Modified Car Alternator Powers Speedy DIY E-Bike – Hackaday

Your garden variety automotive alternator is ripe for repurposing as is, but with a little modification, it can actually be used as a surprisingly powerful brushless motor. Looking to demonstrate the capabilities of one of these rebuilt alternators, [DIY King] bolted one to the back of a old bicycle and got some impressive, and frankly a bit terrifying, results.
We should say up front that the required modifications to the alternator are quite extensive, so before you get too excited about building your own budget e-bike, you should check out the previous guide [DIY King] put together. The short version is that you’ll need to machine a new rotor and fill it with the neodymium magnets salvaged from hoverboard motors.
Once you’ve got your modified alternator, the rest is relatively easy. The trickiest part of this build looks like it was cutting off the bike’s rear wheel mount and replacing it with a plate that holds the alternator and a pair of reduction gears pulled from a 125cc motorbike. Beyond that, it’s largely electronics.
Naturally, you’ll also need a pretty beefy speed controller. In this case [DIY King] is using a 200 amp water-cooled model intended for large RC boats, though interestingly enough, it doesn’t seem he’s actually running any water through the thing. He’s also put together a custom 1,500 watt-hour battery pack that lives in a MDF box mounted under the seat.
To test out his handiwork, [DIY King] took to the streets and was able to get the bike up to 70 km/h (43 MPH) before his courage ran out. He thinks the motor should be able to push it up to 85 km/h, but he says the bike started wobbling around too much for him to really open it up. In terms of range, he calculated that while cruising around at a more palatable 30 km/h (18 MPH), he should be able to get 100 kilometers (62 miles) off of a single charge.
If you like repurposed motors and suicidal bike speeds, you’ll love this build that uses a washing machine motor to push a rider to a claimed 110 km/h. If you’re not worried about speed or range, then this supercapacitor e-bike is certainly worth a look as well.


