I've been trending a relatively new blower here at our plant, installed in March 2006...Already had an unexpected bearing failure on it in 12/2006...No warning beforehand, running in the green.

This blower is trending up again on the drive end bearing. The 3VACC & 3HACC...Amplitude
3.17g's...It alarms at 4.00g's

As you can see from the attachment the whole spectrum is pretty much flat line, except for the 38X peak.

I ran my bearing fault frequencies over it, and none of the frequencies line up. Sort of scratching my head here. All the velocities are well below the alarm stages and looking good.

Has anyone had a similar looking spectrum where there was just one huge peak out there like this, and if so what did you find?

What's really strange is that this blower is really running smoothly to the touch...!You can stand a dime on it's end...

Thanks

Hi Rod,
3g is a lot. I see that you are using a lot of averaging in your spectrum calculation. Do you have a non-averaged acceleration time signal we could take a look at?
Tom Murphy

Tom,

all of our routes are set up to take 8 averages. I could change that for only one I guess, and retake the readings. Is that what you mean?

quote:

Originally posted by Tom Murphy:
Hi Rod,
3g is a lot. I see that you are using a lot of averaging in your spectrum calculation. Do you have a non-averaged acceleration time signal we could take a look at?
Tom Murphy

Just out of curiosity, I went into the motor spectrum and noticed the same 38X peaks in that. I'm wondering now if the motor isn't feeding the blower?? Looking at RB freq, and poles passes now.

All four points on the motor bearings all have dominant 38X peaks. I'll have to go out here get some high resolution low FFT readings I guess...Some of these things are really becoming a challenge for this old Level 1, hehe...

quote:

Originally posted by Rockin' Rod:
Tom,

all of our routes are set up to take 8 averages. I could change that for only one I guess, and retake the readings. Is that what you mean?

quote:

Originally posted by Tom Murphy:
Hi Rod,
3g is a lot. I see that you are using a lot of averaging in your spectrum calculation. Do you have a non-averaged acceleration time signal we could take a look at?
Tom Murphy

Rod,

Honestly, I thought of this possibility right away after reading your original post. The clue - a single 38x peak. 3.17 g peak coverts to 0.08 ips peak at this frequency. This magnitude is feasible for some motors with eccentricity problems to produce. And of course vibration gets transmitted via common structural elements. It could be amplified by resonance in the blower. I suggest checking for resonance and trending 38x peak rather on the motor.

David

Me too Dave...When I got back in and started looking at that blower data, I couldn't help think it looked similar to a rotor bar freq. Can't imagine that being the case just after 2 years in service...Just took some high resolution motor data 3200 lines @ 15K...

Thanks,

quote:

Originally posted by David_G:
Rod,

Honestly, I thought of this possibility right away after reading your original post. The clue - a single 38x peak. 3.17 g peak coverts to 0.08 ips peak at this frequency. This magnitude is feasible for some motors with eccentricity problems to produce. And of course vibration gets transmitted via common structural elements. It could be amplified by resonance in the blower. I suggest checking for resonance and trending 38x peak rather on the motor.

David

Rod,

Sorry for thetypo. I meant of course rotor bar defect.
Could you please post motor vibration data and possibly a trend?

Dave

Rotor bar frequency in a 1250 Hp motor on a hoffman blower. This vibration was more of nusance because of the noise and after the motor was changed to the 1500 HP spare the rotor bar frequency disapeared.

Hoffman_Blower.doc (25 Kb, 28 downloads)

RPM=3584
Slip=14rpm
PPF=28rpm

Looking at the spectrums, I'm getting pole pass frq (28rpm)at running speed, and 2X Line freq 7200 with pole pass side bands, but the amplitude is pretty low. Same symtoms of an essentric rotor, variable air gap, but low amplitudes. All below .100 g's

I don't know if that would be sufficient enough amplitude to throw that blower that much out of whack, you think?

quote:

Originally posted by David_G:
Rod,

Sorry for thetypo. I meant of course rotor bar defect.
Could you please post motor vibration data and possibly a trend?

Dave

Here is the trend on the worse point on this 600hp motor...Not too bad

Dave did this create any vibration in the machine? Whats odd in this blower here is that all the velocity reading on both drive and driven are very low. This problem just seems to be affecting the ACC's...

Rod,

Thanks for the plots. As far as trend is concerned I actually meant waterfall diagram to watch 38x development.

Otherwise you may have symptoms of a rotor bar problem. I suggest taking a hi res low freq vibration spectrum and also a current spectrum to confirm.

Dave

Okay, I looked at those early this morning I get those here, hang on..I was curious if that peak had always been there, and I found it has..

Here is one of the waterfall plots going back about 6 months. I noticed these peaks go back even when this machine was new.

I did take high res data this morning as per the TA recommendation...3200 lines @ 15K FFT Those were there on the previous spectrum so I could ID the slip and Pole pass freq...

Is this a rotary lobe blower? If so, how many teeth are on the timing gears?

No, all our blowers are Hoffman multi-stage. I attached a photo of one we have here.