Alan,

The fault had become pronounced after about 10 months of commissioning the Plant. The 300 Hz and multiples were present right from day one but, the amplitude trend began to escalate. When the Motor had tripped, 12 hours later at midnite, after diagnostic checks were completed on the drive controller and verifed OK; the amplitude weren't all that high either...if memory serves me right, it was below 3.5 mm/s Pk. The Motor "Hum" was quite audible at a distance (>150m); with other machines operating.

Cheers...Rajan Muthukrishnan

Rajan,

* Do you mean that no fault was found in the drive/control?
* Was 360HZ peak trending up consistently, significantly, and for how long, which ultimately prompted you to make a call?
* How long did it take to fail after the call?
* How did humming audible noise manifest itself in the vibration data?

I realize it took place long time ago, but may be still saved in the long term memory.

Thanks,
David

David,

* Not until the failure/trip of the Motor, 12 hrs prior a full blown diagnostics was done on the controller system and reported that all is working fine. I had to take the chap out to site and when he heard the "Hum"...he said "This is not normal but, my check does not reveal any fault".
* Over the course of 4 weeks, the overall level went from 1.2 mm/s Pk. to 3.5 mm/s pk and so did the 6X but, unsure at what amplitude.
*Exactly 4 weeks and after numerous reports and threats to Contractor; since the initial call.
* The only thing I remember, as a trigger, was the base of the 6X and multiples were broad and as the "Hum" got worse the broad base widened; like some form of energy impact...I suppose it could have been resonance.
Hope this helps.

Cheers...Rajan Muthukrishnan

Way back (circa 1998) while working for a manufacturing plant with hundreds of DC motors (including few 800 HP), I made a presentation at the SMRP on DC Motors that documents all of the problems we encountered using an integrative approach. It should be available through their website. What we found out was that:
1) 360 multiples although normal are indicative of bearing failure when they "grow" and should be used as a troubleshooting tool,
2) we never found 360 multiples related to SCR's problem or failure (against all vibration literature that points to this as the source of the problem),
3) the "normal" progression of the 360 multiples is bearing failure (aka flutting) typically not seen because very seldom one uses frequencies high enough to see them,
4) the humming noise is the most conclusive evidence that the motor bearing is in advanced failure mode....it could be days or hours. If you waited until this point, the damage could be extensive
5) while brushes and commutators could play a role on the vibration spectrum, we never saw anything to relate them to the FFT's. For commutator and brush problems, we used different technologies which provided better information than vibration
6) We installed shaft grounding devices but I left that plant shortly after so I couldn't really tell you their effect on the vibration readings....although I think it was a positive action to reduce bearing failures due to flutting.
7) when it comes to DC Motors, their failure is never linear, in other words, good today, gone tomorrow is a possibility you must keep in mind. I had two motors failing after being fully tested and analyzed the day before. Catastrophic failure does occur and you need to keep that in mind when you do analysis and ensure that you have all the documentation to support your analysis and conclusions and more important, to learn from that non linear failure.

Sergio Rossi

All the comments are valid and thoughtful. He may have called bearing defects based on the harmonics being higher than the 360 Hz peak. I think before calling a bearing defect, trending the waveform for increases in impacting energy would have confirmed a bearing defect. We have motors like this and that's how we evaluate them. The noise floor is busy, but I would need to know more. Pay attention to BSF harmonics which could indicate EDM fluting in a bearing race. Voltage leakage caused a bearing problem for us. Good Luck.

Kris,

Out of curiosity....What are the findings on the Motor that was removed from site for inspection based on bearing defect recommendation??

Thanks....Rajan Muthukrishnan

Very interesting discussion…

SR,

360Hz is not a “text book” example of drive problem. A six-pulse rectifier will produce “ripples” in the output current at exact this frequency. So this is electrically related but does not have to indicate drive problem. The pulsation in the current produce pulsation in the motor torque. Now the question is can it be seen in vibration data? People I spoke with told me that they observed 360Hz at a low level for many years and it was of no concern. You experience is very interesting. I have been thinking about it and somehow cannot relate bearing problems with 360Hz. But you have proved it…

Raj,
I was informed that the motor will be send to a different location. I have asked my people to follow up on this. I want to know the condition of the bearings. However, it may take some time. Once I know more I will post a follow up here…

Dear Kris
Definately, a concrete frequency component in the spectrum rarely indicates a bearing problem. Friction and other surface flaws generate components of random frequency and not steady amplitude, except when rolling bearing inherent frequencies immerge, which is usally shown in the envelope spectrum. additionally, it happens that the refered frequency is 6 times the line frequency (If it is 60Hz, is it?. Concequantely, it is showing rather to the direction of electrical source, not necessarily a problem. It is the ripple frecuency which sits on the rectified DC curve. If it has been for long without progressing, it is not to worry about.

Kris,

Thank you for your feedback. I do understand that a drive that uses SCR's will have a 6 "ripples" on its output as can be seen on an oscilloscope hooked up to the armature leads. These electrical pulses are seen as a 360 hz spike on the vibration signature (FFT). My 2nd point was intended to say that I have never seen a drive problem evidenced on the vibration signature as some literature from a very well known vibration training group states. I was trying to clear that issue among the other ones to give a wider perspective when troubleshooting DC motor problems.

Now, on my first point, I meant to said that we should look at the 360 hz as a troubleshooting aid in the event that it "grows". It may be time for your vibration expert to expand the frequency range and use a higher bandwith. What I have experienced is that there is a time based relationship between changes in the 360 hz and bearing problems. Bearing problems are bearing problems and they will evidenced as such (multiples of bearing natural frequencies). Also keep in mind that twice in my career, there was no indication whatsoever of bearing failures but they did occur and one of the bearing was totally destroyed....just the day after we tested it!

Sergio Rossi

quote:

Originally posted by SR:
...
6) We installed shaft grounding devices but I left that plant shortly after so I couldn't really tell you their effect on the vibration readings....although I think it was a positive action to reduce bearing failures due to flutting.
...
Sergio Rossi

I worked at a facility with a few DC motors. This frequency was very apparent on most of our DC drives. We didn't change the motor until we knew what the problem was. Even when the bearings were a problem (fluting), we struggled to determine what to do.

There are options... Install an insulating shield between the bearing and the shaft. Also, using a shaft grounding system. Many are available.

When we installed Shaft Ground Systems on our DC Motors, we noticed the SCR 360Hz issue come down dramaticaly. The reliability of the motor bearings also went up dramatically.

My solution is either replace equipment on your DC Drive or drive system to eliminate the SCR firing frequency, or install an SGS. Either way, if you eliminate the cause, the effect goes away.