unfortunate thing is the access doors on the drive side are mostly stuck which adds to the burden. rgds
OLi

Sr. Member
Registered:
Posts: 1,913
Reply with quote  #10 
You are lucky having such service in place or do they work a lot in that area so that's the reason for the service? As a consultant you are basically on your own here, maybe a waterfountain in the control room if you are lucky. I found another suction cylinder driveshaft spline seize last week and it had all the indications with axial vibration both ends. On a yankee machine, same as before. I am curios why that started to happen 1-2 times per year. It has not done that so often before.
__________________
Good Vibrations since early 1950's, first patented vibrometer 1956 in the US.
http://www.vtab.se
KVib

Member
Registered:
Posts: 7
Reply with quote  #11 
When trying to determine when to change a bearing i put a lot weight on what the waveform trend is showing. If the trend has made an significant change and i have detemined it is a bearing im probably going to replace it.
__________________
Regards
stace1g

Sr. Member
Registered:
Posts: 104
Reply with quote  #12 
Early stage deterioration may be indicated by Gse, envelope, high frequency g's spectra say out to 20kHz dependent upon mounting. May only show up as a broadband 'hump' in the high frequencies with nothing indicated in the envelope except a raised noise floor.  Generally nothing in the usual low frequency velocity spectrum. At this stage the problem may well be stalled by the addition of lubrication in the case of greased bearings and there may not be any 'visible' damage to the bearing. 

As the fault progresses the 'hump' in the HF data may move down in frequency, it may not, and you also may start to see peaks over the hump spaced at a particular bearing fault frequency being harmonics of this frequency exciting the high frequency resonances (accel, bearing housing etc). This information will be pulled out by the envelope, demod etc and you may well see a definite impacting pattern in the TWF (accel)  Again nothing in the velocity spectrum  however you may be able to spot some bearing tones if you swap to a log scale

Further deterioration leads to an increase in harmonic activity in the demod data and the bearing tones will be more apparent in the velocity spectrum even with a normal linear scale. Swapping the velocity spectrum to g's if there is the facility to do so will tend to show up harmonic series. With continued deterioration harmonic activity within the velocity spectra will increase and other bearing tones may start to appear along with modulation, harmonics etc. You may at this point start to see harmonics of 1x and an increase in the noise floor. Acceleration levels may well at this point drop off if not sooner. The bearing will at this point be appreciably audibly noisy and is beyond the point at which change is recommended 

this is what I have seen on numerous occasions but not all bearings follow this classic textbook type failure for example the failure of a cage can lead to very rapid deterioration which may not generate the stages seen above. So yes you need to look at both velocity and acceleration data to include demod and TWF in order to make a judgement as well as assessing the trend development.

Gary 




Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.