Go Down

Topic: (gauge) stepper motor longevity (Read 1 time) previous topic - next topic


For a project Im working on, Ive been using x27.168 gauge stepper motors with a suitable microstepping controller (AX1201728SG).  This combo works great; its really smooth....


...for a few days. And then the stepper seems to lose torque and starts skipping. In fairness, I am driving them hard while testing, accelerating as fast as they go more often than not,  and they are cheap as chips, but Im working on an instrument for a sailplane, and I need this to be far more reliable.

So a few questions; does it hurt these steppers to be forced against an endstop, deliberately causing them to skip? I use this to calibrate the needle position on every power cycle. I thought steppers couldnt be damaged this way, and I dont see how else you can do it, but now Im wondering.

Some automotive gauge cluster repair sites claim these chinese knockoff are no good, and the original switec one's are far more reliable. Im inclined to believe this for once, after all, the gauges in my car lasted 10+ years, but where do I buy original ones (Im in the EU, no amazon) ?

Does anyone know an alternative stepper motor thats suitable for driving a needle? It obviously doesnt need to be very strong, but it has to be reasonably fast (say 600 degree/sec) and with very fine steps -
 current one has 3 steps per degree (and 12 using microstepping). And of course, small enough to fit in an airplane instrument.  NEMA 17 would be a real squeeze, but NEMA11 and 08 should be doable.


On second thought; the problem might be me, not the steppers. My vario doesnt have quite the motion range of the stepper (~230 vs 270 degree?), so on every power cycle, I bump the needle against the edges to calibrate/center.  In my defense, the stepper itself seems to do this too, it doesnt seem to have internal endswitches, so I assumed this was fine. But now Im guessing this is causing me to strip the plastic gears. I'll need to add a microswitch or optical switch or something to detect a known needle position, and use that to calibrate its position.  If anyone has suggestions for something *SMALL*, Im all ears.

Meanwhile, I'll put one stepper at work non stop without the LCD cover, so it cant bump in to anything, and see how that fares.


But now Im guessing this is causing me to strip the plastic gears.
Your Original Post said nothing about stripping gears - is that actually happening?

I don't have any of those little motors but I always assumed they are designed to be moved to their limit at one end to identify the zero position. Are you doing that move slowly?

The QRE1113 is a small optical detector that is still big enough for hand-soldering

Two or three hours spent thinking and reading documentation solves most programming problems.


The stripping was just a guess, as I couldnt imagine anything else failing so quickly. I just opened it up, the gears actually look fine to me, but its really hard to see :

(full res: https://i.imgur.com/spCwTrC.jpg )

Id show you how it sounds and works now, but taking it apart, I seem to have killed it :). It just sounded "rough" and would skip real bad.

As for the speed, when Im zeroing deliberately, Im doing it roughly at half speed. But I must have done it at full speed quite often too by accident when Im overshooting my target and hitting the edges.


Hi Vertigo72,

I am on the same page where I zero stepper on purpose. Were you able to find the cause of steps skipping? I experience it a lot, I am not even moving fast but overtime it causes stepper to get stuck of miss steps etc. My project is batter operated and can't use optical sensor as it will consume milli amps. I might look for some feather touch mechanical switch as end switch but not sure whether zeroing it to the end is really causing the issue or it is just the bad quality of steppers...



How are you driving them? What pattern of steps?
Steve Greenfield AE7HD
Drawing Schematics: tinyurl.com/23mo9pf - tinyurl.com/o97ysyx - https://tinyurl.com/Technote8
Multitasking: forum.arduino.cc/index.php?topic=223286.0
gammon.com.au/blink - gammon.com.au/serial - gammon.com.au/interrupts


Hi Polymorph,

I am directly driving them from pins. This is example I am following without potentiometer:




Losing torque, and starts skipping, maybe heat related. Damaged gear teeth, in particular if bits floating around, can give intermitant overload, leading to mis-stepping. If the torque is reduced, and there is slipping, then the nylon? gear has come loose on the motor shaft, or some other shaft. Using one of these to wave a needle around in an instrument, is not the same as expecting it to do real work. For your application, what is it's advantage over the very common type of rc servo's?


These motors need careful driving if they are not to miss steps . There is a thread on here somewhere about developing a library for them.
I've used the Half stepper library which is "ok" if not properly correct ( they are noisy on it ) .
Bear in mine they are only meant to drive a lightweight needle, but should be very reliable.


Oct 02, 2019, 11:36 am Last Edit: Oct 02, 2019, 11:37 am by vertigo72
Ive postponed/ abandoned the project some time ago, but I do not recall having that same problem with a different batch of steppers and while using a lighter needle.  But I also switched from arduino to an esp32 and used interrupts to drive the microsteps. I also made some changes to the switecx12 library acceleration values.

Which of those cured it, I do not know or remember, but if it helps, I posted the code here:


From what you say, I  doubt you have problems with stepper motor, and I suspect your best bet is using a microstepper controller.


Ok, thank you all! It seems like I need to reduce the weight of the 3d printed needle, My 3d printed needle weighs lot more that usual plastic needle weight. I will do some tests with with lighter load and hopefully it should resolve the issue.

Thanks all again very much! You guys are the best!

Go Up