Ground Fault, or NOT?

[tankfarms]

Recently we have a 500HP, 4160V, 3-PH motor (drives a water pump) that keeps tripping "Ground Fault" via a digital protection relay on motor start-up. Below are some parameters first (also see drawing illustration):

Ground Fault CT type: zero-flux CT (around 3 phase conductors)
CT Ratio: 2000:1
Ground Fault trip at: 2 amps primary
Ground Fault trip delay: 1 second
Motor FLA: 60amps (500HP, 4160V)

Electricians observed that motor was able to start and come over the inrush for the first few seconds--motor amp pegged out first and then dropped to around 15 amps on each phase, then after 3 seconds or so, motor tripped on "Ground Fault" through the digital relay.

This is a fairly old type digital relay, so I'm not able to see exactly how much the "ground fault" current was.
Motor was megged from the starter end with 2.5G ohms on each phase to ground.

Questionable point to me is that how could the motor amps get down so low?. Also, would this ground fault CT configuration also trip on imbalance? The CT is vectorly summing the three-phase current, so any imbalance would cause the summation to a value other than zero?

At this point, I'm not sure this is truly a motor issue. Pump/motor was coupled up with another steam drive (dual drive set-up); however this problem occured when the steam drive is uncoupled, so just motor driving the pump.

Any ideas fellas?

 

[jim dungar]

Something doesn't seem right.
Your really cannot accurately measure ground fault with that set up. 2000:1 with a 2A trip point means you would need to see a 4000A ground fault, which is unlikely on a motor that only draws 60A at full load.

Digital relays have been known to nuisance trip at levels of <10% FLA.

 

[tankfarms]

Something doesn't seem right.
Your really cannot accurately measure ground fault with that set up. 2000:1 with a 2A trip point means you would need to see a 4000A ground fault, which is unlikely on a motor that only draws 60A at full load.

Digital relays have been known to nuisance trip at levels of <10% FLA.

Hello, thanks. The 2 amp is the PRIMARY current, so the relay would see 1/1000 or 1mA of ground fault current.

 

[jim dungar]

Hello, thanks. The 2 amp is the PRIMARY current, so the relay would see 1/1000 or 1mA of ground fault current.

Still not good, most CT's do not 'work well' below 1% of full load.

Why are you trying to trip on mA of fault current? My starting point might be about 5A on a 60A motor. The purpose of this relay is to minimize equipment damage as the motor fails, it does not provide people protection.

 

[tankfarms]

Still not good, most CT's do not 'work well' below 1% of full load.

Why are you trying to trip on mA of fault current? My starting point might be about 5A on a 60A motor. The purpose of this relay is to minimize equipment damage as the motor fails, it does not provide people protection.

Let me provide a bit more background information which I should have included earlier. Our system is a a 4160V wye with neutral resistor-grounded. The resistor limits the ground fault to 5 amps, and 2amps is the ground fault trip level that we have been using in the past.

 

[jim dungar]

Let me provide a bit more background information which I should have included earlier. Our system is a a 4160V wye with neutral resistor-grounded. The resistor limits the ground fault to 5 amps, and 2amps is the ground fault trip level that we have been using in the past.

I would have considered using a 50:5 zero-sequence CT. It is not uncommon for me to use a 5:5 ratio on the grounding circuit.

But still, I think your relay is reacting to noise.

 

[brian john]

The CT ratio as noted is in my opinion way to high.

Several thiings I would do.

1. Test the GFPE relay and CT.
2. Megger all components of the motor circuit.
3. Use a properly rated PQ analyzer and capture current at time of trip.
4. Utilize a CT more in the range of the current values desired. Several manufactures have CT and relays designed as a package for low lever ground fault operation.

 

[tankfarms]

interesting development

interesting development

So, finally, it turned out to be a non-electrical issue. We temporarily disabled the ground fault trip setting and let the motor run. The pump which the motor drives had some "air bubble" issue as my mechanical inspector found out; this caused the pump not able to pump at all, so the motor was not delivering little power (1/4 of its FLA)l to the pump.

Still a mystery on why this would cause the motor to see this "ground fault" current during start-up.

Motor/pump running no issues at the moment.