Ground protection of resistance grounded system

[ish1284]

Hello, I have a system with a 115kV/12.47kV delta wye transformer where the secondary side of the transformer is grounded with a 200A grounding resistor.

I need to set the 50G/51G secondary main protection (i.e. the main breaker which is fed by the secondary of the transformer and goes on to feed a main switchgear). The CT ratio for the secondary protection 3000:5 and it utilizes residual ground protection (i.e. the relay takes the sum of the three phases to calculate the phase imbalance rather than from a neutral CT or core balance CT). If there is a 200A grounding resistor, this means that the max ground fault on the system is 200A correct?

Don't I want the instantaneous of the relay to pickup at 200 * 5/3000 = 0.33, and then the long time pickup of the relay to pick up at some value much less than that? However, this isn't possible because the relay is just not that sensitive with the given CT ratio. Or perhaps my understanding of resistance grounded systems is wrong. The system was not designed by me, I'm just performing the protection settings of an existing design.

Your help will be appreciated!

 

[gray.one]

Hello, I have a system with a 115kV/12.47kV delta wye transformer where the secondary side of the transformer is grounded with a 200A grounding resistor.

I need to set the 50G/51G secondary main protection (i.e. the main breaker which is fed by the secondary of the transformer and goes on to feed a main switchgear). The CT ratio for the secondary protection 3000:5 and it utilizes residual ground protection (i.e. the relay takes the sum of the three phases to calculate the phase imbalance rather than from a neutral CT or core balance CT). If there is a 200A grounding resistor, this means that the max ground fault on the system is 200A correct?

Don't I want the instantaneous of the relay to pickup at 200 * 5/3000 = 0.33, and then the long time pickup of the relay to pick up at some value much less than that? However, this isn't possible because the relay is just not that sensitive with the given CT ratio. Or perhaps my understanding of resistance grounded systems is wrong. The system was not designed by me, I'm just performing the protection settings of an existing design.

Your help will be appreciated!

I alway use a definite time element. The ground fault current is going to be 200A without much variance. The resistance is big compared to any fault resistance. You would set the pickup at a value below 200A so that you are sure it picks up - 150A.

Set some time delay and coordinate that time delay with any downstream GF protection. Don't worry about being aggressive with the time. Low resistance Grounding allows the fault to last a while without doing damage (10sec by memory).

Your calculation for pickup is right, just set it lower.




Sent from my iPhone using Tapatalk

 

[Ingenieur]

Yes, gf will be clamped at 200 steady state

All our systems use a ngr, usually 25 A
we generallt trip at 50% or 12.5 A with a few sec delay
your ct 3000:5 or 600:1 abs is likely not accurate enough
we usually put a zero sequence at 200 or 250:5 or the same on the resistor

 

[Bugman1400]

It would be nice to know the model# of the relay or, at least, the pickup range. Gray is essentially correct with setting the pickup lower than the 200A max current. The lower pickup provide some margin. IOW, if you have a bolted ground fault that is 200A, if your relay is also set at 200A, you may not actually trip because the relay is actually measuring 197A (measurement error). The question is how much margin to provide. If you have to set both elements, it is typical that the 50G have a pickup of around 150A (with a 2 cycle delay) and the 51G around 50-75A with a clearing time of 500% @ 30-60 cycles.

As you surmised, it would have been best to include a CT (50:5) on the grounding resistor in the design.

 

[ish1284]

It would be nice to know the model# of the relay or, at least, the pickup range. Gray is essentially correct with setting the pickup lower than the 200A max current. The lower pickup provide some margin. IOW, if you have a bolted ground fault that is 200A, if your relay is also set at 200A, you may not actually trip because the relay is actually measuring 197A (measurement error). The question is how much margin to provide. If you have to set both elements, it is typical that the 50G have a pickup of around 150A (with a 2 cycle delay) and the 51G around 50-75A with a clearing time of 500% @ 30-60 cycles.

As you surmised, it would have been best to include a CT (50:5) on the grounding resistor in the design.

It's an SEL751A relay with a 5A input card and the pickup range is 0.5 - 16 x the CT ratio.

And is my understanding correct that the entire downstream system ground fault will be limited to 200A until you get to another step-down transformer (i.e. a 12.47/480V delta-wye transformer)?

 

[Ingenieur]

It's an SEL751A relay with a 5A input card and the pickup range is 0.5 - 16 x the CT ratio.

And is my understanding correct that the entire downstream system ground fault will be limited to 200A until you get to another step-down transformer (i.e. a 12.47/480V delta-wye transformer)?

