3D Printing a Functional Wankel Engine

air it's everywhere and yet so useless that is until you compress it meet my 3d printed compressed air powered winkle engine it's an incredibly compact complex and powerful engine and bring this creation into fruition was not easy even the best fdm printers are inconsistent at best and of notoriously poor accuracy so designing a precision engine to operate despite these poor tolerances required a lot of work but i'll get into all that and all the different ways i tried to solve it in a bit first some background the winkle engine was designed by the definitely not narcissistic german engineer felix wenkel almost a century ago and to this day it remains one of if not the most powerful internal combustion engines by engine weight so why did i want to 3d print one of these uh mostly just because they look cool unlike traditional piston engines which have a piston sliding actually in a tube michael engines work by using this eccentric rotating dorito let's call it as this dorito rotates around you'll notice it creates constantly moving air pockets that are expanding and contracting and this is what's used to drive the engine now in a combustion powered rotary engine fuel would be injected here as this volume is expanding it's uncompressed a spark plug ignites it the mixture explodes and the expanding gases push the piston further creating torque and allowing it to continue running now there's a lot of neat advantages to this design of the engine but that's for another video what i care about is how to make one using my fdm 3d printer so without any further delay here it is and believe it or not this was my first attempt okay fine i may have made a few attempts before this one but that's how you learn so before we look too closely at the final working version let's take a look at all the different ways i learned how not to make a winkle engine first up we have jerry jerry's actually my second design but jerry's such a jerry he made me lose my first now this design uses little notches printed onto the piston which are designed to open up ball valves at the little crutch of the motion here as this piston spins around the idea is a little notch pushes the ball valve open allows air in expands spins cool idea didn't work so i ditched it next up is this a total redesign and it's actually inspired by tom stanton's most recent air engine it uses these little air flappers which work very similar to the way his piston engine works to get a better understanding of how my design works let's take a look at the cat so this includes just about everything except for the top engine cover and the pneumatic fittings for the air intake looking from the front we can take the cross section which exposes all the interesting parts so those valve flappers i mentioned previously those are the two green pieces then we have the dorito aka piston aka rotor in purple the eccentric shaft in cyan the main housing in yellow and each of the valve housings in red additionally there's two yellow balls and those are the bbs that will seal against those black features which are all the o-rings now taking a closer look at one of the valves it'll start off like this with the bb sitting in the o-ring which will seal it no air can come in as the piston rotates around it will push in the valve flapper now at this point it will seal off that first o-ring so no air can enter the main cylinder and then at this point it will now open up that one-way valve and allow air to pressurize that large chamber in the red housing but because that first o-ring is still sealed it can't get out then the piston continues around it will reseal the intake so no more air from the air supply can come in then shortly after it will break the seal on the second o-ring and allow air to enter the cylinder at which point the piston will be propelled by the expanded gases and the cycle can continue now this means that i need to have two o-rings which are both somewhere in here uh guess how i got them in [Music] yeah i didn't here we have this one as you can see it didn't last very long but it was the first attempt at using a two-piece assembly this allows me to get those little inner o-rings installed much more easily and then just requires some post-assembly good solution and i'll use it going forwards but this engine's toast next we have this one the first successful engine and when i say successful i mean it's fun it looks like it's spinning nice and smoothly but unfortunately this piston has terrible seals so this engine spun for one reason only and that's because the little valve flappers designed to control the air intake into the engine were pushing against the piston and the fact that it sealed so poorly meant it was very low friction and thus was able to spin around so basically this engine would run just as well without a top plate on not efficient not cool moving on i should mention that that previous engine also only worked because the valve stayed open for random amounts of time and that's because the little balls in the valves are free to float just about anywhere they'd like the result is valves that stay open way longer than they're supposed to and way more than the amount of air that is supposed to be contained in these pockets actually going into the cylinder so for this version i added some pegs inside which i could then attach some 3d printed springs to now these springs worked great unfortunately this also exposed a pretty big flaw on my engine the piston seals still suck here we have yet another total redesign of the engine but good news this time i actually designed it properly i master modeled it it came together way nicer all the parts are in sync it's so much easier to edit and it still doesn't work womp womp here we have basically the exact same engine as the previous one but i actually took my time and used silicone uh but then i didn't wait for it to dry before i ran it so it leaked ah and then i hit enlightenment what am i doing i need to fix the pistons so i finally