SEL-352

[Mbrooke]

When can a substation get away without having breaker failure relays?

 

[davidbeach]

Breaker failure relays or the breaker failure function? The SEL-352 can do far more than just breaker failure. We have one of them, at one of our plants. We have also implemented breaker failure protection in pretty much every scheme we've installed for many years now; each relay does its own breaker failure protection.

 

[Mbrooke]

Breaker failure relays. Don't all new relays like SEL-421 or ABB come with their own breaker failure capability? A separate relay seems redundant, especially when ring buss applications appear to call for one SEL-352 per breaker.

FWIW, as a side, if you have two breakers in series my understanding is you can skip breaker failure protection?

 

[davidbeach]

Used just for breaker failure the 352 seems to be gross overkill. The major transmission provider in my part of the world installs the 352 to do point on wave switching control of single pole breakers. That's much more demanding than breaker failure.

In the newer relays with built in breaker failure logic we use that, but in the 311 family, when connected to a single breaker, it's real easy to implement breaker failure in logic.

Tripping two breakers in series should drastically reduce the chances of breaker failure but I'd still be prepared for it. The fault could be between them so a single breaker failure still needs additional tripping. Simultaneous failure of multiple breakers can, and has, happened. One large company discovered that the hard way. Eventually they determined that the wrong grease had been used during breaker maintenance and a whole bunch of breakers had frozen up.

 

[Mbrooke]

What benefit does point on wave switching have? (just curious) Is it to decrease contact ware or part for power stability reasons? I did research on the 352, and it appears to also store breaker duty, event history and comes with options like sync check:

https://cdn.selinc.com//assets/Lite... Sheets/352_DS_20160122.pdf?v=20160209-151256

But personally for my applications it would be a significant over investment that is unlikely to pay back at first glance. Currently all breakers I deal with are single 3 phase units commanded by two sets of relays (primary and secondary backup). If both have BF equipped and enabled, I am already covered with a stand alone unit having no merit. I did find this paper on the subject, very helpful:

https://cdn.selinc.com//assets/Lite...Failure_BK_20110217_Web.pdf?v=20151124-160507

Good point regarding fault in between the breaker, although Id imagine multiple breaker failures in a substation would require something far more complex like remote tripping than simply opening all the breakers on the buss as current BF relays do in my setup.

The station that had the seized breakers, was this the double breaker double buss station you mentioned a while back lol?

 

[Mbrooke]

And ohh, from the paper:

"First, a distinction needs to be made between static and
reconfigurable buses. In reconfigurable buses, a given breaker
may connect a given network element (e.g., line, transformer,
capacitor bank) to multiple bus sections. Examples of
reconfigurable buses include a double-bus single-breaker or
main/transfer bus. This dynamic association of breakers
complicates the BF tripping logic because the BF protection
system must know which breakers are connected to the same
bus as the failed breaker. For this reason, the BF function is
often integrated with a low-impedance bus protection system,
or the bus protection system receives the BFT signals from
external BF elements and routes the BFT command
adequately to the appropriate breakers"

I am a bit confused by what they mean in regards to main and transfer buss as my understanding has always been that main and transfer buss substations are always designed and relayed as a single simple buss having no dynamic zones of protection. Unless they are referencing a concern regarding the transfer buss breaker (I treat the transfer buss as a transmission line)?

 

[davidbeach]

Point on wave switching minimizes switching transients. Think of a 500kV as a capacitor; just like cap switchers do point on wave closing, line switching can avoid the charging inrush. Likewise, if a transformer is deenergized and reenergized in a controlled manner the inrush can be significantly reduced.

In a number of our older bulk power stations there's a 230kV M&T high side with a single main bus that includes two 230/115kV transformers. On the low side there's a single aux (transfer) bus but three segments to the main bus. Each transformer is on an end and the tie (substitution) breaker is on the middle section by itself. The two sectionalizing switches are interlocked so that they can't both be closed at the same time. This way one tie breaker can serve two otherwise independent buses. BF for the tie needs to know which bus lockout to trip.

 

[Mbrooke]

That makes sense now regarding point on wave switching. Ditto for dynamic zone selection. In my case there is only one main buss, so its static. I think I will skip out on the SEL351 for now, but its certainly one of those relays to consider should the need arise in the future.


In any case I want to give you a big Thank you. Your knowledge on the subject is worth its weight in gold.

 

[Mbrooke]

Ok, so I researched this more and need some real world experience/feed back. Having a separate breaker failure relay increases security against false tripping in nearly all scenarios I've evaluated. One the other hand having breaker failure protection via the normal and backup line protection relays greatly reduces cost and eliminates inadvertent tripping probability due to testing. Win. However, should any relay malfunction, it is possible for something like an SEL411L or 421 to clear an entire station where previously this risk did not exist since these new relays will have access to both bus trips. Granted its documented these relays have a MTBF of 300 to 400 years, but give a 14 bay station with 28 failure points total risk goes up, at least Id think.


