Building an Electric Bike Without Electronics

Big thanks to KiwiCo for sponsoring this video! Get 50% off your first monthly club crate with my code TOMSTANTON at www.kiwico.com...
Get Started in Onshape for Free at: Onshape.pro/To...
Enjoy my videos? These are made possible due to help from my Patrons. Please consider supporting my efforts: / tomstanton
My 3D Printers
Formlabs Form 4: formlabs.com/u...
Formlabs Form 4L: formlabs.com/u...
Formlabs Form 3: formlabs.com/u...
----------------------------------------------------------------------------------------------------------------------------------------
My Other Equipment:
Main camera - amzn.to/2vlvlC6
Main lens - amzn.to/2gMrhru
Main tripod - amzn.to/2tqRjBt
Secondary Tripod - amzn.to/2t1NkMh
Microphone - amzn.to/2uuv9n0
Audio recorder - amzn.to/2v3mjcG
Banggood affiliate: www.banggood.c...
Twitter: / tomstantonyt
3D Printer filament sponsored by 3D Printz UK: 3dprintz.co.uk/
----------------------------------------------------------------------------------------------------------------------------------------
#3dprinting #physics #engineering #magnet #electromagnet #science

[@mrharvest]

I'm glad to see that you haven't broken the laws of thermodynamics. But at the same time I'm slightly sad you haven't broken the laws of thermodynamics.

[@SentientTeapot2444]

Every engineer who saves energy in one part of the system only for it to pop up as extra effort in another part:

[@JustGoAndFly]

The trick is not uploading the perpetual motion or unlimited energy device so the shadow people don't come for you.

[@anwin85alon]

5~30V boost converter PCBs ok la

[@rekire___]

Young man in this household we obey *THE LAW OF THERMODYNAMICS*

[@christopherleveck6835]

The laws of Thermo whatever's can clearly be bent.

[@paulhaynes8045]

My dad worked in the factory that pioneered reed switches, back in the 60s. He was a toolmaker draughtsman and his input was the part of the machinery that held the reeds in exactly the right position while the glass cooled. A much more tricky operation than it sounds, as the quality control was extreme on those switches, if the gap or alignment was out by just a thou or two, they were rejected. As a result, I had dozens of 'failed' reed switches from his testing, which are actually perfectly good switches and still work today. I don't have many left now, but when my youngest daughter (2nd marriage) was doing a project in primary school, I realised that it was perfect for a reed switch. She was then about the same age as I had been when my dad was running those tests, so that reed switch she used was over 50 years old, and still worked perfectly. Not bad for a reject!

[@cadecarp11]

Why was it necessary to tell us that she was from your second marriage?

[@buttcracker]

@@cadecarp11probably not, but still a cool story nonetheless

[@fr33dumb0]

​@@cadecarp11divorcees are a proud bunch.

[@bewilderbeesty]

@@cadecarp11flavour

[@thehandlesticks66]

ukraine will never be in the eu

[@Thingsthatgopew22]

You could add a 17v rated zener diode and a power resistor. When voltage goes over 17v it dumps the power into the resistor, heating whatever, instead of blowing up your supercaps and potentially neuter you.

[@SailingSoWhat]

But a diode would be cheating, the point was to do it without electronics.

[@justinc2633]

@@SailingSoWhat you didnt finish the video?

[@Eduardo_Espinoza]

In parallel right?

[@justinc2633]

@@Eduardo_Espinoza no, in series

[@S54]

it would all dissipate on a diode not on resistor

[@Nighthawkinlight]

This so cool. This could have been half a dozen videos about each component, and I honestly wouldn't be unhappy if you still made those videos. The motor is awesome

[@henrymach]

For those who deal with old cars, your switch is basically called 'points'. You can buy one from car parts stores

[@cooperised]

Yes but no - points aren't magnetically actuated, and they also carry less current than this which is why they don't wear as fast (though obviously they still do wear). Not sure how well a set of points would work as a base for this reed switch.

[@dustintaylor9157]

@@cooperisedthe points I’m familiar with are opened by a cam on a the distributor shaft.

[@greatPretender79]

​@cooperised wouldn't the fixed contact of the point interfere with(if not cancel out) the magnetic field attracting the movable arm and contact since there's a ferrous metal between the two?

[@LuMaxQFPV]

Good point.

[@greatPretender79]

Also that spring is pretty strong, at least in all my cars.

[@Madkite]

So, I used to make tungsten contacts for obsolete ignition systems. You have to silver solder them. A small blowtorch will do it. As for the sparking. Well, that's exactly what ignition contacts do. If the condenser (capacitor) fails. You have to have a high rating as peak voltage can be very high, and the value can be trial and error. It has to be matched to the system. Not too big, not too small. But it eliminates almost all the sparking. Which is all inefficiency as its not just turning off but wasting energy on the plasma. You may then be able to massively reduce the mass of the moving switch. Silver solder on the end of the tungsten rod. The cut it off with a grinding disk. Grind and sand back to a nice thin disk. So low mass. The other contact can be a disk as well then so it has more area to conduct. The lower mass on the Reed switch will mean it can function at higher frequency.

[@The7Tycoon]

Hope Tom will see this, this sounds really insightful!

[@bikefarmtaiwan1800]

It is good info

[@robgerman7105]

Thank you. I was about to make one of those "kids these days" comments about resistive spark systems, but your comment is more informative and less "get off my lawn."

[@PJ818]

There's a German bicycle headlight company called Magnic. What's interesting about their bicycle headlights are that they don't use physical contact between the dynamo, but instead has a rotor with magnets on it that is put in close proximity to the rims of the wheels. The rim rotating past the magnet closest to it on the rotor has a magnetic eddy current induced in the metal of the rim, the eddy current creates a magnetic field, and the magnetic field of the eddy current drags and spins the rotor for the next magnet to do the same thing, etc., spinning the rotor with magnets past some could wire, generating enough power to light some efficient LEDs. Yes, there is some resistive friction between the rim and rotor with the interactions between the magnets and the eddy currents, but it has to be significantly less than those from the much more physical friction of an old-style "bottle generator". I've been interested to see if this could be upscaled enough to charge a cellphone, or power a Bluetooth speaker.

[@metamorphicorder]

A few years ago there was a company called rev lights that used the same principle.

[@metamorphicorder]

My mistake revo light

[@stefanmargraf7878]

Thanks! I was working for years on that. And i am German too.

[@charlcoetzee93]

Hub dynamos work great too

[@semanticks]

Thats such a cool idea

[@TCSC47]

Excellent project! As a very old dude, I am amazed at what we can do with 3-D printing nowadays. In the dark ages, I had to send my drawings down to the workshop and await a much less detailed prototype mechanism to arrive a week or more later. Then I would have to buy the technician a few bottles of beer for prioritising my needs! Thankfully nobody else knew this trick!

[@BodywiseMustard]

To further the explanation at the beginning: it's electric, not electronic

[@ethanpschwartz]

Coincidentally, it's also electrifying.

[@anwin85alon]

5~30V boost converter PCBs ok la

[@joachimkr]

It does the job, what was achieved. Now, when you see the limits of electromechanics, change the challenge: make it reliable and efficient: use electronics.

[@taliru8846]

its electric, boogey woogie woogie

[@CareliFernandez]

Hello how do you make such monthly??

[@Andrew-se9be]

Use an interposing rela/contactor. The reed switch powers the relay, the relay powers the motor. This will save your switch contacts from wear and offloads the high current duties to something designed to switch power.

[@ekalb666]

dang maybe you could combine that full bridge rectifier and some transistors to control the current to the motor all on one circuit board we could call it a motor controller

[@aleatza]

the relay is way too slow

[@takix2007]

@@aleatza A MOSFET would be a little bit cheating, but it is "soft electronics" (as in the definition of "soft magic" vs "hard magic") compared to ICs.

[@aleatza]

@@takix2007 yeah, a mosfet is cheating :'). But the relay will simply be always on or off on that frequency. I don't know, maybe create a circular pattern with copper like a series of notches so that 2 graphite brush can run on it and the notches can close the contact between them. In this way you can avoid the magnet weight on the contact, the eccessive noise and can achieve higher frequency (equals more speed). But it will wear of more rapidly and it wil spark. It will spark like crazy.

[@eekee6034]

It would be cheating by the literal definition of "no electronics, but a single mosfet is not even beginning to be a smidgen of a ghost of a possibility of being as complex as a motor controller. Anyway, a relay is an interesting suggestion. I suggested it too, in a round-about way, mentioning that relays have an inductive load rating and to look out for that. I wonder how well (or badly) it would work.

[@andrew-vl7gu]

at 12:30 you missed putting a Mehdi with FULL BRIDGE RECTIFIER 😄

[@aasherahmed4389]

i heard it echo in my head

[@MeepMu]

Second this!

[@ChaosPootato]

We all heard him in our heads, he doesn't even need a shoutout

[@p529.]

I was honestly anticipating it and was quite sad when nothing happened

[@Deja117]

He also said "free energy". I'm aware, he didn't mean it that way and instead meant it's a human powered induction generator... But I'm scared mr electric man might come after his capacitors.

[@Meower68]

When you can put the magnets on the rim of the wheel, and create a section of "fender" which has coils on either side of the rim, to do your acceleration / regen breaking, you'll be onto something. No belt necessary. The rim becomes your armature.

[@andypanda4756]

Add a basket attachment for the screws, nuts, metal road debris the wheel's magnetic feild whips around in a rotating blender effect in the air. That sounds awesome! Paper clip mini tornados flying around your wheels.

[@ferrumignis]

Putting weights on the rims turn the wheels into great flywheels, hard work to get go and hard work to stop.

[@hoodieguy9279]

I love how you explain everything you make and really make it easy to understand, its awesome

[@janhanchenmichelsen2627]

As boys back in the seventies we made wooden soap box cars with old school 12V DC car dynamos as motors (via a fan belt to a pulley fixed with wire on one of the rear (pram) wheels) and 6V lead batteries as power sources. One battery: Very slow. Turning a home made bar & bolt switch, some sparks went flying - and two batteries were connected in series. Speed level: Slow. Happy days!

[@francisgoldstien6153]

You can use a capacitor to take the heat out of the contacts

[@uptowndisco2]

I took the cheap way and just got a friend (mug) to push 🤣

[@janhanchenmichelsen2627]

@@francisgoldstien6153 I guess your comment was to the creative mr Stanton, not to my childhood memories. But maybe, like the cap protecting car contact breaker points? I still guess that running these far larger currents through points exposed to air will cause rapid pitting and wearing anyway.

[@ferrumignis]

@@francisgoldstien6153 A capacitor can reducing arcing, but it won't help with resistive heating.

[@huzudra]

Regarding the arcing, try adding a capacitor across the switch to maybe help control some of the surge of current on make and break. In older cars they would use this on the ignition to control arcing across the distributor points as they make and break for firing the ignition coil. It was called a condenser. It's still used on modern cars but to absorb electrical noise from the ignition coils.

[@MiesvanderLippe]

That’s missing the point somewhat.

[@Woowoobang]

Motors are basically inductors. Use a flyback diode or switchback diode I've heard it called both. Give a path for the current to reverse back around the inductor

[@tiger12506]

@@MiesvanderLippe He's already using diodes in a full-bridge rectifier. And using coils of wire which are inductors. And super-caps. Might as well use capacitors, which are just plates of metal.

[@huzudra]

@@MiesvanderLippe How so? en.wikipedia.org/wiki/Capacitor#History Is the 1750's too new for this project?

[@CCW1911]

As an old mechanic I was thinking the same, anytime the condenser failed the points would burn up quickly.

[@msf60khz]

I electrified an ordinary bike in about 1975 by using a car heater motor and a car battery under the cross bar. I tried a belt drive to a big pulley (actually just a channel bent into a big circle) on the rear wheel, and also tried a friction drive. Both these worked to some extent, just riding up and down my road. Thank you for the video.

[@HL65536]

You could even build a mechanical throttle: make the reed switch position variable: If it is closer to the magnets, it's on more often, if it's further out, it's on less often, varying the effective power. But yeah, electronics are used for a reason. Even just replacing the reed switch with a simple sensor-transistor combo would make this thing a lot better (also needs a freewheeling diode so the transistor does not blow up).

[@notconnected3815]

You where thinking about opto-electronics? Or hall-Sensor?

[@bosstowndynamics5488]

That wouldn't really result in throttling, it would just cause the switch to either be always on (killing power and shorting the supercaps since it would be pulling backwards as much as pushing forwards on the rotor), always off, working correctly or sporadically and randomly in between, without any kind of consistent increase or decrease in duty cycle

[@nickwinn7812]

If you use a brushed DC motor, you can control the speed using switching which connects groups of your battery cells in different ways to give different speeds. For example if you have a 48V system, in 8 groups of 6V, For low speed your switch connect all 6V groups in parallel, medium speed you connect into 2 sets in series at 12V and then those sets in parallel, and for full speed connect all 8 6V groups in series. This, or a version of this was how EVs were controlled before electronics were available.

[@someonerandom704]

does this feel any different to the driver compared to electronics using PWM to control power?

[@TimInertiatic]

You should put the tensioner on the return part of the chain/belt, or its directly under tension which risks snapping it off and increases drag. Mtb idlers are very beefy and quite large for this reason

[@mike_oe]

Correct. Also make it push on the belt, thereby increasing the contact area between belt and gear wheel.

[@leftaroundabout]

Pretty sure he did this intentionally. The belt's primary role is regenerative braking, not acceleration - the opposite of a normal bike drivetrain.

[@Mueller3D]

At the end, it looks like he found a belt with the correct length (or pulley of the right diameter) to not need a tensioner.

[@Mike-oz4cv]

The motor is used for accelerating *and* decelerating. There is no unloaded side.

[@ElectroBOOM]

Hey Tom! Awesome build as always! I have a question, wouldn't using a core for your windings significantly amplify the magnetic field and increase the torque, or generated energy? I'm not super familiar with the type of topology you are using, maybe I'm missing something important. Have fun! [edit] Oh I see you explained why you used this type at 9:05, to remove the force of regular motors against turning when motor is idling. Although I'm not convinced that force exists at higher speeds. I think it is there only when you want to start rotating them. Maybe I should experiment on this a bit...

[@TomStantonEngineering]

Hey Mehdi! Yes, the lack of iron core was to reduce the resistance. From my experience of using BLDC motors on drones and ebikes, there seems to be a resistive force when spinning. The force is definitely higher when starting, but it's also still there at higher speeds. I've had my ebike speed controller fail before, and it was exhausting riding home with the motor drag 😅 (all 3 motor wires unplugged too). Thanks!

[@iaadsi]

That microswitch wasn't rated for 15 A :-) that number was just for AC. For DC (your case), it's just 0.6 A. AC crosses zero so is much easier to disconnect, the arc is self-extinguishing. A DC arc keeps on burning and pitting the contact surface.

[@pusnirizda5481]

How did you calculated 0.6A from it?

[@Nerd3927]

@@pusnirizda5481 It is printed on the switch.

[@web1bastler]

@@pusnirizda5481 Datasheet. Switches have different load ratings depending on their usecases. In this case it would be DC-23A or DC-4 as all these coils are a highly inductive load and it's switched very often. Per the datasheet these Omron switches can only handle 0.6A at any DC voltage up to 125V. It's even written on the switch case. DC-Switching with contactors/relays is the bane of every battery/HVDC Engineer. Often enough the answer is to look into railway equipment manufacturer catalogues at which point projects can get very pricey.

[@pitaeata8493]

@@Nerd3927 2:47

[@randomname4726]

​@pusnirizda5481 lol it was right there!

[@RossGirven]

13:32 it’s better than free energy, it’s negative cost energy as it is reducing your need to replace break pads 😂

[@F8Tributo]

Brake

[@ninefingerdeathgrip]

Brake

[@RossGirven]

@ thanks for the spell cheque

[@bastienx8]

It's more or less the same thing on a hybrid car, some energy is stored in a battery when the car slows down, reducing the load on the brakes, and this energy is used when the car accelerates again.

[@Katvanished]

they're break pads because they're subject to entropy lol /s

[@diamonddogie]

Bike light generator( dynamo) can be mounted on a moving mount operated by cable and lever(like a brake lever). Put It together in parallel with brakes so that by slightly pressing the lever It engages the dynamo, and full press ads the brakes.

[@dimitarnikolov3527]

To reduce arcing you can add a diode in reverse direction across the coils. If you consider that "electronics" because it's a semiconductor, you can also try a resistor, but that would generate losses. Maybe a capacitor could work too but it's tricky, or a combination of resistor and capacitor in series. This is called a snubber network and can be placed either across the inductor or across the switch. That's why some comments suggest a capacitor across the switch, that will decrease the peak voltage when the switch opens but it will also increase the peak current when the switch closes because the capacitor would discharge directly through it.

[@miguelJsesma]

Old car points with a capacitor would work way better. But brushed motors are already invented too.

[@Deja117]

For the reed switch, you could make a thin acrylic box and fill it with an inert gas like argon. It should stop it sparking. This will also dampen the noise and should protect it from dirt and water as added bonuses. Really cool project, I hope you revisit this as it's definitely a viable cost saving alternative to what your average ebike costs these days, a little less powerful maybe sure, but it's worth the extra £500+ you could save, at least in my book. 😅

[@matthewday7565]

Sulphur Hexaflouride would be better

[@moisejf8331]

Hello Tom, About the spark after breaking the switch ... Did you know about the free will diode? Every time a self is unpowerded it realease the actual magnetic energy through a spike . You can short this spike with a single diode mount in reverse, or even better, keep and redirect this energy for your capacitors. Just saying. May be it can increase efficientcy a bit Sorry for my bad English, sir.

[@notconnected3815]

Yeah, preventing arcing on switching inductive loads is a huge topic in electrical engineering 😉 Companies also put a lot of development in the material of the electrical contactor. Also: If you want the switch to operate fast, then make it light wight. The little magnet on the switch contact adds too much inertia. Try hot soldering the tungsten with silver solder

[@bbrockert]

In english often called a "flyback diode". It is the right answer, but I think a diode qualifies as "electronics" and so isn't right for this project.

[@Cartilog-z4f]

A diode is electronics, and he is specifillay not using that for the motor. That's the whole point of it all.

[@olavl8827]

@@bbrockert It's also sometimes called a freewheel diode, from which I can see "free will diode" as an understandable misspelling.

[@bbrockert]

@@olavl8827 Good point!

[@azlandpilotcar4450]

Nice project! Old gasoline engines used ignition points that were resistant to damage from spark, Also spark corrosion resistant were ignition rotors and distributor caps. Maybe one of these would be a source of more robust contactors? Truly mechanical switching might be done with iginition points actuated by a rotor-mounted cam... Before there were transistors, there were amplifier vaccuum tubes. One of those might also substitute for your reeed switch. Running your reed swith in an airless environment would also cut down on corrosion and wear from spark. Looking forward to your next video!

[@kreynolds1123]

A hallbock array help close the air gap through a portion of the magnetic circuit and contributes to the overall magnetic field in the air gap. A steel plate also helps close the air gap through a portion of the magnetic circuit, and yoking magnets with iron or steel also increases the strength of the field in the air gap. Maybe a steel plate might not be as strong as a hallbock array but you'll get very similar results for much cheaper.

[@Mostafa-vs8bd]

It will be heavier

[@diamonddogie]

Steel plate may introduce some eddy current which increases drag during freewheeling

[@kreynolds1123]

@diamonddogie steel plate to yoke magnetic field, remove the air gap on one side between magnets with an unchanging magnetic field which means little to no eddy currents.

[@kreynolds1123]

@diamonddogie if we were changing the magnetic field inside steel, as would be the case with steel inside the coil, then yes the motor would experience Eddy currents and hysteresis losses due to the retentivity and coercivity of the steel

[@kreynolds1123]

@@Mostafa-vs8bd maybe, but not much heavyer than trying to close the airgap with magnets to make a hallbock array.

[@Madpegasusmax]

you can simply adapt automotive ruptor contacts and a condensator (to reduce emf noise from the arc) from a spark plug distributor . a middle 0 needle amp meter is a cool addition to show how much charge /discharge is happening

[@cliveso]

I'm not convinced that this is in any way better than a DC carbon brush motor.

[@teamsafa]

No it is not better... but i guess it would not be as fun doing the video then.

[@specialagentdustyponcho1065]

A brushed motor has a lot more internal friction, and is much less energy efficient because of it. If Stanton could get this mechanism working more smoothly, it would basically be a mechanical BLDC motor controller, somewhere between the efficiency of an electronically driven BLDC and a brushed DC motor.

[@teamsafa]

@@specialagentdustyponcho1065 The problem is that people have been working on the mechanism for over 100 years and the best mechanical solution they have come up with is the brushed. So to do better you have to go to electronics. On the other hand, if you put the same amount of permanent magnets in a brushed motor you could get pretty good efficiency too.

[@cliveso]

@@specialagentdustyponcho1065 What good is efficiency if you can't harvest enough power in the first place? I'm also very sceptical that, for the same power output, an off-the-shelf brushed DC motor would be less efficient that Stanton's generator.

[@ferrumignis]

@@specialagentdustyponcho1065 A commutator and carbon brushes add relatively little friction as graphite is an excellent dry lubricant.

[@kensmith5694]

serpentine coils are great for this sort of project. You can wind them with heavier wire and they take far less work to make. You end up with effectively all the coils in series rather than parallel. Basically: you use a bit of string or something to find the total length, wind on a jig that lets you just loop the wire and then form it to fit and put the end cap on to hold it all in place.

[@sparkplug0000]

I see you used your time machine to pop back to the industrial revolution to grab a voltmeter. Nice!

[@gaffalover]

Electrical engineer here: You should add a diode antiparallel to your switch to suppress the discharge spark induced by the demagnetizing coils. The spark is generated by the collapsing magnet field of the coils when the switch disconnects. The collapsing field is inducing a voltage, which discharges via arcing across the switch. The Diode limits this discharge voltage and dissipates the energy of the magnetic field.

[@TheStuartstardust]

7:39 - Tungsten should be handled similar to carbon fiber when filing or standing. Be aware of the dust - special devices are normally used to sharpen tungsten under warter to avoid hazzard dust when using tungsten as tig welding tips.

[@timmontano8792]

A couple of years ago I started building something not exactly like your setup but, it was close enough in terms of generating DC power to charge my E-Bike's spare battery while riding with my primary battery. Once my primary battery was depleted to a certain level, I did still have to stop and manually swap batteries, but the intent of my project was to allow me to have unlimited range of sorts, using two batteries: one being charged while the other being drained via my riding my bike. Mine was mounted in the small bicycle trailer that I stow my fishing gear, small ice chest and my spare battery. I rigged up a miniaturized CVT type of transmission so that my actual charging system would still spin fast enough to produce power even when riding at a very slow pace.

[@amarug]

Wow "Stuff Made Here" and now you upload - what a treat!

[@nobodysbusiness87]

I love electromechanical engineering. I would have probably called this video "analog E-Bike" but your the successful youtuber so I'm sure you had good reasons. You always have awesome projects. Keep up the good work.

[@Giuliana-w1f]

Instead of reinforcing the reed switch, you could use a relay. That, combined with the bridge rectifier could give a better result. Another option would be skipping the reed switch entirely and using a commutator.

[@8001010]

commutator is right choice

[@fronthole_guacamole]

It would be difficult to make the relay work. It has an operating time of perhaps 25-50ms which you couldn't compensate for without building something analogous to a centrifugal timing advance system to move the switch as the rotational speed changes.

[@charlieray1891]

The relay solution is a great option, if for some reason we are stuck on the idea of reed switch, the problem caused seems to be the same problem that spark plugs have, so what about using the materials used for spark plugs? Iridium, Nickel, Platinum, etc. there only needs to be a small mount over the contact area. I do believe the relay option is better though

[@Giuliana-w1f]

@@fronthole_guacamole i think fixed advanced timing could work (a smaller reed switch could also improve response time, even if it gets worse in total), but a commutator would still be better

[@KaufDirGeld]

relay is too slow, you need a mosfet.

[@specialagentdustyponcho1065]

I think you should look more into this concept. A mechanical BLDC controller could have some practical use. How about instead of reed switches, you use a set of cams? This would allow much stiffer and heavier-gauge conductors, with a longer throw and faster deactivation to prevent arcing. You could have multiple throttle settings by selectively engaging cams to change the duty cycle.

[@xcoder1122]

Or you can just use a motor with brushes, then you can plug a battery directly into it and it will just work. It also basically uses "switches" inside (the strip that the brushes contact acts like a switch) and yes, the brushes will degrade over time, but good motors allow you to replace them and are built so that the brushes degrade faster than the rest of the motor, so you can use a motor like this for many years and only have to replace the brushes from time to time. It's a lot less of a deal than most people think. After all, if you have a car with a gas engine, you have to change the spark plugs from time to time.

[@igorok1362]

I would like to point out that a car generator also has brushes that need to be changed sometimes.

[@xcoder1122]

@@igorok1362 Most people never learn about that since usually the brushes outlive the engine. Which is another proof that brushes are not as fragile and not such a huge point of failure than most people think. Until recent years, it was normal for motors in washing machines to also have brushes. So the fear of using a motor with brushes just because brushes are a wearing part are quite exaggerated.

[@5467nick]

@@xcoder1122 Brushed motors are generally less energy efficient and less suited to electronic control. Modern washing machines and their variable speed nonsense work much more efficiently with brushless three-phase motors than they could with brushed single-phase motors. I agree that brushes tend to outlive most consumer applications. I don't recall hearing talk of brushless motors saving us from heavy wear of brushes. Any time I see appliances or tools with brushless motors being advertised, it's for higher energy efficiency, variable speed, or higher power, or lower weight for the same power. I agree that OP's concern of brush wear is probably not an issue for such an application.

[@alexkram]

My first ebike was like this. It was a Zap brand bike with two 12V car radiator fan motors, one on each side of a friction roller over the rear tire. A switch on the handlebar toggled the motors from series (slow) to parallel (fast) configuration. It used a 7Ah 12V lead acid battery. It had 8 mile range and worked great for my commute to college.

[@markae0]

The motor is pulling on the belt and the slack removing pulley on an angle. That vector is absorbing a lot of energy instead of pulling on the wheel IMO. Great project! That is a lot of work to do. I hope you enjoyed it.

[@cooperised]

Not sure about that. The force being applied to the tensioning pulley is doing no work. You'll lose a tiny bit of energy in the pulley bearings and a tiny bit more through repeated flexing of the belt but it won't be much.

[@cwt5654]

Even with the standard chain drive there is a rearward vector through the bottom bracket/chain stays when under load. Raising the drive line angle offsets part of this into the vertical direction compensating for the loss at the rear sprocket.

[@ferrumignis]

The only losses a tensioner adds are the friction of the bearing and the energy lost in flexing the belt itself, both quite small. If the "vector" caused energy loss then a serpentine belt on a car engine would never work. That said a tensioner is better placed on the unloaded side of the belt to reduce belt and tensioner loads, but the belt can be loaded in either direction in this case. Seems like Tom realised this as he used a shorter belt and removed the tensioner at the end.

[@durfkludge]

When I saw your nickel contact arcing problem I immediately thought "he should yank some silver contacts out of a switch" and then that's exactly what you did, haha. Excellent work!

[@ShydenPierce]

Lets go, needed something to watch, then you uploaded. Your videos are always a pleasure.

[@a_n_1-yt]

man, always loved the way you visualize the thinking\problem solving process. It absolutely helps me remember the right approach, thanks for that.

[@FixAndModGarage]

That's a cool-looking bike! I'd like to see more of this in the future. Also, I noticed at 7:52, you used solder on the tungsten rod for the makeshift reed. If you plan on using this long-term, solder degrades from vibration and might shake loose resulting in sparking and degradation over time.

[@mrinalbahuguna7415]

Hi tom! Great effort, FYI i had made a bike which works on same principle a year back, i also works with two switch. By the way i have tuned it in a way that it gives back the conserved energy in form of solid boost to the rider, if you want we can discuss the project.

[@realvanman1]

This is a perfect application for something called a commutator. Well designed and well implemented, they work really well. MUCH better than many people who have never studied them think.

[@keithsyers5833]

Oil bath the reed switch and increase your capacitors capacity for voltage and amperage. That's the trouble with permanent magnets the faster you go the more voltage and current and it causes drag. Have an exciter circuit to energize the fields for charging and powering the bike along. Food for thought 🤔

[@OpreanMircea]

9:45 so, it's KERS for bikes

[@gordonbush2054]

Exactly my thoughts

[@EXPLODINGETDOOD]

You could use reed switches to control MOSFETS that handle the higher power current. This way you could use beefier capacitors that draw and supply a lot more current to your motor/generator. Using more current you might be able to get better performance

[@tsbrownie]

To reduce arcing, add a neon bulb across the contacts.

[@robertMichaels-px3gt]

If I may suggest...Add a capacitor,(perhaps a .02uf), across the contacts of the switch to absorb the arcing across them, much like a set of Points and Condensor in a car's ignition system...This will greatly increase the longevity of the contact switch.

[@esra_erimez]

4:15 not to be confused with a Hollaback Girl

[@snakezdewiggle6084]

Carbon Arc Lamps are still in use in some Cinemas. Best used with HVDC. Double up the Diode Full Wave Bridge. Larger diameter motor, or smaller diameter rear gear.

[@Ghosty2k77]

1:45 Hector salamanca in that one episode

[@ultimate.t7835]

"Ting" Ting" TING TING TING TING TING!!!

[@ferrumignis]

@@ultimate.t7835 BOOM!

[@curry6138]

Hi, Tom! . I think if you create a form of small slip-on commutator to turn the AC into DC (with some diodes for safety) to send to a DC motor. That way you can both recharge your power-bank, and discharge it with the high-torque of the DC motor. This will cause more battery drain, and I suggest the slip-on commutator to be a form of spring based system- in order to prevent any parasitic consumption from commutator to the motor.

[@peppi0304]

11:30 one thing that always amazes me is how much noise cars make. just listen to to it. Constant noise pollution...

[@Dougerro]

2 tons and 100 km/h

[@southtexasspecials175]

Omg this one was amazing. That tungsten rod contact invention was brilliant. N capacitor set up. the motor omg 🤯🤯🤯. Props bro. Genuis

[@pcfreak1992]

Wouldn’t a brushed DC motor also count towards your definition of having no electronics? I mean it’s a cool project to build a brushless motor without a controller, don’t get me wrong, but I feel like a brushed motor with a clutch, to avoid the resistance when unpowered and enable regenerative braking, would have been easier and more powerful 😅

[@SkyhawkSteve]

Agree... a brushed motor seems like the obvious choice. I'd like to hear Tom's thoughts... did he reject brushed DC for a reason, or was this just an extension of the BLDC project?

[@notconnected3815]

I guess a brushed motor has too much drag due to the brushes constantly pushing against the rotor.🤔

[@qo92]

See 9:07

[@SkyhawkSteve]

@@qo92 what losses is he talking about? If the motor is disconnected, there is very little drag... just some eddy current losses in the core material. The attractive force between the magnets and the core does exist, but there aren't any losses associated with that.

[@qo92]

@@SkyhawkSteve I had assumed he was talking about those eddy current losses which are minimised in his design. Perhaps a brushed motor constructed with more plastics/nonferromagnetic materials would solve the drag issues his design seeks to overcome?

[@xremming]

Something like this integrated in a production bike could be nice for city rides. Energy for small lights, a USB port for charging a phone, and small extra boost for uphill sections is what I really want from an e-bike.

[@geek9642]

Got to say taking the contacts out of an existing switch was a genius move thinking outside (or maybe inside the box for this one 😂)

[@gameboi30]

Adding iron core will only slightly increase the resistance- if done correctly. But it can greatly improve the Torque. Changing the winding can also help.

[@aaryanmakwana515]

7:03 one solution might be to use an optical isolator and operate the contacts at much lower voltage to reduce the arching problem, the switching signal can be amplified by either an mosfet / optocoupler or even an IGBT. this will reduce the stress on the reed switch and even a off the shelf reed switch can be used.

[@_.madpie._]

Well, optocouplers are semiconductors

[@valtarijunkkala]

mosfet stands for metal-oxide-semiconductor field-effect transistor mosfets are semiconductors too, and even specifically transistors.

[@Sulfuron41]

I absolutely love electromechanical, analog devices. They all have a simple beauty that I really appreciate. I wish I had the knowledge to experiment and build like you do. It's always so fascinating to watch. I really wanted to get into engineering at some level, but I was utterly miserable at math, unfortunately. I could totally pass the math courses, but the amount of time and effort I needed to pour into just that area became unsustainable. I eventually had to face the music and changed majors.

[@MattC626]

Yes! Time to edge to another Tom Stanton video.

[@mikaelfransson3658]

❤WAOOOW!!! What a nice concept! First thought why not two motors or three! One on each wheel to! And now this is much stronger, the need of battery can be as we have in electric drills! Now the concept can get stronger lighter and a delivery unit can be assistent for a greater purpuse! ❤ 🎉 /Mikael

[@kistuszek]

Maybe if you want a regen brake you could mount a regular dynamo like a break pad. That is a cheaper simpler and easier method. You could even operate it with the brake lever, if you fiddle with it so that it engages before the regular brake. :)

[@tanalson]

I see that such set up will be an excellent way to regen your battery for your existing ebike. Just connect to a much bigger capacitor to capture more energy and you can instantly extend the range of your battery

[@MadZax33]

Founder of WaveDrive here! Happy to see something like this, this is close to my invention, but I still use electronics, with a light based analog computer that replaces the computer chip/software in all other BLDC motor drivers. I call it the CBMD (Codeless Brushless Motor Driver) Patent Pending Unlike your system, there are no sparks, and the switching is still done with mosfet's, with full control over waveform timing and duty cycle. This allow the almost exact same performance as highly tuned digital BLDC drivers, without a complex computer chip. I'm also sure you could also make the CBMD pretty easily if you had the designs! but I'm not giving those away yet :)

[@bradleyhodson9388]

Hello, I'm Bradley. I've been working on power generation for 25 years and have some ideas, but I don't know who to contact for a patient. Do you have any suggestions, please?

[@gigacooper]

So are you open sourcing your patent or are you just here to shill it?

[@MadZax33]

@@gigacooper I'm not going to open source a patent worth billions 🤣

[@MadZax33]

@@bradleyhodson9388 just look for a patent lawyer, and see if you can get free consultation from some, and find someone who understands what you made!

[@dann8749]

I really like that all the working parts are visible! I was also thinking about reducing drag in an e-bike motor by making the rotor magnet free and using either just metal sheets in a pattern to be attracted by the stator (reluctance motor) or an active rotor with coils powered wirelessly like... wireless charging 🤔

[@cedricsielens5268]

Only 52 seconds in and these bots already taking over the comments

[@danielb.5373]

Thats what a bot would say.

[@dmac716]

Build that into the rear tire, and you will have a product you could sell. You are limited by the gearbox on the bike. If you installed another set of gears on your alternator, you might be able to get more rpm. Your rectifier is reducing your output energy. The belt drive is suboptimal for powering torque. You might be able to add a hub motor to the front Tire so that you could use the stored energy there.

[@MindCrime550]

1:38 a relay wouldn't violate the "electronics" rule

[@labbertubes]

Much as I love this the spirit behind this experiment, and the fact that you actually implemented it, one has to concede that it is a powerful argument for the use of electronics. (It's also a damn good ad for your 3d printer)

[@trekintosh]

Why did you use an AI picture of a circuit board at 0:19? Surely that’s something you could have gotten a stock photo of for just as little effort as typing a prompt.

[@IamtheWV17]

Why?

[@maxthehuman004]

Or take a photo, really

[@amaljoe367]

Why not?

[@c.jishnu378]

Idk copyright or something. But some images don't have copyright.

[@jaydeep-p]

Why not?

[@guilhermebarros7145]

You could replace the material on the oversized Reed switch with parts of a contact braker from a car distributor. They are purposely made to withstand constant arcing and not instantly deteriorate.

[@emaglott]

I've lamented for years the loss of energy slowing my bike and wished I could use that to get going after a stop. Excellent work on this!

[@tibo786]

Would be cool if you could also make a mechanical system that automatically boosts or regens depending on your acceleration/inclanation. I'm thinking maybe a sort of pendulum that actuates the boost/regen switch. The pendulum would swing backwards or forwards from its inertia depending on if you're accelerating/going uphil or slowing down/going downhil.

[@juliannojungle]

Just use the reed switch to drive an optocoupler, to then electronically toggle the current (maybe with mosfets or something). This way you overcome the voltage limit from reed switches and totally get rid of the sparkling issue.

[@MattOGormanSmith]

You could have the reed switch energising a higher power relay, but you'd still get arcing in the relay unless you use the capacitor others have suggested. What I'm really commenting for is you've inspired an idea of an air-cored generator which pushes iron cores into the coils at higher speed. This could make a good governor for a home made wind turbine but would be mechanically complicated if the coils are the stator.