[Thanks to Mike for the tip.]
i wonder what makes it wobble at speed. 43mph happens to be the fastest i’ve ever gone on a bike and my bike gets real stable & easy to handle anything over 10mph. but i’ve seen little kids taking downhills where their bike picks up a definite wobble, i just don’t know what causes it on some bikes but not others.
Probably un ballenced wheels. At > 50 kph my nice road bike wheels wobbled due to the weight of the valves
Speed wobble is the subject of much spilled ink:
https://www.sheldonbrown.com/shimmy.html
https://selleanatomica.com/blogs/homepage-blog/the-speed-wobble-surviving-the-terrifying-cyclist-experience
Why a specific bike starts to oscillate at a particular speed is unclear though, lots of variables…
My experience is that extra spokes add to stability. I have 36 spoke wheels, a little heavy but great for urban potholes. I’ve hit 55 mph on a long downhill with no stability issues at all.
“Old-fashioned” steel frames have optimal dynamics for rider comfort by reducing vibration, maybe also contributes to stability. Aluminum and carbon fiber are too stiff, transmit vibration to the rider.
Steel bikes were designed to have fairly stiff frames but compliant forks to give the sort of ride you’re talking about. It wasn’t inherent in steel, just that’s what racers liked. Carbon frames are designed to have stiff forks as well, because that’s what’s in style now. If you put a long rake thin steel fork on a carbon frame it’ll give you much more classic grand tour ride characteristics.
That’s the cool thing about carbon: you can give it any stiffness characteristic you want by varying stackup direction.
I’ve had 24 spoke wheels up over 100km/h a few times.
Wheel balance is important. Not hitting the fork resonant frequency is extremely important. The fork resonant frequency is affected by its stiffness, the head angle, and the rake. But just the rake/head angle/loading can also lead to an oscillation condition, particularly with touring bikes that have heavy loads.
As noted above, it’s a really complex system, and it’s rarely only one thing.
It should also be noted that what people call a shimmy could be an unbalanced wheel, which is felt as a radial disturbance, or a classic shimmy, which also involves yaw, or I’m not sure what the name is but a straight front/back oscillation of the fork, that has no yaw (a little bit of pitch) and feels similar to an unbalanced wheel but is much more dangerous. High speed fork oscillation often leads to fork failure and just appallingly bad crashes.
The wheels are unbalanced and need to be taken to a tire shop, as they have the machine and weights to correct it. Though it’s doubtful it will work on bicycle wheels.
You might check with the bicycle shops that specialize selling those bicycles used by world class cyclists.
There are probably other issues since your typical el-cheapo bikes were not engineered to be ridden at freeway speeds.
If it were me, I’d scrap this approach and buy a used scooter or 125cc street motorcycle.
If you check out the geometry of the front forks on the bike. There is very little (or no) rake. More rake would dampen the oscillation out. I have a lot of experience at speeds in excess of 60mph (100kph) on a road bike. You get vibration from the unbalanced wheel, but no wobble.
p.s. My speedometer stopped at 72mph, only exceeded it a couple times. Speed Thrills!
60mph! Wow. You must have even more Olympic medals than this guy:
https://mobile.twitter.com/chrishoy/status/668112825894748161?lang=en
There’s a *big* difference between getting up to 80mph on flat ground through muscle power alone and getting that speed because you’re converting gravitational potential energy to kinetic (i.e. going down a steep hill.)
yep – I regularly used to do 80km/h dwon a big local hill (on a reasonably good bike). Not a problem and very stable – I could go faster but I don’t have the right gears… It seems way faster than when you do 80km/h in a car..
On flatground I would have absolutely no chance at all of doing that..
Mind you, I’ve done 220km/h on the flat on a motorbike and that feels faster than 300km/h in a car (which I’ve also done).
Fascinating deep/wide dive on the subject at https://marginalgainspodcast.cc/bifurcation-and-marginal-gains/
Weirdly enough in this case it probably due to weight distribution/shift and the tourq from the motor.
My brother recently ran into the same issue with a cargo bike that he’s been working on with. Had re-balanced to wheels etc… but in the end it was due to how he was concentrating the weight on the bike near over the rear axel.
In this bikes case the woble is probably being caused because more weight/mass needs to be shifted towards the front. As the tourq plus air resistance of the dude on the bike is making the front wheel lift.
If the battery was mounted on the top bar, between the seat post and the handlebars. It’d have better weight distribution.
My fat tire ebike hasn’t wobbled yet, and I’ve done downhill bombs at 54mph! Maybe bigger is better?
It’s quite obvious from the picture showing the mount place of the alternator of why the bike wobbles at higher speed. It’s not centered thus more weight to one side of the bike. I bought a 1000watt ebike rear hub motor kit from ebay and purchased a 52volt 20ah triangle battery- placed that in the triangle of my frame to balance out the weight! Starting out full charge on a flat road, I can achieve 43mph by not hooking up the two blue wires on my controller! I can use throttle only keeping at around 250watts ( about 15mph) and travel on a flat road or trail over 40 miles! with out pedaling!
haha we have almost the same setup! Mine’s a Violamart 1500w 48v rear hub motor, UPP 52v 20aH triangle batt. Bike hauls @$$! 40mph cruise, 45.6mph max. Ebay’d a 17″ Raleigh aluminum frame, repurposed/rebuilt a 00’s Manitou susp fork, $15 bucks at a bike shop. Transferred a lot of parts from other mountain bikes I had laying around. I figure I have under $1000 in it, yet it blows the doors off everything factory short of a Sur-Ron Lite Bee (which I’d love to have also). Not a wobble, rock solid. Just a matter of good wrenching and keeping an eye on things. I’m an old pro shop bike wrench, so it was an easy build for me.
Some older motorcycle s have what you call a strating damper. It is a small shock absorber mounted from the bike frame to the strating front forkes. This will stop the bike from wabbling at higher speeds. I hope this information helps you . A friend John ansted.
So, are those Rent-a-scooters that have been littering our city sidewalks for the past several years, another good source of neodymium magnets?
B^)
My vote for wobble cause is rear wheel alignment. Had this happen on motorcycles if the rear wasn’t pointed perfectly forward.
That’s a lot of torque from a small diameter motor. I would have thought it would only be useful with a planetary gearbox to match the load
The starter motor alone can pull your car out of a ditch if you have a stick shift, esp older models with direct drive starter motors. Should only be attempted with an old beater, could strip flywheel teeth, maybe burn out motor if it gets stuck. Also not so nice on your battery.
“been there done that”
