Ground fault trip on motor starting

[msb10]

One of my clients has a new installation where an MCC feeder circuit breaker is tripping when a 250 HP vacuum compressor starts. The trip unit indicates that the breaker has tripped on ground fault. The trip unit is a GE MVT+, and is set to trip at 200 amps with a .1s delay (GE Spectra SKH8, 800A frame, 700A sensor, with GFPU of 0.25, I^2t out, delay band 1).

They tell me that they can reset the breaker and try again, and it typically starts. They indicate that it always trips the first time and sometimes on their second attempt, but that they consistently get it started by the third, and that it always trips on ground fault.

I have questions for the facility that I am trying to resolve now, but my question for the board is, what is an expected value of ground fault current during motor starting? Does anyone have any experience or maybe a reference that they can point me to?

Thanks,

Matt

 

[davidbeach]

The current sensors used by the breaker saturate during start up; the trip unit sees more unbalance than really exists. One thing to try is to turn the I²T on, that will give you a bit more time for the sensors to recover from saturation without tripping.

 

[dpc]

There's probably no actual ground current - it's just error in the current sensors as David Beach stated already.

I would try increasing the pickup setting of the GF trip. The delay could also be increased, but this will likely cause coordination problems.

The ground fault trip works by adding the three phase currents together. During starting, the motor inrush (dc offset) will not be the same in all three phases. And the current sensors can saturate as David mentioned.

 

[msb10]

I'll ask that they try that (increasing the pickup as well as turning on the I^2t).

I appreciate the input.

Matt

 

[electricpete]

I am not familiar with that style of protection. I would definitely defer to the comments of others others that adjustment of the settings is the first logical step.

Brainstorming what else might be in play in absence of setting problem:
1 - capacitors connected to ground in an unbalanced configuration.
2 - A little further out in the realm of unlikely stuff - an actual fault might possibly be burning moisture away from a faulting area after the first start?

In absence of some weird explanation, to what do we attribute the increased probability of successful starting on the 2nd and 3rd starts? If in fact it is saturation of the CT's causing uneven sensed current, what changes? A trip during a failed start attempt interrupts a larger current (LRC vs running current) and might possibly leave the CT in a different state than if it had been secured normally ? But if anything, I'd have thought the CT would be more likely to have higher residual magnetism (leading to unevern CT saturation) following a failed start than following a normal stop. Any thoughts on why this behvior?

By the way, what voltage?


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

 

[msb10]

As it happens, I'm going out there in the morning to look at this. My first position was to be in the "there's something wrong" camp, but I can accept that current sensor saturation may be causing a false trip. Having said that, I'm going out there to see what is really going on. I'm somewhat suspicious about the "always holds in on the second or third attempt" business. Maybe just the planets are lined up right by the third attempt. I'll report back.

This is a 480V, 250 HP motor, started across-the-line. Seems a little big to start DOL, but this is what they've got.

Thanks for everyones input.

Regards,

Matt

 

[RRaghunath]

480V, 250HP motor is likely to have flc of ~225A. An earth fault pu of 200A is pretty high for this rating motor to trip, that too when there is time delay of 0.1sec associated with the unit.
We typically provide 5-15% of flc, 0.1sec for motor earth fault protection when it is conncted through CBCT and 15-30% if it is through residual connection of phase CTs. The protection is always stable during start.
I think there is need to investigate further, including the integrity of the earth fault protection unit in the breaker.

 

[msb10]

We made the changes I described earlier (increase trip point, add I^2t), and the motor is starting without incident. We also measured current on the equipment ground at less than 2 amps, for what that's worth, with the motor running.

Thanks for everyones input.

Matt