I want to make a remote controlled ball.  It sounds easy but it's not as easy as placing a small car inside a ball because the car's wheels will drag, it will flip up side down and wabble around.  I also thought of a one wheel contraption with a bunch of support caster type wheels and the one wheel could rotate to turn but what would keep the outer fram from turning instead of the wheel?  So I came up with this design.  The opposing wheels will ensure a turn inside a ball.

I made this little two wheeled contraption with pingpong ball stabilizers.  The wheels are driven by servos modified to be continuous rotation.  See my Servo Modify page for that info.  They are mixed using a V-Tail mixer so steering is accomplished by slowing only one wheel.

I started with a hamster ball. I sanded all the riges out of the inside so it will have a smooth surface to run on.

Here it is going into the ball.

It worked okay but not as well as I wanted.  I think the stabilizaion wasn't very good.  It seemed to wedge it's self a lot and was slow.  So, I made Revision #2. It was the same concept but better stabilization and powered by geared airplane motors.  So, I had to use speed controls with each of them.  It just weged its self harder.  Back to the drawing board.

At one point you can see it turn inside the ball. The apposed wheels DID work well for that.

New Concept.  MASS DISPLACEMENT.  I figure if I can make a 2 axis contraption that is attached to the inside of the ball with a weight on it then I should, theoretically, be able to move that weight anywhere in the ball I want which means I could roll it anywhere I want.  Let's see.

First I made the attaching point to the ball.  That's a servo horn screwed to the inside of the ball.

The second attachment point to the ball needs to pivot freely.

FIRST YOU MUST MODIFY YOUR SERVOS TO CONTINUOUS ROTATION.  I used Hitech HS-55's because they're small and strong.  Yeah, yeah, I know ... everyone, including the manufacturer, claimes these servos can't be modified for that.  So, for those of you who are up to the challenge, here's a link of how to modify a HS-55 to continuous rotation.  It's not easy, but it's worth it.

Here's the contraption I made.  It was just a quick put together to see if it works.  as you can see the aluminum frame (I call the chassis on axis X) will be attached to the ball and spin with a servo.  The inner piece, I call the horshoe on axis Y, will spin with the second servo.

Here it is barely fitting in the ball.  If this works I'll make a more balanced contraption with NiMh cells instead of the LiPo pack because it will be more versitile

I was  happy with it's performance so I will build a balanced more sturdy one.  I stripped the gears of the servo so I will use less displacement weight in the horse shoe.  I will also get a new ball so I can see inside.

Here's the new chassis, horse shoe and pivot pin all welded steel with ball bearing pivot points. It's all measued and designed so the spinning parts are exactly perpendicular.

This is the ball bearing for the horse shoe with the set screw on top to keep it from being pushee out.  The inner hole is 1/18"

Here's the bearing for the pivot pin that will be attached to the ball.

And here's the entire rotating group assembled without servos

Here's the rotating group assembled with the servos, receiver, and batteries.  I ballanced the X and Y axis by placing the bateries strategically.  If it's off balance it will "warble" instead of roll straight.

Here it is all ready to go.

I'm pretty happy with how it turned out but it's very hard to control.                     7/1/08 Luckily I had most of this stuff in my garage.  All I had to buy was the ball and some steel.  Doesn't everyone have V-tail mixer and supplies to convert servos to continuous rotation laying around?

7/19/08 Well, it's VERY hard to control.  I can fly 4 chanel airplanes with ease and I can even fly helicopters but this thing is a whole new "Ball" game.  Haha.  The servos have a tendancy to creap which shouldn't be a big deal but because they are continuous rotation they no longer have "fighting tendancies".  A normal servo knows where the arm is and fights to keep it there.  So, in this application the weight can pull the servo around and that hurs the opperation stability too. I gave it a lot of thought and experimented with it and decided to change the inner servo back to standard.  This made a HUGE difference.  My girlfriend (who's never opperated anything RC was able to drive it around the leg of a table and back to us!

Click here to go to RC Ball Revision 5 The best and last revision.

Back to my home page

If you have any questions please feel free to email me at prabbit22m@aol.com