that’s the go to redneck way of moving onto a trailer when your car has broken its timing belt.
Also works for getting to the gas station when you run out two blocks short…
From the video, it sounds like he was surprised by how well it worked too.
Even motorcycle wobble at high speed such as what some people attempt, when you reach speeds on the Salt Flats when attempting to break speed records. Some of it is due to the lack of aero down force where the tires actually start to lift off the road surface more of it is simply due to the flexing of the tires as they are not designed for high speeds and they lack surface contact area of the tread. I cases where the wheel is small spokes the spokes actually stress and do not remain perfectly round wheels. There is also buffeting of the entire package of the bike and the rider from the air moving over and past the bike and rider. The faster you go the more you are subjected to these forces, all of them. I went 140 mph once on a Kawasaki super street bike fitted with a full fairing and didn’t notice any wobbling but I think most of the reason was on account of the fairing and design of the bike, it was engineered to go fast. Bicycles are not engineered to go fast, they do not come equipped with
Road bikes are most certainly designed for fast riding, check out the downhills on the TDF, 100 kph on windy French rural roads with bad pavement. Check out the ratio of the top gear! Check out the aerodynamics of the helmet and the bicycle and the clothing, even the shoes are tested in the wind tunnel. Bicycles are most certainly designed with speed in mind.
yep, a lot of ppl don’t realize the speeds a bike can attain. I once picked up and drafted a semi truck at the top of a mountain grade on my Cannondale R1000 road bike. Hit 62 mph (99.7 kph) before I chickened out and pulled off lol! The sound of those skinny 24c tires whining was terrifying…only a meter off the bumper!
Track bicycles can easily reach speeds of 110kph. Again, every part of the bike is specifically designed for these speeds, nothing in common with a standard bicycle.
Also possible that the rims loose their shape at that speed and begin to oscillate. A motorcycle has either solid rims or beefy spoke wheels because of the bike weight, so not as susceptible to that, although they too will shake if not balanced correctly. He should strap a high speed camera on the side and see if he can see deformity at higher speeds in either the value stem spot or just outright tire shape itself, could also be micro misalignment on the hub too I suppose.
I’ve got a standard Sun 36h front rim on a standard Shimano oversized front hub- my DIY e- rig cruises at 40mph, rock solid. Just a matter of maintaining a properly tensioned and trued wheel. Unpowered road and mountain bikes reach speeds of 40+ all the time, on downhills.
Why would you need to replace the rotor? If you just send a controlled current through the field winding, with a simple PWM controller, it should work without any other modifications, except the replacement of the rectifier diodes with a 3-phase controller. A car alternator is basically a SepEx motor and can be driven as such.
From what I’ve read it isn’t as efficient or as powerful. The rotor poles are shaped like triangles not bars, the stator poles are. Probably to make it quieter and not whine. Delcotron’s had non-evenly spaced fins on the pulley fan, for less pitched noise I assume. After all there is gobs more power available so it slips more and doesn’t cause as much cogging load on the fan belt.
With that 4:1 gear reduction, that alternator is only doing 2500 rpm at 70 km/h, nowhere near its peak power — it’s burning lots of amps to get the desired power, just heating up the motor windings. That I^2R term kills you.
Sure, this is HaD, but it’s painfully clear the thing was just hacked together. Not actually, you know, engineered. It would be more efficient AND more powerful to gear it to something like 8:1 or 10:1 to let it wind up to 8000 rpm or so. Motor heating losses would drop 75%.
It’s true alternators are usually geared up quite significantly on cars, 4:1 sounds very believable and that’s on engines that redline at 6-8000rpm.
However, I don’t know if that would translate to best efficiency when used as a motor – usually the gearing on an alternator is so that it can provide a good amount of power while the car’s idling in traffic with the AC going full blast and the engine cooling fans spinning.
In the 70s I routinely hit 60+ mph downhill on a Royce Union 10 spd that by today’s standards would be considered junk. I experienced no stability issues what so ever. I could have easily gone faster but had reached the limit of my legs to keep up with the pedal rpm.
lol…ppl haven’t lived until they’ve drafted a semi truck going down a mountain grade! I jumped one on my 18lb. Cannondale R1000 road bike- Chickened out at 62mph…the combination of silence and increasing whine of my tires became terrifying, tucked up behind the trailer bumper…I had to let off, he didn’t know I was there, so…super dangerous if he just tapped his brakes!
Performance appears similar to my 1500kW ebike. There’s either a massive efficiency difference between this hack and an engineered product or a problem with how they were measuring the power. Would be neat if they used the peak power for six minutes to see if the battery was dead.
My previous comment was deleted, apparently for pointing out this same thing. I was even careful to avoid bad words, so I really don’t know why the editor felt it necessary to censor me.
This design only uses a 4:1 reduction, so the motor only runs at around 2500 rpm at “full” speed. It would get more power, more efficiently, if it were allowed to run into an 8:1 or 10:1 reduction, dropping the motor current in half for the same output power, and the motor resistive losses to just a quarter. The OP even states the motor is very hot at the end of a short run.
I built a DIY e-moto/bike that cruises comfortably at 40 mph and has topped out at 45.6 mph. It’s steady as a rock at those speeds because it’s purpose built. It’s just a 26″ wheel Raleigh 17″ aluminum mountain frame with a repurposed 00’s Manitou Pro suspension fork up front. (at motorcycle speeds, at least front suspension is a must!) Inexpensive but perfectly adequate Cheng Shin tubed combo tires with double liners, thorn resist tube in back, slime in both. Standard 36h Sun rim with V-brake up front, Violamart 1500w 48v hub motor in back with Shimano disc, 12ga spokes. 20aH 52v battery gives me the range and power to have this rig be my daily commuter and also haul 80lb. trailer loads of groceries, propane, and water 8 miles back with 2k feet of altitude gain.
That alternator build is probably wobbling due to the offset mount location causing all kinds of off center rotational forces, not to mention the static weight offset alone!
Please be kind and respectful to help make the comments section excellent. (Comment Policy)
This site uses Akismet to reduce spam. Learn how your comment data is processed.
By using our website and services, you expressly agree to the placement of our performance, functionality and advertising cookies. Learn more

source

Post a Comment

%d bloggers like this: