Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

MMC Design (For HV applications) - a question of max ratings

Joined
Dec 11, 2011
Messages
4,364
Points
83
Hey everyone,

I'm gearing up to build a SGTC with the NST I grabbed from Grenadier, and I've run in to a little question. Sometimes things are so simple that you doubt if you've got it right, this is one of those cases, and I figure the best way out of verify it with an outside source. I'm not new to HV, but I am new to TC's. I'm used to CW based HV stuff.

Anyway, when designing a MMC that you're matching for resonance (or just above resonance, as with a SGTC, to prevent voltage runaway during the event the SG fails to fire) you need to overshoot the mark in both capacitance and maximum voltage rating.

My NST has an output impedance of 250mOhms so the MMC will be resonant at 0.0106uF, and needs to withstand a repetitive Vpk of 10500V. I'm shooting for 0.011uF and 12kV for posterity. I could go for 15kV but then the cost rises significantly.

The question then is... which voltage rating to use, the Vdc or Vac one? I thought, it must be obvious.. use the Vac one as your tank cap is across an AC source, right? Yet I can't find polypropylene film caps with a higher Vmax (Vac) than 500Vac! This is very low, wth is going on? This is why I started to second guess myself. The same caps which are rated 2000Vdc are only rated 500Vac.

I'm currently looking at these:
940C20S22K-F Cornell Dubilier Polypropylene Film Capacitors

So, did I goof up, or am I only finding shitty caps? At $2/cap if I have to make a MMC out of 500Vac caps and a Vpk target of > 12kV I'l cost $576!!! (288x $2 @ 12 strings of 24caps to get the needed >0.011uF @ >12kVmax). With 2000Vac rated caps it would only cost $36 (18 x $2 @ 3 strings of 6 caps).

If everything is set up right the energy across the tank cap should be dumped in to the primary coil every cycle, with nothing left over (that is, if I match the primary to the tank cap's reactance to achieve resonance), so would I be able to use the Vdc rating since there would be no residual charge at the end of each cycle? If I could use the Vdc rating I'd have to double my target voltage as Vp-p = 2Vpk, right?


Update; the only thing I can find on the subject is that if you use very high quality capacitors you can use the Vdc rating as long as you include a 100% headroom above the Vpk of each AC half-cycle. There doesn't seem to be anything other than anecdotal evidence to support this, but it is the widely accepted practice. In other words if you series string caps together to get 20kV DC rating you can use it for an AC source with 10kVpk (7kV RMS) safely. Anything less than 1.5x the Vpk is discouraged because of resonant rise problems, even with a larger-than-resonance cap (which reduces resonant rise).

So there's good and bad news in that; YES you can use the DC rating and save a lot of money. NO, you can't just add a few kV headroom, it has to be at least double the Vpk for reliability.

So, if the partial SGTC I'm eyeing/bidding on falls through I'll have to save up to do it all from scratch, the right way, which is about $150. I've also been considering a SSTC kit, but after seeing the streamer shape and size for an equally budgeted SSTC I'm not impressed; 2" streamers which are straight and sword like vs long >5 inch branched streamers that Nikola himself would have been proud of.
 
Last edited:





Joined
Jul 4, 2008
Messages
2,499
Points
113
I'll make this as straight forward as possible.

CDE capacitors for an MMC is the most efficient way to go for cost/nF.

CDE makes 1.6Kv 680nF capacitors. Look at these 942C16P68K-F

20 in a row of these will give you enough for a 32Kvdc rated MMC.

Been running the same cap bank for 4 years, not one failure. All on a 12KV transformer bank.

$8per cap. 20 cost me around $160-180CDN.
 
Last edited:
Joined
Dec 11, 2011
Messages
4,364
Points
83
Just an update; I had gone with the CDE 942C20S47K-F 0.047uF 2kV caps which have hig dv/dt and good (low) ESR for my MMC shortly after I last updated the OP. They were both the best in terms of reliability for TC applications and best price per capacitance I could find available online. Everything else was a compromise in some regard (mostly dv/dt or ESR) or expensive, either by way of price, or the fact it was not stocked. I stand by the caps I went with, they have taken lot of punishment from my SGTC without ever getting warm, even when badly out of tune.

I do not doubt the caps S_L has posted are better though, if you can find them available.
 
Joined
Jul 4, 2008
Messages
2,499
Points
113
Just an update; I had gone with the CDE 942C20S47K-F 0.047uF 2kV caps which have hig dv/dt and good (low) ESR for my MMC shortly after I last updated the OP. They were both the best in terms of reliability for TC applications and best price per capacitance I could find available online. Everything else was a compromise in some regard (mostly dv/dt or ESR) or expensive, either by way of price, or the fact it was not stocked. I stand by the caps I went with, they have taken lot of punishment from my SGTC without ever getting warm, even when badly out of tune.

I do not doubt the caps S_L has posted are better though, if you can find them available.

I bought them direct through CDE. I actually bought $400 caps. They threw in some extras.
 




Top