3phase induction motor vibrating at stator slot passing freq?

[ituanra]

We've got a fan 3 phase induction motor generating annoying noise at different frequencies as we increase voltage. In a 3D Spectrum with noise power level in dB(A) (colors) against frequency (x axis) and rotor rpm (y axis) we clearly distinguish:

a)blade passing frequency and harmonics up to the 5th as lines freq=k*rpm/60 , k=1,2...5
b)vertical line at some 300Hz which may be 3*(supply freq) or a ressonance.

Until here, no real mistery. Now what bothers is:
c)a line freq=(number of stator slots)*rpm/60 which is not really noisy.
d)two lines that appear to be the preceding one modulated with a frequency that matches freq=(nº of poles)x(slip frequency)
One of this last two lines is on the left and the other on the right of c) and all three seem to meet at synchronism (slip frequency=0) with freq=(number of stator slots)*(rpm syncronism)/60
The line on the left is the noisiest. This line cuts the x axis at -100Hz and has an expression freq=A*rpm/60-100 where A happens to be close to the number of rotor slots (which confuses me even more).

(everything said in c) and d) are just guesses)

Other considerations:
-excentricities rotor/stator have been measured rigorously and look alright.
-Rotor bar condition has been checked several times and appears to be in good shape.

I couldn't find anything about stator slot frequency excitation, not to mention beeing modulated. Could that do any sense? Or else, am I interpreting things wrong and seeing just coincidences?

I can send spectrum if anybody is interested

Thanks in advance for any help provided

 

[ituanra]

...and by the way, this is an external rotor motor.

thanks

 

[electricpete]

300hz is often a frequency produced by electronics powered from 3-phase rectifier.

Your sidebands definitely sound like pole pass frequency, particularly since they converge at low load.

I personally don't think stator slot pass frequency is a problem in itself. But the fact you have pole-pass sidebands would imply some form of rotor assymetry. It sounds like you looked at the right kinds of things to investigate that. Did you also check for possible excess clearance or bow which might create dynamic eccentricity. Also certainly pole pass frequency can be created by porosity of cast aluminum rotor bars and rings.

Out of curiosity
- What does it mean to use "rotor" as an axis for your "3d spectrum"? I've never heard of that.
- What did you check to rule out rotor bar problems? Visual inspection may not be enough. It can be challenging to find. Single phase test is one of the best tests. Although it is relatively rare to have rotor bar problems.
- What is the horsepower and speed of your motor? Does the load start frequently? How long to come up to speed? (high horsepower, high speed, frequent starting, weak power system and your high inertia fan are all risk factors for rotor bar problems).
- How high is the stator slot pass frequency peak and the sidebands?
- Does the magnitude of the sidebands vary with load?
- what is "an external rotor motor." Do you mean there is an external shaft-driven fan under a shroud (common TEFC design)?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[ituanra]

Thanks electricpete for your interest. Here I try to answer your questions.

- What does it mean to use "rotor" as an axis for your "3d spectrum"? I've never heard of that.

In my y axis I can read the rotor RPM so each horizontal line is a spectrum for a given velocity (from 150rpm to 900rpm). The magnitude of the sound level at given rpm and frequency is given by a color. I can send you a picture if you want.


- What did you check to rule out rotor bar problems? Visual inspection may not be enough. It can be challenging to find. Single phase test is one of the best tests. Although it is relatively rare to have rotor bar problems.

We cut the one of the shortening rings of a rotor and we measured all the rotor bar resistances one by one. Some of them where up to 15% error so we thought that could be the problem. We then built a special rotor with copper bars welded to the shortening rings (also copper made) instead of the cast aluminium. We assumed those were perfect rotor bars. We got the same spectrum, same noises.

- What is the horsepower and speed of your motor? Does the load start frequently? How long to come up to speed? (high horsepower, high speed, frequent starting, weak power system and your high inertia fan are all risk factors for rotor bar problems).

2KW, 900rpm, I don't know how long it takes to accelerate, very quick. It's a small machine so I don't think inertia is a big deal here.

- How high is the stator slot pass frequency peak and the sidebands?
I'll tell you soon

- Does the magnitude of the sidebands vary with load? It's difficult to say since load means in our case blades. The blades amplify the noises generated by the motor. But I'll try to find it out.

- what is "an external rotor motor." Do you mean there is an external shaft-driven fan under a shroud (common TEFC design)?

in an "external rotor motor" the rotor surrounds the stator. i.e. the stator is inside te rotor. It is a convenient disposition for fans since the blades or the impeller can be mounted right on the rotor. I found a couple of examples on the net:

http://lion-ball-motor.en.alibaba.com/group/50043281/External_Rotor_Motors.html

 

[vanstoja]

I am totally confused by your data indicating that an induction motor running with a 100 Hz line frequency has a rated speed of 900 RPM but also operates at speeds down to 150 RPM. A 900 RPM synchronous speed rated induction motor should be operating at 60 Hz line frequency with 8 magnetic poles. Stepdown to 150 RPM would involve line frequency reduction to 10 Hz with a proportional voltage reduction to maintain a near-constant volts per Hz ratio to limit overheating.
After diligent searching for a plausible cause of electromagnetic stator slot passing (better called "loading", SSL) vibration in the open literature, I found nothing to believe and so consider such a motor noise source to be highly improbable. From actual experience, however, I know that aerodynamic or hydrodynamic SSL is indeed possible if stator slot closure wedges or coil tops are recessed below the magnetic iron core inner diameter so as to produce pressure pulsations in the airflow or waterflow through the motor's magnetic gap. Higher SSL vibration levels will occur with more viscous waterflow than with airflow in the gap. Recession depths need only be of the order of 0.002 inches per inch of stator inner diameter to create appreciable SSL vibration peaks in water-cooled motors. If the motor rotor is translating radially or whirling while rotating, then magnetic gapflow pressure perturbations become amplitude modulated and many rotational (1X) sidebands may straddle each SSL harmonic peak.

 

[ituanra]

My motor runs at:
line freq: 50Hz
rpm with load: 880rpm
synchronous speed: 1000rpm
I regulate speed through voltage regulation under load.
I never said otherwise.

I think the stator slot passing freq is a fake path. I think I should ask first:

- How do I diagnose my motor problem through vibration, sound level or current analysis?
- All the information I found about this regards internal rotor motors. Does it all apply for an external rotor motor?

thanks