0
\$\begingroup\$

I want to eliminate the wrong direction torque of stepper motor during super fast decelerating.

The wrong direction torque is that: When you suddenly decelerate the input pulses of stepper motor's controller, and if the load on the shaft exceeds the maximum electrical torque, the stepper motor will "slip"(lost sync with the drive circuit). Once the motor has rotated past the current "step"(the motor slips), the current driven into the coils will try to push the rotator ahead toward the next step, at this time the torque is in the wrong direction as you want to decelerate the motor. If the motor is turning rapidly, the torque in the wrong direction may cancel out some of the effect of torque that would be in the right direction(which is to decelerate the motor), so you should not suddenly decelerate the stepper motor controller's input pulses.

But I think with positional feedback, the controller can eliminate this wrong direction torque of the stepper motor in super fast decelerating.

The positional feedback can tell the controller the absolute position of the rotator. Once the motor has rotated past the current "step"(the motor has slipped), the controller can change the current driven into the coils so it won't push the rotator ahead toward the next step, it can only continue to stop the rotator when it has passed the so called "current step", it will continue to try to stop the rotator with its electrical torque even the motor has slipped.

There are many closed-loop stepper motor in the market, the vendors didn't explain the details of operating principle, only said:

  1. Due to closed loop control, the ES series easy servo systems can always implement 100% torque of the motor, and do not need the huge 50% torque reservation in normal open-loop stepper systems.
  2. The built-in rotor position detection sensor constantly monitors the motor movement. If synchronism is about to be lost, closed loop control is used, eliminating concern over loss of steps.

Can those closed-loop stepper motor in the market eliminate the wrong direction torque during super fast decelerating?

\$\endgroup\$
  • \$\begingroup\$ The closed loop controller, when properly set up and tuned, should not allow you to deaccelerate fast enough to lose steps so the problem should not come up. I would think that if you applied an external force to slow the motor down faster than the controller could slow it, that it should behave gracefully but you never know. Think of ABS braking on your car- it supposedly doesn't let you get into a skid-- by preventing you from applying the brakes too hard. \$\endgroup\$ – Spehro Pefhany Aug 22 '14 at 15:24
  • \$\begingroup\$ Specifically refer to your last line, "Can those closed-loop stepper motor...during super fast decelerating?". Yes, if you choose large enough motor torque and wattage rating with respect to physical load. Panasonic (is it ok to name company here?) makes industrial servo motors, about 100 models, 50 to 2000 watts, suiting different mechanical speed, torque, inertia, etc. Work principle is optical rotation encoder, 5000 to 20000 step per revolution, highly refined controller, PWM current output to 3 phase motor stator coil, permanent magnet rotor. \$\endgroup\$ – EEd Aug 22 '14 at 18:27
  • \$\begingroup\$ The controller drive 3 phase coil to produce a 'vectored' (precise angle) magnetic field to hold the rotor at user-commanded angle with great precision (thousands step per revolution). Optical encoder detects if error in angle, then, then command the 'vectored' field to move the rotor back to desired angle. If small mechanical load is applied to try to move the rotor, drive current is small. For high load, current auto adjust upward. I believe these AC servo system (motor and controller as package), is what the original poster refers as "closed-loop stepper". Expensive, work excellently \$\endgroup\$ – EEd Aug 22 '14 at 18:35
2
\$\begingroup\$

The torque on the motor is a function of the current through the coils. Back EMF created by the rotor moving causes a voltage. Whether this voltage turns into a torque depends on what is connected to the coils. If left largely on its own, like if the coils are shorted, the back EMF causes current that opposes the rotor motion. For example, electronic braking of permanent magnet motors is often done exactly this way, by shorting the coils. Your premise, and therefore your question, make no sense.

\$\endgroup\$
1
\$\begingroup\$

The simplest way to figure on using a stepper motor with a position sensor is to say that whenever you want the motor to accelerate clockwise at any given moment, you drive it 90 degrees clockwise from where the position sensor says it is right then. When you want it to accelerate counterclockwise, you drive it 90 degrees counterclockwise from where it is right then. When you want it to coast, you float the leads (clamping to the rails). During maximum braking, the motor will "slip", but when the motor's rotational phase gets to be within 135 degrees past where you're driving it, the sensor will cause the driving phase to be advanced 90 degrees so the motor will only be 45 degrees ahead of where you're driving it.

Note that when braking in this fashion, the motor will turn backward after coming to a stop. This behavior may be avoided to some extent by advancing the drive phase a certain amount of time after the motor has advanced by a phase (so that one will try to drive the motor to its current position, rather than the previous one). If the motor hadn't yet reached the centerpoint of the next phase, advancing the phase may apply torque in the "wrong" direction, but the fact that the motor hadn't yet reached the next phase would imply that it was traveling sufficiently slowly that even with such acceleration it would still end up stopping at that phase. Effectively, this behavior will mean that instead of going backward by 45-135 degrees of phase angle, it will move forward or backward by 45 degrees.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.