Glad to hear that you managed to fix the issue, though.

Before today, the extruder's tension screw had no effect, no matter how much I tightened or loosened it, even after a disassembly and reassembly. After today's reassembly, I actually had to tighten it for the filament to load.

This is an unsatisfying resolution, but at least it's a resolution. Thank you both for your time.

And it turns out that every instance of the motor getting stuck since I disassembled the extruder was because... the controller blocks the motor whenever the temperature goes under 170°C (which happens to be the temperature used for bed leveling with PETG -- which is what I use). As soon as it is 170°C (or higher), the motor is unstuck.

When I keep the nozzle temperature above 170°C, the motor does not seem to get stuck when it has no load.

Test load/unload filament which forces to do a fast rotations.

No issue with those. (Though once the motor stops moving, asking for a load/unload does not make it move either.)

Check if the connections in the connectors in the cables are tight ( if you plug/unplug them they may become loose and don't have a proper contact with the pins in the board.

They seem fine. I unplugged and replugged a few times, no change.

Extract the extruder motor and try to rotate it with a fingers, can you feel any irregularities/obstacles when doing it?

When unpowered? Yes, I think so.

If yes then maybe motor is damaged physically - in best scenario disassembling it and cleaning the bearing would be possible, in worst case get a new motor.

I'm considering it. I'd just like to make sure the issue is in the motor and not the controller.

Try to swap plugs with another motor, for example X axis with extruder, and see if the X motor is reacting properly - this way you can detect if the fail is in the motor or in the electronics on the control board.

Is that safe? According to https://github.com/prusa3d/Original-Prusa-MINI/blob/master/DOCUMENTATION/ELECTRONICS/mini-motor-kit.pdf the extruder motor does not have the same specs as the other ones.

It's also pretty hard to try because this only happens during a print (so the printer has to be assembled for the pre-print calibration to work), but I can't let it drive the X motor thinking it's the extruder because it would crash into the end of the axis. So I would have to disassemble the X axis while it is printing (and quickly enough to be done before the motor gets stuck).

For what it's worth, I finally managed to interpose a breadboard between the controller and the motor, to measure voltage. Once the motor is stuck, voltages on the two coils are 3.7V and 0.2V; and never change until I reboot the printer. Even pausing the print does not change them.

Check main board for damages or signs if the wear out, for example swollen tops on the capacitors.

Good call. It looks fine.

Check if the cables are not tangled with the other cables, which can cause interferences of the signal. Check if the connections in the connectors in the cables are tight ( if you plug/unplug them they may become loose and don't have a proper contact with the pins in the board.

Extract the extruder motor and try to rotate it with a fingers, can you feel any irregularities/obstacles when doing it? If yes then maybe motor is damaged physically - in best scenario disassembling it and cleaning the bearing would be possible, in worst case get a new motor.

Try to swap plugs with another motor, for example X axis with extruder, and see if the X motor is reacting properly - this way you can detect if the fail is in the motor or in the electronics on the control board.

Check main board for damages or signs if the wear out, for example swollen tops on the capacitors.

General neglect, dust and cobwebs.. Basic cleaning, lubrication and general maintenance have all been missed.

See: https://help.prusa3d.com/en/article/regular-maintenance-mini_133222

The motor is clean on the outside, and does not seem to be designed to be opened.

None of these mention maintenance required by the extruder motor. And I do mean the extruder motor, not the rest of the extruder.

A dirty print bed. Give it a really thorough clean and avoid touching the print area.

My problem is not about bed adhesion

Damp filament. Look up threads here about drying filament especially if it has been left open.

The motor gets stuck even when it is not driving anything. So it can't be a filament issue. The filament I used was brand new and sealed anyway.

General neglect, dust and cobwebs... Basic cleaning, lubrication and general maintenance have all been missed.

See: https://help.prusa3d.com/en/article/regular-maintenance-mini_133222

A dirty print bed. Give it a really thorough clean and avoid touching the print area.

Damp filament. Look up threads here about drying filament especially if it has been left open.

Cheerio,

Since then, the extrusion motor often stops during prints. The printer does not detect it is stuck.

This happens even when using .gcode generated with the older PrusaSlicer, and even with the extruder completely disassembled (ie. the motor does not drive any gear).

Then, once stuck, the motor rarely starts again until I restart the printer. It managed to unstuck itself once after pausing/resuming a print, but usually does not.

Unplugging and replugging it from the controller board does not help. Though even when stuck, it does spin just a little bit the moment I plug it back to the control board.

While stuck, it is still pretty warm, and emits the same slight white noise it does when not stuck.

Should I replace the motor? The controller? Both?

Is the 3kg spool a red herring, or should I avoid those in the future?

Thank you for your answers