Number of SEL487Bs for dynamic 2 zone protection

[mbrooke]

How many 487B relays are needed for 2 dynamic zones of buss protection with 14 potions, 3 phase tripping and 3 phase re-closing circuit breakers? And roughly how do I wire them? I am familiar with GE's B90 which requires 1 relay per phase and another relay for disconnect monitoring giving 4 interconnected via fiber optics but totally lost on SEL.

 

[rian0201]

14 potions? The manual provides a very detailed idea on this one. It depends on what part number you have. You can put all three phases on 1 relay provided the number of cts fit. If not then, you have to per phase per relay.


Sent from Mars

 

[Bugman1400]

I think he means 14 positions.

The utility that I mentioned before with the line PTs on this double bus scheme has the exact same setup with the B90s. Because of the limitation of the B90 to 24 CT inputs, you'll need to use a B90 for each phase (3 total). I don't think the bus can be protected dynamically during switching from one bus to the other without opening the breaker first. Otherwise, the bus diff sceme will have to be disabled during switching. However, there may be some special logic that can be used to allow switching dynamically, but it most likely involves a lot if I/O design.

 

[mbrooke]

]I think he means 14 positions.

Yup, positions as in bays or elements. Auto correct often gets me at the worst possible time :lol:

The utility that I mentioned before with the line PTs on this double bus scheme has the exact same setup with the B90s. Because of the limitation of the B90 to 24 CT inputs, you'll need to use a B90 for each phase (3 total).

Yup, exact same set up for me.

I don't think the bus can be protected dynamically during switching from one bus to the other without opening the breaker first. Otherwise, the bus diff sceme will have to be disabled during switching. However, there may be some special logic that can be used to allow switching dynamically, but it most likely involves a lot if I/O design.

My understanding is that with 4 b90s you can get dynamic zones where closing 2 disconnects produces a single zone and opening the other will put the closed disconnect into the correct zone. But I am unsure how many are needed for an SEL487B.


This utility with the line VTs, does it have position switches on the disconnects to help assign zones? Or is it done by hand?

 

[mbrooke]

14 potions? The manual provides a very detailed idea on this one. It depends on what part number you have. You can put all three phases on 1 relay provided the number of cts fit. If not then, you have to per phase per relay.


Sent from Mars

Only 6 for SEL487B and about the same for B90. At minimum you need 3, however I am unsure if SEL calls for more. Or how they are wired.

 

[epimetheus]

487B can do 7 elements in a single relay (all 3 phases in the same relay). If you have more than 7 elements, you can do 21 elements with 3 487B's. The 487B's can be set up for dynamic bus replica protection if the disconnects have auxiliary switches. I recently designed a 69kV GIS substation that was double bus, single breaker with bus ties and couplers, protected with 487B's using dynamic bus replica.

Sent from my Pixel XL using Tapatalk

 

[mbrooke]

487B can do 7 elements in a single relay (all 3 phases in the same relay). If you have more than 7 elements, you can do 21 elements with 3 487B's. The 487B's can be set up for dynamic bus replica protection if the disconnects have auxiliary switches. I recently designed a 69kV GIS substation that was double bus, single breaker with bus ties and couplers, protected with 487B's using dynamic bus replica.

Sent from my Pixel XL using Tapatalk

Neat Any fiber optic links between the 3 relays like with a GEB90?

Your disconnects had 3 wires and 2 switches per disconnect wired something like IN101 & IN201 for disconnect 1; IN102 & IN202 disconnect 2; ect?


And (hypothetical) can the disconnect switches be substituted to mimick board switches provided proper verification is followed?

 

[epimetheus]

No fiber between the relays required. You just parallel the inputs and outputs of each of the 3 relays. Each relay protects a single phase. Best to wire an 89a and 89b contact from each switch into the relays, but it can be done with 89b contacts only. You could use mimic board switches instead, but I wouldn't recommend it. Much higher chance of mis-operation.

Sent from my Pixel XL using Tapatalk

 

[mbrooke]

No fiber between the relays required. You just parallel the inputs and outputs of each of the 3 relays. Each relay protects a single phase. Best to wire an 89a and 89b contact from each switch into the relays, but it can be done with 89b contacts only. You could use mimic board switches instead, but I wouldn't recommend it. Much higher chance of mis-operation.

Sent from my Pixel XL using Tapatalk

Very good, makes perfect sense. In regard to the mimick board (and disconnects for that matter), a "check zone" is also doable? The idea is that the relay would break apart currents per zone, and if any are incorrect, it would indicate so without tripping like it would without the check.

 

[epimetheus]

A check zone is doable, and recommended. Any time both disconnects on a line terminal are closed, either intentionally during a hot swap, or unintentionally due to switching error, the zones are combined and the check zone trips immediately upon detection of differential current. The 487B application manual goes into this in great detail.

Sent from my Pixel XL using Tapatalk

 

[mbrooke]

A check zone is doable, and recommended. Any time both disconnects on a line terminal are closed, either intentionally during a hot swap, or unintentionally due to switching error, the zones are combined and the check zone trips immediately upon detection of differential current. The 487B application manual goes into this in great detail.

Sent from my Pixel XL using Tapatalk

But the check zone doesn't actually trip anything, just alarms that a trip would occur if the relay went back to having 2 zones?

 

[epimetheus]

No, when both zones are connected via something other than a breaker, i.e. both bus disconnects on a single line terminal are closed, and a bus fault is detected in either zone, the check zone trips both normal zones.

Sent from my Pixel XL using Tapatalk

 

[Bugman1400]

Not sure. I'll see if I can find out.

I am still having a tough time understanding how a dynamic bus zone can work. It seems to me that the switching would cause an issue. The only way I could see the switching being done without the bus diff being disabled is to create a giant bus diff zone that included both buses. Once the switching was done, you could go back to 2 bus diff zones.

 

[mbrooke]

Not sure. I'll see if I can find out.

I am still having a tough time understanding how a dynamic bus zone can work. It seems to me that the switching would cause an issue. The only way I could see the switching being done without the bus diff being disabled is to create a giant bus diff zone that included both buses. Once the switching was done, you could go back to 2 bus diff zones.

Thats how it works, as the switch closes it becomes one zone, when the other completely opens its back to two zones. This explains the concept a bit better:

http://www.gegridsolutions.com/smartgrid/sep06/busbar_protection_paper.pdf

https://cdn.selinc.com/assets/Liter...tection_SC_20130912_Web.pdf?v=20150812-080535

 

[mbrooke]

No, when both zones are connected via something other than a breaker, i.e. both bus disconnects on a single line terminal are closed, and a bus fault is detected in either zone, the check zone trips both normal zones.

Sent from my Pixel XL using Tapatalk

Makes sense. There is no way to configure a "test" zone?