Sunday, March 13, 2011

p0w3r Drive: Sexy Things that Spin

Believe it or not, but I am halfway through revision 2 of the hubmotor. You ask, "what happened to the first version?"

Simple answer: I got lazy and the aesthetics bothered me. I mean, the motor was 1.625" wide, with only 1" of filleted tread. The 1/4" endcaps looked bulky and unattractive but I couldn't add an edge chamfer because the inner diameter features were already finished. Additionally, I neglected to add sufficient wiggle gap to the magnet ring rings, which made it extremely difficult to get the magnets in place. One formerly excited afternoon transformed into a painfully disappointing sanding feat. Once I had ONE magnet in the spacers, I knew that this was not worth the effort.

So with all these changes to be implemented, I elected to simply not finish version 1. Here is version 2: incredibly badass and sexy.

The wheel is now only 1.25" wide, with a 1.5" hubshaft (more wire space, finally!). I will still be using 1/4" endcaps, but I will turn down the edges and internal features so I can sink the endcaps further into the wheel. This also allows me to rid of the polycarbonate spacers, which served no purpose whatsoever. The mechanical lock nubs were replaced with precision eyeball machining aka I will be relying on a frictional fit from the endcaps and the OD of the magnet ring. This might seem like a bad idea, but It is better than trying to use 4-40's to lock the wheel and any slipping might act as a neat clutch feature. I also added .016" of space to the magnet gap so the magnets just drop in. No file work required.

To make the endcap features, I will be adopting a trick I learned from Shane Colton. First, machining will begin with a waterjet rough circle cut. A severely undersized internal bore and the seven through-bolt holes made. The bore will have a capscrew through it, accompanied with several nuts. These nuts will then be clamped in the chuck of the lathe, and all outer features will be finished. Thereafter, the bolts will be removed, and the bore will be opened with a boring bar. More work certainly, but I expect a much finer product.


  1. Awesome!

    One caveat - I wasn't using the screw/nut to center the rough circle since the OD turning was the first operation. If you waterjet the bolt pattern first, it defines the center and the screw and nut rig might not be good enough. You could probably pressure-turn it with a live center into a waterjetted hole instead?

  2. So do you mean I should waterjet the circle and create the bore by clamping the circle on the outer edge and turning it from the inside? I thought about that, but I wanted to make some edge features to the disk, and the lathe chuck couldn't handle the small 1.125" bearing bore. As a result, I had waterjet a center hole in the same operation I made the radial holes. My hopes were to spin the disk using the screw method (hopefully perfectly centered), make the edge features, and then finish the bore for the bearings.

    Idk if any of this makes sense. I am pretty terrible at night time writing. But I promise the next report will have pictures and explain the process in better detail. Just have to wait until after midterms :(

  3. The part I was worried about was getting the disk, held in the lathe by the screw/nut, to be perfectly centered with respect to the 7-hole bolt pattern that was cut on the waterjet. If you can do that, the rest should go as you said.

    Otherwise, if you leave off the 7-hole bolt pattern and do that on the mill after all the lathe operations, that's sure to work, because you can find the center of the bearing bore.

    I did both ways and they both worked. But my air gap is also huge, so tolerances were pretty loose.

  4. Yup. I totally understand what you mean now, haha. My second endcap has a bit of wobble in it but I think I have a remedy. Perhaps I can use a flathead screw instead of the socket cap, and countersink the center hole. Theoretically it should self-center right?

    My spring break starts next week. I have additional material so I'll probably give it a try. The reason why I dont want to add the holes later is because I dont have a rotary table and I'm lazy :P

  5. Hi Jamison,

    Do you have an email I can contact you on? Have a few questions. You can grab me on