Motor life reduction when exceeding limits

[cokeguy]

This has been discussed a few times, but now it is happening to me and would like to hear opinions and (especially) experiences. We have a 460 VAC 250 HP non-inverter grade 1.15 SF motor driven by a VFD. We know SF should be assumed as 1.0 when using VFDs, especially with non-VFD grade motors, but right now we are right at the motor´s limits, and probably 10% of the time amps are above the motor´s 291 A limit, closer to 320 amps. It hasn´t tripped because we raised OL limits at the VFD to 320 amps.

The manufacturer obviously and understandably doesn´t endorse this situation and will not make any guarantees or service life reduction estimates. My question is, anybody out there has run a motor this size under these conditions for a year or more? Or else, has anybody actually correlated and corroborated premature motor failure under this conditions? I normally don´t push a motor too hard, but in this case changing the motor is not an option because we are still on the experimental stage of a process change, so we would prefer to push it over its limits (risking premature failure) if there is a real chance that we can get away with it at least for a few months. However, if there is consensus that driving a motor this hard will almost certainly cause failure within a few weeks or days, then we will look at other options. This motor in particular comes from a very reputable manufacturer, Emerson-US Motors, so I would expect them to be very conservative. Thanks a lot for any advise or comments or experiences.

 

[edison123]

Does your motor have any winding temp RTD's to show the actual operating temp at the overload ? The winding life is dependent on the operating temp and if yours is within the limits (120 deg C for class F machines)at the overload, then you can probably run the motor continuously.


* Basically, I would like a full-time job on part-time basis *

 

[cokeguy]

No, we don´t have any means to measure winding temp, but it "feels" Ok externally, we constantly monitor temperatures "by hand" and as long as it doesn´t become too warm we don´t even bother to measure the external temperature periodically.

By the way, we are also overspeeding the motor by 10%-20%, close to 75 Hz and 2000+ RPMs instead of the nominal 60 Hz and 1780 RPMs. Therefore, we have been toying with the idea of reducing the driven pulley diameter in order to operate the motor closer to its nominal speed at full power. However, could overspeeding the motor actually help us in this case because of the increased airflow from the motor´s fan?

 

[dpc]

The continuous higher speed (and temperature) operation could also cause problems with motor bearings, not just the windings.

As edison says, the main concern with the windings is temperature, not amps. The old rule of thumb is that every 10 deg C increase in temperature reduces the insulation life by 50%.

 

[ozmosis]

What is your ambient temperature surrounding the motor? The higher the ambient, the less headroom you will have to play with on motor limiting temperature.
The fact you have a 'non-inverter grade' motor seems to me to be asking for trouble to be honest. Is this Class B insulation? If so then the temperature limit is lower (100~110degC) and again, subject to ambient temps. Plus your motor will have the additional heating effects and higher peak-peak voltage (dv/dt) from your PWM(assuming) VFD.
If it is the cost of motor you are worried about then first look at the process you are running and evaluate the 'down-time' question: "If the motor failed and the cost of down-time to the complete process is > that of purchasing a larger motor" then you have your answer I would say. Another factor to consider is that if your motor did fail due to insulation breakdown, then there is no guarantee that your VFD might not come away unscathed. So you could potentially have even greater costs.

 

[jraef]

cokeguy,
You have several areas of concern, but edison123's comment still reigns supreme. Since you don't have a way of monitoring heat directly, I would look into an infrared scanner that can be used on both the motor and the bearings. Just because you don't feel it on the external cooling fins of the motor does not necessarilly mean that the windings inside are not overheating. VFDs tend to find the weakest spot in the windings, so your hand touch methodology is reliant on finding that spot. An infrared scanner can find the hottest spot more easilly, and then you can monitor that spot on a regular basis.

You also don't mention the distance from the VFD to the motor. If under 25 linear feet, probably not too much of a concern, but if over that, make sure you have output filtering on that VFD or you will damage the motor windings from high voltage standing waves (lots of papers and articles published on that). Even that 25ft rule is debateable in some circles!

Bearing races can also be a problem because ground currents generated by the PWM signal through the stator to the rotor flow from the rotor through the shaft to the frame can cause pitting and somewhat rapid failure.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[petronila]

Hello Cokeguy,
At operating from a VFD and the increased current, both will contribute to higher winding temperatures. For your conditions you will expect the winding temperature to increase approximately by 25 up to 30 degrees C. That reduces in theory the winding life by half for every increment of 10 degrees C. That minds the expected motor´s winding life of 10 years will decrease to approximately 2 to 1.25 years by the combination of increased load and the VFD.

