Dear Sirs,

Due to inertia of motor/load the amplitude of
modulation is inversely proportional to the frequency at the same modulation force i.e. the more frequency ,the less we see modulation at the same modulation force (see att.files)
How this fact is accounted in MCSA-technology?

Victor

Great questions and discussion

I will tell you my limited understanding. For low inertia loads, we expect roughly equal upper and lower pole pass sidebands around line frequency in current. For high inertia loads, the upper sideband decreases and lower sideband increases by the same amount. In the extreme of a very high inertia load, the upper sideband is non-existent and the lower sideband is twice as much as would be for very low inertia load.

There was a paper that presented some theory behind this, although I haven't looked at it closely yet. Also myself and some others have corroborating experience that for motors driving very high inertia loads (in our case huge flywheel), the upper sideband is non-existent.

Link to that discussion:
http://maintenanceforums.com/eve/forums/a/tpc/f/7161085...481072103#7481072103
(the relevant discussion starts with post by electricpete 22 February 2007 08:36 AM )

Dear electricpete, thank you very much for reply.

I must say, that it is the first answer on that "simple" question in essence.
Before that people or don't answered this question at all, or answered "I do not understand", or answered even so: "So what is your question about MCSA? What does "How do MCSA accounts this fact?" mean. I still have difficulty in understanding what you are asking. If I ask you the question that "the color of the snow is white. How do you account for that fact?" What would be your answer?"

Where we can read about influence of moment inertia on PPF sidebands?

It's very important, but here we say about MECHANICAL moment modulation.

Simple example:
Please, place the door between fingers. The more the frequency of hand movement,
the less the response of the door. The heavier the door, the less it
response...

Sorry for this big picture in my previous post (I wanted to add a little link to it) and sorry for word "freqyence". Frequency, of course...


Victor

I am in 100% agreement with your comments. The first result of a broken rotor bar is torque oscillation. That torque oscillation may or may not tranlsate into speed oscillation, depending on the inertia (no speed oscillation for high inertia loads, lots of speed oscillation for low inertia loads).

Attached is an article which was published by Thompson in the 32nd Texas AMU Turbo Symposium (2003). I downloaded it when it was available for free on their website. It is no longer available, as they only keep the last three years posted.

On the bottom of pages 155 and top of page 156 (ending in equation A-7), he describes how the mmf harmonics results in a current lower sideband at F-2sF = pole pass frequency below line frequency. This is the end of the story for very high inertia loads, since there is no significant speed oscillation. As I mentioned before, we see with our reactor coolant pumps with a very high inertia humongous flywheel that indeed the lower sideband is the only one present... no upper sideband at all.

Now apparently the upper sideband only appears when speed oscilation appears. Thompson says this:
"Speed and torque oscillations occur at 2sf1, and this induces an upper sideband at 2sf1 above f1."
So it is only with typical low-inertia loads that we have speed modulation which causes the upper sideband. I have never figured out the exact math explanation for this (maybe you can help us figure it out?), but it sounds plausible to me.

TurboTAMU_RotorCaseStudiesThompson_T32pg145.pdf (343 Kb, 19 downloads)

Dear electricpete, thank you very much for reply.

We see the influence of moment inertia on mechanical faults and on rotor bars faults.

Does it mean, that absolute values (dB on characteristic frequencies) is not so important
and MCSA is valuable mainly for trending of faults?

Do you have information about it influence on another faults (eccentricity etc)?

Victor