Building a more efficient bike

the bicycle is the most efficient form of human-powered transportation and today we are going to attempt to make it much more efficient the conversion of power from our legs to the wheels is where things can improve in a typical crank operation may not be designed to be as efficient as it could be so let's take a look at a normal crank operation here we have a readout of angle and torque now if we add a mask to the system and pause at 45 degrees we should be producing half the total amount of torque available in this system and if we pause at 90 degrees we should be producing the total amount of torque available this is because as we rotate our lever arm is increasing or decreasing in length and a longer arm translates to more leverage now if we take a look at this graph of torque versus angle you can see that the Pica monotorque is only available for a short amount of time at 90 degrees and it quickly drops off from there now what we're aiming for is to have the most available torque in the system at any point in the rotation so graph may look something like this we also need enough space in the frame to house any sort of design components that we come up with and as it turns out my dad has an old mountain bike that has just the frame we need now before we do any designing we're going to measure this bike and bring it into CAD so that we can use the model as a reference but measuring by hand seem to leave way too much room for error so I tried using some free software that lets you scan in objects from The Real World but it's pretty terrifying but it does seem to be just accurate enough for what I need so we're going to give it a shot [Music] and now with everything scanned in we can add whatever we need to to the scene but that's coming later and I'll start by cleaning up the geometry following the scan of the bike and then I can bring it into my CAD software and start building out the model without interfering with any parts on the virtual bike frame [Music] [Music] now even I was confused with everything going on here so I made a quick mock-up to demonstrate the main function [Music] alright so in this model there are two plastic rollers that follow a plastic track and so the shape of the track here is really important we can think of the track here as a set of Hills and Valleys and these straight areas of the track are where we should get a constant force which should translate to constant torque and the rounded areas are just too smooth the transition in between now if you imagine wrapping that flat sign pattern around a cylinder we can now transfer the linear motion of the rollers into rotary motion one concern that I have is that these rollers stay perpendicular to the track on a flat surface but once you roll that track around a cylinder the track stays perpendicular with the cylinder but the tangency point between the rollers and the track change so what you're left with are these small contact points between the rollers and the track which could cause some weird things to happen but we're going to worry about that later Now by keeping the variables the same we'll just swap out the crank for our cam now as you can see when we reach 45 degrees the torque is already at the maximum level this continues to 90 degrees and even up past 110. if we take a look at the graph for this model we can see that the peak torque is reached very quickly and maintains this for more than half the rotation and comparing this to our typical crank arm graph you can see what the benefit might be but before we get started I got to prove this out in the real world which after looking around in my dad's junk I found this alternator which I'm going to screw to this wooden board and then tie to the bike using a pulley and hopefully we can light up a light bulb more torque should generate more speed and therefore translate into a brighter or dimmer bulb depending on how fast we were going but I guess I'm not an electrician so we switched it up so my brother figured this out but sound should be a direct representation of the speed of a wheel and the more torque we have the faster the wheel should spin now you should be able to hear the speed change pretty clearly but what's really cool is we can track our torque curve right across the audio waveform now even though that was cool I still need something a little more data driven so I hooked up a scale to the drive shaft or the crankshaft on the bike and with this we should be able to replicate the animation that we saw earlier now if you are following along at home I don't recommend this green horseshoe it may be a perfect shape to hold onto the pedal but it was designed for horses and backyard games not pedals after collecting way too much data I moved into Excel where I plotted the X and Y values all by myself and it does look like it's matching our prediction now we have a lot of work to do before we get to any other experiments so let's get started foreign [Music] [Music] oh and now that we've got it working let's explain how it works we'll start with this cam Drive but we'll need it to transfer power to the wheels and for that we have a set of minor gears that will change the direction of rotation outputting power from the drive gear to the crank gear using this chain and the whole thing is powered by your feet while gliding along these linear shafts with these cam rollers mounted to these linear bearings and finally everything is held in place with these frame mounts now before I had a chance to break anything I did another test with the same setup as before and after plotting everything once again in Excel we are left with the result that closely resembles our prediction of course we had to break something luckily this was fixable with a little bit of super glue and some JB Weld which should hold up fairly well next I had to figure out how to tension the chain so first I tried using this plastic spring-loaded piece but I needed the chain to be loaded in both directions so this gear seemed to work out pretty well for that [Music] oh yeah [Music] if you enjoyed this video a great way to support me directly is by checking out my patreon a big thanks to those of you who already do and consider subscribing because there will be more projects in the works [Music] [Music] oh sh well I wasn't wrong about the uh stress under the plastic

Channel: Works By Design

Views: 1,723,203

Keywords:

Id: xev18rBrNr0

Length: 10min 21sec (621 seconds)

Published: Sun Jun 25 2023