E-bikes combine a bicycle with a big lithium battery, a speed controller, and a motor. What you get from that combination is simple, efficient transportation. [Tom Stanton] wanted to build an e-bike himself, but he did it without any of the fancy electronic components. But the real gem? The weird janky motor he built to run it.
The concept is simple. An e-bike is electric, in that it has an electric motor and a source of electric power. However, [Tom] intended to eliminate the electronic parts—the speed controller, any battery balancing hardware, and the like. Just think no transistors and microchips and you’ve got the right idea. Basically, [Tom] just built an e-bike with motor weak enough that it doesn’t need any fancy throttle control. He can just turn the motor hard on or off with a switch.
The bike is built around a reed switch motor. This uses magnets on a rotor, which interact with a reed switch to time pulses of electricity to coils which drive the motor. [Tom] wound the coils and built the motor from scratch using 3D printed components. The project quickly ran into problems as the reed switch began to suffer degradation from arcing, which [Tom] solved with some innovative tungsten contacts.
Controlling the bike is pretty simple—there’s just a switch connecting a capacitor bank to the motor to provide power on command. No electronics! However, [Tom] has also neatly set up the motor to charge a bank of supercapacitors when coasting downhill. In this regard, the bike can store power on a descent and then use it for a boost when required later on. Between the weird motor and the weedy capacitor bank, it doesn’t do much, but it does work.
If he’s looking for a more potent power source, perhaps the answer is already out on the street — in the form of a battery pack salvaged from the cells in discarded vapes.
Wait hold on… He FILED grooves in the tungsten rods? Did he…get these tungstens on…… Amazon?
Interesting choice of commutation too, maybe I should shut my mouth and make my own YouTube video with the ol’ carbon brush, but maybe that’s too arcane.
Sharp File on it’s edge will cut the Tungsten.. Just not for very long, till the points are gone from the File.. Grinder works well..
High temp Solders will work, however the Strip will be warped Badly..
Thinking using a Hall-effect Sensor hitting a IGBT to drive the Motor..
Once you add the Charging Diodes, the ‘No Electronics’ Banner has been tossed.. However Simple Electronics are ‘Allowed’ :)
Cap
Tungsten carbide dust is nasty stuff.
IIRC about 2% of the population has a bad reaction to the cobalt binder and will die from very small exposures, the rest of us can stand more but it’s still bad for you.
TLDR: You don’t want the dust in your lungs.
https://www.casmetcarbide.com/images/Casmet_MSDS-WC.pdf
Also: Grinder does not ‘work well’. Unless you’re using diamond impregnated wheels.
Yes, I’ve cut custom step drills carbides on a surface grinder w spindex.
He used tungsten electrod rods made for TIG welding, which are not the same as tungsten carbide and won’t contain cobalt. Like pretty much any metal, you wouldn’t want to breath much tungsten dust, but it’s not inherently toxic. And the hardness of Tungsten is high (Mohs scale 7-9) but a jeweler’s diamond file (pretty common in machine shops or in small-part workshops) would work just fine.
Tungsten != Tungsten Carbide
In practice, it mostly is.
Pure Tungsten is very brittle and the ‘carbide’ part is commonly dropped.
No, not if these are tungsten metal rods for welding. There’s various kinds, many of which have a couple percent non-tungsten, but they’re not secretly made half out of carbon as they would be if they were actually tungsten carbide.
https://www.jasic.co.uk/post/tig-tungsten-electrode-types
Isn’t plain old metallic tungsten only Rc 30 or so? Sure, it’s not highly machineable, but it’s not like it’s harder than hardened steel in that case.
He’s using purple which is about 95% tungsten with some cerium oxide and yttrium oxide. I use the exact same tungsten. It’s a master of none tungsten. You can easily file tungsten if you haven’t used it to weld yet. Before being heated up it’s pretty soft and you can very easily bend it by hand. I don’t have a rockwell measurement of soft tungsten, but it’s well below O1 or 100MnCrW4 hardened to 58HRC. During use it can heat up to 3000C at the tip if my memory serves me right, and after cooling down it’s hardened. You can’t file it anymore when hardened and you can’t bend it anymore. If you try to bend it, it will take a lot of force (with pliers) and it will eventually snap off. I use fiber discs on an angle grinder to reshape the point when I end up dipping it in my weld pool again.
Tom said in the video that the front light dynamos generate resistance to pedaling even when the light isn’t on.
