I have simple device that measure vibration and gSE (just number with out specter). When I measure acceleration and gSE to control bearing I have only simple scale of numbers whey bearing is good and alarm number.

Searching on internet gSE is more reliable instrument then acceleration and in the practice I can take more reliable gSE reading then acceleration, what you think?

My question is how to know with my instrument when I need to replace bearing and how to estimate how long bearing can make?

I am wondering if anyone has better scale of acceleration/gSE to diagnose bearing?

Thanks in advance...

quote:

Searching on internet gSE is more reliable instrument then acceleration and in the practice I can take more reliable gSE reading then acceleration, what you think?

I would think the opposite is true. More must be know; RPM for example. gSE works best with higher RPM machines - acceleration works with all.

With your limited instrumentation you'll need time to trend your numbers; but never think you can achieve what FFT analyzers and PC software can attain.

How does your particular instrumentation work; filters, fmax, averaging, ???? and etc... You'll have to evaluate and determine where you want to go and cost justify your decision.

gSE is designed for bearing control and think so is better, but I hearted it is also pure technology.

It measure overall gSE with 5kHz filter and I measure acceleration with 2kHz filter. I started with bearing 6322 when is replaced it gSE was 0.06 and on another bearing it was 1,3 and the one that we replaced it was 3,5. RPM is 1480 rpm. What would be scale for acceleration or gSE for mentioned case to call of the bearing? And, ENTEK manual alarm for my case is 0.7 gSE...

Anyone willing to share acceleration scale for bearing

Please, where I can get bearing damage acceleration severity scale...

Have you tried Entek/IRD for the information on Entek/IRD technology?

quote:

Originally posted by Enrique BL:
Please, where I can get bearing damage acceleration severity scale...

I think most will agree that the pattern and the trend are most important.

But if you are looking for a number to judge bearing severity, a very good starting place imo is the true peak (or true peak-to-peak) acceleration.

Some great thumbrules were published in "DESCRIPTION OF PEAKVUE AND ILLUSTRATION OF ITS WIDE ARRAY OF APPLICATIONS IN FAULT DETECTION AND PROBLEM SEVERITY ASSESSMENT" by Robinson and Berry (Emerson Process Management Reliability Conference 2001 October 22 - 25, 2001.).

Available here
http://www.compsys.com/drknow/aplpapr.nsf/06b6f5a4de2ea...05fa041?OpenDocument

The ability to sense the true peak is limited by the sample rate (higher is more sensitive). Peakvue has the ability to detect those true peaks based on the hardware sample rate.

This message has been edited. Last edited by: electricpete,

Thanks for direction, I will check with IRD, I addressed to them one but not got back answer...

I will examine article....

Enrique

There is no one number to give you for gSE. It will depend a lot on which transducer you are using.

Dave

New moments,

I lubricate looking damaged bearing and after measurement of acceleration and gSE it decreased 10 times!! Now it is hard to call bearing, I have to lubricate first but I am not sure when will I recognized bed bearing even it is properly lubricated...

Monitor it while you lube it.

Danny thanks but I do not understand you... you suggest to monitor it while I lube it in run... I understand the concept but it will come to point I think even I lubing it it will not decrease accelerations because bearing is damaged?