Why not have 2 bakups for Protective Relays?

[jeffreed]

I was wondering why utilities don’t have 2 backups to their primary relay. If one relay had to be pulled out for whatever reason they would not have to replace it because they would still have 2 relays. Wouldn’t this save them money? How often does a relay get pulled out? What would the cost be to install the 3rd relay? The commissioning time would be a lot less. Thanks for your input.

 

[dpc]

Why not 3 backups?

I'm not following your logic. If a relay has to be pulled out, they still have functioning protection. Assuming a failure of the remaining relay while one is out of service is a double contingency and normally no one can afford to design for that.

The commissioning time would be greater because there would be 50% more things to test and 50% more problems to troubleshoot.

Also, you are assuming that the only failure mode is the relay. This is not the case. For true redundancy, each relay would need its own CT and PT inputs, its own breaker trip coil and its own battery. Outside of nuclear power plants, this is not practical.

Breaker failure schemes provide backup protection, although it is slower than the primary protection.

To be honest, a lot of protective systems don't even have one backup and just depend on the next device upstream to eventually trip.

David Castor

http://www.cvoes.com

 

[jeffreed]

If they pull a relay out, do they not put in another relay to replace that one while it is out, or do they just wait untill the relay comes back to put it in?

 

[waross]

I worked on a 13000 Volt switchgear lineup where there was an overlapping transfer switch ahead of and behind the breaker for each outgoing feeder. The transfers were to two buses that were joined by a service breaker. Any breaker could be bypassed by the bypass breaker and then removed for servicing. The protective relays had their trip outputs connected to the bypass breaker. I don't remember redundant protection relays though. Protection relays are pretty dependable.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[slavag]

Hi.
Its depend on the voltage level and countries rules.
Its depend on the protected object.
For example:
lines 220/330/500kV one main protection and two back-ups protection.
big transformers, one 87T , second 50/51 and next 87N.
But for the MV, waht David and Bill saied...depend

Good Luck.
Slava

 

[marks1080]

Most utilities no long use 'back-up' protection as a source of redundancey. Back-up implies that that the back-up relay will only operate when the primary fails to. This leads to long clearing times. On high voltage bulk electrical equipment we should use (as a minimum) two redundant protective systems. Usually referred to as an 'A' and 'B' scheme these protections will race each other to clear the fault. It is also common to use different manufactures equpiment for each scheme, separate power supplies, etc...

Back-up protections are still common to see, but we do not rely on them to provide redundancey.

Mark

 

[cranky108]

Actually many of use what we cann primary and secondary protection, from the same manufacturer.

The problem with duplicating things is the maintenance and cost. Having two circuit breakers, is expencive. Having two batteries is expencive and ups the maintenance. Having two PT's also increases your chances of failure.
Having two communications schemes becomes problomatic, expencive, and also ups the maintenance.

There are limits, and we prefer to make the limit the impact on the system, and cost.

 

[jeffreed]

Thanks!! You all have cleared up some misconceptions that I had.

 

[marks1080]

cranky: I agree with your statment. Duplication of cct bkrs is not a good idea. We duplicate our protection schemes, not the elements themselves. When possible we will use separate trip circuits for breakers, but we don't duplicate the ITs or breakers.

There are two philosophies on wheather or not to use the same or different manufactures for duplicated systems. I think I do prefer using different manufactures when it is available, but arguments can be made for both sides.

Mark

 

[cranky108]

The problem with using different manufacturers for relays is it means we have to understand both. There are two manufacturers thing method, or so it appears. One is to copy another manufactures as much as possible (could be a common mode problem). The other is to make a totaly, as much as possible, different design.

The issue with common mode failure is where do you draw the line? Few people will install totally redundit systems as they cost to much.

 

[davidbeach]

Charlie Henville gave an interesting presentation on protection system redundancy at WPRC in 2008. What gets left out of most discussions of common mode failures in protection systems is that there are many more things to fail than the relays. Same protection engineer develops the settings - big common mode problem. Same design engineer lays out the wiring - big common mode problem. Better to have one relay that is really well understood than two different relays only adequately understood.

 

[marks1080]

I still think it is better to have two relays that are both very well understood

That being said, different situations call for different approaches. When dealing with critical system elements redundancey is a bigger issue than when dealing with the LV side of a station.