medium voltage instantaneous trip

| I have a typical 12470v dual service line up. Both feeders can never be | paralled. The mains and feeders are protected with ABB 51y very inverse | relays. The feeders supply mutliple transformers in sizes from 500 to | 2500 kva. Maximum load on any line is about 200amps. Short circuit | current is about 9000 amps at the mains and about 8000 amps at the | furthest point. For some reason they've been running with the | instantaneous trip set rather low on both the mains and feeders. I | would never use the instantaneous on the mains because there's no way to | coordinate it with the feeders. The instantaneous on the feeders seems | difficult to set because the short circuit current isnt' that much lower | at the furthest point in the system. | | Basically whenever anything trips, they locate the fault, fix it, and | turn everything back on. I'd prefer to get rid of the instantaneous on | the mains and possilby set it on the feeders. | | Any suggestions? | Thanks | Dave
IANAEE ... so this isn't a real suggestion.
I wonder why these things can't be programmed for a finite delay. If the mains were delayed slightly (and I have no idea how much time is appropriate for things like this), and recover if the fault is cleared at the feeders, would that be effective enough at the main without any coordination?
Obviously it wouldn't be an instantaneous trip. But neither is coordination when the downstream is saying "hold on, I can clear this". If that delay needed to be different depending on where the fault is, then it does get more complicated.

A coordinated system is what you want difficult is standard. Call an engineering company and have a coordination study done. CH- SqD-Seimans-GE all have people that do this for a living. I am not nuts enough to just start changing settings on medium voltage equipment. You might even get a hold of the power company they might help you find some one in your area. It is to their benefit for you to have a controlled system.

Stacked 50 functions can't generally be coordinated. You are going in the right direction--check to see if the 50 functions are not already disabled on the main-breaker 51Ys.
See IEEE Std 399 {brown book}
--s falke .