What do you guys think about the idea of using a brushless DC motor or a stepper motor that has a low number of steps per resolution in a screw drive linear motion system?
Look up closed loop servo motor control.
Not worth the effort…
Closed loop is useful for the brushless DC motor option, but less important if short circuit breaking is used.
I feel that motors with few steps per rotatation should have good speed and power. I suspect that the many steps per rotation decrease the maximum speed. I do not think you should need many steps per revolution for a system that you know will have the speed mechanically reduced.
For a stepper motor, you just trust that you did not miss a step.
A brushless motor must recieve the pulses at the right time to make the motor move further and should be essentially a be a 4 wire 4 step motor. A person should actually be able to lock the motor in one of those 4 positions by maintaining the voltage to the last selected position.
Yeah, you could use a 7.5 degree tin can stepper or similar to drive a screw. It would work exactly the same as a normal hybrid stepper but with coarser steps/mm. If the coils are low inductance it could move reasonably fast.
Hybrid steppers actually have the highest torque density of any stepper and it’s hard to find them in coarser than 1.8 degrees per step. But if you get the drivetrain configured right so you’re at the peak of the power curve, a short-pitch screw with a coarse-angle stepper could run perfectly well compared to a long-pitch screw with a 1.8 degree stepper.
The main benefit to a tin can stepper is much lower cost because they’re easier to manufacture. They’re also available in smaller sizes. Otherwise they’re basically not as good as hybrid steppers on any technical metric.
@NathanielStenzel but how do you know how far it actually traveled. You will provide a rotational force. That force will cause an acceleration related to the friction and loading at that particular time. So the speed that the rod turns at is some predicted value plus noise. Then some time later you provide a locking short after a time has passed and you have traveled X=(Vestimated+Noise)*T.
Because of the nature of these errors (sticky spots, not being level, thermal expansion/shrinkage) I would bet that they add, not just cancel. Steppers over come this because they have hard lock points ever X° so you can assume that you are re-zeroed then. DC motors don’t do that.
@Camerin_hahn brushed DC motors can run out of control speed wise. A brushless one gets current for each quarter turn. It would tend to keep in synch with the current pulses.
Regarding the brushless DC motor, the magnet poles and the electromagnet poles will want to stay in synch. The only question in my mind is if you will have to set an accelleration limit to help the poles stay aligned. That could be figured out during print testing, just as it would with a stepper motor. As I said, I plan on using it as a stepper as opposed to trying to sense what positiin the motor is so that it could speed up more and more.
Theoretically, any sensorless brushless ESC knows exactly at which position the motor is at what time and should be able to keep track of the number of rotations etc as long as the motor is spinning. Sensored ESCs and motor combos even close the feedback loop and are able to keep track of positioning even with the motor disabled.
So theoretically, you’d only have to modify the ESC software and you’d have yourself a super-low-torque, but super-high-speed and super-low-resolution servo, which unfortunately is exactly not what you want for a 3D printer axis. It sorta works for gimbals (which close the loop with a accelerometer) because they are a) balanced to get away with the comparatively low torque of a BLDC and b) profit more from the low-resolution, and therefore smoother torque from the BLDC.
Btw, permanent magnet (including hybrid) steppes and brushless motors are practically identical, they are wired slightly differently, but work on the exact same principles. One is optimized for positioning, the other for power density.
@Thomas_Sanladerer I would never use the high speed motor for anything else than a screw drive which would take care of the torque problem.
ESC = electronic speed control
An issue that has not been mentioned is that the non-accumulating error in each step is rated as a percentage of each step, and this is probably significantly greater (if they even bothered to measure it) on a BLDC motor than on a stepper.
A 3-wire brushless motor with the right ESC firmware to hold positions like a stepper will probably still have a significant amount of play in its position. These motors are not designed to hold precise positions with minimal error, but rather to move as smoothly as possible through a series of steps controlled by pulses from an ESC that might not have the best timing (carried by inertia), so I would expect them to build more play into the system.
They’re probably built to be more “squishy” than a stepper. This means that while it’s holding position, you may very well be able to turn the shaft a dozen degrees or more in either direction before getting significant resistance, while I would bet that a hybrid stepper would have comparatively less play relative to its step size.
Since BLDC motors use near-identical operating principles as steppers (but with different phase control / commutation schemes) I think it’s probably safe to assume the load-induced error / step size behavior will be similar to a stepper.
@Ryan_Carlyle but most of the BLDC motors I’ve seen have rotors and/or stators that are at least slightly helical, so that the poles don’t align perfectly. This seems to be for the purpose of reducing detents, but it will also add the “squashiness” I described, as well as resulting in a general reduction holding torque.
@Whosa_whatsis good point, so like steppers (load error directly related to step size) but some unknown amount worse.