contact rating for pulsed current

[RyreInc]

Looking for a relay/contactor to direct pulse current to one of two fixtures. Pulse is sinusoidal, 8kA peak, 2300V initial, 100us pulse width. 30 sec. cycle time, so small average current.

Would a relay rated at 100A continuous work, and if not 100A, then what? Is a relay a wise idea at all?

One option (I know the voltage ratings are probably too close):

http://www.components.omron.com/com...455A9EC85257201007DD583/$file/G9EA-1_0209.pdf

What are the failure modes of such a solution? Cold welding? Magnetic interference from pulse through contacts?

(Just so you know, I'm also considering more reasonable options like a second ignitron or thyristor, but this could potentially be the cheapest solution, and wouldn't require opening up the machine.)

 

[VEBill]

The business of electrical contacts is a science unto itself (materials, surfaces, etc.). All that carefully-engineered science will be vaporized with 8000A on a contact rated at 100A. Then you'll be back to banging two bits of metal together and hoping for the best, with a commensurate lack of reliability and lifespan.

How about finding something from the Power Transmission industry? They switch such levels every day.

 

[RyreInc]

Thanks for the response.

One thing I didn't mention is that the relay needs to be less than $250 to be price competitive. I haven't found anything less than $300, and even these have current rating of "only" 150-300A, and usually too low of a voltage rating.

I should probably note as well that the relay will already be closed when it is introduced to said pulse.

 

[VEBill]

"...the relay will already be closed..."

I now noticed that you hinted at that in your first post.

That's a much easier requirement. 100uS is such a small time that the contact's thermal mass (almost certainly) won't allow any damage from heating. Unless it's localized to such an extent that the peak temperature causes problems.

You should estimate the voltage drop for a range of contact resistances (from new to old).

That you mentioned a price it seems that this is for production, so that raises all sorts of interesting angles.

Give this thread some time to see what others can offer.

 

[ScottyUK]

Most contactors designed for service on 690V would probably survive the voltage pulse. You might even find a standard like the appropriate secion of EN 60947 actually defining pulse withstand levels for the utilisation classes. I'd check for you but my copy is at work.

Current-wise, contactors are generally designed to form part of a type-tested combination starter, which essentially means that the contactor will be fit for continued service after a through-fault of less than a certain magnitude. 8kA is a lot of current, but it's actually a pretty small fault level in utility terms and you won't need a very big contactor to survive it for 100uS. 100us is much shorter than the clearing time of a circuit breaker or a typical motor fuse.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[VEBill]

Scotty's post reminds me that the SD-QO circuit breakers in my house panel are rated for "10,000A" faults. They're of fairly lightweight construction and sell for about $10.

 

[ScottyUK]

A further thought - although you likely won't find it among the ratings - is that if you put multiple poles is series you are greatly increasing the isolation distance. This will probably move the voltage problem to the length of the creepage path over the surface rather than a direct flashover through the air.

If you are intending to use this as part of a safety isolation then don't use a relay or a contactor, use a lockable disconnector or switch. Many circuit breakers can fulfil this duty too.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[RyreInc]

That's good news, looks like I can get away with a relatively wimpy relay.

As far as putting poles in series, that probably isn't an option due to contact resistance, which I need to keep to a minimum (good idea though!). I'll just have to find a relay that can withstand the voltage. The option I posted earlier seems to fit the bill.

 

[potteryshard]

I seem to recall seeing catalog listings for radio transmitter antenna remote switching relays, which would suggest designs suitale for high voltage and high instantaneous current levels...