Current imbalance persisting.

[itsmoked]

Remember the Hydrualic Press...

The 2HP cushion pump is still giving me fits. Today I went back and ran the cushion measuring everything.

A-B 492.5V
B-C 493V
C-A 494V

The electronic overload was tripping in about a minute.
Current:
A 1.4A
B 1.2A
C 1.8A

This represents a nasty 33% mismatch.

I rolled the three phases.
Current:
A 1.2A
B 1.5A
C 1.3A
20% mismatch

I rolled the three phases again.
Current:
A 1.5A
B 1.3A
C 1.15A
24% mismatch

I went back to the first roll. It ran for 15 minutes and tripped.

This is a 2.8A motor barely loaded.

I checked the voltage drop from in front of the fuses to the three motor leads leaving the panel. They all had a 1.0V drop. Fuse contact, fuse, fuse contact, contactor, into terminal block, out of terminal block. (1.0V across the board)

In all cases the high current motor lead remained the high current leg.

I think I am down to the motor and the motor connections. I was told the motor's nine leads were hooked up correctly and all the wire-nuts(groan) were tight. I told them lets try re-terminating the motor connections with compression connectors. If that doesn't improve things we need to yank the motor.

Do you guys concur?

If this motor is taken to a motor shop, (it is still under warranty being about 2 months old and never having run for long), can a typical motor shop check for current balance?


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[electricpete]

fwiw, here are my my thoughts (right or wrong... my job is not to select unbalance protection setpoints, so I may be missing something).

We have less that 25% current unbalance using the NEMA definition.

The NEMA derating curve gives a derating by voltage unbalance.

Current unbalance is 5 – 10 times voltage unbalance.

The edge of the curve is 5% voltage unbalance which corresponds to 25 – 50% current unbalance. That certainly envelopes our case.

The derating factor is about 75%.

Apply the derating factor to nameplate current 2.8 amps. 2.8*0.75=2.1

The highest current should not exceed 2.1A on a continuous basis. I can see no reason to trip at 1.8 amps.

This would be a very conservative approach as far as I know (for example, if we assumed current unbalance 5x voltage unbalance, we could have selected 2.5% volt unbalance and 92% derating). I think the unbalance is much less a concern for small motors than for large motors (where the rotor is the concern). The fact that the cause of the current unbalance is something other than voltage unbalance doesn't change the thermal conclusions as far as I know.

Note again that as you increase the load, you can expect the current unbalance to decrease so if the trip wasn't so sensitive you may well be able to operate this motor all the way up full load.


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

 

[flexoprinting]

Thank you Gentlemen, itsmoked thanks for this post I am watching with interest for a conclusion.

 

[edison123]

Keith

"Six inches from this unit is an identical OVLD relay running a 3/4HP fan using three wire loops. We've never had any problem with that one. It has NEVER tripped."

Why don't you swap the damn relays and see whether the problem travels with the relay or stays with the motor ?

 

[electricpete]

In the event anyone is trying to decipher my illogical parenthetical statement, I should correct it. Should have been: "for example, if we assumed current unbalance 10x voltage unbalance, we could have selected 2.5% volt unbalance and 92% derating"

And just to elaborate the two cases, since there is a range current unbalance = 5-10x volt unbalance, if we pick 5x we convert 25% current unbalance into 5% volt unbalance (derate to 75%), while if we pick 10x, we convert 25% current unbalance into 2.5% volt unbalance (derate to 92%)

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

 

[itsmoked]

electricpete; Yes I was trying desperately to decipher your, "illogical parenthetical statement". LOL.

But thru it all I was rapidly coming to the conclusion that the SS OVLD relay may be defective and that I should just swap the damn thing with the adjacent one, even though that is a royal pain in the butt.

Agreed edison123 a splendid idea!

flexoprinting; I'll stick it out and certainly make sure you guys hear the ultimate solution.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[LionelHutz]

Well, If you want to follow MG-1 then the derating is given by voltage unbalance. A derating only comes into effect above about 0.8% voltage unbalance which you do not have.

