Weaponizing Elementary Science Experiments

[austiwawa] was playing around with one of those simple linear motors people build as friendly little science experiments. There’s an AA battery in the middle of a set of magnets. When you put it inside of a spring it zips around inside until you run out of spring or magic pixies in the battery.

Of course, the natural question arose, “How do I make it go fast!? Like fast!” After making explosion and woosh noises for a bit (like any good hacker would) he settled down and asked a more specific question. If I made the coil the barrel of an air gun, and then shot the battery out… would it go faster?

So, he built an air cannon. It took some ingenuity and duct tape, but he managed to line the barrel with a copper coil. After that he built an experimental set-up, because making something dangerous is only okay if it’s science. That’s the difference between sensible adults and children.

He shot three “dead” rounds through the cannon, and got a baseline result. These dead rounds were made so by placing the magnets at the improper polarity to forego the motion-boosting properties. Then he shot three live ones through. It went measurably faster! Neat!

What’s the silliest thing you’ve ever seen properly characterized? Let us know in the comments below.


31 thoughts on “Weaponizing Elementary Science Experiments

  1. For people who want some solid info but don’t want to go through the video:
    “dead” shots: 129FPS, 118FPS and 112FPS
    “live” shots: 143FPS, 150FPS and 134FPS

    A bit too high variance on the shots, supposedly with identical rounds and the same pressure, but there’s definitely a difference with the “live” rounds. I wonder if the “dead” rounds have any braking effects too.

    1. Unpaired student’s t-test gives 97% probability of difference. 98.6% if you use one tailed. Having said that, I am having a hard time believing that a AA battery can add that much kinetic energy during the fraction of a second that it is able to exert its oomph effect, i.e. while traveling down the barrel. I don’t know how much kinetic energy the “dead” shots have, but it has to be a fair amount. The “live” ones have almost 50% more kinetic energy. AA batteries just aren’t able to output that much power.

      1. I get about 60J for the ‘live’ shots and about 40J for the ‘dead’. I made lots of assumptions (batt_mass = 23g, mag_mass = 6.79g * 6, velocities are averaged). So, what you said, about 50% more kinetic energy. I wonder how this would scale? For example, what about 4 magnets on either end? What about bigger batteries? Could we switch the coil and the magnets and fire a copper coil-slug?

        I don’t have time now, but this might make a fun weekend project.

        1. I find it intriguing that the battery in a system without air hardly comes out of the coil, but with some air it suddenly is responsible for an increase of 20J. 20Joule is a lot of energy and if this system is capable of pushing it out of a single 1.5V battery in a tenth of a second, well then I guess we have to nominate the designer of this contraption for a Nobel prize.

          I don’t want to be negative here… but this simply can’t be right.
          If we never here from this project or it’s results again well, that can mean 2 things… The experiments and it’s calculations suck OR the experimenter has been kidnapped by a weapons manufacturer (or the government) and is working in a high security top secret environment and is not able to publish any results.
          In both cases… not good. Anyway, because I didn’t want to be negative I guess I have to say, please take care, be scared of vans with darkened/blinded windows. Never so out alone and always carry a cellphone to call for help, and please be careful.

          As a reference: http://hypertextbook.com/facts/2001/KhalidaNisimova.shtml
          Although far from perfect, it gives you an idea of the energy stored inside a battery and batteries can only gives their max. amount of energy is spread over a long period of time days… not fractions of a second.

        2. Neglect relativistic effects. The independent speed of the dead magnetic bullet is roughly 120fps. The live bullet is roughly 140fps. Separately a live magnetic train went down the coil on it’s own at about 2fps. This should really be simple additon, shouldn’t it? The numbers on the trials are too far apart to make sense.

    1. Yeah, this is a mess. The test order isn’t randomized, the experiment is wildly uncontrolled, and just turning the magnets on one end around doesn’t make it a control case – it introduces new possibilities of error.

      I guess for me, the kicker is to look at how slow it is moving through the wire in the beginning. Looks like maybe 1 FPS. How does that suddenly get amplified to an increase of 20 FPS in the air gun. Doesn’t pass the smell test at all.

    2. The magnetic train, by itself goes down coil at about 2fps. The average of the first three trials (“defective”) is 120fps. The average of the second three trials (“effective”) is roughly 142fps. Shouldn’t this be simple addition? On it’s own the battery should add only about 2fps. Instead we see 22fps increase. Something smells.

  2. “[B]ecause making something dangerous is only okay if it’s science. That’s the difference between sensible adults and children.”

    At what point does this site start to incur liability when they KNOW the dangers of builds combining PVC with air cannons but they dismiss it or actually openly mock those very well known and very real risks?

    1. Calm down chicken little.
      It’s a site whose aim is to deliver hacks, use of materials outside of their designed use. Every hack has the implicit disclaimer that trying to replicate it may risk property and person. Everyone knows PVC isn’t pressure rated for air, no need to constantly nag everyone that they risk their safety by doing this.
      I’m gonna need you to turn in your Safety Police badge on your way out,

      1. Hacker or not, people are still subject to the laws in their area. Some politicians are so afraid of anything that even looks like a gun that there are laws on the books concerning anything air powered (like BB guns).

        From Wikipedia: “Two states (New Jersey and Rhode Island) define all non-powder guns as firearms; one state (Illinois) defines certain high-power and/or large calibre non-powder guns as firearms; three states (Connecticut, Delaware and North Dakota) define non-powder guns as dangerous weapons (but not firearms)”

        Yes, this is the world we live in — some people are so busy trying to make people safe that they criminalize nearly everything.

