Gearless Magnet Bike

if you drop a magnet down a copper pipe it seems to experience a resistance force acting against it despite the copper being non-magnetic this is caused by the magnets generating electrical current in the copper called Eddy currents and these Eddy currents create a magnetic field within the copper as oppose the permanent magnet almost like an invisible no contact frictional force acting between the magnet and the copper now my plan is to use a metal disc and a disc full of magnets mounted very close to one another to potentially create some kind of magnetic clutch to replace the gear system on a bicycle I started by cutting two discs on my CNC router one for aluminum and one from copper the reason for this is I'm not sure which is the best material for this application my intuition would say copper as it's far more conductive than aluminum but if you ask Google it claims aluminum is the best and has a research paper to back it up however the majority of any current demonstrations use a copper pipe but maybe this is because copper pipes are easier to Source from plumbing stores in a recent video uploaded by veritasium about the world's strongest magnet he drops a sheet of copper and aluminum at the same time within a strong magnetic field and the aluminum sheet seems to fall far slower indicating a greater resistive Force so I've 3D printed this axle as part of a test rig to allow the discs to freely spin like a then mount a set of magnets either side of the disc with without making any contact then I can hang a weight via a length of string to apply a constant torque on the disc which should provide a reliable way to measure the resistance produced by the magnets for reference this is The Descent rate with and without the magnets the reason the non-magnet test is still slower than freefall is because it has to overcome the inertia of the disc but how does the aluminum compare to the Copper the copper clearly has the best braking Force as a disc spins 38 slower than the aluminum disc which makes sense as copper is about 54 more electrically conductive over aluminum but the reason why the copper sheet fell faster in the veritasium video is probably due to Copper weighing over three times more than aluminum so although the resistive force in the magnetic field might be a bit stronger the weight of the sheet skews a results in fact even if I run my test without the magnets at all the copper will still win due to the extra inertia of the heavier disc because of this I only measure the RPM of the disc at the end of The Descent when the disc has finished accelerating and is spinning at a constant speed and it's clear that with the grades and materials I have available copper is definitely the winner at least in terms of braking performance because at nearly six times the cost and three times the weight maybe aluminum would be better for other applications the next factor to consider is the magnet orientation in this first test I have all the magnets mount hunted with their poles aligned so the North Face is pointing towards the camera but what if I flip one set of magnets so we'll have a north south north configuration this slows the disc even further because the Eddy currents in the copper disk are produced by a change in magnetic field so having the magnetic field change with the flip magnets generates stronger Eddy currents in the copper and therefore more resistance I next wanted to test these smaller Cube magnets as I can pack them tighter around the perimeter of the disc and these produce so much resistance that I think we need to increase the weight I also tried the cube magnets in a checkerboard configuration but this seemed to perform worse since I've been using magnets I had laying around from older projects I need to find some new magnets for this setup the best option I could find online were these 10 millimeter Cube magnets which should outperform the previous eight millimeter magnets so I ordered 200 of them but once mounted on the test rig they performed very similar to the smaller magnets despite being 200 larger in volume which may be considered the quality of these magnets are skewing the results of this test after a quick Google search it seemed there are many grades of these neodymium magnets and these seem to be on the lower end so I had a search around to find a reputable seller of n52 grade magnets which appear to be the strongest they were double the price but it's clear why attaching the cheap magnets to my bench vise I was able to suspend a single 1.2 kilogram weight below it but with the n52 grade magnet it could hold about two and a half times that so I then printed amount for these new magnets and as you can see they produce so much resistance that the disc barely moves so now we know the combination of these high quality magnets and this copper disc produces such a high resistance force how are we going to fit this on a bike my initial plan was to mount the copper disc to the rear wheel spokes as I've previously mounted pulleys for electric motor drives but one issue is the spokes aren't designed to be a Precision mounting point so it will be very difficult to get the copper disc spinning perfectly straight to avoid hitting the small gaps between the magnets the only other option is to mount everything to this thread on the Hub which is designed for just a single sprocket so we need to fit a copper disc a disc full of magnets and a sprocket into this 12 and a half millimeter wide thread which also has two different diameters and thread directions the largest of which is apparently a standard bicycle Imperial thread of 1.37 inch diameter and 24 threads per inch