spent some time and played around with all these different piston designs to figure out just why my engine had such terrible ceiling now at first i tried something like this which uses thicker o-rings at the apexes so that they should theoretically have more squish and allow them to seal better without applying as much pressure but unfortunately i found that these caught in all the little nooks and crannies from the intakes and exhausts on the inside of the engine bummer so then i figured hm why not go bigger and it was worse so then i said hmm why not go smaller and this actually worked surprisingly well after a little more tuning of the engine dimensions considering these apex seals are only one millimeter it required much more accuracy printing the piston and the housing but surprisingly this design worked pretty well and actually managed to create some pretty bad but non-zero seals that makes sense you'll also notice something else unique about this piston it has these little cuts in it this was a pretty cool idea i had which should allow the air to be injected into the engine quicker the left side shows with this new cut and the right side without notice how much earlier in the piston cycle the air gets injected into the cylinder so then we get to this engine it has an even bigger piston head reservoir it's using the new piston with better seals it was assembled properly i was patient and it didn't run what was my problem this time actually it didn't really have a problem it just didn't have enough power to turn over and sustain itself i bet if i added the second intake onto the other side of this engine it probably would have run just not very efficiently oh but the reason i didn't do that was because i accidentally printed this part in a rolled back state so it didn't print the exhaust vents and i had to drill them out by hand and i didn't really put them in the right spot my bad and next up we have this engine i fixed it by printing it with exhaust vents exciting and i also added both sides because i figured why not go all out and unfortunately it just didn't want to run this was the last version before i made some tweaks to the internal dimensions of this cylinder wall and it was just a little too tight in some spots too loose in others didn't seal well enough but was too high friction and wasn't able to turn over and finally we ended up here this is the final version and it has one critical feature which none of the others have it works well at least eventually it did made a few big changes with this version of the engine which needed some fine tuning to get working right so the biggest change with this version is that has a piston that actually makes a good seal surprising it took me this long right well no ceiling of the piston is basically the ending problem with these engines and it's why in real life these engines drink oil like nobody's business so how did i solve it well just a lot of iteration really and the result is an engine that is actually driven by the expanding air rather than those earlier protos which were just driven by that valve flapper pushing against the piston speaking of the valve flappers man what a pain they were you see this one this is what used to be in the engine when i first built it up and then i swapped in this one and to my surprise [Music] it suddenly worked and do you want to know what the difference was 100 microns yep i also did something with this version of the engine that i never do i printed the piston on a raft and why uh once again chasing the micro printing directly on a build surface will imprint any non-coplinarity into that first layer because that first layer of hot plastic is gonna just be eating it up whether it's 20 microns or 200 but by printing on a raft that raft will eat up any of that coplanarity error and the part itself will be printed on top which means the two faces should be nice and coplanar now i could have just leveled my printer bed better but yeah okay great so now you know what i've been doing for the past two months but let's get into the real meat and potatoes the reason you click this video [Music] and small aside yes i'm aware the engine cover is not made on an fdm printer the engine does work just fine with an fdm printed cover but then you won't get to see stuff like this [Music] hey look i put a propeller on it and speaking of propellers i also 3d printed one of tom stanton's earlier prototypes just as a comparison now as a quick disclaimer i did contact tom for a collab earlier on in this project he still hasn't gotten back to me so i can only assume that he was so flattered by my offer he was just too nervous to reply but that's alright let's take a look at the competition dang yeah that's um that's impressive so that was with about one liter of air at 40 psi and for comparison here's my rotary with the same air supply that's it yep now as you could probably tell my engine does have a lot more power but it just kind of releases it all at once yeah mine's definitely a lot less efficient and that's okay i didn't think it was going to be efficient i just wanted to make one that works and it does so that about wraps up this project thank you so much for watching huge thanks to all my patrons your support is really helping so much to make these videos possible if you enjoyed please like the video subscribe comment below if you have suggestions on projects you'd like to see and until next time

Channel: 3DprintedLife

Views: 249,087

Keywords: 3d printed wankel, wankel air engine, 3d printed rotary, 3d printed engine, rotary air engine, air powered wankel, air powered rotary, wankel engine, 3d printed air engine, custom wankel engine, air engine, tom stanton air engine, 3d printed project, hackathon project, compressed air engine, engineering, mechanical engineering, 3d printed, robotics project, capstone project, air powered engine, 3d printing, rotary engine, 3d printed projects, engineering explained

Id: S6cQpoUxeC8

Length: 13min 15sec (795 seconds)

Published: Sun Mar 07 2021