        So, making stuff is cool, but making stuff in prison is far more challenging. Everything depends on the definition of “non-powder gun” in your particular area.


        1. I dont see how using gun powder to propel a small mass at high velocity is different to using any other method to propel a small mass at high velocity. The end result is exactly the same so why should they not be regulated the same way? Obviously things need to be tiered based on muzzle velocity or projectile kinetic energy but the method used to produce those factors is unimportant.

        1. So the morale is to not drop a massive weight from a very high height on a very cold tube of PVC?
          What if it’s like summer and I’m pressuring it to 40psi and not impacting it?

    2. Have you ever sat down and, for yourself, done the math or even tested the dangers of compressed air in PVC? Judging by your comment, you have not. That is not to say that there is not danger there, but not nearly so much as is claimed by some.

      Also, note that this is HACKaday, not INTENDEDPURPOSEaday – people are here fore the hacks, the misuse of existing stuff to do even better stuff than was thought could be done with it.

      1. Actually, yes. Plenty of people have “done the math” and tested it. In part due to the litigation risks and in part because of known history of failures. We know pretty well how PVC fails under air pressure. It’s been known to industry for well over 40 years now!


        Plastic Pipe Institute, in its Recommendation B dated January 19, 1972, recommends against the use of thermoplastic pipe to transport compressed air or other compressed gases in exposed plant piping.

        Charlotte Pipe: Testing with or use of compressed air or gas in PVC / ABS / CPVC pipe or fittings can result in explosive failures and cause severe injury or death.

        GF Harvel: “GF Harvel’s PVC and CPVC piping products are “rigid” thermoplastic materials. As a result, GF Harvel does not recommend the use of PVC or CPVC piping products for the testing, transport, or storage of compressed air or gases.

        Warning! The use of rigid thermoplastic plastic piping in compressed air or gas applications can result in severe bodily injury or death.”


        I am not advocating against hacking parts and things to discover novel applications. I am however advocating that people utilize ANY OF DOZENS OF KNOWN SAFER MATERIALS when doing so with respect to compressed air containment. That is all.

        I also am calling out the writers and editors for deliberately stirring the pot despite the known dangers that this usage pattern entails in order to drum up more controversy and thus more site traffic.

        1. one shouldn’t use any brittle material for pressure where one can avoid it, it is the failure mode that is dangerous as far as i know and that is why if one took the proper precautions or had a sufficient safety margin there shouldnt be a problem, i am surprised to see ABS on your list considering that ABS is frequently used in pressurized systems, http://www.epco-plastics.com/abs_technical.asp.

          duty cycling and wear is something to consider but that is true of anything.

  3. I’d like to see an additional control: Dead battery, or otherwise broken circuit. Even if your magnets are incorrect, moving your magnet through an inductor might be causing some braking due to an induced current. Maybe put a piece of tape over one of the terminals. This braking force is powered by the air cannon, so it has the potential to be more powerful than your motor/train. (unless I’ve got this all wrong).

  4. +40% increase in kinetic energy! We’re talking about what, ~50J of energy in that battery?
    The battery’s ESR would surely limit it’s energy output to 5W. 5W over 10mS is about 14uJ?
    Is that correct?

    1. Yes. Better way to look at it is that the battery would have to supply >5kW over 10mS to increase its speed by that amount. Speed increase is definitely not because of the battery.

      1. I betting the magnet being forced through the coil (circuit completed by the battery) is actually causing drag in the control group. This effect would be reduced in the “correct battery” group. What he is likely measuring is the difference in braking force.

  5. Should test “shuttles”: “normal”, with magnets installed normally, one battery set reversed (neutral), and “normal” shuttle reversed in the tube.

    Also, silly thought maybe… instead of just wrapping the copper around a “dowel” or whichever, wind it around a nylon / FRP threaded rod of suitable diameter so the distance between adjacent wraps is more consistent.

    Also, maybe try tinning the wire its entire length with a lead-based solder to take advantage of the softer lead content and its lubrication effects.

    Finally, don’t fire into towels. Maybe instead fire into an enclosure of suitable depth and filled with “packing peanuts”, or other suitable, deformable packing… loosely packed shredded newspaper maybe?

  6. This needs to be extended into a significantly larger matrix of samples (tests).
    Assuming the same projectile…
    Using –
    – a repeatable trigger valve (solenoid)
    – different winding spacing (distance between turns)
    – different wire gauges (current drawn vs generated field strength)

    Another step could be looking at manufacturing your own ‘cartridge’ – to vary the magnetic effects on the body of the payload etc.

  7. As a check, I’d like him to make an additional three trials with the “effective” bullet, but put BACKWARDS downs the barrel. As a first order apporximation, I’d expect the velocity to show a substantial DECREASE from the “defective” speeds he recorded. (Similar in magnitude to the INCREASE in speed that he attributes to the “effectivie” magnetic bullet over what he terms a “defective” bullet.) Substantially, the major velocity of the air shot, being opposed by the minor velocity of the magnetic bullet action.

    I’d also like to see additional trials with a completely dead battery with with the magnets in the “effective” and “defective” orientations. Or a completely “open” circuit battery.

    With that said, I agree with others that the amount of additional kinetic energy that he thinks the “effective” bullet is gaining from the battery is inconsistent with the capability of the battery and the time spent in the barrel.

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.