So You Want To Try This Yourself? Well...

Over the last few months, I've gotten a number of requests from readers for help motorizing their own bikes, and most seem to want plans and parts lists for whatever scheme they read about here, but there are no plans and no parts lists--it's all done on-the-fly, with whatever I have on hand, whenever possible. Since I doubt most people will happen to have the exact same bike or set of parts available as I did for a particular scheme, I don't think knowing what exact parts or having exact plans of what I did will help them much. If what I show and explain on these pages isn't enough to get the basic idea across enough for someone to build their own out of what they personally have access to, they'll probably not be able to build it from plans and parts lists either--simply because it's highly unlikely they'll be able to find the exact same parts, or that they'll fit on their bike exactly the same way, and if they can make it work anyway with what they *can* find, well, they could have done it without the plans, too. :-)

So this post is a ramble on what is needed to do all this, sort of a more in-one-post gathering of some of the thoughts and choices I've made along the way. By no means is it complete (or it would be almost as long as the rest of the blog, by itself), but it might help those interested in doing this gather their thoughts and figure out what they want to do and can do.



First, you're going to need some kind of battery system, no matter what motorization scheme you choose.

The cheapest one is SLA, or Sealed Lead Acid. Scooters and computer battery-backups use these types, though the ones from scooters could generally be capable of higher output power for a longer time, often they are identical batteries, at least in the larger battery-backups. It's hard to find good used batteries, though, so you will likely have to buy them new, which can cost upwards of $40-$50 each for 12V/12Ah, which is the smallest size I'd recommend if you want any range, and you'll probably need at least two and probably three of those, to get 24V or 36V, to get the kind of power you need out of your motor (this depends on your motor type/etc). They're heavy, because they're made of lead, and don't hold a lot of power for all that weight, plus a few other disadvantages like slower charge times than other battery types, etc.

There's other kinds like NiMH that are much better in capacity vs weight, but also much more expensive. Li-Ion or Li-Po or Li-Fe-Po (like A123) are better than NiMH, but cost even more, and have a few other drawbacks, such as complexity of chargers and battery pack design, if you intend to roll-your-own instead of buying prebuilt packs with chargers.


Then you'll need some kind of motor, preferably high-torque and low-speed, but you can use a high-speed low-torque motor if you have something to convert the speed into torque. That means gears (like car transmissions use), or sprockets and chains (which bikes use). Little bitty gears or sprockets on the motor, and really big ones on the wheel.

The most effective way to do it that I've found so far for the low weight and lack of complexity is a friction drive with something really small on the motor shaft transferring power directly to the tire's outer surface on the wheel, because the smaller the diameter of the motor end of things vs the greater the diameter of the wheel end of things, in such a situation, the greater the change from speed to torque you get.

It's difficult to get anywhere near as high a ratio as that in *one stage* of reduction with gears or sprockets that are of a size or weight that's practical on a bike. Multiple stages of reduction are possible, but that adds weight and complexity, which means less range and more chance of something breaking that you can't fix on the road.

Best is to get a motor that will already do low speed/high torque, like a treadmill motor or some other kinds of load-bearing motors, some of them industrial. Permanent Magnet DC motors are best, because they use less power than Shunt DC motors (the other option for a reasonably small DC motor). Then there's two kinds of PMDC's, the Brushed and the BrushLess--the latter of which will need a more complex controller, but is more efficient. Some AC motors can be converted to DC, either by wiring them as Shunt motors if they already have a set of coils for the rotor and a separate set for the stator (like most of the hand drills I've taken apart), or replacing the outer diamagnetic ring with permanent magnets, making them PMDC motors (like I did in the Cieling Fan experiment). But easier to just find the kind of motor you want to have and leave it as it is.


Then you need some kind of motor speed control, which is best done with an electronic version, or ESC. The ESC can be built from scratch easily enough (there are basic schematics around the web, though rarely with values matching what you'll need, thus you'd have to adapt them to your purposes), and fairly cheaply if you can find enough old motherboards or computer power supplies that use high-enough voltage MOSFETs on them to parallel enough MOSFETs together to get the current load divided up between them enough to keep from blowing them up. And if you do parallel them, make sure you have a few extra in there, or some sort of current limiter to shut off power if one blows, otherwise the extra power being drawn thru the rest of them after one dies will quite possibly catastrophically destroy the rest of them in less than a second. Probably so fast you won't even know that's what happened--there'll just be no motor power anymore, and maybe some smoke (not always even the smoke). Probably all the parts you need can be found in old circuit boards, if you have enough of them and/or a place to look thru old discarded electronics for them.

