Search all of my sites with Google

Tuesday, January 15, 2008

Of magnets, wire, and gears

I have done a bit more work on converting the cieling fan motor into the hub motor, mostly figuring out what calculations I have to worry about, and researching magnets and whatnot that I'll have to buy someday to finish it. Unfortunately there's nothing I can find to take apart that has the magnets I need in it, which have to be the rare earth Neodymium types, and of the type that are polarized thru their thickness, so the north or south pole of each one faces away or toward the windings/motor axis. I can get them as cheap as a buck each, but I need at least 12, and preferably 36 of them, plus whatever shipping costs (typically $5 or $10 if they can be shipped ground, but way more if they have to go by air due to extra shielding needed by law).

The harddisk magnets I have now are nice and powerful, but totally wrong polarization and shape. I *can* still use them for a motor, if I can get enough identical ones (so far I have no more than 4 of any one kind), but it is a very different design of motor, with the windings still in a radial pattern on the fixed portion, and magnets still on the rotor outside that, but in a different configuration. The windings would be done in the plane of rotation, instead of perpendicular to it, arranged rather as an orange cut equatorially looks, with each orange interior segment representing a separate winding. The magnets, instead of running around the outer circumference, outside the radius of the windings, would be placed in the circle to either side of the windings, and have to be on *both* sides, needing twice as many magnets as the other design. Since they would then be segments of a flat arc, though, with poles along that arc, the harddisk magnets would be ideally polarized already for such a design.

The catch there is that I would have to machine a rotor for each side to mount the magnets on, and also a support plate for the windings, along with core laminations for each segment, all of which must be precisely machined and well-balanced. I have no tools to do this, and would first have to build a lathe and a drill press, and then a balancer for measuring and correcting the imbalances in each part, so the motor wouldn't shake itself (or the bike) apart as it spins.

It also is not in theory as efficient as the other design, but the ones I have seen use no iron cores in the windings, and that severely limits the field strength of the windings, compared to the other design, making for less torque (though possibly higher speed because of less cogging caused by magnetic attraction to the cores from the rotor magnets). With a core in each winding, it would improve the torque a lot, but also increase cogging, probably dramatically since there will be opposing magnets on each side of it.

Right now, I can't build *either* motor, because I either need to buy magnets, or I need to machine precision parts with tools I don't yet have. So I'm again looking into alternative existing motors.

A friend is going to haul a Toyota Tercel to the junkyard soon, so he is letting me part out whatever I want off of it first. I already got the small fan motor it has, which is a highspeed low-torque motor, which if I can build a reduction gear set for it (it's too big a speed difference to do it with chain and sprocket--even if it didn't break something it'd be VERY noisy and dangerous to leave exposed to potential chain contact with anything) might actually be able to drive the bike a little, though not as much as I need, I think. One really nice thing about it is that it's a sealed motor, so it's essentially weatherproof--ideal for the bike for that reason, at least.

I also got the car battery, which while old does hold a charge. Even after only a couple hours of charging it would run one of my higher-torque motors well enough that I could not hold the spindle stalled with pliers after power was applied, so it will be useful for testing stuff "in the lab" so to speak. It's too heavy to use for actual bike power, of course (I think it weighs more than the whole bike; I can barely lift and carry it). I'd like to get the alternator and starter off, too, since they have wire on them I can use to rewind onto other motors, but I would have to jack the car up to do it, since they're *under* the engine and behind it, since it's a transverse-mounted engine. I'd also love to take out as much of the engine compartment wiring harnesses as possible, since they have "weatherproof" connectors on them, but I didn't have time to do that, and might not before it is hauled off.

I also have a vacuum-cleaner motor, from a Dirt Devil that was donated to me since it has broken plastics on it. Motor works fine, and since it's a shunt motor, it will even run fine on DC (battery) power, even though it's from an AC-powered vacuum. It just takes a lot more power than a motor with permanent magnets in it would to do the same work, since the "magnets" are being created by the shunts using lots of extra battery power, in addition to that in the regular windings. Again, it's a highspeed, low-torque motor, and not nearly as suitable as the fan motor would be, either, since it's an open-cage motor.

The problem with these highspeed motors is that I must build a reduction gear set for them, and I really have no easy source of tough gears--just plastic ones from laser printers and the like, and even there, most of them won't mesh with each other in the ways I need them to for the right ratios.

