Cardboard Robot Arm's Stepper Mounting and Mechanical Advantage

Martin_H

I've watched the cardboard robot arm video:

and taken a look at some of the static images:



I found his method of linking the steppers to the cardboard disk interesting. What he's done is attach thread to each end of the disk, wrap it several times around the stepper spindle. The friction from the thread on the spindle will be quite high as long as there's no slack in the line. So effectively it is like a belt drive. I think this should give the stepper a fair amount of mechanical advantage over mounting it at the center and acting on the disk axis directly. It also removes the need to make the stepper shaft and disk center concentric.

I want to try to build a stepper controlled lazy suzan using my ebay Chinese stepper motors. They're geared down, so the might have enough torque to act on the center, but acting on the edge of the disk might increase that force. It might also be easier as long as the thread has no slack.

But I figured I would ask people with more mechanical experience what they thought of this step up?

erco

SLOW. Very low RPM of the Chinese SMs will be even slower geared down using cables. Cables will eventually stretch & slip (undoing all calibration work) using just a friction wrap. Better to use a long shaft, run the cable through a hole through the shaft (like soap box derby steering) and put many turns on either side to reduce/eliminate slip. But this will limit how much cable travel you can get. Life's a compromise...

Martin_H

What about attaching the stepper shaft to the center of the lazy susan. What do you think about that option?

erco

Depending on the weight (moment of inertia) of the loaded turntable, might work fine. If you want positional accuracy, you'll need to ramp up & down more with a bigger load.

Martin_H

I called my local hardware store and they have lazy susan bearings in stock, so I might try this tonight. BTW I found a web page http://www.alibaba.com/product-gs/573569488/Stepper_Motor_for_Instrumentation_equipment_Automatic.html which says these steppers are intended for use in a "Automatic vidicon Money leveling machine Camera Air conditioner", I wish I knew what that was.

Martin_H

I cut a disk for the turntable and assembled the lazy susan fairly easily. I also mounted the stepper in the middle with its shaft going into the center of the turntable. But testing showed some binding, so the stepper shaft is not completely concentric to the lazy susan bearing. It's close, but not dead center which is what I feared.

I'm thinking that I might be able to resolve the binding by letting the stepper floating around bit on the mount. As long as it doesn't turn a bit of play shouldn't reduce accuracy that much.

The other problem to solve is creating a secure connection between the turntable and the stepper shaft. One thought is to crimp a tube so it matches the flattened shaft and the epoxy that in the center of the turntable. The other is filling the hole in the turntable with epoxy putty and pressing the shaft into it. The idea is that the epoxy putty will harden into a shaft profile and make a secure connection.

Martin_H

I used a flexible shaft coupler made out of aquarium tubing to couple the stepper to the turntable. This did the trick in terms of providing enough flexibility to handle the lack of concentricity. However these small steppers don't have the torque and it stalled too easily. I was energizing two coils at a time too.

So I'll need to try some sort of belt drive which means that these motors are likely too slow. I have some other scrap steppers recovered from old printers which might be acceptable instead.

Martin_H

Finally, some modest success. The ebay stepper does have enough torque to turn the turntable by pulling on a string wrapped around the edge. Here's a video

I think that this proof of concept has gone as far as I can take it. I would need a faster stepper, better belt, and a turntable with a grooved edge to hold the belt to go further.

erco

Nice demo, pal! The steppers are an interesting experiment for certain precision applications (small mobile robot, etch a sketch, maybe even a rapid prototyping machine) but they won't replace simple DC gearmotors for simple, general motion anytime soon. Steppers require more power, I/O pins, and processor time.

Martin_H

erco wrote: »

Nice demo, pal! The steppers are an interesting experiment for certain precision applications (small mobile robot, etch a sketch, maybe even a rapid prototyping machine) but they won't replace simple DC gearmotors for simple, general motion anytime soon. Steppers require more power, I/O pins, and processor time.

Thanks. One thing steppers have going for them is their accuracy without the need for encoders. An H-bridge controlling a DC motor uses two or three pins, plus another one or two pins for a shaft encoder. So the pin count isn't so bad and with a Propeller you can dedicate a cog to pulse generation.

But an RC servo wraps all of this up in an easy to use package controlled via a single pin. It's no wonder that I keep going back to them.

garyg

Martin_H wrote: »

Finally, some modest success. The ebay stepper does have enough torque to turn the turntable by pulling on a string wrapped around the edge. Here's a video

I think that this proof of concept has gone as far as I can take it. I would need a faster stepper, better belt, and a turntable with a grooved edge to hold the belt to go further.

Lazy Susan bearings are pretty sloppy.
I mount them upside down from the instructions, so that the weight of the lazy Susan tightens the bearing.
I use a power screw driver to drive my fishing line lazy susan turntable.
My problems with driving things with motors, is that the motors are usually too fast.
The approx 10:1 reduction of your turntable is a good thing for my applications, and a bad thing for your stepper motor application.