I read the recent thread about a "new" device (another optical trip unit) and it got me to thinking.

Mechanically, about the best I've seen is that with vacuum interruptors, 3 cycles is pretty typical and once in a while you can find a 2 cycle unit. There are current limiting breakers as well which helps alleviate the true "top end". The fastest interruptor I've seen is the S&C "electronic fuses" which use an electronic trip unit and a capacitive trip device to trip a fuse, which gives circuit breaker accuracy and fuse speed. You also get fuse-level energy handling which is another issue with vacuum circuit breakers (the only alternative is SF6 interruptors). Unfortunately obviously when it trips though you have to replace the fuse link.

I noticed that some years ago S&C was experimenting with an electronic circuit breaker that could trip at fuse speeds because it is literally an IGBT based circuit breaker. If one builds a "simple" SCR based "switch" (because SCR's are cheap and can handle very high power levels) it makes it possible to build some very large units in a fairly compact space. IGBT is even better because you have so much more control but costs become an issue quickly. Either way though with SCR technology you can trip in 1/2 cycle, and with IGBT's, it can get down to potentially microsecond level switching.

Either way, it gave me a third idea. One problem with either one is the losses, heat management, and noise. What about combining these with the arc terminator concept that Square D has? In this case, it's a matter of outfitting a standard circuit breaker with an electronic switching system that simply unloads the arc. In this scenario for instance, one have SCR's mounted in such a way that unlike normal SCR configurations, they are configured to intentionally short circuit the phases. In this scenario, the electronic system would fire the SCR bank to intentionally shunt the arcing fault, terminating the arc by providing a lower resistance path to ground or phase-to-phase. While this is going on, the "normal" electromechanical trip unit (SF6 or vacuum) opens. Once it has opened, the solid state system shuts down, stopping all flow of current.

The advantage of this system is that the breaker can potentially get very small because the vacuum/SF6 system does not have to be particularly fast, nor does it need a particularly large amount of interrupting current capability. These are provided by the switching system. Depending on the solid state configuration, startup (inrush) currents could also be limited analogous to a soft start. The mechanical breaker is simply there to overcome the poor performing state of the solid state system (during "normal" operation at full conduction).

Anyways...just some thoughts. Technically this system should even be fast enough to avoid any need for high speed fuses since the power electronics should be operating at a speed which is fast enough to be able to shut down in the event of a damaging amount of current, though obviously this is only if it's not just a shunt...presumably the remaining vaccuum contactor can't tolerate high fault currents either.

Arc eliminator

You can also find a fast arc mitigation system for MV
here. http://www.safearc.se
jankar

The web site says nothing about how it works. It says that it's an "Earth Switch" which sounds and looks like the Square D "short it out" product. But some things don't add up.

The quoted speed is 5 ms. It is clearly designed to be triggered by the VAMP fiber optic arc detection system. I recall VAMP claiming something in the 2-5 ms trip speed.

So how fast does the switch itself trip? Is the total tripping time equal to the grounding switch PLUS the relay (VAMP) trip time or am I missing something?

The brochures make a lot of claims about safety but are severely lacking in detail.

Arc Eliminator

The Arc eliminator uses a Thomson coil which also is known from military applications. I made a sketch to better explain how it works. The mechanical unit will require 3.5ms to close and before that the protection relay will examine the trip signal and analyze if there also is a high current. This takes a little more than 1ms. But the arc elimination is ready before pressure, heat and toxid gases had time to develop in any dangerous amount. The existing equipment is for MV. If there is a market for a LV type I have applied for a patent for an even faster one.
The cotact travel can be shorter.

Jankar