I expect its just an average cheaply built small motor. Almost all motors show a fair amount of unbalanced currents at low loads so what you are reporting is not odd to me.

Another interpretation of MG-1 used is to take the largest difference in phase currents and divide by the rated current to get the imbalance.

Check the connections for fun but you will most likely have to change the motor to make that overload work. The simplest is to just change the overload to one that does work.

 

[itsmoked]

Thanks guys! This is one of those annoying ones. I called the plant guy today because they were going to redo the connections this morning and then yank the motor if the current stayed greatly imbalanced.

Today's response: "Ah.. They said they needed the parts run so I told the operator to go try it. He ran the full 4 hour job and it never tripped once."

So I will order a new SS Overlord ~$120 to have on hand for that panicky call I'll get in the near future.

Can you run a 2HP motor safely with just delay fuses? Seems a bit big for that. Type

http://www.bussmann.com/pdf/1001.pdf

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[waross]

Hi Keith;
Back in the 70's the preferred protection in a lot of plants was the melting alloy type of overload relay. No differential protection at all. Many industries ran on these without differential protection and had years of dependable service.
In the event of a complete phase loss the current would usually be enough to trip off.
The calibration of the bi-metal type of overload was not very good and it was not uncommon for motors to fail even though the protection was nominally correct.
One of the fuse manufacturers was promoting dual element fuses for motor protection.
I used to have a sizing chart for dual element fuses for the protection of motors.
I believe you can still meet code with fuse protected motors.
Respectfully

 

[electricpete]

Just to follow up on LionelHutz' comment. Even though the NEMA derating curve applies for voltage unbalance vs current unbalance, I believe the NEMA curve has relevance to this discussion for two reasons:

1 - A smart current-sensing relay providing protection for unbalanced currents will presumably implement derating roughly equivalent to that given by NEMA MG-1. It doesn't know the difference between current unbalance caused by the motor or the power system. So bringing the NEMA curve into the discussion gives us a rough idea when we might expect the relay to trip (regardless of the source of the unbalance).
2 - If we were not concernred about the relay characteristics, but only protecting the motor, we would want to understand the thermal effects of the current unbalance. I believe the thermal effects are in generall not too tremendously different whehter the current unbalance is caused by a power sytem voltage unbalance or may a magnetic circuit unbalacne.

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

 

[LionelHutz]

I looked at MG1 1998 section 14.36 and it covers derating due to voltage imbalance. I didn't see anything specific about current imbalance in MG1 but it's a big publication and I didn't spend too much time looking.

So, it seems to me that an overload using current imbalance as part of it's algorithm has had the current imbalance part derived from reading MG1 and converting the section about voltage imbalance to watch current imbalances instead (using the typical 6-10x current imbalance multiplier). This is mostly done because the overload is not built to read the phase voltages but also to provide some form of unbalance protection for the motor. However, this method is not spelled out in MG1 so it's the manufacturer that has decided how to do it.

If I missed something then let me know the section because I'd like to read it.

I guess what I'm saying is that if the voltages are balanced then you can theoretically just set the overload to work with no concern for current imbalances. The motor manufacturer could theoretically build their motors so they run at rated load with a 20% current imbalance but still meets MG1 operational specs. And, with the cheaply built small motors these days, they probably do. Watching current imbalance is probably a good idea for a big expensive motor but may not be for a cheap little 2hp motor.

In the end, the motor is running under rated load and all of the phase currents are below the rated current so it is unlikely the motor is going to be damaged by the current imbalance being read.

 

[electricpete]

Lionel - I agree 100% with everything you said. I also don't think it contradicts anything I said.

Before I thought perhaps you were disagreeing with the way I had referenced MG-1 voltage unbalance derating with the 5-10x conversion but apparenlty you didn't.

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