Cold temperature affect motor encoder?

Regarding # 172:1 Metal Gearmotor 25Dx71L mm MP 12V with 48 CPR Encoder

I’ve been using this motor for nine months with no problems. It opens and closes an outdoor gate. An Arduino Nano with a motor driver controls the motor. The Nano counts the encoder pulses to control gate position. The motor is near the top of a waterproof enclosure that is open on the bottom, and there are baffles that prevent wind-blown rain and snow from reaching the motor.

I went on vacation in late January and the gate was completely not used for two weeks, during which time the temperature dipped to -15 to -20 deg F (-29 deg C) for several nights in a row.

After I returned, the gate no longer returns to the proper position. It is off by 15 degrees or so. Prior to vacation and the cold spell, the error would be no more than 1 or 2 degrees.

Temperature are back to “normal”: +15 deg F at night, high 30s in the day.

Seems unlikely, but would the cold temperature and super dry humidity have affected the encoder? Any other thoughts?


It is not something we have characterized, but I am skeptical that the temperature would have that kind of effect on the encoder. My recommendation would be to do some tests in simplified setups to isolate exactly where the error is coming from (e.g. is it a mechanical problem with the gate, an issue with the Arduino, a power supply problem, a motor problem, or an encoder problem?). As a first step, can you take the motor out of the gate and look at it on a scope to see what the encoder signals look like?

- Patrick

Thank you for the reply. Yes, I didn’t think temperature would have that effect, but it’s the only thing that has changed that I’m aware of. Seems very unlikely to be the motor, a mechanical problem, or the Arduino. Maybe wiring, though it’s hard to imagine how two weeks of inactivity with cold temps could have affected wiring.

Unfortunately, removing the motor from the gate enclosure is not a simple task, it’s painful (cold) to access it “in place” in the winter, and I don’t have an oscilloscope!

I do have a spare motor, and I have a Vernier LabQuest Mini with voltage probe capable of 100,000 sps (about 50 times the rate of one channel of the encoder), so come springtime I’ll take a look with that and maybe try the spare.