Restarting a coasting motor 2

[TestBeforeTouch]

How long do you need to wait to restart an induction motor that is still coasting? I have a 1000HP 4160V motor for a large blower that will coast for 10+ minutes after it is shutdown. In this case, if the motor is off for more than 5 minutes the plant must go through a very lengthy startup process that they would like to avoid.

 

[itsmoked]

DOL?

Probably 10 seconds. Call it 30 if you can. You are waiting for the magnetic field to decay which should only be a few seconds.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[Skogsgurra]

Keith asks DOL? And for a very good reason. If you have an inverter on the motor, it may be able to do a flying start. Some inverters have a flying start that senses if the motor is running and also if the rotor has demagnetized or not, locks to actual motor speed and accelerates the motor in a controlled manner from there.

If you don't have it, it is a good idea not to wait too long. Restarting when the speed is high reduces heat build up in the rotor. If there is a volt meter on the motor side, you can check how long motor voltage stays when the motor is shut down. Or connect one to the hopefully existing VT.

Rotor flux usually decays in less than ten seconds. So I think that you can restart after that time.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Marke]

It is common for the rotor flux to decay in less than a second, so restarting in ten seconds should not be a problem.

Some soft starters using torque control should not be restarted on a rotating motor, but other starters are not a problem.
Some drives have a flying start and can restart on a spinning motor, but some do not and should not be used to start a rotating motor if you want to keep the smoke in.

Best regards,

Mark Empson

http://www.lmphotonics.com

 

[electricpete]

Agreed with the above comments.

Beyond a few seconds (or certainly a minute), there is not a concern for closing out of phase with the residual voltage since it's gone.

Whether the motor coasts to a stop or not is not a big difference. As far as sleeve bearing wear goes, better for them to restart while still rolling.

There is another issue to consider for the long term health of the motor. Given this is a large motor, imo you should not perform the process of restarting shortly after shutdown on a regular basis because the rotor doesn't have a chance to cool down.

The motor should be rated for a minimum one start hot or two starts cold per NEMA...so you are taking it to it's limit of one start hot. I wouldn't want to do that limiting cycle of one start hot more once or twice a year myself.

We just had a discussion for a large motor at our plant whether we could incorporate a certain sequence of shifting that require securing and restarting a particular pump on a regular every two weeks (for plant reasons that are too complicated to explain here). I decided if we were doing it this often, we should let the motor cool off before restarting. I looked at the stator winding temperature trend (I realize stator is not the same as rotor, but it's the closest thing I've got). It takes this motor about 16-24 hours to get to it's steady state temperature of about 20 degrees above ambient (space heater). I was surprised it took so long, but then I remembered the cooling is dramatically reduced once that motor stops rotating. I provided the feedback that we could do that on a regular basis as long as we waited a day before restarting. That advice wasn't particularly well received.


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

 

[Skogsgurra]

Right Pete!

That's exactly why I think that the restart should be done as soon as possible. If the speed hasn't dropped more than a few percent, then the rotor heating will be neglible. Letting the motor stop means that you will heat it a lot when starting again.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[waross]

If you have direct connected capacitors, the voltage decay may take much longer.
respectfully

 

[electricpete]

skogsgurra - you're absolutely right, sooner is better for this case.

Now that you mentioned it, I thought about my motor that I am trying to manage. If it had a long coastdown time like the one in this thread, it might meet our needs. Unfortunately, out motor when coupled to this pump coasts down in < 5 seconds - we happened to measure that for troubleshooting another problem on a sister pump last week, otherwise I never would have known.

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

 

[electricpete]

Gunnar - I vote you a star for pointing that out.

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

 

[asekhar]

For process industry applications, we normally specify that "the motor be suitable for restarting against residual voltage of 100% nominal voltage, 180 degrees out of phase with supply voltage". Incase the motor is specified like this, then there is no harm in restarting without any delay, if not it is safer to wait for few seconds to let the residual voltage die down. In any case a delay of 3-4 sec should be sufficient. Advantage of quick restart is that the motor will draw lesser starting current & that too for much shorter time because the rpm is already there.

Anand Sekhar
Power Systems Consultant

 

[TestBeforeTouch]

The motor is started across the line and there are no capacitors.
There are actually three of these motors in service and the problem is from time to time these motors blow a 9R fuse in the starter. Usually, but not always, the fuse blows when the motor is started. I talked about this in thread 237-178916. I just replaced the mechanical overloads and relays in each of the starters with a Multilin 469 so I could understand what is going on.

When I started up one of the machines for the first time after installing the 469 I got a single phasing trip with currents of A = 624A, B = 622A,C = 0A. The trip was 5 seconds after starting (the motor takes ~12 secs to get up to speed). The 469 trips on single phasing after two seconds so it looks like the C phase 9R fuse blew after 3 seconds. The other two motors started up with no problems this time.

I measured the resistance of the A and B fuses with a digital low resistance ohmmeter (I am not certain how good this is for determining the health of a fuse, but it was interesting.)The A phase fuse measured 1.8 milliohms, B phase fuse measured 4.4mohms, and a new fuse measured 1.8 mohms. Since the B phase fuse resistance was 2.5 times a new fuse I replaced the B phase fuse along with the blown C phase fuse. After this the motor starting up fine and is running now.

The problem is that 624 amps in 3 seconds should not have blown the 9R fuse. So I now suspect that the fuses are blowing or are severaly damaged because the operators have been restarting the motors too quickly. I have set the Multlin to prevent a restart for 2 minutes.

I appreciate your help and look forward to any other thoughts.

 

[jraef]

It appears as though your issues are thermal rather than based on residual voltage or regeneration, but since I already wrote this (I took too long in proofing etc.), I thought I'd pass it along for future reference even though it is now a bit off topic.

When designing DC injection brakes a few years ago, we had to know what amount of injection delay (dwell) time to build-in as a minimum so that users could not turn it down to a point that would allow injection to occur when the motor could still regenerate and destroy the brake module. We found this excellent article and partially based our design on his findings.
Dip-proofing Technologies Article Link
Original mfr's link

It should be noted (as we discovered later) that the author works for a company that makes products primarily for small motor loads. We had a few problems on the beta units on large HP LV motors, which we attributed to extra residual magnetism in the larger mass of steel, and/or additional interrupt times on air-gap contactors; some brands of large contactors can take upwards of 250ms to completely interrupt current flow. So we extended the delay times to 1 second on brakes over 100HP and had no more problems (from that anyway). if I'm not mistaken, IEEE Red Book states that motor fault contributions should be based on 2 seconds of regeneration, so I would say that any more than 2 seconds would be superfluous and may present that additional heating risk that Gunnar and Electricpete pointed out.