Again, I'm stuck with having to precision-machine parts (gears, in this case, even harder than the other stuff), or go buy some from somewhere. I'd love to take the transmission of the Tercel apart, but I know I don't have the kind of time it would take before it's gone, and I also couldn't lift the engine out to do that. I *might* be able to get the differential apart, if it has one, but I think this is a front-wheel-drive, and I've never taken one apart to see what kind of gearing it has in it. I don't even think the rear differentials have the gearing I need, so probably front ones won't either. I really need to get a pile of tough gears of many ratios that all have the same tooth size, so I can use any pair of them together for various motor experiments. Just haven't found a good source yet that is as close to free as possible--most want more than ten bucks *per gear*, which is ridiculous for used parts rusting away in a junkyard. :(

Ah, well, I'm learning a lot, and it's fun, if a bit frustrating due to insufficient resources for experimentation. It's actually made me more thoughtful and inventive in my scavenging, as I am finding parts sources for things I would not have imagined before. For instance, TV's have a lot of loosely wound copper magnet wire in them, on the yoke of the CRT. Old CRT PC monitors, too. The bigger the tube, the more wire (and sometimes thicker, too). Most AC motors are loosely wound, and easy to get apart. Tranformers are often also this way, and big ones are found in many large UPS's, though there aren't many turns of wire on them, apparently, from what I'm finding. Thick, though. (one of the Tripplites I took apart has a transformer with magnet wire more than 1/8" thick, and square in cross-section, stiff enough I could make bracelets out of it!).

I also will have some more bikes to experiment upon tomorrow, as I finally got first dibs on something from the Freecycle mailing list (after a bit less than a couple months of trying). Unfortunately someone else I was to go get some old (but working) celphones from tomorrow morning already gave them away to someone else, after telling me I could come get them. At least she did let me know they were gone tonite, rather than letting me go tomorrow and find them gone when I got there. Can't complain much, though, since it's all free stuff someone else is giving away because they don't need it anymore. :)

The irony of much of the Freecycle stuff is that if I had a working motor on the bike, I'd have the range and speed I need to get more motors--there's been a number of people on outlying areas that have had things I could use to finish a bike with, but I would need a finished bike to go get them, unless I feel like pedalling out there for 5 or 6 hours, and then the same back again. I'd have to leave in the dark in the morning, and come back in the dark at night. :(

However, perseverance and patience are my friends, so I will get what I need eventually. :)


  1. Gee, why don't you just buy a Cryalyte motor for $250 and be done with it?

  2. Well, I suppose that Anonymous here that posted this comment hasn't actually read much of this blog, if any of it. Especially the parts that talk about not buying anything for it wherever possible due to lack of money and the goal of trying to use 100% recycled components. :-P

  3. Well, it might be totally outdated , but your article is pretty interesting, have you done any more changes?

    curious to read/watch

  4. On this cieling fan motor, I have gotten it rewound with heavier wire in a more useful pattern, and I have a few of the magnets I need (but not all of them), out of scrapped bicycle hub motors (similar to the misspelled Crystalyte motor the first Anonymous on this page refers to).

    The bike project itself has lots of updates, and has spawned one fully operational version of a friction drive using radiator fan motors, on DayGlo Avenger, though it is no longer in use due to breakage of one of the motor shafts.

    Another recycled motor, from a powerchair, drove CrazyBike2 (a custom-built semi-recumbent, also of recycled parts), for almost a year before various problems I had caused me to give up on it till I can fix them permanently.

    Until then, it is running on a recycled hubmotor and recycled NiMH batteries (as is DayGlo Avenger).

    More can be read on the projects as they are now over at under membername Amberwolf. I'd put a direct link to the various threads for the bikes, but Blogspot doesn't wrap or htmlize links in comments, so they dont' work. Short ones like the above I can manually htmlize.


Alternate suggestions or improvements to anything that's been posted is very welcome, and extreme detail is preferred to brevity.

Keep in mind that unless you leave an email address in your comment, I haven't any way to reply to you except to reply to your comment here. That means if you want a reply, you'll have to come back to *this* blog entry and it's comments to see my reply to you, unless you leave some method of contact within your comment.