Variable low speed motor, how much can I vary it?

[Habeebe]

I'm looking at getting a motor to drive a micrometer. It's on an interferometer mirror for the curious. I'll need around 1RPM or so, but I want to be able to vary the speed, because I'm pretty unsure about how fast I'll need it to go for my experiment to work. I'm thinking a range of like 1/4 to 2 RPM would be pretty safe. I won't need much torque, as the motor will just be turning a micrometer knob.

So I'm looking at geared DC motors, but I don't know how much variation in speed a given motor will have, and what specified speed I should get. I don't mind building a PWM if I have to, but it'd be really swell if all I had to do is hook it up through a potentiometer. Should I go for the highest RPM I might need at the rated voltage? Will that let me go down in speed by a factor of 8 or so?

 

[skeptic2]

I once built a PWM for an O gauge train set and it would start the train moving as slowly as you wanted. What about using a worm gear to reduce the speed along with using a PWM.

 

[Baluncore]

Using a DC motor without feedback, to regulate speed you must control terminal voltage, to regulate torque you would regulate current.

You want to set the speed to a slow but reasonably constant value, independent of torque. For that you need to drive the motor with a variable low voltage from a low impedance driver such as an emitter follower. The efficiency of a PWM is probably not important for such low power.

If you were to use a series potentiometer, that would effectively make a constant current source, the motor torque would be maintained constant, the speed could vary widely. That is not what you want.

Select a higher voltage DC motor, then run it on less than 5 volts, you will get a proportionally lower speed, largely independent of torque.

The power required to drive a micrometer is very low. A geared motor can produce enormous torque, it would be good to use a rubber band or sleeve as the drive connection to protect the micrometer from high torque when things go wrong.

 

[skeptic2]

The problem I was trying to overcome in my application was not inefficiency but the lurching from a dead stop that is typically seen in toy trains. There is a significant breakaway torque needed to start the train moving and once that torque is exceeded and the train begins to move, it continues to move at significant speed. The advantage of using PWM is that with a voltage slightly higher than that needed for full speed, when the train is starting to move, the voltage of those pulses is enough to supply the break away torque without making the train lurch. It will be impossible to adjust a micrometer if your motor lurches every time you reverse direction.

Due to the smoothing effect of the inductance of the motor, if you use a high frequency PWM its effect will be very similar to DC. The frequency must low enough to allow the pulse to overcome the breakaway torque. I used a signal generator pulse output to determine the frequency that would start the motor at the lowest voltage but still run the motor without a noticeable pulsing. In my case that frequency was 70 Hz.

 

[Baluncore]

@ skeptic2. My reply was to the OP, not a criticism of your anecdote. The “lurching” you describe is characteristic of a poorly regulated voltage or a high series resistance circuit.

The OP requirement is I believe, to sweep the micrometer at a reasonably constant rate with a DC motor. That requires a low impedance voltage source, not a series current limiting resistor as considered by the OP.

It does not matter how the low impedance voltage is generated, it could be a linear regulator such as an LM317 or a variable PWM supply. They both solve the problem.

There is no doubt that a significant gear reduction should be used. Windscreen wiper motors are way more powerful than needed to drive a micrometer, they usually wound for a low RPM and have a worm reduction, giving about 30 RPM. Running a 12V DC motor without significant load on 1V will reduce the speed to about 2.5 RPM.

 

[Habeebe]

Thanks, you've been super helpful.