Shooting ping pong balls at Mach 1.2


Next time you’re in a Nerf gun battle, you better hope you’ve got this absurdly powerful ping pong ball gun. It shoots common celluloid spheres at over 400 meters per second, or Mach 1.2.

This ping pong gun is the work of [Mark French], [Craig Zehrung], and [Jim Stratton] at Purdue University. As you would expect, the gun is powered by compressed air housed in a length of 3 inch schedule 80 PVC pipe. One end of the pressure vessel is sealed with a PVC end cap, while the other is closed off with a doubled up piece of duct tape to contain the pressure.

The interesting bit of the build is a de Laval nozzle between the pressure vessel and the barrel. Just like a rocket engine nozzle, this bit of machined PVC compresses the air coming through the burst duct tape seal and allows it to expand again, propelling the muzzle-loaded ping pong ball at supersonic speeds.

The guys have written a report on their gun, you can grab that over on arxiv.

48 thoughts on “Shooting ping pong balls at Mach 1.2

  1. As you would expect, the gun is powered by compressed air housed in a length of 3 inch schedule 80 PVC pipe.

    As you would expect, this device should NOT have been made out of PVC. It’s like HaD actively encourages this.

    1. Do you really think that three professional mechanical engineers would use pvc if it wasn’t safe? I am quite certain that they made the appropriate calculations to ensure safe operation.

        1. Although pressure rating varies by diameter, sch 80 pipe, even PVC, has a publish WOG (Water, Oil, Gas) rating, so there is known values for compressed gases, in this case compressed air..

    2. Oh, great, another HaD comment complaining about PVC. If you want to see PVC replaced as the material of choice for air cannons, include some (any) information on materials that are a better choice. If you really want to see a “safer” air cannon, you’d find a material that’s as easy to source as PVC, build a cannon with it, and submit it to HaD.

      You’ll undoubtedly find that there is no drop-in replacement for PVC in simple, cheap, air cannons. So, what you might want to do, is build something that can fit over a PVC pressure tank, out of a material that is capable of containing any explosion.

      It’s all for nought, though. PVC air cannons have filled the minds of generations of tinkerers, and aren’t going to go away any time soon.

      1. So in your mind, the only people qualified to point out an obvious, fairly serious, and commonly ignored safety problem are people willing to completely rebuild an unsafe project in a safe manner?

        If I’m at your house and I point out frayed electrical wires sticking out of the walls, you’re going to ignore me unless I can tell you exactly how an electrician would fix the problem?

        Yes, it’s great to say “X is unsafe, use Y instead” but often the *most* important thing is to point that X is dangerous before someone gets hurt because they’re ignorant and/or cheap and/or stupid.

        There’s no “drop-in replacement” for cheap, simple black powder bombs, either, but that doesn’t mean it’s a good idea to build one because you read about it on the Internet.

      2. Even regular steel plumbing pipe and fittings is probably safer than PVC. I’d imagine that it’s much less likely to shatter into shrapnel in the event of a failure and probably has a much higher pressure rating.

      3. I’ll just leave this link here:

        Also, when someone I knew was making a potato cannon out of PVC, he wrapped the pressure chamber with several layers of fiberglass reinforcement. Pretty simple to do, and at least improved the safety. Though the other thing I convinced him to do was make sure to always fire it from a considerable distance away.

        1. So I read these and watched the kid.

          Three are about the same event, where the lead guy glued a joint, and said to wait a day to test. Two other guys ignored that and tested it an hour later. Nothing said why the PVC pipe burst, or where, or the size of the pipe.

          Another had a PVC pipe burst near welders working on something different. Nothing on why the pipe burst – ie did a spark hit it?

          So far, 4 of 4 have zero real information other than a PVC pipe burst. Zero technical analysis reported.

          The last is at least two kids (and an unknown holding the camera) doing something really stupid – putting black powder in a plastic pipe and lighting a fuse. Actually, it is hard to determine if it was the plastic pipe, or the metal pipe on the top.