To those who went from separate stand alone relays to an integrated approach, what pros and cons have you witnessed? Any misoperations? Or simply stop worrying enjoy the benefits?

 

[DTR2011]

In my neck of the woods, a separate BF seems to be preferred. It ranges from a SEL 501 to a SEL-352 to SEL-451 in newer installations. Most applications are ring bus. In all cases, the BF relay trips the bus diff lock out (86B). I have seen some applications of a dedicated 86BF device, but not in newer designs.

I have also seen SEl-421/411L/311L 321/311C schemes where 50BF & 79 is implemented into the relays to trip the 86B/86BF. The advantage I see from this scheme is that an entire protection system (Primary or backup) can be taken out of service for maintenance / modifications, without the loss of BF protection. It makes things a bit easier to explain to system operations that only the primary, or back up protection will be out of service, when starting work. Generally with the Transco this is straightforward, but if dealing with other entities such as Genco (Nuclear especially), having all parties aware that all functionality is in place is a statement most like to hear.

As far as reducing testing errors, yes it can occur, but if a BF trip from an IED that does it internally, or one that just initiates a dedicated relay, in either case the schematic was not analyzed and all trips or initiates were not pulled.

 

[davidbeach]

To clear an entire station you'd have to have a double bus-double breaker configuration and have the relay erroneously decide that both breakers have failed. Or a ring of less than five positions. With breaker-and-a-half you can't clear more than one bus and one bay; that only takes out two positions.

Over the years we've seen more examples of inputs picking up falsely than we have of outputs closing falsely. Based on that, I'd say that internal BF is more secure than a separate BF relay.

Different utilities look at identical problems and come up with very different solutions; there's no single right answer.

 

[Mbrooke]

Yup, double breaker double buss. Same concern in theory for single buss. Any false close will clear both these schemes.

Where inputs have failed to pickup correctly, was this via incorrect settings or a defect within the relay itself?

 

[Mbrooke]

In my neck of the woods, a separate BF seems to be preferred. It ranges from a SEL 501 to a SEL-352 to SEL-451 in newer installations. Most applications are ring bus. In all cases, the BF relay trips the bus diff lock out (86B). I have seen some applications of a dedicated 86BF device, but not in newer designs.

In my system breaker failure energizes the same master trip circuit that buss differential energizes, unless of course its dynamic zones, which BF simply tells the buss diff relay to trip the buss in question by its own logic. FWIW I do not employ lockout relays since buss faults have the option of being reset by SCADA (yes I know that may turn heads lol). But in any case what matters is having the BF mechanism close its contacts only when a breaker has actually failed.

I have also seen SEl-421/411L/311L 321/311C schemes where 50BF & 79 is implemented into the relays to trip the 86B/86BF. The advantage I see from this scheme is that an entire protection system (Primary or backup) can be taken out of service for maintenance / modifications, without the loss of BF protection. It makes things a bit easier to explain to system operations that only the primary, or back up protection will be out of service, when starting work. Generally with the Transco this is straightforward, but if dealing with other entities such as Genco (Nuclear especially), having all parties aware that all functionality is in place is a statement most like to hear.

Same here, my thinking precisely. Having all functions in service without any portion being taken out during maintenance gives the biggest appeal for me ditching separate relays. IF security can be met, its a no brainier.

As far as reducing testing errors, yes it can occur, but if a BF trip from an IED that does it internally, or one that just initiates a dedicated relay, in either case the schematic was not analyzed and all trips or initiates were not pulled.

I run my trip circuit through the test switches, and plan on doing the same for the breaker failure circuit coming out of the relay.

 

[Mbrooke]

Ok, so I am going with an integrated BF function on both primary and secondary relays. Everything is set, however one thing has me rather lost. For a buss fault accompanied by a stuck breaker, how do I send a DTT to the remote substation for the stuck bay?


My thinking: program SEL411L and SEL 311L (and the like) so that internal BF is automatically initiated when a dedicated input is energized (by the buss trip relay). This way each line relay will try and re-trip its breaker along with initiating the BF timer. IF CTs do now drop below 100amps after a 8 cycles a DTT signal will be sent to the remote station via fiber optic path (OPGW) supporting normal current differential.


For transformer relays the same input will take place, however instead the relay will trip the low side breaker (as will probably happen even for a normal breaker response).


I've checked the SEL manuals and nothing stands out regarding this.

The other option is simply clearing via zone2/zone3 which is intend only as an absolute last resort.

How easily doable/common is this?