Keyed through-axle hubmotor conversion...

[Kodin]

Been thinking about 15 and 20mm through-axle hubs lately and it got me wondering - What stops someone from making a solid 15mm shaft, popping a hubmotor shaft off a hub, installing a key into the keyway of the stator permanently, and cutting a keyway slot in the shaft so it can slide in and out of the motor effectively without causing issues with the bearings? In order to center the stator, add tight-tolerance tubes that can be snap-ringed either side of your bearings, and butt up to the stator bore. If the keyed shaft is ~0.005mm undersize for the bores, (or thereabouts,) it might be possible to retain serviceability while having a heavy through-axle design available. As far as dropouts go, make them clamping with keyways as well, and just pop in keys when you install the shaft. If anyone likes the idea, I'll draw up some diagrams later to explain the idea.

 

[spinningmagnets]

Doctor bass had a shaft custom-machined to convert a MXUS 3000W to a through-axle. Apparently, all it takes is money...

 

[Kodin]

I'm thinking far simpler than that; make the motor built on a tube with internal keyway(s), where the key(s) are mounted to the motor itself. Then have a shaft with a keyway cut all the way along it's length act as the "through" axle. Keeps any keys from restricting the axle's movement while still allowing for insertion/removal in a relatively easy way, and the key restricts rotation. From there, either cut a keyway into your dropouts, (hopefully also clamping,) or add torque-arms that have keys. I'd have suggested the axle be splined, but that seems overkill for the application, not to mention it's less likely to be retrofit-able with existing through-axle dropout designs. I'll have to look up specs on keyed shaft torque limits to get a better idea, but at first glance it seems doable.

 

[Punx0r]

Wouldn't there be nothing but the force of the magnets preventing the stator sliding about on the shaft in use?

 

[amberwolf]

In order to center the stator, add tight-tolerance tubes that can be snap-ringed either side of your bearings, and butt up to the stator bore.

 

[Drunkskunk]

Actualy, the axles are splined, that's what keeps the stator on the axle.

With enough money, you could make this work. But generally the reason a keyway is used is so you have a "fuse" in the design. the key shears off before the system is damaged. You wouldn't want that with a hub.

There is no reason the axle needs to be round. There's no reason the torque arms needs to attach to the axle. if you set out to redesign the axle, there are a world of possible ways you can make it better.

 

[Punx0r]

Thanks AW, I didn't get that earlier.

 

[Kodin]

Well there's two ways it could be done - One: single large-diameter hollow shaft, which the keyed shaft slides into. Two, and the way I'd prefer: swap bearings in side covers out with ones that have larger inner diameters. Then add tubes that go through the bearings and butt up to the stator shaft bore.

Since my CAD isn't done, here's an MSPaint drawing to explain.

Idea is that with separate tubes you can maximize the diameter off the solid shaft for strength, while still allowing accurate centering of the stator to the side-covers with as much or as little preload in or out as you'd like. Also makes motor disassembly easier, though you'll probably need to put a 45 degree chamfer on one end of the shaft as the stator will want to jump to the magents when the shaft is removed. Worst-case you could hollow the shaft slightly to allow for a longer small-diameter bolt to be placed in and used to pull the shaft into place for installation. Interestingly, phase/hall wire routing could also be done via the same keyed channel if practical.

What do you guys think so far? I'd like to know of any and all pitfalls that can be found.

 

[teklektik]

Coming soon at a theater near you....

You can also see that the motor has a hollow aluminum axle. This is 20mm ID and allows the motor to be used with thru axle suspension forks, or with stub axles for supporting from a single side only. For normal forks, we just stick in a small insert to each end that makes it quick release compatible. And the hexagonal shape on the end of the axle is for an integrated torque arm. So there will be no spinout forces present on the fork dropouts.

from: A Freewheeling Stoked Edgerunner Hauling and Touring Mac

 

[Kodin]

Yeah, I saw that as well; I really do like the hexagonal torque arm engagement, as it's simple to fabricate on the end of the builder. Only question I have is say you mount this to a front fork, how do you install or remove the motor? Not all forks have the play to stretch the dropouts over the hexes. I guess it would work well if you had removeable torque-arms, but that ends up being a custom machining job for each and every fork tube diameter and design on front fork applications. Then again, running a keyway broach through clamping dropouts for an aluminum dropout fork may weaken them too far, so my argument may be moot. Sure does solve the stator-jumping-out-of-alignment-when-you-remove-the-shaft problem, but you may or may not need to enlarge the bore of the stator to make that work. Definitely an option, just not one that will work with all configurations. Still trying to factor this into a design that could also be adapted for cassette side-cover rear motors as well with smaller diameter axles.