Re: [SeattleRobotics] Re: Microstepping [was: Engineering problem for discussion (organ swell servo), hijacked]
- Why "microstepping"? The sort of resolution you're needing is pretty low.
I'd have thought just ordinary full steps would be fine.
> How long can the cable be between control and motor?What matters to the motor is current rather than voltage. So the length of
the cable (i.e. its resistance) isn't too important.
Your motor controller has to limit the current going to the motor to ensure
that the motor doesn't overheat. The current tht the motor will "try" to
take varies according to what it's doing - turning, standing still, applying
a torque; the motor acts as an inductor and you're driving it with pulses;
blah, blah, blah. The bottom line is that you try to give it a fixed
The resistance of the motor will be from a few ohms up to, say, 20 ohms. So
long as the resistance of the cable isn't way more than that, you'll be ok.
How many cores does the cable need? Probably 4. Typical 2.5amp 4-core mains
cable will have a resistance of around 40ohms per kilometer. (Remember that
the current has to go out and back so you should think of it as 80ohms per
So I'd say, if you're using less than 50m you shouldn't have any problem.
> the optical rotary "pot" is very attractive for its longevity."Optical pot"? Do you mean an absolute position encoder?
Optical encoders like those on a mouse don't give you the absolute position.
They just tell you how much more you've moved. I suspect that the first one
that Dave suggested
does that but also has an output to say when the shaft is in the 'zero'
position. That may not be what you want.
Sine/Cos encoders and magnetic encoders aren't optical but will definitely
do the job. For a price!
As your foot pedal moves less than 90deg, there are other ways to go.
In the past, I've used crossed polaroid films, a light and a photodetector.
As the polaroid films go from parallel to crossed, the light through them
drops to almost zero. It's reliable, cheap and there are no electrical
contacts to wear out. I've even used a light attached to an arm and a
photodetector attached to the base - as the light get's further away, the
photodetector output falls. It's non-linear but you don't particularly care
about that - if you must, you can can correct for the non-linearity in the
I think you could also look at 'tuning capacitors'. Search eBay. I've never
heard of anyone using a tuning capacitor as a rotary encoder but it makes a
lot of sense to me. They're built to last.
Capacitative encoders are rather trendy just now. As they're so new,
industrial ones are a bit expensive but it would be trivially easy to build
your own that's as accurate as you need. A couple of pieces of tinfoil and
an AD7746 chip would give you an excellent position encoder.