That is the same relay we often use
yes 200 on the 12.47 kv sec

fyi ngr value = 7200/200 = 36 Ohm
it will dissipate almost 1.5 MW, so it can't carry too long lol

good primer on ngr systems http://www.postglover.com/wp-content/uploads/downloads/2014/04/NG112-06_Technical_Information.pdf

 

[ish1284]

That is the same relay we often use
yes 200 on the 12.47 kv sec

fyi ngr value = 7200/200 = 36 Ohm
it will dissipate almost 1.5 MW, so it can't carry too long lol

good primer on ngr systems http://www.postglover.com/wp-content/uploads/downloads/2014/04/NG112-06_Technical_Information.pdf

This forum rocks, thanks all.

So it sounds like the ground protection with this 3000:5A CT is essentially useless. One bus down there's another main with a 2000:5 which is basically useless as well when considering the relay/CT accuracy.

 

[GoldDigger]

That is the same relay we often use
yes 200 on the 12.47 kv sec

fyi ngr value = 7200/200 = 36 Ohm
it will dissipate almost 1.5 MW, so it can't carry too long lol

So if you set the trip at 50% you could still be dissipating ~400kW in the resistor for a non-bolted fault, yes?

 

[Ingenieur]

You may consider this
a pt/59G across the ngr X0 to gnd
pt 7200:120
2 benefits
you can trip with it, set at 60 v sec (100 A x 36 Ohm = 3600 v prim)
and it will detect an open ngr (lost ground) and trip if the correct relay is selected

 

[Ingenieur]

So if you set the trip at 50% you could still be dissipating ~400kW in the resistor for a non-bolted fault, yes?

Yes
but if there is a time delay the ckt will still see 200 A for the delay period
we use ngr's rated for continuous (15 min iirc)
IEEE has various ratings

we use 2 relays
current and voltage
the current is set at 40% x ngr and <3 sec delay
the voltage at 50% and 1 sec plus the current delay
bear in mind though full ngr i flows during the delay period

 

[ish1284]

It would be nice to know the model# of the relay or, at least, the pickup range. Gray is essentially correct with setting the pickup lower than the 200A max current. The lower pickup provide some margin. IOW, if you have a bolted ground fault that is 200A, if your relay is also set at 200A, you may not actually trip because the relay is actually measuring 197A (measurement error). The question is how much margin to provide. If you have to set both elements, it is typical that the 50G have a pickup of around 150A (with a 2 cycle delay) and the 51G around 50-75A with a clearing time of 500% @ 30-60 cycles.

As you surmised, it would have been best to include a CT (50:5) on the grounding resistor in the design.

Any suggestions for how to set an alarm relay? The transformer secondary neutral also has an SEL387 with a 200:5 CT input but it doesn't perform a trip function. I assume it's for alarm purposes.

 

[Ingenieur]

Any suggestions for how to set an alarm relay? The transformer secondary neutral also has an SEL387 with a 200:5 CT input but it doesn't perform a trip function. I assume it's for alarm purposes.

that likely is the primary gf protection
perfect range
set at 100 A (2.5 at relay)

without knowing the specific pn the relay may be overkill

 

[ish1284]

Yes
but if there is a time delay the ckt will still see 200 A for the delay period
we use ngr's rated for continuous (15 min iirc)
IEEE has various ratings

we use 2 relays
current and voltage
the current is set at 40% x ngr and <3 sec delay
the voltage at 50% and 1 sec plus the current delay
bear in mind though full ngr i flows during the delay period

Thanks for the tip. If it were my design I would consider it.

 

[ish1284]

that likely is the primary gf protection
perfect range
set at 100 A (2.5 at relay)

Based on the relay and metering diagram, it serves no trip function. It shows the differential element performing a trip, but not the 50G/51G. Perhaps it's a mistake in the drawing which I plan to sort out with the design firm.

 

[Ingenieur]

Based on the relay and metering diagram, it serves no trip function. It shows the differential element performing a trip, but not the 50G/51G. Perhaps it's a mistake in the drawing which I plan to sort out with the design firm.

are there 2 ct's ?
one on the ngr
the other?

it looks like it can be set up to trip on diff and/or gf

 

[ish1284]

are there 2 ct's ?
one on the ngr
the other?

it looks like it can be set up to trip on diff and/or gf

It is in fact set up to trip on differential which should protect the transformer and there's another differential which will protect the switchgear.

I was thrown off because the relay and metering diagram calls for 50G/51G protection as well but in reality, an adequate design was not provided to actually provide 50G/51G protection.

 

[Ingenieur]

It is in fact set up to trip on differential which should protect the transformer and there's another differential which will protect the switchgear.

I was thrown off because the relay and metering diagram calls for 50G/51G protection as well but in reality, an adequate design was not provided to actually provide 50G/51G protection.

post the diagram
if there is a ct on the ngr then that relay will give gf trip
look at the relay manual