It turns out those two situations are identical, because “locking” the motor just maintains the current at whatever it was for the last (micro)step while it was moving.
In industrial applications, stepper motors run at high temperatures with no problem at all, because it’s well within their rating. When they’re attached to a diode laser frame, that same temperature seems scary-hot.
The temperature depends directly on the motor current and, in some cases, the default current setting is somewhat higher than strictly necessary. The current determines the motor torque (higher current = more torque) and higher current can eke more performance from a smaller = cheaper motor.
IMO, manufacturers set a timeout value to forestall panicked tech support calls.
If the motor gets too hot for comfort, you can probably reduce the current until it’s comfortable, although you’d want to test the performance to make sure it’s still doing what you want.
That makes sense. I think in my case it wasn’t so much the shaking as the tension on the belts simply pulling the head ever so slightly. The machine draws circles and shapes very accurately to the point where I’m satisfied that the belt tension is well adjusted and grub screws are tight.
Like Lizzard said, it would be very easy to chase a non existent mechanical gremlin in pursuit of a solution.
I adjust my belt tension as follows. Put the Y axis to the front of the machine. Tilt the front of the machine up in the air. The Y axis should slowly slide from front to back. If it doesn’t move the belts are too tight. If it slides and bangs into the back of the machine the belts are too loose.
Next put the X axis all the way to the left of the machine. Lift the left side of the machine up in the air. The X axis should slowly slide to the right of the machine. If it doesn’t move the belts are too tight. If it slides and bangs into the side of the machine the belts are too loose.