Design flaw pinion gear on #4751 19:1 Metal Gearmotor 37Dx68L mm 12V with 64 CPR Encoder

I might be missing something but there appears to be a design flaw on this motor where the pinion gear (if that is the name of the helical or worm gear comiing out of the motor) slips down the shaft towards the motor so much that it no longer (or only partially) engages with the gearbox first gear. This has happened now to me with 2 brand new motors after a few minutes of use on a cart pole balancer robot.

photos are at https://photos.app.goo.gl/6QpFxATGK8JsPF7w8 . You can see how the gear has slipped down the shaft so that it only engages the tip of the heleical gear with the gear box.

The result is that the motor jams and gear teeth on the worm gear break off.
The problem can be alleviated by opening the motor and prying up the gear nearer to the tip of shaft (or just past it), but it is not clear if it can be fixed in place using some thread locker compound.

I think it is a flaw in the design not to have a bushing under the gear to prevent this happening.

Hello.

I am sorry to hear you are having trouble with your gearmotors. We have not seen this type of behavior with these motors before, nor are we able to reproduce it with units we have here. Are these units you ordered from us directly, and if so, can you email us your order information and some pictures of your overall setup? Please also include additional information about how you have been using them, such operating voltage, speeds, and loads.

- Patrick

Thanks very much Patrick. I added photos and video to the album in first post of the cartpole robot (from aliexpress, https://www.aliexpress.com/item/32947544982.html?spm=a2g0o.productlist.0.0.350f75a1ydztIv&algo_pvid=5a32e1c4-1806-47f3-9eee-14637c5dc9c5&algo_expid=5a32e1c4-1806-47f3-9eee-14637c5dc9c5-0&btsid=0ab6d59515832278851835402ee33f&ws_ab_test=searchweb0_0,searchweb201602_,searchweb201603_

I ordered the Pololu motors after burning out the stock motor on the robot. The wiring is different but was easy to sort out and rearrange on the encoder/motor connector. Drive is 12V and load is as seen, not sure how to judge it. The motor frequently reversed and driven nearly at full power to drive the belt-driven cart.

on the first Pololu motor, after about 10m of operation, a grinding noise started. After disassembling the gear box.I saw the pinion gear had slipped down the shaft until barely engaging, and that caused some teeth to break off. I was able to replace it with pinion gear from stock motor that fit snugly and perhaps would have been fixed in place with the Loctite threadlocker I used. But then during debugging I switched to 2nd pololu motor. Again, after a few minutes (maybe at most 30m) the same thing happened. The gear slipped down the shaft to the point that it was not engaging fully with gear box.

Hope this helps. To reproduce, maybe drive with some load at medium power with frequent reversals.
Tobi

Hello, Tobi.

Thank you for the additional information. We checked units here and verified that the pinion gears really are quite tight on the motor shafts (we could not make them slip on the shaft using hand tools). Additionally, they are mechanically at the limit of what’s possible without risking cracks to the pinion gear, so we cannot make the fit any tighter.

I think that rapid back-and-forth oscillation in your untuned configurations is just too much for these motors to handle, and unfortunately, we do not have a solution for making that better on the motors themselves. Adding a bushing as you suggested would not address the fundamental problem of the pinion moving relative to the motor shaft because once that connection starts giving way, the wear will quickly accelerate and lead to one failure or another anyway.

Is it an option for you to address this at your application level, either in software by implementing acceleration limits or in hardware by adding something between the motor and the pulley that can dampen the torque spikes? The video on aliexpress shows a well-tuned system with relatively smooth motor motion, and I would expect our gearmotors to tolerate that well. Your videos show a high degree of jitter that I expect is largely responsible for the problem, so perhaps this could be mostly addressed through better tuning of your system.

- Ben

Hi Ben, Thanks very much for having a careful look at the problem. The reason I think it is a shortcoming is that the stock (super cheap, no brand visible) motors that come with the robot don’t seem to have this problem. I’m not 100% sure about this because we managed to burn them out, but they never had a gear breaking problem like the Pololu ones before we burned them out.

The problem is not that the gear slips OFF the shaft, it slips too far ON to it. I believe the mechanics of the helical drive probably cause this.

We will try to see the effect of strong threadlocker and will have a look if a bushing can prevent the gear from disengaging. The option of limiting control is not an option for us at this point since we are developing learned optimal controllers and the control is, well, uncontrolled right now.

Again, I appreciate your suggestions and we will let you know what we find out.

I understood what you were saying about the gear slipping axially. My point is that blocking that axial movement is unlikely to be a complete solution because the pinion gear will still be slipping rotationally relative to the motor shaft, and that will just eventually lead to a different kind of failure.

Is it an option for you to lower the supply voltage or limit the maximum duty cycle? With different motors failing in different ways, reducing the power could reduce the probability of all those failure modes.

- Ben