I’m changing the motor because the gears on the smaller one continuously break…I’m overloading them. The other reason for changing them is in order to increase the overall size of the device to make it closer to myoriginal design , which is not just a matter of appearance, but by increasing the size I hope that the robot can increase it’s range of impact. When bigger, it can still be cute and appealing, but may also be threatening…the smaller version cannot be the latter.
I have found a new motor which I’m going ahead with testing and so forth. It’s a decent size (not too big, not too small), is a decent speed for handling (50 rpm), has good torque (yet to measure), has a good gearhead (looks robust), runs at 4 Volts (my desired operating voltage), is affordable (about $10 each), and is in my hands today (important when on a tight schedule)! I bought it at Creatron. I looked up the manufacturers. The motor is a Mabuchi rk-370c, which is apparently a 24 Volt motor! This is confusing because the motor was sold as being a 6 V motor, which you can also see on the label in the photo below. They seem to run well at 4V, my desired power. The gearhead is by Shenzhen Pengguang Electronics Co. Ltd, but good luck finding out about them online.
The new motor measures 2.11″ long (not including shaft) and .98″ in diameter. It’s about twice the length of the original.
The shaft is 0.158″ or 5/32″ in diameter, and .379″ long, with a 0.095″ long and 0.274″ diameter bushing between shaft and motor. It has a flat for a set screw.
It draws ~0.7A at 4V when stalled, and ~1A at 6V. Don’t know torque yet.
It has 7 gears (6 steel, 1 brass) in the gearhead, each measuring 0.365″ in diameter, and ~0.07″ thick. Which seems much more robust than the originals…the breaking of gears was a major problem. I should find out more about material properties.
But what is a good size? The original module was 2 inches on the short sides, with the 1″ motor placed in the apex the facet. If I scale up the model as is, the module will grow quite a bit: 4″ on the short sides, and 5.5″ on the hypoteneuse. That seems awfully large to handle. To keep it smaller, 3″ on the short sides, 4″ on the hypoteneuse, the motor would have to moved off center, which would require (sigh) a redesign of the module. Here’s a sketch of possible and approximate module sizes with the motor in different positions.
If the motor is moved off center, than gears are required to transfer the rotation from the side where the motor is located back to the apex of the facet. I looked up some gears but not much luck yet. They are too expensive and the materials present issues (steel is heavy, and other materials are perhaps not strong enough). I looked at Internal gears and geared collars on Smallparts.com (great site), but spur gears may be more in order…will do more searching.
Moving the motor also frees up the center for other features. Off the shelf slip rings could potentially be used, which would probably work a lot better than the ones I designed on the first version. All the same, I did look at ways to improve the DIY slip ring. I used music wire as the spring pins the make contact between disks. These pins were not uniform, and didn’t all make continuous contact. I’ve been thinking to use roller levers instead, as I mentioned in the last post. Buying just a lever from digikey is ridiculously expensive ($20 or so each). It’s cheaper to just crack open a $1 switch and fish out the lever. They come out in tact.
The all-metal versions are conductive (think they’re steel). These ones can pass 1A of current easily, and start to warm up at 2A. May be possible to use them as part of the power bus, especially if several are used on the power rings of the slip disk.
Some power considerations:
The new motors stall at .7A. So let’s say there were 16 modules in the snake. That means there could be up to 12A on the power bus in an extreme situation (all motors stalled). Also I haven’t measured power spikes when the motors start up, which might be more. Still, these rollers should be able to handle that much amperage in short bursts. Under normal circumstances I would imagine there would be less than 5A on the bus. Modules should be able to coordinate in order to distribute and reduce load. They should rarely be all maxing out at the same time.