0
$\begingroup$

I was successful in running a 28BYj-48 stepper motor with 4096 steps per revolution. I am happy with the result. But there is a free movement of the motor shaft itself - you can move the motor shaft slightly with your hand without making any steps. Feels lose, but event the brand new ones has that motion, so I am sure its manufactured like that.

Is there any way to reduce that unwanted free motion? I am trying to add a hand with the motor and now that small movement became quite a noticeable gap.

This is my first question on robotics - so please be pardon any mistakes.

$\endgroup$
1
$\begingroup$

It's a geared stepper, which means that it's a stepper (64 steps, I gather) followed by a 64:1 (approximately) gearbox. What you're seeing is almost certainly backlash in the gear train (see where I "suspect a poor mechanical joint..., below). Any gear motor, stepper or not, is going to have backlash. Because there's a tradeoff between the precision of the gears and how much backlash you can get out of the assembly before the thing binds up, a cheap motor is going to have more.

If you can preload the motor with a constant torque that only points in one direction, then you could reduce the backlash (although maybe not eliminate it, given the amount of gear reduction and the cheapness of the unit). If you need more precision that that, then you're going to either need a direct-drive stepper (which, thinking about, is going to be hard to find with that many steps), or possibly a geared stepper with a much higher precision gearbox, or you're going to need something like a bazillion-pole brushed or brushless "regular" motor, position feedback, and a controller.

(Former answer left in because it's valid for a direct drive stepper, if you find one with 4096 steps).

How much motion are you seeing? Just by the way that the magnetic circuit is designed, you should expect some motion within a step (in an unpowered motor it's not terribly inaccurate to think of the rotor as being connected to the frame through a spring, at least until you overcome the motor's holding torque).

If there's more slop than that (which works out to less than 1/10th of a degree in your case!) then I'd suspect a poor mechanical joint between rotor and shaft, which is just wrong. Did the motor come from a reputable source?

$\endgroup$
  • $\begingroup$ The motors are from seeed studio. I have 3 of those motors and all three behaves the exact same way. To be honest, it looks like that the motion is unavoidable because of the mechanical design. But is there any way i can reduce that motion by putting in more weight or something ? $\endgroup$ – NH Evan Dec 25 '17 at 12:26
1
$\begingroup$

Adding a worm gear to connect the output shaft to the load you are trying to move might solve your problem, since worm gear arrangements are "self locking".

Here is a reference for you: http://mechteacher.com/worm-gear/

By the way, what the others have said is 100% correct. It seems you might be wanting some way to minimize the inherent play that is always present in a gear train. Also, you might want to look on the net for different methods used to control backlash to get ideas on how to control the movement.

Also, is the motor powered or not powered when you detect the movement? Whether or not you can keep power applied will make a difference, too.

$\endgroup$

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.