Setting Breaker Failure Logic 1

[Mbrooke]

Would anyone have guides or advice on how to set the breaker failure logic for an SEL387 relay used in a transformer differential application controlling two circuit breakers? The user manual briefly mentions breaker failure and SEL makes note that it can be created using word bits and variable timers however none appear to go into detail regarding that.


The desired outcome is to have independent BF for each 33 kv primary and 11kv secondary breaker, in addition to being able to initiate BF logic for each breaker (retrip + current timer / BFT) via an INXX1 and INXX2 contact respectively.

......

A question on the side- how necessary is it to initiate BF when a primary or secondary breaker is tripped via a relay outside its zone of protection? I ask because theoretically over current built into the XFRMR differential would pick up a fault outside the zone of protection.

Second, while I normally do not do this for line protection- would you guys agree in using 52A by itself as an indicator of breaker failure? The scenario that I have in my mind is a 33kv L-G bus fault being cleared via bus protection while having a 33kv XFRMR breaker stick. The stuck breaker CTs would not measure much current, but the 33kv buss would be kept in a state of over voltage due to the delta primary.

Here is the single line of the substation at hand:

 

[Mbrooke]

These did not attach in the first post, but here they are:

http://files.engineering.com/getfile.aspx?folder=4e6aa69f-86a7-4566-aa3c-b67cee06ee30&file=txd1.jpg

http://files.engineering.com/getfile.aspx?folder=f50b6127-8b1b-4ed5-9a1f-6b9df5f1d4ee&file=txd2.jpg

http://files.engineering.com/getfile.aspx?folder=b02ab820-61b4-4474-947d-ee9b25fb0711&file=txd3.jpg

 

[cranky108]

It sort of depends on how you want it to work. To latch the inputs, or not, and if you want retrip.

And SEL should have papers on this, but in general it would be something like:
a timer ANDed with an overcurrent, or I would use a phase and ground overcurrent ORed, and not a single overcurrent. INPUT AND SV1T AND (50phase OR 50ground)
Where SV1=INPUT.

 

[Mbrooke]

Hopefully lol. I hate building Boolean logic circuits.

 

[cranky108]

Learn to love them.

 

[Mbrooke]

Nooooooooooooooo... LOL.

 

[davidbeach]

Very similarly to what was outlined in thread238-414092. There are three sets of SVs in the 387, so the names will be a bit different and the overcurrents have to match the winding overcurrent names.

 

[Mbrooke]

True, but how do I get 52A also play a role instead of just CT magnitude?

SVn = (SV1 + TRIP) * (50Px +50Gx)
SVnPU = 10 (or what ever value you want for your BF timer.
OUTnnn = SVnT.

SVn = (SV1 + TRIP + INnnn) * (50Px +50Gx)

 

[cranky108]

Many of us don't trust the 52A, so we don't include it.

The breaker may open, but the contacts may still carry current.

 

[Mbrooke]

Neither do I- but under certain conditions (I am assuming) a faulted transformer can pull current below the 50BF pickup. Right?

 

[cranky108]

So how low can you go on your 50BF? We set a lower limit of somewhere above the load current, so if it is triggered by mistake it won't make a big mess of outages.
For a fault to be as low as load current, it would be a wimpy fault.

Besides a phase to phase or three phase fault would normally not have a huge impedance to ba a wimpy fault.

A ground fault can be a wimpy fault, which is why you might want to think about putting your 50BF pickup for ground faults lower.

 

[Mbrooke]

Where would a turn to turn fault stand though?

 

[davidbeach]

On generators we include the 52a in the equation; some people also include it on transformers but we don't. In either case it's to deal with faults that might be below the current detector threshold.

Just go with SVn = (SV1 + TRIP + INnnn) * (50Px + 50Gx + 52A)

Once initiated, either by a trip or from a lockout relay, it will time until there's no current and the breaker mechanism indicates that it's open.

 

[Mbrooke]

Thanks, easier then I assumed.

Once a trip is initiated through an INXXX (ie, IN202 for LV BF) it will time the LV breaker, and if the LV breaker does not trip within 8 cycles (selected value) it will then open the HV breaker only, correct? If the HV breaker then does not open within 8 cycle it will then lockout the HV bus. Same will take place visa versa, ie HV to LV. To confirm, these equations will take care of this sequence?

 

[davidbeach]

When the SV times out you have a breaker failure condition, you figure out what to do with it in your wiring and logic.

For a transformer application we trip the transformer lockout, and initiate breaker failure on a lockout trip since there are other devices that trip the transformer lockout. Breaker failure of any given breaker then trips the lockout on the other side of that breaker (which initiates breaker failure protection for those breakers). So, a transformer trip with failure of the low-side breaker would start out with the transformer lockout, tripping both sides; when the BF timer on the low-side times out the low-side bus lockout is tripped.

 

[Mbrooke]

When the SV times out you have a breaker failure condition, you figure out what to do with it in your wiring and logic.


For a transformer application we trip the transformer lockout, and initiate breaker failure on a lockout trip since there are other devices that trip the transformer lockout. Breaker failure of any given breaker then trips the lockout on the other side of that breaker (which initiates breaker failure protection for those breakers). So, a transformer trip with failure of the low-side breaker would start out with the transformer lockout, tripping both sides; when the BF timer on the low-side times out the low-side bus lockout is tripped.

Correct, for a transformer fault that is the plan and understood.

However for an 11kv bus fault the plan to have the 387 relay trip the HV breaker only if the transformer LV breaker fails to open. The idea is to keep the transformer in service even after a bus fault because the SST (station service transformer) taps directly off the 11kv secondary.

 

[davidbeach]

Makes sense.

 

[Mbrooke]

Good we can agree on this

However {SVn = (SV1 + TRIP + INnnn) * (50Px + 50Gx + 52A)} one initiated via IN201 is sufficient by itself to be configured to only trip the HV breaker (while also initiating its BF) and not immediately lockout the HV buss?

My apologies ahead of time if my description is off.

 

[davidbeach]

Yep.

I notice that it starts off SVn = (SV1 + ... where it should be SVn = (SVn + .... We use SV1 and I must have not made the equation fully generic at some point. The SV is included in the equation to latch if the trip or the INput goes away before timing out.

 

[Mbrooke]

Thank you and thanks for clearing that one up! The rest is self explanatory from here on- thanks again for your time