In order to Avoid early bearing failure you could use a High Temperature Grease (220 C).

Regards,

petronila

 

[DickDV]

By all means, continue to operate the motor in overspeed. That alone is probably responsible for the temperature control you have. Motor life will be longer, not shorter for running above 60Hz.

There is, of course, a limit but, I would say, up to 80Hz, you are doing the motor a favor.

 

[edison123]

cokeguy,

The touchy-feely stuff with the body temp will in no way indicate actual winding temp. I have seen motors with winding RTD's indicating 130 deg C while body temp was measured with contact and non-contact meters at 60 deg max.

Yours is probably random-wound, impregnated stator with no possibility of installing RTD's in slots unless you go for a rewind. I would suggest install winding RTD's in winding overhangs to monitor the actual winding temp.

As DickDV says, operating at higher speed will benefit the cooling. Just make sure your bearings are rated to run at the higher speed (see skf website).



* Basically, I would like a full-time job on part-time basis *

 

[waross]

Another factor to consider when changing pulleys is the added torque required if the motor has to produce the same power at less RPM. More torque means more current.
Until you run out of voltage, changing the pulley to increase the speed will be more beneficial than slowing the speed. You may be able to boost the input voltage a few percent to help the Volts/Hertz at the higher speed. Check the specs to see if your VFD input voltage is at the maximum allowable.
respectfully

 

[waross]

A reality check on the bearings. I got lost on the Emerson web site, but Baldor lists bearing part numbers by frame size, not speed. At 3600 RPM the bearings should be serviced twice as often as at 1800 RPM. At 6000 RPM and above, contact Baldor for bearing service intervals and information. 2000 RPM doesn't look like a problem.
respectfully

 

[Marke]

Hello cokeguy

As has been expressed by others, one of the issues is the operating temperature of the insulation. If your ambient is b elow the rated ambient of the motor, you will have more head room. Operating in overspeed will increase the cooling of the motor so for an equal current input, you can reasonably expect a lower temperature rise provided that the voltage waveform is sinusoidal. The problem with operating on a VSD above rated speed, is that the output voltage of the VSD is limited by the input voltage and so the output voltage waveform applied to the motor will be flat topping, introducing significant 3rd and 5th harmonics which will add to the motor (and particularly the rotor) heating.
At the end of the day, it is a complex question to answer. I would suggest to monitor the temperature and compare the surface temperature at rated conditions and see how much of an increase you get at the elevated condition.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[DickDV]

Marke, I don't believe that you will get "flat topping" on the output of a PWM drive. You will get a properly formed sine wave in the PWM pattern but the frequency will be above 60Hz. At least, that's how ABB's drive output behaves.

 

[cokeguy]

Thanks all for your input, especially the comments about overspeeding, obviously with the bearing constraints in consideration. We will hold on the pulley change until further assesment. On the other hand, we´ve been told to use reactors on the drive´s output to smooth out the output waveform (this is a saftronics GP10 drive, I suppose it is PWM), and that way we could run safely up to the 1.15 service factor amp levels. Is that really a good option, or are there any caveats to it? Thanks again everyone...

 

[jraef]

An output reactor is never (or at least rarely) a bad thing on a drive, it has several tangible benefits, one of which is to reduce the likelihood of motor winding damage.

The '...safely up to the 1.15 service factor amps levels" is, however, entirely debatable. Even without a VFD, the meaning of Service Factor operation of motors is in and of itself debatable, and has been discussed here several times. At best, you may be able do it continuously, but even NEMA states that you can expect shorter motor life when running above 1.0SF. Add to that the fact that the power will be coming from a VFD, with or without a reactor, and you can pretty much count on a shorter life. How much shorter is going to be really tough to determine.

Eng-Tips: Help for your job, not for your homework Read faq731-376

 

[electricpete]

I don't think there is a big hazard to the bearings from overspeeding in this case. Especially if you monitor the bearing housing temperature and are below perhaps 150F.

If it were a 2-pole motor in this size range with greased bearings I would expect hot bearing temperatures even at base speed and aggravated by overspeed.

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

 

[electricpete]

Forgot to say I agree with those that recommend to continue overspeed based on slight winding cooling improvement.

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