It's way easier to just buy an ESC, and the ones from http://4qd.co.uk are a good choice, depending on what you want out of your ESC. Just remember that if you think you're going to run a lot of current, get an ESC that can easily handle *more* than that current at it's running stage, not just at it's *peak* stage, or else you'll eventually burn out the ESC, especially if you put a lot of strain on the motor. Don't forget that you need to choose your battery pack voltage before you choose your ESC, or else you could blow up the ESC by using too high a voltage on it.

It's possible to get by without an ESC, and just use a switch to turn power on or off to the motor, but then you only have two speeds--pedal, and full-motor-assisted. :-) It's kind of hard to control acceleration that way (I've tried it). A slightly better way is if you have battery packs with taps at different voltages, such as 3 12V SLA batteries, you could set up a switch or set of switches to engage just 12V for low speed, 24V for medium speed, and 36V for high speed. Again, not much control, but better than none, though I can say that having tried it this still doesn't provide a really workably safe solution.


After you have these three things, you'll have to actually get them mounted on the bike, and work out all the little things you didn't plan for when considering what you'd need originally. :-) If you do it like I have, with simply whatever is available at the time, it's harder to predict all of what's needed till you're in the middle of it, but it can be more fun this way, if you like the challenge. More frustrating, as well, when you realize that because the bolts you happen to already have are 2mm too short, you can't use them and have to either go buy some that are long enough, or rethink your mounting strategy, or something else entirely. :-( This has happened a few times, with different kinds of parts, including on my current homebrew recumbent that's still in-progress.

If you're not just using whatever you already have on hand, and are going to go out and buy whatever you need for it, you'll either want to keep costs really low by planning every single thing out first and making sure you find the three most expensive things recycled somehow, or else just go and get one of those cheap conversion kits, probably a few hundred dollars at most, and save yourself the trouble and money of figuring all this out after having spend more than a kit cost on the parts you need. (If I'd gone out and bought new all the parts used in the last scheme I've used so far, I'd've spent enough to buy one of the low-end kits that might do the same job or better--but I've spent virtually nothing on it so far).


If you're wanting to build one of the same ones I've built, reply to this post with which one, and I'll make a new post with all the details of what I used to make that, and as much of how I did it as I can explain. There is an Instructable in the works for the last friction drive one (that I'm still using), but I don't know how long that will take to finish (weeks, probably, since I'm occupied with the recumbent as well, and it's associated Instructable).

If you're just wanting to build one generally like what I've built, well, it depends on what resources, tools, and skills you have available as to which version you'd be best off with. Also, you'd be pretty much on your own as to how to adapt it to your bike, since your parts would be different from mine. Replies to this post with questions will still be answered, but it's going to take a lot of back-and-forth to figure out exactly what's needed, most likely.

If you are very mechanically handy, and have lots of good tools around (including welders, lathes, drill presses, and other machine-shop tools), with easy access to whatever parts you might need, it's pretty easy to make a number of different kinds of motors work, such as the two different kinds of radiator fan motors I have shown on this blog.

If you're very mechanically handy but don't have much in the way of tools, say just hand tools, then there are a lot of things you won't be able to try. At least, not without a lot of patience, and/or alternate ways of doing some of the work, potentially building some tools to do the work first, before you can build the actual devices. That's pretty much where I started with this, but now with only a few power tools, like an angle-grinder and a welder, I have a lot more options than I did before. With the lathe from a previous post now nearly usable (made a belt for it), I have even more things I can do, especially since there are ways to use it as a drill press.

If you are also electronics-saavy, you can probably build your own ESC from scratch, which will save you money but might not be as good as the ones already out there (and I'll tell you from experience that having your ESC smoke out on you when you're 3 miles from home with a heavy load you had expected the motor to help you pull is no fun at all. :-(


Access to parts can mean anything from having a bunch of stuff already stashed away in your storage areas or garage or whatever, to lots of yard sales or thrift stores nearby, or even just being on the Freecycle.org mailing lists for your area to be able to pickup (over time) whatever you might need for free (which I do reasonably often). Almost everything ends up on Freecycle eventually, if the groups you are near are large enough. There's also Craigslist, and if desperate, Ebay. :-) I prefer to use Freecycle and yard sales because that way I'm recycling things that probably would have been thrown away and wasted otherwise. Also because Freecycle is free, as are all items posted on it, and often yard sales are cheap or free for some things, especially if they see my bike and I talk about what I'm doing with it, and how I'd be using the things I want that they have.


So, questions and comments are welcome as always, though I don't get too many. :-)