The Angular Momentum Roller

angular momentum rollerI had a thought a while ago that a device could be built with no external moving parts controlled by conservation of angular momentum.  Angular momentum is the tendency of an object to stay at rest in a circular direction if no other forces act on it.  It’s why a large spinning wheel is hard to stop and why a skater can spin much faster by pulling her arms in.  It’s also why you don’t start spontaneously spinning for no reason.  However, if, say for instance, you had a large wheel that started to spin very fast because of a motor or other reason, your body would have the tendency to rotate in the other direction.  This is the whole premise behind this “vehicle”.

My first experiment with this was a strange and fairly crude object that I called the “angular momentum device”.  Although the science behind it was rather dubious, I had hoped to induce a spin about the center axis by spinning the two wheels on the sides.  In the video, it seems to work, but I think it was more a matter of vibration.  The idea was that something like this could be caused to turn by accelerating both masses in different directions.  It didn’t work with the “roller”, and after more thought on this, the “science” behind it doesn’t really make sense. If spun in opposite directions, each wheel’s angular momentum would cancel the other one out. Nonetheless, for some reason, which I’m not sure of, it seems to work:

After this seemingly successful experiment, I still thought some sort of locomotion was possible using angular momentum.  After some thought, forwards and backwards seemed likely, but turning less so.  So I started building the strange item that you see in the pictures below.  This was pretty much thrown together, so I won’t lay out the details right now, but basically it has a PVC pipe frame, masses and wheels that I made with a hole saw out of 1/2 inch plywood, a small radio control receiver, and two motors out of an automatic air freshener – I used the sense-and-spray version as they were cheaper and have similar motors as the other kind. Apparently everything I make these days has air-freshener parts in it somewhere.  Update – when this device was made, I didn’t have my Zen Toolworks CNC router at the time, but it would have made the “wheels” much easier to make.

pvc cutout with motorspvc with wooden wheelroller with RC

The results of this build were pretty neat.  The motors spinning up did cause the device to lurch one way or the other as I had hoped.  The device isn’t perfectly balanced, so it tended to start rolling faster when the heavier side was up. This usually meant it was already coming down, so reversing the masses during this condition caused this burst of speed. This looks pretty cool in the video below, so I’m not too disappointed with how it turned out.

roller cutout

One can’t get continuous motion with this device as once the momentum is balanced, it tends to roll to a stop.  If, however, the masses weren’t balanced as they are, this may be possible.  I milled one end off to make them non-concentric, but so far this hasn’t worked all that well.  I believe a better control system is needed for this to work correctly, and probably more weight.  Here are the results of what I have so far:

In the video, I put it in a tennis ball container to try to show that it wasn’t powering itself by touching the ground with the masses, although one wheel still sticks out.  The goal would be to get a device like this to turn on it’s own as well as go straight, but that would require some clever mechanics to keep everything in the right plane.  I think it would actually require another twisting mass parallel to the floor.  There is definitely a solution (and I have a few ideas of how to do it), but that is for another post.

Oh and I’m proud to say that, as I own the world’s most advanced tripod mount for the ENV2 phone, I now own what I believe to be the only remote control tennis ball sleeve in the world. Feel free to correct me on this if you own either one.

[ad#Google Adsense-text leaderboard wide not home]


  1. Angular Momentum Experiment - Hacked Gadgets - DIY Tech Blog - pingback on March 11, 2011 at 8:08 am
  2. Robot gets around on lopsided wheels - Hack a Day - pingback on March 18, 2011 at 7:00 am
  3. Hi,

    Nice idea.I was able to follow everything in this post till the point where the chipping of the flywheels would help to perpetuate the motion forever .Could you please explain this part so that I could get a better scientific picture.

    • Thanks for the response! I suppose I didn’t really explain that part very well. Maybe I didn’t even think it through very well before I hacked the wheels up. Part of my thought process was that if there was some vibration it would kind of sling things forward. Now that I think about it more, there would have to be some sort of controls on this to start and stop the motors at just the right time.

      The other thought that I had about using a offcenter mass is that if the offcenter mass could be controlled correctly, it could actually drive it continuously because one end could always be made to be heavier than the other. I think this is more realistic, and I have some plans to demonstrate something similar, but a better control scheme would be needed to do that for this device.

  4. Appreciate your prompt response.Thanks

  5. Hi,
    Nice plan. I was able to shadow everything in this post till the tip where the chipping of the flywheels would help to be responsible for the motion forever .Could you please explain this part so that I could get a improved technical image.

    • Hi, glad you enjoyed it. As far as put angular momentum goes, I don’t really think it helps – more of an experiment. I think it could help if there was a better way to control it though.

      If both the side wheels could be coordinated, one side could be made heavier than the other causing it to roll continuously (not that I’ve proven this). I’d like to try something like this, but haven’t gotten around to it yet.

  6. Nice idea as far as i learn this site it’s really help me more to of an experiment..Thank you:)

    • Thanks! If you ever come up with any experiments from the stuff you see here, feel free to send me some pics or a link and I’ll put it up!

  7. The Creepy Glowing Remote Control Box | JCOPRO.NET - pingback on May 21, 2011 at 11:02 am
  8. Great idea for this experiment and a very detailed report.
    In the first experiment I was wondering why you thought it was vibration effecting the spin?
    I did not look like that from the video.

    • Yeah, I didn’t think so at first either, but if you look at the mechanics of everything, the two rotating masses should be cancelling each other out and inducing a “twist” on the center of everything. If they were parallel to the ground instead of perpendicular, that would be a different story.

      I might be wrong, it’s been a long time since I’ve taken physics. It does seem to actually be spinning though, so maybe there’s another explanation.

  9. Thanks for answering that.
    I do think if it was perfectly balanced you might get a better outcome, or at least be able to tell if vibration has anything to do with it.

    • You’re probably right.

      I had some other ideas that would use this principle, but for better or worse there’s always something else on my project list!

  10. Great use of air fresheners, very creative thinking!!

  11. i like this experiment this actually we want to use as our experiment in school project in physics and we will do our project proposal but we couldn’t find purpose of this experiment. if you would like i want to have that immediately and its procedure so that we will know that its successful experiment.

    I was expecting your answer by tomorrow thank you:)

    • There wasn’t really much of a purpose, just to see if it could be done. Maybe the purpose would be to see if a vehicle could be made to roll with only its internal angular momentum?

Trackbacks and Pingbacks: