I need to ask this question. It may seem crazy but I was thinking about this last week when troubleshooting that last motor.

Lets say I go out and take some readings on a motor at 6400 lines of resolution, and my FMAX is 12,000cpm. Trying also to pick up side bands around running speeds.

I come back in and look at the readings, and there are no peaks associated with any motor problems at all...Is this what I should expect from a good motor, or do ALL motors show pole pass freq. and side bands and 2X line frequencies even when they are okay?

I hope I ask that ok...And when using my SKF CMVA 65 which we have, isn't there a formula or some way to tell the amplitude of these side bands on the 2X line freq, and pole pass freq. My machine does not read in db's only g's and ips...I know the PDMA test use db's and htz between the peaks.

I think someone mentioned the other day there was a way to determine how bad the rotor or stators were doing some sort of averaging. Bottom line is I'm trying to determine when enough is enough, and it's time to pull the motor...

Thanks.

At 12,000 cpm Fmax, you're not looking out far enough to see rotor bar or stator slot energy, or for that matter most rolling element bearing frequencies. I routinely look out to 70X turning speed when collecting data to insure I cover these possible faults. If you're looking specifically for turning speed sidebands and such your set up should work fine but then you'll need a second data set to see the higher frequency stuff.

A dB is another way of expressing a ratio: dB = +/-20*log(Ratio) (changing the sign is equivalent to inverting the ratio). If you have a main peak at magnitude 5 and a sideband at magnitude 0.05, then the dB difference (or "dB down") comparison of these two peaks is 20*log(5/0.05)= 40dB

Yes I normally do both. I go out to 240,000cps even 300,000 if need be. I was looking for the peaks with the PPF back at 1 & 2X. You know the Fs & Fp sidebands which show up in the 12K and below range.I get the 2XLF further out like you say.

Thanks

quote:

Originally posted by Joe Gurney:
At 12,000 cpm Fmax, you're not looking out far enough to see rotor bar or stator slot energy, or for that matter most rolling element bearing frequencies. I routinely look out to 70X turning speed when collecting data to insure I cover these possible faults. If you're looking specifically for turning speed sidebands and such your set up should work fine but then you'll need a second data set to see the higher frequency stuff.

I want to go back to my previous question. When a motor is good there are no peaks at 2XLF PPF etc, correct...

I also refer to the Technical Associates wall chart as well. And when motor problems begin to occur then come the peaks? Right? Sort of a strange question, but I assume that when a motor is okay like a new one, that there are no peaks.

quote:

Originally posted by electricpete:
A dB is another way of expressing a ratio: dB = +/-20*log(Ratio) (changing the sign is equivalent to inverting the ratio). If you have a main peak at magnitude 5 and a sideband at magnitude 0.05, then the dB difference (or "dB down") comparison of these two peaks is 20*log(5/0.05)= 40dB

Sorry Rod, I didn't understand what you were asking. Yes, I don't think you would see rotor or stator sidebands (at any significant amplitude)around turning speed unless there was an issue, but I may be wrong. That's my understanding about it anyway. Someone else may have different information but I only have found these sidebands on problem motors.

Thanks Joe. I know it's sort of an off the wall question, but I was assuming motor problems would be the same as any other.

Just like NOT having a 1X peak when there is NO imbalance problem would be a good excample.

I didn't want myself to get in the mindset of thinking all motors are having rotor and stator problems when seeing those peaks. Or I guess another way to put it would be on a new motor install, there are no fault peaks...

I am not suprised to see 2xLF or RBPF on "good" motors. You should trend the amplitudes of these peaks to see if they are getting higher. I don't expect to see Pole Passing Frequency (PPF) sidebands on a "good" motor.

R-Rod,

I would not be surprised either if a new or rebuilt motor exhibitted a slight or moderate 2xLF peak or other electrical frequencies. You have to always ask yourself "Will this motor last me X number of years?", "X" being the number of years YOU expect your new motor to last or in between scheduled overhauls.

Just as a trending tool limit, I use 0.1 IPS at 2xLF.

DAN V.

Since I've started getting into motor troubleshooting I've yet to see the PPF sidebands get up to .100ips This last motor was around .06 to .07 Not sure about the 2XLF. I'll look tomorrow and see where they have been running.

I just pulled out another 350hp blower motor. It's been trending up slowly the past several months. This motor has been in there running since 1991...Looking at the spectrum it appears to be mostly rotor inbalance. At least the dominant peaks are mostly 1X...Maybe a couple of 2XLF as well...Need to look closer at this one.

Regardless, it's due time for an overhaul. I can feel and hear a knocking sound from inside...It was in the red alarm as well, even when uncoupled from the blower. Just like the last motor.

quote:

Originally posted by DanV:
R-Rod,

I would not be surprised either if a new or rebuilt motor exhibitted a slight or moderate 2xLF peak or other electrical frequencies. You have to always ask yourself "Will this motor last me X number of years?", "X" being the number of years YOU expect your new motor to last or in between scheduled overhauls.

Just as a trending tool limit, I use 0.1 IPS at 2xLF.

DAN V.

I generally agree with Steve's comments about electrical frequencies in vibration.

To put my own spin on it, I would be least concerned about RBPF +/-2LF pattern.

I would notice 2*LF more than RBPF, even though still not particulalry concerned unless the overall gets higher than I like. It may indicate the motor was not assembled perfectly.

Now as for ppf. I have to say that I have several machines that have ppf sidebands around 1x and 1x-harmonics which are very noticeable even on a linear scale in vibration and I live with them because they don't show up with any large sidebands around 1x on current signature (we don't look at higher frequencies in current like some do). Certainly it is less common than 2*LF. But as long as the 1x vibration is reasonably low and not a lot of harmonics, I have a hard time mustering the courage to recommend to pull the motor based on pole pass sidebands... particularly when I don't know what the cause is. I have the general feeling it is some form of dynamic eccentricity. Anyone agrees? Checked runout on the shaft extension of one of these while secured and didn't see anything.

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