Supplier for 15kV/250nF or similar pulse caps

[daqq]

Hi guys,

For a project we need caps with a total capacity of 250nF @15kV. They must be able to provide a short pulse of around 15Amps pulse, with a duty cycle of 0.1%. Any ideas on where to start looking? It must be a normal supplier, not ebay NOS.

Thanks,

David

[Marco]

Just from hanging around on a laser mailing list the following names come to mind, Maxwell, Aerovox, General Atomics.

PS. given your location Russian suppliers might have some nearby distributors as well, but no idea where to start looking for them. Google is always a good start though

[TerraHertz]

http://www.highenergycorp.com/
High Energy Corp. | PO Box 308 | Lower Valley Road | Parkesburg, PA 19365 | phone: 610-593-2800

The HV ceramic disk caps from Nitto Denji (Japan) are probably all too low in capacitance. Hard to tell since their site doesn't seem to have an English version.

Gosh, googling "high voltage pulse capacitors" turns up some pics of spectacular looking gear.

[Marco]

Forget what I said, you don't really need a pulse capacitor. Those are hopelessly overspecced for your relatively low current pulses.

[Arjan Emm]

Pulse caps are rated in the kA ranges. Way too heavy duty.
Maybe a couple of IGBT snubber caps in series? I know WIMA has 3kVdc 2,2uF ones. 5 in series might be close enough. They are probably overrated too, current wise.

[T3sl4co1l]

Snubber caps I don't think come high enough in value, but DC link (bypass) caps should do.  They come up to 875 or 1100V, something like that.  Maybe 20 in series, so values of 5uF, which are easily found.

Tim

[daqq]

Hi guys,

First of all, thanks for the information. The reason for the pulse caps was that this should be quite fast - very low rise time ( <100ns preferable) and the actual on time should be around 2us, with a repetition rate of 500Hz - this gives a fairly low average power, but a large pulse power. As such parasitic inductances could cause problems. The best parameters I've seen were in various pulse caps.

The reason for the relatively high capacitance is that the voltage should drop as little as possible during the actual pulse.

But now that I think about it you are right, a bank made up of a series-parallel combination of good fast (but not too pulsy) caps should do the trick, assuming some charge balancing resistors are used.

Is there any cap you would recommend for this? I'll scour the obvious suspects (EPCOS, ES-OSTRAVA, AVX) tomorrow.

Thanks,

David

[Marco]

Maxwell will gladly sell you something for 1000s of dollars which will undoubtedly do what you want.

I'd just get something like this and try it out (pretty sure that's a CJE capacitor by the way).

[T3sl4co1l]

Stray inductance isn't a concern.  15kV and 15A is an impedance around 1kohm, so you're looking at a time constant below t = L/R = 100ns, or L < 100uH.

Tim

[daqq]

Thanks. Stray inductance is also a bit of a concern due to what happens after a quick change of current - you can get massive voltage spikes.

[ElectronicTonic]

I used polypropylene pulse caps for my DRSS Tesla coil in series-parallel combination. Bought them from here: http://easternvoltageresearch.com/flexibrute_capacitors.html. They are also available from Digi-Key, Allied Electronics, etc. The biggest cap in the series is 0.15 uF, 2000V, Conell-Dubilier 942C20P15K-F. It has 5 m-ohm ESR, 32 nH ESL, 2879 V/us dV/dt, 432 Amp peak current.

I used the 0.10 uF caps in combination to get 0.025 uF, 24 kV. Here they are in my Tesla coil if you're interested: https://www.youtube.com/watch?v=HdwcZX07NFk&list=PLmj4qMKo6w6dj18EPKZKn10BVotjBsFHY&index=1. Skip to 8:17 to see the caps.

Hope this helps.

[LaserSteve]

Just from hanging around on a laser mailing list the following names come to mind, Maxwell, Aerovox, General Atomics.


Care to share the list in a PM?

Steve