Caps In Series

[B-BOMB7]

I want to build a coilgun and have come up with a setup for my caps that I think will work but I wanted to make sure before I wire it all together. I was thinking of putting a 35,000V .003uf Maxwell Pulse Cap in series with a 20V 1farad Speaker System Cap. Using the Product Over Sum method for two series caps, .003uf X 1000000uf/.003+1000000, I found that the overall rating for my caps would be 35,020V 333.33uf with an output of, .5 X .00033333 X 35020 squared, or 204,398.0227 Joules. I feel confident that I did this part of my calculations correctly however, I'm not quite sure what my charging voltage should be and I've been having some trouble finding this out so I was hoping you might be able to tell me. Also, if you've found anything that you think may be wrong with my calculations, I be happy to know what they are!

Thanks,
Bryan

 

[fadY]

It seems to me that 0.003*1000000/(1000000+0.003) is more like 0.003µF and not 333.333µF...

 

[bogdanfirst]

hey, with 200J you can kill averybody on this forum......check the multiplers and submultiplying calcs.

 

[mechie]

Series Caps

I think there is a fundamental flaw in your idea that one high voltage cap and one high capacitance cap will give you the best of both worlds ...

If you charge two caps in series the smaller one will charge (and discharge into the load) faster than the other. The power will therefore be a sharp spike from the pulse cap, followed by a-forever decay of the 1F cap.
Given the power storage of the 1F cap I wouldn't dare let it see more than its rated 20v unless it were in another country :shock:

If you try to charge caps in series you should put a resistor in parallel with each so there is a DC path to all the caps (a voltage divider), a MegOhm or more is OK, just to equalise the voltages.

The pulse cap is designed to cope with fast discharges, the 1F unit won't be - this is important - RIPPLE CURRENT - check its rating :!:

Can you make the coilgun low enough impedance to run from the 20v available from the 1F cap? thought not.
You are stuck with banks of similar high voltage caps in PARALLEL to give the capacitance at a workable voltage.

Sorry :cry: