Stepper motor issues

[xnicke]

I have an application where I use a stepper motor to rotate an object plus/minus 90 degrees.
The motor is set to rotate a number of step that are slightly more than 90 deg and stop against a mechanical stop. However, it has happened quite often lately that it "jerks" back a lot (everything form 1 deg to ~50) once it hits the stop.
I was hoping that any of you have any suggestions of the cause (and fix) since this is way out of my area of knowledge.
The motor is connected to a planet gear. I can not feel any play in the shaft and it rotates smoothly when I turn it by hand.
Note: I have not seen this issue when I continue to rotate once it has reached the stop (i.e. run it several times in a row at the same direction)
Note 2: This happens both CW ans CCW.
Could the issue be resolved by settings in the controller?

 

[MikeHalloran]

Maybe.
You may have to venture farther into the bizzarro world of steppers to find out.
Start by forgetting everything you know about other types of motors.

Steppers will normally 'ring' upon reaching the next step unless some damping is provided.

Mechanical damping is not durable enough to deal with steppers, so electronic damping is often provided within the power driver circuit for each coil.

Alternately, some controllers may be capable of adjusting the step timing to take advantage of the natural ringing tendency to prevent the ringing from happening.

( For background on steppers from a completely practical perspective, search for stuff by Al Leenhouts, sales of whose masterwork constituted much of his retirement, last I heard. )

( For background on steppers from a first principles/ theoretical perspective, search for stuff by Taft / Gauthier / Harned at UNH. )


But I'm puzzled by the behavior you report; normal errors amount to a step or two, not fifty degrees. Perhaps you have gotten the controller confused. This might be a good time to do a factory reset and reprogram it, one step at a time, following the user manual.

Commodore floppy disk drives used no position sensor on the head drive stepping motor. They would just bang the head assembly against a mechanical stop for a number of steps greater than a full stroke, then count back from there to figure out where the head was. It was very hard on the stops, especially when a diskette got flaky and the drive had to 'recalibrate' like that many times.
Consider adding optical interrupters to your system, just 'inside' the mechanical stops. They are much cheaper now than they were back in Commodore's day. Your controller can surely accept input from them, and they'll do a lot less damage than the mechanical stops. ... which you should keep but relocate 'outward' a bit.



Mike Halloran
Pembroke Pines, FL, USA