My What A Large Capacitor Bank You Have

[Daniel Eindhoven] put together this 11,344 Joule capacitor bank that he says would be perfect for weapons such as a rail gun, coil gun, or electrothermal-chemical gun. He machined a couple of aluminum plates to act as a positive and negative bus. The two are separated by a denuded sheet of PCB (making us wonder how he got the copper to peel off like that). Once charged there’s the little problem of how to discharge the system without getting bit, which [Daniel] solved by building a pneumatic switch. We didn’t find the test-fire footage very interesting but we did embed the demonstration of his switch after the break.


[Thanks Kurt via Hacked Gadgets]

43 thoughts on “My What A Large Capacitor Bank You Have

  1. This is precisely the kind of project I should never attempt. I almost killed myself fixing an old TV once…I should not be allowed around capacitors.

    Awesome power, though. I want to see it zap something.

  2. The bounce from the pneumatic switch worries me a bit. I think a good addition would be to notch the sides of the copper block, and to have some sort of spring loaded clamp that catches the block. The arms on the clamp could be electrically conductive, or just made out of teflon.

  3. @Wes: I have the same problem in the latest version of Ubuntu/Firefox… haven’t dug into it, but it works on windows and Fedora

    @ReKlipz: Yeah, I’d agree the bounce is a bit problematic. I don’t see why he’s using a pointed contact though… He’s got a fairly precise, well-aligned mechanism there, it we seem to me that he should use all the surface area he can.

  4. Oh… if you’re reading this Daniel Eindhoven, I’d also not use aluminum as a contact. Fixing it could be as simple as adding a small piece of copper to the end where you make contact.

    Al Oxide (Al2O3 I think?) is a pretty good insulator. You’ll still make contact, especially with that much contact force, but your switch performance is probably going to be a lot worse than it should be.

  5. Aluminum capacitors shouldn’t be put in short circuit as the leads (and the connection to the inside sheets of the capacitor) are not designed to manage the immense current. Some capacitors are manufactured for high current mode, like those for xenon flashes, even for those direct short circuiting should not occur.

    Besides that, the pneumatic switch is nicer than useful, there is no reason to minimize the contact point, on the contrary you should try to make it as big as possible, and avoid the bouncing at all, as it will generate more arcs and reduce the switch life dramatically. The contact materials are also a bad choice.

  6. omfg this is the shit I don’t want to see. When HaD takes a turn, I turn to HackedGadgets. It’s like Engadget redirecting to Gizmodo…
    HackedGadgets is more of an original “Hack a day” than HaD is. They may only post one hack a day, but at least their shit is original. Another example of HaD starving for content.

  7. It’s silly projects like this that really make me wonder. What’s the point? There are so many other better ways to do this, if there is a point. There’s just no application that needs a bunch of crappy aluminum capacitors. It’s not even cost effective if you have to replace your crappy capacitors all the time for some application that doesn’t make since. Why aluminum? Terrible conductor.

    I’m guessing this was just for the “wow” factor, because “wow”, people don’t get it. I’d leave things that are high power, high voltage to the pros, before someone gets hurt…

  8. >Why aluminum? Terrible conductor.

    For solid blocks or thick plates, it doesn’t matter.

    And for foil capacitors: Have you ever seen devices that use anything else? (apart from antiques)

    > 0.2F
    This is a lot, but more important is the total stored energy and how fast it can be discharged (low inductance). Paralleling small capacitors is a really good means for low inductance.

  9. Anon complains “omfg this is the s**t I don’t want to see. When HaD takes a turn, I turn to HackedGadgets.” and this exact same project was posted there July 16, 2010. ROFL

  10. I was digging through my garage and found a 2500V capacitor rated at 3F. The first thing I said was ‘Sweet – I could toast some stuff with this!’ The second thing I said was ‘This could be dangerous – If I weren’t so lazy I’d think of some safety precautions’.

    Anyway, this pneumatic switch seems a bit overkill. I just used two broom handles with the contacts on the end of each. We let it rip across wet grass and it made the dew evaporate and the cells in the plants burst. IT WAS AWESOME!

  11. Bandwidth Limit Exceeded

    The server is temporarily unable to service your request due to the site owner reaching his/her bandwidth limit. Please try again later.

  12. >For solid blocks or thick plates, it doesn’t matter.


    As far as low inductance, the real trick is low ESR, which adding giant thick plates of aluminum won’t help.

    >And for foil capacitors: Have you ever seen devices that use anything else?

    Are your talking about other applications, or only devices like this? Because there are tons of other capacitors types.

  13. What’s all this every contact nonsense starting and ending with a pinpoint contact? The reality of the situation is a hideously complex electrodynamic plasma physics problem. Brush discharges and corona are a good examples of arcs not being pin point needle like things. It would be accurate to say, that when sufficient volts are present, all mating connections start and end with corona.

    Anyhow,I’m a plasma physicist so may have a biased interest in this field.

    3F at 2500V..? I think not! That would weigh about 20 tonnes. We used to use something like that to energise field coils for an experimental fusion reactor (Compass in Culham science centre).

  14. What’s up with all the aluminum hate? Aluminum beats out copper in cost, corrosion resistance, and conductivity to weight ratio. Copper wins in conductivity/volume, and solderability, but when one is dealing with large pulsed currents a purpose built solution is practically unavoidable. Soldered connections are a bad idea in this case, and aluminum is far cheaper and easier to procure than copper. If it’s not conductive enough, just make it bigger. It’s not like a transformer, where you’ve got a limited winding window.

    Although I am impressed by the low inductance construction of thew capacitor buses, the switch has WAY too much bounce. A possible solution would be to quick press two pieces of roof flashing together. Plate with silver, as it corrodes slowly and the corrosion products are conductive.

    — = aluminum flashing
    I = dielectric
    < = external connections


    Make the dielectric thick enough to avoid breakdown, and quickly clamp the plates together with padded wood blocks to fire. You could slide a spare piece of dielectric in the gap as a safety.

  15. Revised ascii art, as it was distorted in the previous post:

    – = aluminum flashing
    I = dielectric
    < = external connections
    * = air gap


  16. > the real trick is low ESR, which adding giant thick plates of aluminum won’t help

    It will. It’s called a ground plane in its simplest form.

    Also, paralleling many small capacitors is usually better than one single big capacitor, because for film/foil capacitors, ESR is dominated by the external wiring.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.