Box ‘o Bangs, A 2,180J Capacitor Bank

What happens when you wire up 16 capacitors? Sixteen 2500V 40uF capacitors to be precise… [Lemming] calls it the Box ‘O Bangs. Theoretically it outputs 4000A at 2500V for a split second.

They bought the capacitors off of eBay, and they appear to be good quality BOSCH ones, straight from Germany. They were apparently used for large-scale industrial photo-flashes, but who knows since they’re from eBay.

Soldering it all together proved to be a challenge, as once they realized just how many amps this thing was going to put out, they needed some thick wire. It looks like about 2ga wire, which, spoiler alert, still isn’t enough for 4000A — but since it’s only for a split second it seems to do fine.

Once everything was built, it was time for some scientific tests — what can we put between the leads to explode? Stay tuned for some slow-motion glory.

First up, welding a bottle cap instantly.

Now, how about a graphite pencil?

And perhaps the most interesting of all… a CCFL tube backlight module from a laptop screen… completely unexpected result!

While impressive, it’s no 24,000J capacitor bank…

35 thoughts on “Box ‘o Bangs, A 2,180J Capacitor Bank

      1. That’s why you leave things like this charged when you live in the ‘hood. A crackhead breaks into my place and sees copper wire?!? Have fun “twatwaffle”…

    1. It did solidify, it was quite amazing, it was also cold to the touch almost immediately.

      We did keep it but unfortunately some twatwaffles broke into the space a few days later and broke it in the process :-(

  1. Not sure if I should really get excited at the title (2180 J..) I’ve seen on some chinese site selling super capacitor in range of 1000 F (???) though max voltage is less than 3 volts. So a 1000 F @2.5V would store what around 3000 J ? (Yeah I know its difficult to believe chinese sites)

    1. The problem with supercapacitors (or double layer capacitors) is their internal resistance. You can’t discharge them quickly enough to get the absurdly large currents required for these experiments. They are more suitable for energy storage.

    2. A 1000F cap @ 2.5V will store 3125 J ( E=0.5*C*U*U ). The internal resistance depends on the type, see: http://www.maxwell.com/products/ultracapacitors/k2-3-series for a 0.27mOhm ESR.

      The bigger problem is that the 2.5V severely limits you in terms of what you can do and what kind of speed you can achieve. At such voltages and high currents (say, 500Amps) cabling resistance becomes a limiting factor. It’s easier (and faster) to do a full discharge of 2000J @ several kV rather than @ 2V – this is important for stuff like can crushers, high voltage pulses as was seen etc.

      1. exactly. i have a 433 farad @15 volt bank made from maxwell 2600 farad caps. with solid copper bus bars and thick cable you can make some nice sparks and gouge metal, but that’s about as exciting as it gets at this low of a voltage.

        1. That only works for very short pulses, and you need a way to trigger a pulse on the capacitor side which gets difficult to do when you’re talking about this sort of amperage.

          1. The flash tube or device under test becomes the “switch. Divide the caps into two banks with the flash in between. Parallel the flash with an inductor. It will charge through the inductor which will block the pulse and force it through the flash – after breakdown (or arc used as a switch). A plunger switch starts the voltage to drop on one bank but nothing happens until breakdown of the gas in the flash, which gives time for the plunger to get good contact, then super fast high voltage high current through the plasma. I might find a link to an example if anyone wants.

          2. Well I would like the link very much; I.am readiing.’Handbuch Tesla Experimenten’.(only in German.)
            The book starts with some theory, but contains good
            Example projects as.well

            The.concept of swichtimg large.energies. as fast as possible, s a one.the.crucial element in the projects;
            However not many differrent switching methods are shown.

            So.extra imput is welcome’!

            Kind.regard

    1. You can find large laser caps out there. They are big and heavy and often in a rectangular packages to pack into racks well. Mainly they have two features: Very low internal resistance and a geometry and materials for very fast discharge. They are not safe by any means since they are also high voltage. They were built for large flash lamp arrays – usually nitrogen in quartz tubes to get loads of UV. Maybe Xenon. There is a relatively easy to built pulsed UV laser that uses a single panel of double sided PCB plus a good big fast high voltage cap in a design that IIRC does not need mirrors.

      At Stanford a physics grad student was once killed when he was showing the labs to a girl friend and absent mindedly sat on a bank of charged laser caps.

      1. He died from shorting out a capacitor bank with his a$$?
        Darwin award or myth.
        I found one that is similar http://stanforddailyarchive.com/cgi-bin/stanford?a=d&d=stanford19751104-01.2.6&e=——-en-20–1–txt-txIN——-#
        It’s actually really hard to find causes of death; they keep quiet.
        I tried to make a cap bank with those 300V ones in disposable cameras, but it kind of just resonated and sounded strange instead of setting off 4 flash tubes like I wanted. The failure convinced me to read about HF circuits.

        1. My friend was a student of Schiff (who passed away during his time) and he was one of the founders of the Stanford Graduate Student’s Union and very interested in preventing cover-ups of this kind of thing and getting recognition for work performed by grad students working for faculty and conforming with labor law. Grad students were basically indentured servants before that. Note in the article the, investigation depth being determined by employment status! Not to mention potential insurance problems.

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.