Trying to design a robotic joint.

[dennis_n]

Hi everyone. This is my first post on the forums.

I wanted to ask your expertise on a project I'm working on.
I'm trying to design a setup where I rotate a ∏ shaped frame that has a 1kg load on the horizontal bar. The space between the frame should stay empty.
The length of the side rods may be up to 1 meter, with the ability to slide the horizontal bar up and down.
I need precise rotational motion so at first I thought let's go simple and attach a servo directly to the joint. Then I realized that the torque requirements would be tremendous at one meter away.
I'm on a budget here so I don't want to spend too much on a servo unless I really have to.
So what would be a setup that would minimize the torque requirements on the servo while keeping the precision? Timing belt and pulley? I'm not sure how this setup would work.
The weight (horizontal bar) has to rotate from 0 to 90 degrees around the joint.
It is important that the breaking between rotations, doesn't produce any bouncing.
I'm also trying to keep the machining requirements minimal cause I don't have much else than a drill and a saw.
Any ideas? Thanks!

 

[Unrest]

Can you explain a bit more clearly?

If it's rotating about a vertical axis then it may only need a very low torque. That also depends on the acceleration. If it's a horizontal axis, calculate the torque required to overcome gravity. Once you have some estimated numbers you can compare to shafts, belts etc. to get a feel for what kind of drive might work.

"precise" could mean a lot of things. Maybe you need feedback somewhere else on the frame to allow it to bend?

Doesn't really seem like ANSYS would be much on a simple frame structure like this.

 

[dennis_n]

Can you explain a bit more clearly?

If it's rotating about a vertical axis then it may only need a very low torque. That also depends on the acceleration. If it's a horizontal axis, calculate the torque required to overcome gravity. Once you have some estimated numbers you can compare to shafts, belts etc. to get a feel for what kind of drive might work.

"precise" could mean a lot of things. Maybe you need feedback somewhere else on the frame to allow it to bend?

Doesn't really seem like ANSYS would be much on a simple frame structure like this.

The whole setup is to to do spherical photography. Here is a quick model I made of the whole setup as I was thinking it originally.

 

[Unrest]

Because of the limited range of motion, you could also use a big pulley with a steel wire or chain fixed to it. Wind the other end of the loop around a drum on the servo/motor.

If you're not worried about speed you could also put a counterweight on the arm and that eliminates the torque problem. Drive it with anything depending on how accurately it's balanced.

A cheap way to prevent overshoot may be to use a mechanical or magnetic catch to grab the arm when it's at the end of it's path. Then release the catch and move again after any vibrations have died away.