          I would need to research PVC pipe, but the pressure part is schedule 80; the barrel schedule 60. They paid attention to material strengths.

          Yes, taking materials to their failure point is dangerous. Shucks. If you are going to post “this is dangerous”, at least have real engineering data to back up your statements. That is why it is called science. Not that such things concern lawyers.

          1. Not to mention they’re engineers. At Purdue. Give a little credit where credit is due!

            Hack Man, where (and what) are your credentials? I trust these guys know full well what they’re doing. I agree with bandit that, without a detailed study, we can mostly size up your evidence as acts of blatant stupidity. ;)

  2. For a commercially available version, check out Airburst rockets. You can see a youtube video here:

    They go shockingly high. Not mach 1.2 but surprisingly fast and high for something powered by a bicycle pump. No de Laval nozzle, but I am tempted to machine an adapter with a nozzle to see if I can get better speed.

    1. The reason for using diaphragms like that for firing such a thing is that it opens effectively instantly. If you used a ball valve instead, the ping pong ball would already be blown out of the gun (at a much lower speed) well before the ball valve finished opening. If you want to control the firing time closely, you use a second tank that will fill the pressure chamber quickly, and use a valve of some sort there.

      This whole system is very similar to the technologies we used for hypersonic testing in the engineering lab I worked in many years ago. The basic device is called a “shock tube”, because it sends a supersonic/hypersonic shock wave down the tube when the diaphragm bursts.

      IIRC, we used various forms of acetate and mylar for the diaphragms, but it’s been 15 years and I don’t remember that clearly. You want something that bursts at the right pressure, in any case.

  3. Ok being the horrible nerf nerd that i am usually i would condone the use of pvc mainly because the pressure needed to propel a foam dart isnt even close to the maximum amount of air pressure pvc can withstand but they are operating at 620 Kpa (not exactly sure about the unit of measurement but if thats kilograms then they are lucky it hasnt burst yet) for something like this i would highly suggest machined aluminum.

  4. Nice!

    I’m wondering how much energy they’re losing by blowing the seal at the end; the professor claims the air that slips past the ball is compressed and that is what causes the diaphragm to burst (implying the ball does not actually hit the diaphragm), but this compression in front of the ball must also exert a force on the ball, decelerating it in the process, because the pressure is presumably higher than the pressure now behind the ball.

    I’d think an possible way to improve this would be using some kind of material that isn’t actually attached to the barrel, but only held in place by the vacuum, so that it doesn’t require any energy to release. This wouldn’t be a complete solution, of course, because the seal has to either move out of the way (good luck with that), or be punctured by the projectile, which pretty much doesn’t change anything compared to the current solution.

    But maybe they’re just not losing a lot of energy in this diaphragm.

    As for the “trigger” diaphragm, they could use a stronger diaphragm and puncture it mechanically, but they probably don’t care about this bit level of control.

  5. Showing something this cool, and showing how to build it… Too bad for the disclaimer, I understand how dangerous this is, but its soo cool and easy! How can I use this to launch small rockets and still be safe? Both vacuum and compressor are easy to do with a bicycle pump.

  6. It is impossible to go past the speed of sound in a medium while using that medium as a compression based propellent. By definition the speed of decompression is the speed of sound itself, therefore the maximum speed you can achieve compression is somewhere less than mach 1.

    You can think of it as the ‘c’ of compressed air.

    Essentially I am politely saying that they need to spend more time in class and less time in the back yard. Though I will admit I tried doing this years ago but had the savvy to do the math first.

    1. Read here:

      On de Laval nozzle (the type of nozzle used here):

      “Its operation relies on the different properties of gases flowing at subsonic and supersonic speeds. The speed of a subsonic flow of gas will increase if the pipe carrying it narrows because the mass flow rate is constant. The gas flow through a de Laval nozzle is isentropic (gas entropy is nearly constant). At subsonic flow the gas is compressible; sound, a small pressure wave, will propagate through it. At the “throat”, where the cross sectional area is a minimum, the gas velocity locally becomes sonic (Mach number = 1.0), a condition called choked flow. As the nozzle cross sectional area increases the gas begins to expand and the gas flow increases to supersonic velocities where a sound wave will not propagate backwards through the gas as viewed in the frame of reference of the nozzle (Mach number > 1.0).”