which obviously makes things a little difficult as I can't exactly machine this thread but fortunately we have a solution as form Labs have sent me their form 3 plus printer and some of their rigid 4K resin this 3D printer works by curing a resin with a UV laser and this allows for much more precise and hopefully stronger Parts than my other 3D printers once the print is complete it needs to be cleaned in isopropyl alcohol before going through a final UV cure to make sure the part is as strong as possible then simply break off the support and we have a mount for the copper disk with the fine Hub threads already printed in once threaded onto the Hub it's clear that this makes for a much straighter Mount but the question is is it strong enough well I know how we can find out as I also printed this rear sprocket from the same resin to see if it can handle the torque through the chain without stripping the thin threads that lock it to the hub and surprisingly it worked quite well for about a mile of mixed flat riding and slide up hills until this happened big problem for the project however it gave me some useful information as we can see from the footage the crack started from the sprocket teeth and moved inwards towards the Hub suggesting that the Hub wasn't the weak point but instead the design of the Sprocket was as it can't be thicker than three millimeters and still mesh with the chain so the next step is to cut a large copper disc that will be one half of the magnetic clutch and we will directly Drive the rear wheel then we need a lot of magnets mounting to a freely spinning hub that will be driven by the chain but how many magnets do we actually need I went back to the test rig and used this container of water to vary the weight hanging from the axle which directly varies the torque on the disc and this showed that the RPM and the disc is proportional to the torque applied so if we plot a graph of torque versus RPM we get a nice straight line the next thing to test is how the RPM is affected by the number of magnets so each test had a different number of magnet sets with each set containing four magnets as I have two on the front and two on the back and this result wasn't linear so increasing the number of magnets didn't reduce the RPM proportionally but from the first test we know that the weight is directly linked to the RPM so we can actually calculate the required weights we need to reach 100 RPM for each set of magnets this now gives us a constant RPM value with a varying amount of torque and can be used to calculate the dissipated power which if we plot a graph of the power at 100 RPM we get a straight line so this means the number of magnets is directly linked to the power transfer through the clutch and not the RPM I then found this calculator online for working out the power required to ride a bike at a given speed so let's say we want to ride at 10 miles per hour we need a power output of about 68 Watts from the test I know that five sets of magnets can produce 7.8 Watts at 100 RPM this means that I'll need 44 sets of magnets and because there's four magnets per set that's a total of 176 magnets but there's a catch this 100 RPM is the speed difference between the copper and the magnets which at 10 miles per hour the rear wheel and therefore the copper disc will be spinning at 135 RPM so the magnets need to be spinning 100 RPM faster than that at 235 RPM this might sound like a lot but actually through the sprocket ratio on this bike I only need to Pedal at about 84 RPM to maintain the speed difference which is definitely achievable I then realized it was cheaper to buy these magnets in packs of 100 so I ordered 200 just to be safe which seemed to wipe out their stock and having 200 magnets means there's 50 sets which should fit perfectly around the perimeter of the copper disk I then press fit some bearings onto an adapter that matches the smaller thread on the Hub then these bearings can be pressed into the rear sprocket that also has mounting holes for another disc that will be used to attach the magnets which I cut from an aluminum sheet once everything is on the bike it's easier to understand how this will all work the chain spins the aluminum disc which will eventually hold the 200 magnets and it isn't linked to the rotation of the rear wheel or copper disc at all due to the bearings so the pedals and rear wheels spin completely independent of one another and the only way to drive the rear wheel is through the magnets inducing current in the copper disc I then 3D printed a magnet bracket using clear resin on the form 3 and these turned out really nice the magnets can then be pushed into position with the flipped polarity configuration and attached to the aluminum disc with a couple bolts by spinning the pedals it's clear the magnets have an effect on the rear wheel which is very promising as this is just 10 of the final number of magnets but there is a slight issue the chain sticks to them to stop this from happening the magnets need to be spaced further from the chain which meant redesigning and printing new Hub parts to move the clutch towards the wheel as well as a thicker aluminum disc to prevent it flexing under the load of the magnets also because the magnets stick to the spokes too I had to squeeze the spokes together with cable ties to move them away from the clutch which isn't structurally ideal but this is a prototype and now the magnets are able to spin freely without sticking to the chain spokes or steel bike frame and we can get a sense for how this clutch both accelerates the wheel when the pedals Spin and applies a braking Force when the pedals are stopped all we need to do