MMC layout, 630 or 1600V capacitors?

[AndreiRS]

Ok I kinda gave up on the home made capacitors, they are not working.

Found some 630V and 1600V capacitors on a store. I could use 14 0.1 uF 1600V capacitors in series, and 7 strings in parallel. Or I could use 44 2.2uF 630 on a single string. Both would give around 50nF. But one layout is just a big series connection, which sounds a bit bad to me, and the other has parallel connections too. Price would be around the same on both cases.

These caps are not those meant for high pulse, but for 2 seconds I think they could work without exploding.

[Mads Barnkob]

You should definitely, of the two choices, go for 14x 0.1 uF 1600V capacitors in series, and 7 strings in parallel.

One long single string of 44 caps could have problems with bad voltage sharing, high combined ESR and low current capability.

[AndreiRS]

Nice, thank you Mads. I will ordem them next month, I just hope they work. They are not those capacitor with big leads, but actually the thin ones.

These, 20 or 25mm.



I was looking for these, but they cost around 10 dollars each.



980 dollars is a bit out of my budget.

[Nijin]

To build a good MMC, you can use 942C capacitor serie from Cornell Dublier or C4AS serie from Kemet. For my DRSSTC I use C4AS 680nF 1200V capacitor, they have a better Irms in 4 pins configuration than 942C.

[Hydron]

Are you planning on using these in a DRSSTC, or a spark-gap coil?

The 942C capacitors mentioned are very good, but are rather expensive and unnecessary for a DRSSTC MMC (which are probably less stressful on the capacitors).

[AndreiRS]

On a rotary spark gap. I don't have how to buy these ones. Our mail system here is being privatized or something like that, and they are not delivering international stuff in time. I will just go with the cheap ones.

[Hydron]

Having 7 strings in parallel will help a lot with current capability, even with the cheap ones. You'll just have to try it and see!

When I first ran my DRSSTC I used a horrible MMC made out of 5 different types of capacitor while I waited for the mail to deliver the good ones - it got a little warm but worked well enough to test the coil.

[AndreiRS]

I don't plan running it for long periods, some 5 seconds "bursts" would be good already.

[AndreiRS]

Maybe tomorrow I can test it. I got the parts today and already assembled the mmc.

[profdc9]

Looks good.  What kind of capacitors are those, polypropylene, polyester, or ceramics?

Dan

[AndreiRS]

Hi Dan, thanks. They are polyester. But I don't know much more than that because I couldn't find the datasheet. It has a logo that is kind of a circle and says HBC.

[Mads Barnkob]

I think they are comparable with the old and much used Panasonic capacitors, you can find some in the list here, also with datasheets available: http://kaizerpowerelectronics.dk/theory/good-mmc-capacitors/

You should make a center tapping point at the end of the MMC where all strings have a even length wire to it, like you have it now the two strings closest to the connection point can conduct up to 40% more current than the string furthest away from the connection point. Inductance matching between strings is very important in parallel constructions where current sharing is critical for component temperature rise / life time

[Hydron]

Hmm, I would have thought that the impedance due to the capacitive reactance would swamp the effect of lead inductance - I'm not sure I agree with Mads on this one and personally wouldn't be concerned with the lead length.

[Mads Barnkob]

Hmm, I would have thought that the impedance due to the capacitive reactance would swamp the effect of lead inductance - I'm not sure I agree with Mads on this one and personally wouldn't be concerned with the lead length.

It is not the lead length of the capacitors, but the black wire ring at the end of the MMC that should be remade to have even inductance for all strings.

[Hydron]

I still think it will have close to zero effect - the black ring will be <1uH which is tiny in comparison to the reactance of the capacitors at TC frequencies. For example at 500kHz (a reasonable guess, if anything on the high side which makes the lead wire effect bigger:

14*0.1uF in series = 7nF, reactance at 500kHz = ~45 ohms per string of caps

With an estimate of 1uH for the black wire(generously large - it's likely a quarter of this or less from the geometry) gives reactance of ~3 ohms @ 500kHz, or less than 10% of the capacitor reactance.

While there will be some effect from different connection points, I don't think it will be enough to change much at all - certainly won't cause enough difference in current to have an effect on current sharing/heating.

[AndreiRS]

Thanks for the link. If I'm lucky it will be close to the biggest panasonic capacitor on that list.

Hmmmmmmm about the leads. I think it will work that way. But maybe I can save them from premature explosions by changing the layout. Not that difficult to change it. It is already 7pm here, tomorrow I will do some testing.

[AndreiRS]

lol

Turned it on. Working good for half a second, lots of capacitors get destroyed, turn it off.

[Hydron]

Sorry to hear it blew up, but good to hear the 1/2 second was OK! I would recommend trying polypropylene capacitors rather than polyester next time - polypropylene is much better for this use.

[AndreiRS]

I don't have where to buy them now around here. Maybe in the future. Thanks for the advice.