SMC-Flex indicating current imbalance

[ddmckee54]

We recently installed a new 845HP air compressor and are using a 1000HP SMC-Flex to start the motor. Under load the compressor will load the motor to a maximum of about 108% of the nameplate FLA of 900 amps. Apparently it's common practice for air compressor manufacturers to run the motors well into the Service Factor when under load.

What kicked over this particular anthill was that the soft start was indicating a current imbalance of about 140 amps between Phase A and Phase C. We immediately had the motor checked out by our local motor rebuilder and it was determined to be OK. We then monitored the line side of the soft start with a Dranetz and the load side with a Fluke, both meters were clamped around all of the parallel conductors. The meters agreed with each other to within a few amps, but there was a large discrepencay between what the meters were reading and what the soft start was displaying. Phases A&B in the soft start were reading high and Phase C was reading much lower.

Rockwell has since replaced both the primary CT's and the and secondary CT's, along with the soft start control module. Phases A&B are still reading high, although now they almost match in value - but Phase C is still reading about 100 amps lower than Phase A or B. About the only thing that we haven't changed in the metering circuit is the wiring harness.

We have several other compressors using SMC-Flex soft starts, so today just for fun I checked the Phase currents on some of them. We have two that are running 450HP compressors and both of these are showing the same characteristic, Phases A&B are about 50-60 amps higher than Phase C. These compressors are running at about half the current of the big one so the imbalance seems to be fairly proportional to the 845HP motor.

I'm wondering is it just me, or is anybody else out there seeing the same issue on their SMC-Flex soft start?

Don

 

[jraef]

Very important:
Does this phenomenon occur only when ramping, only when running, or both? No guessing, this has to be real. If nobody thought to check it while ramping, can it be re-done like that?

Reason: The SMC Flex uses 3 individual power poles, each with it's own single bypass contactor. If this is only happening while running, when the bypass contactors are closed, it indicates a potential problem with one of the bypass contactors. The fact that it is happening on the SAME PHASE on multiple starters might be the result of a very severe fault that took place on that phase while these compressors were running, which has damaged that contactor in all of the units. If so, a higher resistance in that one pole would increase the current in the other two. But if that were the case, the problem would NOT take place while the SMC Flex was still on the SCRs, i.e. during ramping.

The other more remote possibility is that the A/D converter, the device taking the AC current value from the CT transducer and making it available to the mP as a digital number, is defective and not providing the correct values. I've seen that happen on other digital soft starters (albeit never on the SMC), traced back to a bad batch of A/D converters from a very large mfr years ago. It would not likely be that same batch because that was before the SMC Flex even existed I think, but that just means it can happen. The problem with that is, why only and always Phase C? Common mode noise expressed on that phase maybe, so it only affects the A/D converter in that power pole module? Maybe. Can you connect a scope to look at the AC power quality on Phase C?


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[itsmoked]

You've never mentioned the voltages. How close are they?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[ddmckee54]

jraef: We've only got about 20-25 seconds while the motor is ramping up to speed, but the apparent imbbalance is there at that time too. We can, and have, switched the plugs on the control module to see if we can make the problem rotate phases - which it did. We did this at the request of the Rockwell engineers after we had replaced the secondary CT's and the control module, with only a minor improvement. That is what led us to believe that the primary CT's were to blame, the original primary CT's did have two different manufacturing lot numbers.

I don't know if it makes any difference, but we are starting the 845HP motor in the current limiting mode. It's limited to about 3600 amps, which is about all that particular sub-station's main breaker can handle. The 845HP air compressor is a three stage centrifical compressor. There is an RPM range where the compressor will develop mechanical harmonics so we need to get through that range as fast as possible. The other soft starts are on different sub-stations and are starting in the soft start mode with 30 second ramps.

itsmoked: We've got a 480 volt ungrounded system, most of it's been here since I was born and we're both starting to show our age. The phase to phase voltages on this sub-station are about 485-490 volts prior to starting the 845HP motor. They will drop to around 420-425 volts while the motor is starting, which we expected. When the motor is up to speed they will settle in at about 470-475 volts, all three phases are within a volt or two of each other.

We've got less than 25 hours on the 1000HP soft start's clock so I'm not really trying to figure out what the issue is, I'll leave that to the nice folk's at Rockwell. I'm really trying to find out if anybody else hase seen this issue.

Don

 

[jraef]

To that last point, no. I talked to the product manager this morning, he has heard about your issue and there are discussions about it at the factory, but this is not someting that has any internal history.

I don't think they knew this is an ungrounded delta system however. That might make a difference in that any MOVs used to protect the SCRs, if they are referenced to ground, may have failed. I'll have to check on that and post back later today.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[jraef]

Oh wait, you said that your Dranetz readings did NOT indicate a real phase current imbalance, the only issue here is just that the STARTER is showing one, and it doesn't agree with the external meters, correct? It can't be tripping on Phase Current Imbalance, because there is no trip fiction for that, only Voltage Imbalance. So this is basically a metering issue. Forget what I said about the MOVs etc., couldn't be that, I was thinking there really WAS a current imbalance. There is still reason for concern here, because the perceived current imbalance will skew the overload trip curve downward to help protect the motor against excessive heating from negative sequence currents caused by the imbalance, which might result in nuisance tripping. So until this is resolved, understand that this might happen and resist the temptation to turn up the OL trip values.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington