Brett, To rotate my torso, radar, etc. I've been giving some thought to setting up a DC motor with two relays as an H-Bridge (see image) then controlling the motor with a servo. The servo would have to be placed as if it were trying to rotate the part, but it would actually be pushing against small buttons that would activate the motor. I made this quick 3D drawing. I hope that it makes sense. The blue would all move together. In reality, one would have to mount the buttons on something that was spring loaded so that the servo wouldn't strip its gears if it moved faster than the accompanying motor. Once the part catches up to the servo, it will stop. It is something like power assisted steering. I know that motor controllers are the right way to go. I bought them for my drive motors, but I cannot afford to do them all that way. Besides, this has the advantage of keeping track of the position. Maybe this will help you.

Cheaper is always better, and that is a very interesting way to do this. Have you tried it out yet?

No, I haven't. It is just something that I have been kicking around in my head. I just ran across this too. It might solve the same problem.

www.ebay.com/itm/Analog-Rotation-Sensor-V2-Arduino-Free-Cable-/160626748153?pt=LH_DefaultDomain_0&hash=item256619aaf9

@ Danger!

Hmmm.....I like your idea and will consider trying it out.

A couple of years ago I had an idea similar to yours but, I never built it. I am to the point now in my build that it may be a good time to test ideas like these. My version is a little crude but it may also work.

I invite everyone to take a closer look at this and let me know if you think it will work.

Rgordon, It seems like that would work to me. Our ideas are almost the same, except you use latching relays, and hit the micro switches in short clicks, my system would require the buttons to be held until the servo stops, and the torso catches up to it. (In my picture, the buttons look as though they are mounted on a solid piece. I didn't want to get too technical, but they would have to be on something that swings and absorbs the difference in speed between the servo and the actual motor. Once the servo stops, springs would bring it back to center relative to the torso, as the torso aligns itself with the servo). I am not sure if I will actually do it that way or not. It is just something that I was thinking about. My fear is that in practice, the momentum of such a heavy torso will overshoot the servo and there will be a lot of the DC motor jumping back and forth between forward and reverse until it lines up. The rotation sensor that I linked to in my last post might make more sense. It would be more difficult to program. I would like to take advantage of EZ-B's simple servo controls, but I have to weigh it against making the hardware more complex, and the risk of something breaking down. I guess that I will have to experiment a little more. This whole project has been a few years of trial and error so far. If it were easy though, there wouldn't be much of a point in doing it. I would just take up Shuffleboard instead.

Danger! Yeah I had thought that jumping/hunting might be a problem with mine. That was why I chose the micro switch method. So when it hit the stop switch and unlatched the circuit there was a small amount of space/slop it could coast before it hit the other switch.

Then I thought about using a motor that uses a worm gear. I think that worm gear motors don't coast but stop dead in their tracks. That may be the answer to both our designs.

Any ideas on what RPM to use so it doesn't jerk so hard when it stops?

Oh well looks like its time to test it out. Thanks for getting me motivated on this again.

I have a windshield wiper motor for mine. It has a worm gear in it. I have tested them with relays. The H-Bridge does stop them instantly. As far as the speed, Bob May used to fling that torso around like it was a top. I'd love to be able to re-create that. Prudence tells me that would be a bad idea. I'm thinking along the lines of him turning 180 degrees in about three seconds. Speaking of 180 degrees. I would like my torso to make a full rotation. A nice feature of the original robot was that he would turn his body around and drive off backwards, so he would need 180 in either direction. This could easily be done by mounting the servo on top of the horn of another one. Either, or both could turn. I will start experimenting on that. Right now, I am working on the mechanism to bend his waist forward and back. Unfortunately, I work a lot of hours and do not have much time to spend on it.

i seriously need to get some H-Bridges and start experimenting.

Will you need slip rings for the waist? There are some home-made slip ring "how to's" out there I just can't rmember the links. How are you planning to make him bend at the waist? Linear actuator, cam mechinisim?

I too find it hard to find the time to work on Magnus.