# How can I dynamically control the amount of torque necessary to rotate something for a test rig?

I'm looking to build a test rig for a robot that rotates a 1" diameter pipe 180º (roll not yaw or pitch). I am currently testing a motor's performance when subjected to various kinds of PWM (high duty cycle, low duty cycle, etc). I would like to characterize how this performs under various loads.

Is there a simple mechanical mechanism I can attach to a fixture and insert into or around the pipe that lets me control how easy or difficult it is to rotate?

I am thinking of something like a drill bit chuck that fits inside the pipe and expands or a circular clutch that clamps down around pipe to add resistance when one tightens a thumbscrew. I would like to go from no resistance to full stop for a 4N•m motor. I would like to be able to test how 'sticky' the pipe is using a torque wrench.

I imagine this would be very simple but I can't think of something that would do this!

• Commented May 24, 2014 at 0:07

Only rotate 180º, right? Not continuous rotation?

How about wrap a string around the pipe and then hang a weight off the string? As you rotate the pipe it will lift the weight and the torque calculation is straightforward. If you need more torque you can artificial increase the diameter where the weight is attached.

Hmm, looking at your title again you say "dynamically control" ... does that mean you need to change the torque in the middle of one test? One of your examples includes a clutch with a tightening thumb screw, though, so maybe not?

Well, the torque produced by a DC motor is proportional to the current running through it. Simply attach a DC motor to the end of the pipe, and control the current somehow.

The problem with this is that the torque is all in one direction. If you need to test your system in both directions, then you'll also need some way to reverse the direction of the current.

The other problem is that a DC motor's torque isn't perfectly constant as it rotates. If you need perfectly constant and controllable torque, then you'll the following will solve both problems:

Couple the motor to the pipe using a torque sensor. Use a PID controller to allow you to control the torque. This way you can control the torque accurately and very dynamically.