now is Mount the remaining 180 magnets to hopefully increase the torque transfer through the clutch and with the rear wheel locked there is a huge amount of resistance on the pedals but the magnets are still able to spin and with the pedals locked the wheel is also able to spin and experiences the same resistance through the clutch however when both are able to spin the pedals Drive the rear wheel with ease and the braking Force stops the wheel almost instantly I think it's time to test this thing and if we can actually move through the power of magnets let's go okay it's kind of working if it just feels like I'm in a really low gear let's see what happens if I stop pedaling oh I can really feel that resistance through the magnets turn around okay hill now struggling a bit it's still able to move can't get much speed at the moment drop a pedal quicker oh right I mean it works I think I might need to adjust the chain tension because it sounds like it's sticking to the magnets and then starting to jump off the sprocket so let's tighten that a bit in theory the faster the more torque it should output according to my previous tests but like now I'm just matching the speed of the copper and then if I stop pulling oh I can really feel that resistance it's a really weird sensation because it it's uh I'm not directly linked to the to the rear wheels it almost feels like I'm wheel spinning if I put too much force in it's very hard to keep a constant pedal speed with the with the rear wheel okay here we go and lock the pedals so it's not quite as strong as a regular break uh but it does slow down a bit not bad I think we need a larger sprocket on the front to increase the gear ratio the reason why it feels like it's in a low gear is due to the difference in RPM between the magnets and the copper because in order to generate Eddy currents the magnets must always be moving faster than the copper therefore to reach a higher speed I need to adjust the sprocket ratio to spin the magnets faster so I removed the original 45 2 sprocket and machined my own 68 to sprocket which should increase the ratio significantly and has a diameter of 280 millimeters which is the size limit for this bike frame fortunately when buying new chains they are slightly longer than most bikes need so I didn't need to join Two Chains together to make up the extra length this really helped when cycling as I was able to maintain a pretty decent speed when riding on flat ground it's difficult to explain exactly how this magnetic clutch feels as it's so unique when riding on smooth flat ground as long as the magnets are spinning faster than the wheel the toolk at the rear wheel is determined by how fast you pedal also the braking Force seems to be a lot stronger at high speed due to the difference in RPM of the magnets and the copper so it's not great for bringing the bike to a complete stop but if you wanted to bleed up some speed it definitely has some advantages over a conventional break due to its no contact friction which would save on brake pads and that's the summary of The Good features of this system so would I recommend you build this for your own bike definitely not you see when climbing Hills it really struggles the best way I can describe it is it's like riding on Sand every 10 of the pedals requires a lot of force and you barely move forwards it might look like I'm riding in a low gear due to the pedals spinning faster than the wheel but because the RPM difference between the magnets and copper is so large the resistance is also really high so I'm putting a lot of force through the pedals and going nowhere speaking of nowhere I was exhausted after riding up this relatively easy Hill and if you're wondering where all my energy was going well when I got home I took out my thermal camera and hopefully this makes it a little more obvious yep the copper disc has heated up by about 20 degrees Celsius which using the specific heat capacity of copper and the mass of this disc is about 8 000 joules of wasted energy I had hoped it would somewhat simulate a continuously variable transmission where you could pedal at whatever speed you like and the clutch would vary the torque output depending on the inclination of ground you were riding on however the wasted heat energy makes it so much harder to ride and therefore I can't see it having any advantages over a direct chain drive but hey I had an idea and I wanted to test it which was made possible by onshape which is a professional grade computer aided design and product data management tool for businesses and in fact on-shape is used by Form Labs who create the formlabs resin printer onshape was actually created by the same Founders as SolidWorks because they saw the engineering teams needed product development tools that were more modern so they created onshape from scratch and unlike other CAD programs onshape uses Cloud native architecture which allows for real-time collaboration can be used on mobiles and tablets and much more plus it means zero crashes leading to no lost work they continuously add new features every month and one major update coming soon that I'm excited for is the addition of professional grade cam or computer-aided manufacturing as this will make it far easier for me to generate G-Code for running my CNC router overall uncheap allows engineering and design teams to work far more efficiently so if you or your business would like to try onshape for free go to onshape.pro forward slash Tom Stanton or check out the link in the description down below

Channel: Tom Stanton

Views: 5,731,641

Keywords:

Id: Dg8oVR4k5Dk

Length: 17min 56sec (1076 seconds)

Published: Fri Mar 24 2023