Main-Tie-Main differential protection 2

[rockman7892]

Can anyone provide any guidance/recommendations for providing differential protection on a double ended sub with Main, and tie breakers? Both main breakes have CT's and tie breaker has CT's on both sides of breaker. Does anyone know of a good reference for this?

I am also interested in providing 67N directional ground fault protection with main and tie breakers. Can anyone provide a refence for setting up this protection?

 

[slavag]

Could you please attach single line diagram

 

[davidbeach]

Two zones of bus diff. Two relays if using high impedance bus diff. Could potentially do it all in one low impedance bus diff, such as the SEL-487B, if it has enough inputs.

Explain what you want to do with the 67N.

 

[dpc]

As David said, two bus differential zones that overlap the tie breaker.

For directional overcurrent, you generally will need voltage transformers to polarize the relay to respond to current in the desired direction.

 

[rockman7892]

Attached is the switchgear drawing showing the breaker and CT connections. I have come to learn that this scheme is known as a partial differential scheme. I'm not 100% sure how it works though.

In my case the tie-breaker is normally closed and the facility runs with one main breaker closed or the other operating off of the one utility source.

So looking at the drawing Lets assume Main breaker 1 is open and the Tie and Main breaker 2 are closed. For faults on any of the feeder breaekrs the feeder breakers should trip and and the tie or main should not trip. If there is a fault on Bus A then should the tie breaker trip? And if there is a fault on Bus B then the main should trip?

From what I looked at it seems like this scheme combines a main breaker overcurrent settings and tie breaker overcurrent setting into one curve/setting? Normally I am used to having main settings and curves and then a tie breaker setting and curve. Does this scheme allow for only two curves instead of 3 with the tie and mains rolled into one curve?

 

[rockman7892]

In regards to the ground directional relaying I'm not really sure how to approach this. Customer asked about using them but wasn't really sure how to implement them. I'm just curious if they should be used for protection in this application?

Thanks

 

[jghrist]

This looks like a partial differential scheme, where only the source currents are included in the summation. It is more or less like overcurrent relaying that reacts to the sum of all of the source currents and trips all of the source breakers. The time delay has to be long enough to allow the feeder breakers to trip for a feeder fault.

If you had a fault on the other bus, the current in the main and tie breakers would subtract and you would not get a trip from the partial differential. The directional relays on the other bus would trip the tie breaker. This would allow you to set the partial differential and directional overcurrent relays with the same time-current characteristics and avoid having to coordinate between tie breaker and main breaker overcurrent relays.

It's a little difficult to figure out the scheme because of the way the one-line is drawn. Normally, you would have solid lines representing the ac voltages and currents and dashed lines for control. Your one-line looks like there are CT circuits going into lockout relays. There is a directional and a power element shown with no voltage input. It's all very confusing.

 

[rockman7892]

I agree the drawings are somewhat confusing.

So with the relaying schemes shown it sounds like there is not any traditional overcurrent relay setting for the main and tie breakers individually it sounds like the main and tie are set as combination (eg buss diff and directional) is this correct?

Can you explain how the directional overcurrent relay on the "other bus" would tirp the tie breaker? I'm not sure I follow this.

Also there is a relay function shown for a 67N however I dont see any ground CT's at the tie breaker as there is on the mains. Does this mean that ground fault cannot be used in a partial differential or directional function or is it possible that you have to use the residual connection of the phase CT's for ground functions?

 

[jghrist]

The directional element would only have the tie breaker current, not the summation, for it to work like I described. This isn't how the one-line appears, however. But then, the one-line really doesn't show the CT currents being summed. It is drawn to look like the current from the main breaker CT goes through all of the relay elements and then through the bus tie CT. This can't be true for a bus fault where the main CT current is not equal to the tie breaker CT current.

The 67N could use the CT residual current.

 

[slavag]

Yep, its seems as partial differential protection.
Actually, I'm not see any reasons for use 67 and 67N protection.
Only in few special cases, for example as back-up protection of 87L or/and back-up protection of 87B, some kind of directional zone selectivity.