Looking for a "highly resistive ring"

[benmuskler]

Hello! I'm completely new to working with arduino and electronics and am currently working on a little project involving a stepper motor. I need to be able to detect in which direction the motor is pointing. In order to do so, I came up with an idea inspired by how potentiometers work. The idea is illustrated in the figure and basically works by measuring the voltage between a reference point and a conductor that moves along a metal ring with high resistance.
The problem with this idea is that I cannot find anything like the metal ring part. I've seen there are resistance wires, but the "resistance per meter"-value seems to be too low as the ring in this case will have a very short circumference (might be wrong, though).
I would be grateful for recommendations either for parts to use, or alternative ideas.
Thanks!

Image: https://postimg.org/image/49nw2mct3/

 

[berkeman]

Hello! I'm completely new to working with arduino and electronics and am currently working on a little project involving a stepper motor. I need to be able to detect in which direction the motor is pointing. In order to do so, I came up with an idea inspired by how potentiometers work. The idea is illustrated in the figure and basically works by measuring the voltage between a reference point and a conductor that moves along a metal ring with high resistance.
The problem with this idea is that I cannot find anything like the metal ring part. I've seen there are resistance wires, but the "resistance per meter"-value seems to be too low as the ring in this case will have a very short circumference (might be wrong, though).
I would be grateful for recommendations either for parts to use, or alternative ideas.
Thanks!

Image: https://postimg.org/image/49nw2mct3/

Have you looked at how rotary encoders work? You generally get 2 signals out of them, and whether one is leading or lagging the other tells you which direction the shaft is spinning. That is a much more robust setup than having a sliding contact resistance measurement (which will wear out fairly quickly)...

 

[Nidum]

You usually know in advance which way a stepper motor is going to rotate ? With common types of stepper motors the direction of rotation is set by the sequence in which you energise the windings .

 

[Nidum]

If you actually want to detect the rotation then a slotted disc and optical switch is a simple method that you can use .

 

[berkeman]

You usually know in advance which way a stepper motor is going to rotate

Oops, I missed that he said stepper...

 

[benmuskler]

Thanks for the answers! I realize I may have phrased my question a bit strangely. What I meant was that I want to detect the angle between some reference point and the arm in the picture. Nidum, is this what you suggest slotted disc and optical switch for?

 

[billy_joule]

The key advantage of stepper motors is that they have accurate position control without feedback.
If your stepper is missing steps and losing positional accuracy then something has gone wrong (eg poorly selected) and an encoder probably won't help.

 

[berkeman]

The only thing you might consider adding to your stepper motor setup is a "limit switch" type of device, so you can get an absolute position as a starting point. Just rotate your stepper motor through a revolution or less, to trip the switch. That gives you your absolute position reference, and then you can just count your steps CW or CCW from there to keep track of the shaft's angle.

 

[Nidum]

Stepper motors provide accurate and usually reliable angular positioning .

Problem is that they have no initial reference position . To rotate the output to a specific angular position you need to know where the zero degrees position is so that a calculated sequence of steps will get you to the wanted angular position .

Slotted disc is one way of detecting the zero position . For this application the disc just has one slot .

 

[Nidum]

Depends on the demands of the set up though - you can just preset the motor to zero by hand sometimes .

 

[benmuskler]

Thanks for all the advice! I like the idea of the limit switch. I guess, if I want to avoid adding another part to my setup I could manually rotate the motor shaft to a reference position before running the program?

Also, I think the motor may be faulty, as billy joule pointed out. When I use the arduino Stepper library to step 50 steps CW and then 50 steps CCW the shaft doesn't move back and forth as it should. It's like it diverges (quite quickly) from the region of these 50 steps that it should be within. Also, the store I bought it from sold me a non-functional DC-adapter, so I do have some trust issues towards them :C
I'll try to control the motor manually by applying voltages over the coils, to test it further!

 

[Merlin3189]

I have a 360^o continuous potentiometer taken from an old XY plotter. It achieves a reasonable resistance by using not a solid ring but a fine resistance wire wound round a torroid, so that each mm of circumference is perhaps 10cm of wire. I think these were expensive items, so I'm not suggesting you use this.

Maybe you could use a ring of some more resistive material, such as used in carbon potentiometers. The difficulty for DIY fabrication would probably be getting a uniform coating to give a linear change of resistance.
As others say, an optical encoder might be the thing to go for.

Depending on the resolution you require, can I put in a vote for a Gray coded disc. For low resolution, say up to 10^o, this would be easy enough to build yourself. I don't know if high resoluion devices are commercially available at all. It has the advantage that even switching on with no memory of previous operation, you can immediately know where you are. Stepper motors and slotted discs require you to move to a reference point and count the pulses to calculate your starting position.

I wonder whether you could use a rotary version of the variable differential transformer. I imagine two perpendicular sense coils with an energised coil rotating in the same plane, though I understand commercial products have all stationary windings and a rotating iron core. (I haven't been able to fathom out how they do that yet!)

 

[Baluncore]

I guess, if I want to avoid adding another part to my setup I could manually rotate the motor shaft to a reference position before running the program?

You can program the stepper motor to move to a limit or reference position at the start. If you have a reference position detector the controller can monitor motor accuracy and let you know if steps are being lost.

 

[Tom.G]

The usual reason that a stepper motor skips steps is the mechanical load is too high for the motor torque. This could be a static mechanical condition (the shaft being held stationary when stepping) or trying to accelerate a high inertial load (the motor rotor hasn't had time to get to the next step position before the stepping pulse ends.). Try a slower stepping rate and/or lengthening the step pulses. The step rate and pulse width need to be 'tuned' to get maximum performance from a given motor/load combination.

Of course it could be a bad motor too; or an under-powered motor driver circuit.