I've been using CSI PKVue for the past 10 years to detect early bearing defects. I'm a huge fan of it. Recently I've had the opportunity to analyze a high speed spindle (12000+ rpm)that has angular contact bearings with ceramic balls. (NSK100BNR10HTDUELP4Y)
My question is: Since PKVue works by detecting high frequency stress waves, and stress waves are produced by metal to metal contact, will PKVue readings be the same in ceramic insert bearings as they are in all steel bearings?

I can't speak directly to PeakVue, but I do know that ceramic bearings can generate extremely high acceleration levels. Even ceramic bearings in good shape can generate high enough levels to overload 100 mV/g accelerometers with levels of over 70 g peak. In some cases, 25 mV/g or even 10 mV/g accelerometers must be used to monitor ceramic bearings and even these can overload when the bearings deteriorate.

Jon
Spintelligent Labs

Don,

I would think it would do ok. A stress wave is a stress wave. Maybe the Good Doctor at CSI can shed some light on this.

Jim Crowe

I have used PeakVue to monitor the type bearings referenced on multiple drill head spindles, It works very well. Far the really high speed applications, caution would have to be used to avoid overdriving the accelerometer. I used 10 mV/g accelermoters.

quote:

A stress wave is a stress wave.

Good point, but how does one define a stress wave, and why how can one measure stress with an accelerometer?

Stress wave emissions caused by impacting, fatigue cracking, abrasive wear, etc., are short-term (several microsec to a few milliseconds) transient events that propagate away from the initiation site as bending and longitudinal waves at the speed of the sound in metal. The bending stress waves introduce a ripple on the surface that will exite an accelerometer with sufficient bandwidth.

Regards,
David

Knowing Bill, I'm pretty sure it was a rhetorical question. Nevertheless, I think stress wave is just a catchy term to describe the means that vibration from impacts travels from the point of origin to thepoint of measurement.

We don't have any ceramic bearings but it's interesting to hear they cause higher vibrations. Are the ceramic rolling elements harder than the normal steel rolling elements? If so, doesn't that put more stresses on the races?

Pete,

I'm not sure why ceramic bearings generate high noise levels. I've seen a couple papers around explaining the mechanism, but I can't put my hands on them at the moment.

The first time I can across this was on a high speed milling machine. The accelerometer would overload when the bearing started to fail. It was a fairly reliable indication, but telling somebody "When the accelerometer overloads and the data looks like crap....." just doesn't come across that well. After that I saw levels on a number of ceramic bearings in good shape that would overload 100 mV/g accels.

Jon
Spintelligent Labs

Thanks Pete,

What I have suspected is that stress waves is a marketing term. If it is trademarked I would want to give due respect to the term. With an accelerometer (if not an error like some type of acoustic coupling) one measures vibration and not stress. Even with a strain gauge one measures strain and not stress directly.

After reading your responses, something came to mind. Correct me if I'm wrong, but isn't one of the purposes for using ceramic bearings, to maintain tighter tolerances? (and handle the higher temps) That's why their used extensively in high speed precision spindles. If that is the case, then maybe that could explain why the ceramic bearings overload the 100 mV/g accels. The tighter tolerances provide a better transmission path for all those high frequencies. Pete asked "Are the ceramic rolling elements harder than the normal steel rolling elements?" I believe they are. The harder the object, the higher the frequency range. Maybe that's another reason for the transducers being overloaded. I'm just thinking out loud. In any case, you all have provided me with the answer to my question.

Thank you,
Don