    2. Read the references (if you have access). This is all explained in quite some detail. Despite what you’re saying, they show measured results (for real, not doctored, not made up) that demonstrate a velocity of mach 1.23. You can’t just “make stuff up” and then publish the results on It doesn’t work that way.

    3. If you couldn’t be bothered to notice there was a vacuum in the pipe the ping pong ball was in. Someone did this experiment in my sophomore physics class with an evacuated pipe and no pressure vessel (just 1 atm of pressure) and got near c. Try including near 0 atm of resistance next time you are doing your math first.

    4. Umm … they *measured* the speed of the ball. That means they setup a calibrated instrument and determined empirically (fancy word for test in the real world) the speed at 1.23 mach. They were able to do so *because* they got the airflow to go hypersonic speeds via the nozzle.

      Go watch the video.

      So basically, even though they have the real data, you are calling them a liar. Either they did or did not shoot the ball over mach 1.0 – cannot have it both ways.

    1. Terminal velocity is the velocity that an object falls at under constant acceleration once fluid resistance and acceleration have been balanced; ie drag forces equal the force due to gravity. This is based on the drag coefficients, weight, density of the fluid(air), and gravity. For a ping-pong ball it’s about 8.6 m/s.

    1. Their work is quite different from a light gas gun.
      Firstly it doesn’t use light gas. Light gas guns use light gas because the maximum acceleration compressed gas can yield is the speed of sound*. Light gas guns get around this by using a lighter gas than the fluid they are firing into; ie compressed hydrogen or helium into regular air. Light gas guns also differ fundamentally by not having divergent nozzles. They launch a piston towards a cone that narrows to the bore of the gun.
      The work done by the Purdue team stores compressed air in a tank. When the burst disc ruptures the air flows through a restrictive nozzle whose minimum dimension is -smaller- than the bore of the gun. The nozzle then diverges to bore diameter. De Lavalle nozzles work by creating a supersonic shockwave at the focus of the nozzle and then through nozzle geometry directing the shockwave out the nozzle and in this case down the bore of their gun. The ping-pong ball isn’t launched so much by pressure differentials as it is pushed by a shockwave.

      *De Lavalle nozzles circumvent this stipulation by rapidly compressing the working fluid causing it to heat up simultaneously creating a supersonic shockwave which is focused out of the nozzle.

  7. Pray that no one nearby thinks to try steam as a power source for a gun, mwhahahaha!
    It’s one of the few cases where the power source is way more dangerous than the actual slug. It would be cool (hot?) to be able to run a gun all day without gunpowder, though. I suggest titanium or magnetic steel instead of the more brittle anmagnetic steels or heaven-forbid,.aluminum. A pressure cutoff and several other mitigating features could work as a safety system, not that I’m recommending that someone play with this without test dummies in concrete bunker isolated with cameras and remote controls. It’s a bit impractical anyways, but hey someone probably will invent one just because they can so might as well tell them how to avoid becoming steam-cooked and chopped-hamburger. ;)

    Well, the danger with steam is actually not so much the hot gas, as the chance of it going from supersaturated to water in a fraction of a second leading to overpressure. There’s a reason that coal plants are considered way more dangerous than nuclear plants to work at and nuclear submariners worry more about the water pressure than radiation from the generator on board… If you can invent a man-portable pressure vessel able to contain water that would normally take up gallons of space, in a quart, then you should probably be in the submarine/spaceship business.

    On a related not, if you COULD use steam this way safely, think of how much energy it has. It’s actually a pretty neat idea if made practical. We’re talking 150PSI without using an internal combustion engine or similar technology. Think of how much work could be extracted by the time the steam expanded to match the surrounding air pressure. There are in fact vehicles that use water turning into ice as a power source near the Arctic Circle and they have no where near the expansion of steam.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s