High Fault Currents

[txjohnny]

When doing some previous research, I stumbled across an article on solutions for systems with high fault currents. It discussed different ways to deal with the high currents, solutions for lowering the currents, etc. It may have been in an IEEE journal. I did not need it at the time, but now I wish I had saved it. Can anyone point me to a similar technical article or white paper that discusses solutions to this problem? I need to start lowering the fault levels on a preliminary design and I need some guidance on what direction to go. Thanks in advance.

 

[davidbeach]

Raise your source impedance. Work at a higher voltage. Pretty basic physical properties.

 

[magoo2]

Current limiting reactors have been used by utilities and industrial customers alike. What voltage level are you dealing with and what level of fault current?

As an alternative, you might find some benefit in using current limiting fuses and living with the high fault currents.

 

[txjohnny]

Yeah, I am really looking for a reference. Perhaps something that compares different network topologies. I could just increase the impedance as davidbeach said, but that would raise my losses and I am already concerned about voltage drop. I could split the loads down more and divvy it up among a greater number of xfmrs, to keep voltage drop down and lower available short circuit current, but then pricing (and space!) becomes an issue. If I start making a ring-bus I think I could get the losses down (this is another one of my major concerns but that is a different topic)but my available short circuit current goes up again. There are a lot of different ways to go and I was thinking surely someone else has looked into this already. I hope I don't have to be the first to start an in-dept study into the pro's and con's all of my options here! But since you are curious, I have levels of over 130kA on 600V MCCs with the tie breakers closed (main-tie-main aka. secondary selective network).

 

[cranky108]

The biggest problem with most methods of reducing fault current is you increase your losses.
Current limiting fuses is an option that dosen't.
Another is a neutral reactor on your source transformer. But it won't help with phase to phase faults.

 

[davidbeach]

I could just increase the impedance as davidbeach said, but that would raise my losses and I am already concerned about voltage drop.

Yep, that's the rub. A nice summary of what a lot of system design is all about. One way of dropping in a chunk of impedance is through the use of a transformer, with the right ratio you can deal with the voltage drop issue and have have lower fault current.

Years ago I read about, but have never seen in operation, a superconducting bus coupler that at normal currents is superconducting but at much above maximum load it quenches and becomes a rather high resistance. Nice in theory, but so far I've no evidence of it in practice.

Another possibility, if you system topology works, is to use reactors as bus couplers. Say you have two transformers that each feed a bus and then you have the busses tied together for reliability and redundancy; the two busses each have their own load and little current flows through the tie with both sources available. Put a reactor in between the two busses and a normally open breaker around the reactor. Normal operation results in the same voltage on the two busses due to the connection through the reactor. On the loss of one source there will be load flow through, and voltage drop across, the reactor. As fast as the source goes away and the incoming breaker is opened the breaker around the reactor is closed. You have a short voltage transient, but you probably had a voltage transient associated with what ever caused the loss of the source. Once down to only one source the reactor is no longer needed for fault current limitation.