Move until stall, but move if unstalled

[RitchieP]

I'm trying to implement a program where it will run a motor until it is mechanically stalled. But would like the ability to continue running if it is unstalled (I hope that's a word).

Is there a way to achieve this? Currently, my implementation is something similar to this.

if abs(left_motor.load()) > 25:
   left_motor.stop()
   wait(100)
elif abs(right_motor.load()) > 25:
    right_motor.stop()
    wait(100)
else:
    left_attach.dc(30)
    right_attach.dc(30)
    wait(20)

This method works, but I would like to see if there's a better way to implement it. Because when the motor is mechanically stalled, I can still hear the motors buzzing as if it's still trying to move, but realize it is stalled, then only it will stop.

[laurensvalk]

Stall and load detection works faster and better when you use run() instead of dc(), so that could be a start.

Because when the motor is mechanically stalled, I can still hear the motors buzzing as if it's still trying to move, but realize it is stalled, then only it will stop.

A motor experiences load if it is moving and is being slowed down. If the motor is stopped, there isn't such a thing as load, so how would you measure it?

[RitchieP]

Stall and load detection works faster and better when you use run() instead of dc()

I see! Interested in how it works under the hood that makes it faster and better.

If the motor is stopped, there isn't such a thing as load, so how would you measure it?

Yes, that's a question that's also buzzing my mind. I tried to increase the wait(100) to a higher value like wait(1000). And every time when the motor is mechanically stalled, the motor goes quiet for a second, then you can clearly hear the motor creating a buzz noise every second for a short while. (like eeee eeee)

Which makes me wonder, are the motors actually trying to move again, but because it's stalled, then it builds up load and only stops after the measured load is greater than my defined 25

[laurensvalk]

When actively controlled with a method like run(...) it has a sense where it should be and where it actually is. The difference between them is a measure for the load. I.e. it is precisely how hard it has to work to stay on your target speed or position.

With dc(...) there isn't any target. The load is estimated by a basic simulated motor model. It works and it is much better than nothing, but there is a bit less information to go on so it is a bit slower.

I'm not sure exactly how your program works from this snippet, or what the load is in your design. We might be able to help with some more detail of what you'd like to do. Maybe there are other ways than load. For example, you could look at speed. If it is moving it is probably not stuck, but it depends on your design.

[RitchieP]

Sounds great! I'll give run() and using speed as a measurement a try.

I'm not sure exactly how your program works from this snippet, or what the load is in your design.

The code snippet is actually encased in a forever loop (while True:), and my goal is for the motor to run until it is mechanically stalled with the value returned by load(). But if at some point I removed the mechanical stall (I.e. I removed the lever connected to the motor), I want the motor to be able to keep spinning.

My main concern here is that I hear buzzing sounds when the motor is stalled. I'm not sure if it's destructive towards the internals of the motor, as mentioned in this line. (Cause these motors are expensive stuff haha)

Which makes me wonder, are the motors actually trying to move again, but because it's stalled, then it builds up load and only stops after the measured load is greater than my defined 25

[dlech]

Having the motor energized while stalled for short amounts of time doesn't hurt much, but over long amounts of time could overheat the motor and cause damage and will also drain the batter much faster without doing any useful work.

Would it be possible to use some other sort of sensor to detect the lever position instead?

Otherwise, I would try the run_until_stalled() method with a longer delay between retries so that the motor is off most of the time when stalled (at least 1 second).

[RitchieP]

Yeah, at the moment, it looks like there's no real way to achieve what I want without using another input other than the motor itself. But @laurensvalk idea of using run() and speed does help the situation a bit at first glance.

I guess I'll mark this discussion answered for now.

Thanks for all the help and exchange of ideas! Appreciate it!