Hello to all;
We have a verticle mounted VFD motor and fan that is vibrating badly. We have increased the speed incrementally starting at 60%, 65%, 70% and 75%. One direction of the motor outboard end amplitude is 1.3 IPS and 90 Degrees from there it is .285 IPS. As the motor speed increase to 65% speed the vibration level drops and at 70% the amplitude continues to decrease. At the 75% speed the amplitude increase to about what it was at 60% 1.0 IPS. Is this an imbalance issue or resonance? The high directional amplitude and the amplitude decrease with speed could make one believe this to be a resonance issue. To remove this motor is extremely difficult therefore, we want to make the correct call. The help you folks provide is always extremely valuable and I certainly appreciate your input.
Thanks for your time and opinions.

Good day,
Charlie

It sounds like resonance although the two peaks close together is somewhat unusual (maybe two resonances?)

Some things to further evaluate resonance:
- Phase plots and magnitude plots. Phase incease of 90 degrees from low speed up to the magnitude peak would tend to confirm resonance.
- Bump test.
- Check if vibration can be decreased using temporary bracing (although bracing as a solution probably won't work unless you move the vibration out of the operating range).

Out of curiosity how low is your vibration at other speeds... let's say 50%. The lowest number you mentioned is 0.285 ips still seems a little high so even if you do conclude that you have resonant amplification, you may also have a high forcing function: unbalance, misalignment etc. Getting rid of whatever that condition is may also reduce the vibration.

Severe vibration and I'd assume the greatest magnitude is transverse to the flow rather than in-line w/piping.

Sounds like you have imbalance but it may be exciting resonance and probably is at these magnitudes.

Is there a specific frequency that increases and decreases just before and just after passing the greatest amplitude of vibration. How broad is the spike at its bottom at this frequency?

In addition; what is the alignment? I just did a vertical last week but not exactly like this one. Mine required aligning to correct.

I'd shut-down and do a bump test, check alignment (correct as necessary) and I do believe you'll be balancing. Once corrections are made the excitation force may not prove to be a resonance altering problem. But if problem still exists a dynamic absorber may provide a solution to keep you in service.

Do you have data you can present here? Please post so we'll know how to call the shot a little better. Thanks.

Depending on the structure of the duct work, you could have two closely spaced resonances.

Does this logic apply?

If you have round duct work, then if suspended and unattached you would have about the same resonant frequency in all directions. Then you attach it to supports and bends in the rest of the duct and you change the natural frequency because you change the stiffness. But because you change the stiffness in differing directions
you now have different natural frequencies in different directions. Also because you started
with a natural frequency that was basically equal in all directions, you will then have directional natural frequencies that could be very closely spaced together.

Does that sound logical?

quote:

To remove this motor is extremely difficult

Are we safe to assume that, from this quote, the fan is mounted directly to the motor shaft?

Did this problem start all at once or gradually appear or start after a replacement of motor/fan/bearing rebuild or repair?

If imbalance is the problem, can it be balanced at the speed with the highest amplitude in hopes at other speeds the amplitude will be lower also or is it difficult to access the fan for balancing?

Post of electricpete and Sam are very complete



Originaly posted by Danny Harvey

quote:

Also because you started
with a natural frequency that was basically equal in all directions, you will then have directional natural frequencies that could be very closely spaced together.

About the quote, phase angle change data will tell the true.

Charlie, if those are overall numbers what's the dominant frequency (1X, 2X, etc.)? What's it mounted on? What's the service? And I'm curious, too, if it's belt driven (maybe it's close to 1:1) or direct drive.

Need more info Charlie. But, if it's a direct mount to the fan shaft, phase angle across the coupling may indicate an offset alignment issue. Since you question imbalance, I'm guessing, and I think others may be also, that the amplitude is at running speed. Is this a newly commissioned piece of equipment? Can you take phase angle readings? More details....
PS. Those are pretty high amplitudes, so if you end up going into the equipment and solving the problem, please let us know what you find.

Charlie, I know I should not ask, :-) but have you for sure cleaned out the fan wheel from dirt?

Mr. Lindholm;
We are in the initial investigation, this equipment is in a manner that is not conducive to inspection. Basically this fan and motor will need to be removed. I am concerned that if this were an imbalance issue why did the amplitude decrease with an increase in speed. The duct work is verticle about 50 feet tall and round. The dominant peak is a one times running speed.
Thanks for your assistance.

Good day,
Charlie Bryant

Mr. Hand;
This is a direct drive motor to fan and the dominant frequency is a one times running speed. This component has been in operation for about three years and the information I am getting is that there are certain speeds that this fan has not been able to operate. We plan to take additional phase readings to confirm previous data initially the phase at the outboard end compared to the inboard was 145 deg. out of phase.
Thanks for your assistanc.

Good day,
Charlie Bryant

The symtoms sound like resonance as you suspect. The phase angle shift from outboard to inboard could indicate resonance or couple unbalance. Is the fan rotor attached directly to the motor shaft or is there a coupling?

Similar problem we also faced but in a horizontal mounted simply supported fan. The outboard bearing was showing the similar pattern as this case ( vib reducing then increasing as the rpm is increased ). Dominant freq all the times was IX only. On further analysis/interrogation two problems were found : Rotor unbalance & outboard bearing housing cracked in the plane parallel to the shaft. Housing was replaced and fan was balanced. Now it is running fine.
Good Luck

Anurag

I am just quoting a similar situation which we experienced in a vertical pump. The vibration reading (velocity) in one direction at the NDE bearing of the Motor was around 14mm/s(Peak) where the 90degree from there it was around 5mm/s. It shows an unbalanced condition and we changed the Motor. But the newly installed Motor also showed the same type of signature. May be it was a resonance condition. But we have to check whether the foundation is in level so that the Motor is fixed properly.

Charles
This a ressonance, with velocity increase the vertical machine reach the frist mode of vibration and just above, the second mode - their are ortoganals. If you put 80 and 85% the vibration level will drop and late will manifest the third natural frequency with the third mode of vibration.
The your solution is modify the support struture.

Ricardo Góz from Brazil

Thanks to all who responded to my inquiry, you have been very helpful. It will be a few weeks before we can remove this component and make final determination.
Good day, to all.

Charlie B.

It is not unusual to see vibration that is "directionally sensitive". This is usually a matter of differing stiffness in different directions, but not necessarily a resonance issue as is often thought. Sometimes a machine shakes much more in one direction than another "because it can"...

That said, your case does sound like resonance because of the large changes in amplitude with relatively small changes in speed. Since this is a VFD, one of the best ways to really understand is to pick the point with the highest average amplitude and measure vibration and phase at that point as you slowly bring the speed up. This can be done simply in "monitor" mode or by using your balancing software to take "snapshots" every 50 - 100 rpm. Just write down the amplitude and phase readings. Then do the same thing as you decrease the speed, taking readings at the same speeds if possible. Then plot out the results of both amplitude and phase vs. speed. This will give you a great deal of information about the resonances present.