I believe the new hub-dynamos work quite the same way as Tom’s motor, and without the belt resistance. So very little idle-resitance there. Also the “sailing light” kind of front/back light consumes very little power.
I haven’t tried one but I know people charging their phones on such modern dynamos..
i had an old front light dynamo that did not generate resistance when the light was off. the power switch was a physical lever that moved its little roller on and off of my tire. it sounded like it had a flywheel in it, so the light would stay on for a few seconds after you stopped. it obliterated my tire, but it was great while it lasted
You can get a bike dynamo which rolls on the metal rims instead of the rubber tire.
Either way, the old rim drive dynamos were never really built for any proper use. They’re more of a gimmick than anything – they came with the bike as a minimum to satisfy the point of having “a light” and that was about it. It was worse in the past when you didn’t have LEDs, so you actually needed to pedal a good 10-15 Watts out of it to get any light – mostly thanks to the dynamo being so inefficient – and you still wouldn’t see jack s**t with the dim yellow glow that went in and out.
You would rather use a flashlight and burn through expensive zinc carbon batteries than switch the dynamo on, assuming it was even functional. Usually it had water inside, which then froze and locked up the rotor.
I wonder if you could help reduce the arcing of the reed switch the same way as you would with old-school Kettering points ignition: a condenser (i.e. a big capacitor, one of the only contexts in which one still uses the archaic terminology).
https://ratwell.com/mirror/users.mrbean.net.au/~rover/ketterin.htm
Or would that only be practical with high voltage/low current like spark plugs eat
Brushed DC motor with separate field/stator windings can be controlled with a reasonably sized potentiometer. You don’t need to balance a battery that often, and when you do, you can just charge each cell separately to top them off.
(For extra points, hang a weight from the potentiometer shaft and you’ll have it automatically increase the assist when accelerating or going uphill. I have been doing the same with accelerometer for years and it works great.)
Maybe link the potentiometer to chain tension such that pedaling harder calls for more assist.
A pivot mounted, spring loaded idler sprocket deflecting the pull side of the chain under low tension would rotate around the pivot when pedaling harder (more chain tension). The rotation of the pivot rotates the potentiometer.
Tune spring tension, pivot arm length to give the right amount of feedback for a given pedal pressure.
I haven’t a clue how much extra drag this would introduce, or how much you could feel that spring loaded tensioner when pedaling. Would it deflect/bounce/boing and give unneeded boost at the natural points in the crank cycle where the rider has the most mechanical advantage? Could it be debounced with a small shock absorber? Could we make the bike so heavy that it couldn’t be ridden without the motor’s assistance?
I think the bigger issue is how the chain tensioner moves in response to changing gears.
It’s a great idea otherwise.
The first power meters for bikes used this and as it turns out the chain moves around a LOT during even casual riding. It hits the chainstay and flies way up when you hit any kinda debris. This is why so many carbon frames have a metal chainstay protector of some sort.
There was another Tom who used the equivalent of a super cap to store energy on the bicycle he used to ride to & from work,with a hill slope in between his home and laboratory. His last name was Edison, and he used a clockwork spring to store the energy from the down hill ride to make the up hill pedal easier.
This tom built a bike a few years ago with a flywheel and a clutch, seems a lot simpler and probably just as effective. https://youtu.be/gahKxbwUcYw?si=ifkm_EdVqhkrWpEA
Mercury wetted contacts (inside the reed switch). We had to use some reed relays and the contacts caused all kinds of problems. Switched to relays with mercury wetted contacts and problems solved. (Not sure bout being open market though, I was in AT&T labs, and these were AT&T (bell system) relays). I still have a bunch in my junk box.
Reed switch motor are cool, not wierd!
You can throttle one with no electronics, simply by mounting the reed switch on a lever which can move it away from the rotating magnets.
Each stationary electromagnet will be powered for a smaller fraction of a rotation the further away the switch is.
(A small movable permanent magnet near the reed switch might also work as a throttle.)
Use two double pole single throw reed switches for each electromagnet so you can get maximum mechanic power, by reversing the polarity of the coil as needed.
You can buy reed switches with mercury wetted brushes, hermetically sealed in glass, to avoid damage from arcing. Arcing will still happen, and create electrical noise… a capacitor or a transistor may help here.
RoHS says no more mercury. Of course they exist, but whether you’re allowed to buy one anymore is a different matter.