I have a 200Hp motor with a Westinghouse electronic reduced voltage starter. The initial thought was rotor bar failure but upon collecting data, I found 1xslip sidebands around turning speed and harmonics. (2 pole motor) The motor current signature shows 4Hz sidebands around line freq. with an amplitude of 30Amps while line amps are 198.
I suspect a problem in the starter electronics but have never identified this before. Any thoughts?

ckspkg02-08.doc (39 Kb, 129 downloads)

It looks like a bad rotor to me. Why do you think it's not?

I have not had this in my experience but I think in order to blame the soft starter you have to obtain a spectrum and TWF of VOLTAGE output for this soft starter via a step down isolation transformer with good ratio that will lower the voltage to proper values.

Variation at 4 HZ of the Output frequency will point out to the starter.

David

David,

"Variation at 4 HZ of the Output frequency will point out to the starter."

I've not heard this before, can you explain why 4Hz sidebands would indicate a soft starter issue?

Thanks

I think that David was alluding to the fact that we can investigate a current distcurbance by examining the upstream voltage.

If the disturbance is present in upstream voltage just as heavily as it is in the current, it often suggests the problem is with the upstream power supply. If the disturbance exists in the current but not in the voltage, it points to the downstream load. The concept applies for a variety of "disturbances", including unbalance, oscillation, and harmnonics. One complicating factor is that a load-induced current fluctuation can also cause small voltage fluctuation due to series source impedance.

We don't have any softstarters, but I was under the impression that soft starters went to bypass mode once the motor was started.


Was the data taken during steady state?
Do you know if the soft starter was bypassed.

If not a starter issue, you might also look for very strong load oscillation. Look for oscillations in fluid system pressures, flows, valves.assuming this is pump/fan.

I assume pole pass frequency is 1hz?
What is the motor speed?
What load does this motor drive?
Any belt or gear?

Hi
the problem may be a cracked rotor bars

Joe,
Broken rotor bars, from all I've ever seen, ALWAYS give poles*slip freq sidebanding in the vibration and motor current signatures with the current sidebanding amplitude being reduced as you move further away from line (typical sidebanding). This shows NO slip sidebanding and very strong 4Hz sidbands.
Pete,
This starter does not go into bypass from what I can determine. It was operating at steady state full load. The load was also varied and showed similar charateristics.

quote:

Originally posted by electricpete:
I think that David was alluding to the fact that we can investigate a current distcurbance by examining the upstream voltage.

That is exactly what I meant, although in this particular case likely it is not the power source but rather load dependant effect. SBs at 4 HZ in a 2 pole motor at the given speed could not result from a broken bar. In particular, as Pete mentioned, if the softstarter goes into bypass after run-up.

For the sake of investigation it will be also intersting to verify if there also is amperage frequency variation. This could be done from the the motor amp TWF. 40-50 cycles of motor amp TWF should be acquired with very high sampling rate and using a cursor examined
for frequency consistency.

David

David,
I did collect waveform data for motor current but did not post it. Would you believe it modulates amplutde at 4Hz. Surprise surprise right.

Nick,

In the motor amps TWF you posted there is clearly amplitude modulation at 4HZ ( the cause is unknown though ). Out of curiousity I suggested above to examine the TWF also for frequency variation ( modulation). In this case you need a better TWF resolution.

David

Nick,

I misread your first post and thought you said 2X slip sidebands (you said 1X)in the vib data. I've definitely seen broken rotor bar issues where the current data sidebands didn't land where you'd expect to see them. Take a look at the attached current plot. It's similar to yours but not exactly 4Hz sidebands. This was definitely a bad rotor. Sorry, I don't have the associated vibe data anymore.

Pete,

Thanks for the explanation. Since I grew up on the mechanical side, a lot of times the electrical stuff is still magic to me.

Rotor_Bar.doc (46 Kb, 38 downloads) Rotor bar

What is this motor driving? What is the final speed of the driven part?

quote:

Originally posted by Joe Gurney:
Take a look at the attached current plot. It's similar to yours but not exactly 4Hz sidebands. This was definitely a bad rotor.

Joe,

If you remember, what was found to be wrong in this motor? The SBs don't look to me like pole pass frequency and there are so many of them. I suspect it had rather frequency then amplitude modulation. It is possible that it is load related SBs as well.

Could you please post the TWF?

David

David,

I added the waveform to this attachment. I was called to this job because the motor was making wierd noises, and current was flucuating excessively. This was definitely broken rotor bars. The problems were visible when the rotor was pulled out of the stator. I saved this data for all these years because it puzzled me, and have asked many people with more experience to explain why the spacing was as it was. The general consensus has always been that this energy was NOT rotor bar related but after replacing the rotor the problems went away. Unfortunately, I was a contractor at that time and the customer didn't want to pay for follow up testing since the machine had settled down so I don't have any "after" data.

Rotor_Bar.doc (60 Kb, 33 downloads) Rotor bar data

Hello, every one!!
Nick,
After ElectricPete post you must thinking about new soft starter because it should go to bypass mode after motor starting.
Joe,
You added very interesting waveform. Did you remember nameplate of this motor, what is the motor speed?

Thanks and good hunting!!

Hello !
With great pleasure I reading post on this board and I notice that date of registration for all user included in this post, except me and eng.ahmad, is between 20 and 22. february. 2005 !!

I am SKF user and similar spectra I saw in SKF paper “A Summary of AC Induction Motor Monitoring “.I attached page 6 where are two picture of MCSA spectrum. Second picture is spectra of motor with 4 cut rotor bars and broken end ring with enormous sidebands ( 4.5 Hz ) .I asked guys from SKF about formula for prediction number of broken bars and explanation about big difference in sidebands between spectra on picture 2 and 3 with similar faults but I did not receive satisfying answers .

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

Motor_in_lab_.doc (126 Kb, 30 downloads)

This is usually for insulation break-down. Perform PI testing and/or megor depending on where you are and the rules there.

4 Hz sidebands do not sound like twice slip. What is the motor rated speed at full load? You will know for sure from there.

All you are seeing is a 4 Hz modulation, no guarantee that it is electrical. Suggest you look at the coastdown vibration data to see if the sidebands disappear immediately on de-energizing.

Another possibility - is the rotor of diecast aluminium? I have seen in some of those with voids that twice PPF (4 x slip) & its harmonics dominate. These could probably match the 4 Hz frequency.

Sam - I have never seen such sidebands prior to insulation breakdown. If you have some case studies, would really appreciate if you could post them here.

Regards,

Aditya

Panta,

In February, 2005, the former www.reliability-magazine.com forum was merged into this one and that is why so many of us registered during that period.

Nick,

While using vib equipment to see rotor problems, it is not the best tool. If you can get someone to take their PDMA tester to verify what you suspect, that is the way to go.

Regards,