[@MrTheNark]

I think the sparking can be significantly reduced with a flyback-diode across the reed switch. The voltage across a coil is proportional to the derivative of the current ( v = L dI/dt ). So if you suddenly break the circuit by opening a switch, no current can flow and you induce a huge voltage (hundreds of volt) across the opened switch. A flyback diode across the switch (which cathode to the + of the supercapacitor and anode to the - of the supercap) will create a path for the current to continue flowing through the coils, preventing the sparking. It is a well-known trick.

[@beyondfossil]

Interesting project! What we were reminded of with this is how incredibly useful transistors are, specifically MOSFET and IGBT. They can reliably switch in the kHz to MHz range without any sparking using solid-state electronics. They're everywhere now!

[@jlander82]

Great job, Tom! Small suggestion for whatever it's worth: a tensioner usually does better on the slack-side (bottom side in this configuration) of the belt. The top side is going to almost always be in tension, so it doesn't need a tensioner there. On the bottom side, you can even spring-load it to be an auto-tensioner, but on the top side, you would be fighting the spring every time you apply torque to the pedals & tension to the belt.

[@emceegreen8864]

The 1990s ZAP power system was all electric without electronics. Used brushed DC motors with a series /parallel switch for two speeds. Really worked well and they made around 30,000 of them.

[@justsomedudeontheinternet]

if you set up a clutch, you could have the magnets seperate from the coils, then whey you pull the bar, it gets them close enough to charge or brake. This would allow you to have no resistance when you dont want it, and have a much stronger force when you need to brake or speed up. I was thinking something liek oldschool brake mechanism jury rigged to separate and close the "clutch"

[@kitemanmusic]

Tom, you are getting fantastic use out of your 3-D printer! You have some amazing ideas and get a good chance to try things out. I have a bike idea that I need to discuss with someone, which as far as I know has not been created by anyone yet.

[@lilyrooney]

when you pulled out that supercapacitor pack i got really envious lol. this is probably the funnest shot at that KERS bike idea ive seen so far

[@oiooiioioiooioii5400]

This is very cool and I like the ingenuity with the Tungsten and soldering method, but diodes are electronics by my definition.

[@stevergr9039]

It would be amazing to see this used for bike lights! Making use of the energy that otherwise would go into the brake rotors.

[@StrategyYouDidntKnow]

Really amazing video and I always wanted to understand the reed switch. Maybe a bicarbonate solution or even combine with surge arrest.

[@eformance]

Some thoughts: car alternators have an energized core and thus there is only a very weak magnetic field causing drag (this refers to a self-exciting alternator). Obviously the reed method is limited because the opening of the contacts causes an arc. You can use snubbers to try and reduce the arcing, either a flyback diode, an MOV, or a Quencharc "snubber". Many washing machines use 3 phase induction motors, you might consider a similar concept using "upcycled" parts to show how free/garbage can be repurposed like this. I would guess those magnets in your motor cost quite a bit.

[@dancoulson6579]

You could consider adding a capacitor across the switch contacts. When the contacts open, the current will continue to flow for a fraction of a second through the capacitor until it's charged up. This should be enough to stop any arc from forming by allowing the inductive load to dissipate into the capacitor. Considering the large and inductive load you have, I would suggest something like a 220uF electrolytic and another 0.1uF ceramic for good measure. When the contacts close, the capacitor is discharged. However, most of the damage is done on the opening, not closing of the contacts.

[@oasntet]

I suspect if we watched the contacts close in slow motion, it's bouncing a dozen times every time the magnet comes by... Mechanically de-bouncing that would go a long way towards reducing the rate of wear.

[@Rouverius]

Very cool. As for the arcing, I wonder if you had the reed switch powering the coils of relay that in-turn powered the coils. There would be some power loss because of adding the relay but it might make it more reliable.

[@Caluma122]

Honestly, the level of 3D printing engineering going on here is next level 🤯 I can't get over how strong the stuff is and how well it held up. Very good video.

[@bbesser1138]

Regarding the electric arc, which appears between the switch contacts, the arc can be prevented by using a snubber network (capacitor + resistor). Next time, I recommend that you consult with electrical engineers. I really enjoyed your video! Thank you very much!

[@dadsfriendlyrobotcompany]

Im glad you were able to utilize a halbach array for the rotors! In my own tinkering research, i have found that changing the magnetic flux topology will alter the cogging torque and efficiency of the motor within motors with iron stator slots. You can actually achieve low cogging while maintaining the high torque of a motor with iron stator slots through playing with the magnetic flux topology. Play with the width of the poles, and the halbach array magnet orientation. Of course my research was done with modified alternators (radial flux) with permanent magnet cores, and not the axial flux motor/generator in your video. My page has vids and links to my other pages where I house my files for others to use. I can also do custom design for anyone needing it in their own endeavors. Cheers!

[@moonfther]

nice work! How about using an old set a points from a car distributor. make a knuckle on the wheel where the points open and close according to the magnet crossing its point of coil pulse to open and close the points. Run a small capacitor across the points to hold down sparking. It's the way the old cars use to fire the spark plug. Always cool to see your inventive ideas.

[@AK-vx4dy]

Bicycle with KERS ;). If you finally used diodes for bridge, you could use one(s) for damping EMF to limit sparking on contacts (or capacitor ?). But i have mixed fellings if diodes are not electronics ;) Maybe not easily availabe but i'm sure i've seen heavy duty reed switches or maybe you can use it to open and close relay.