Oskar's CVT (Continuous Variable Transmission) - Does this kinetic art have any useful application?
Вставка
- Опубліковано 7 лют 2025
- Print it yourself at oskarvandevent.... Buy at i.materialise.... Oskar's Continuously Variable Transmission (CVT) is based on one-way bearings. Many CVTs are based on friction. A well-know example is the DAF Variomatic ( en.wikipedia.o... ), which uses a push belt between conical pulley plates. This is also used for bicycles by Peyman ( www.peymancvt.... ). Another friction-type of CVT is the Enciolo CVP ( Continuously Variable Planetary, enviolo.com/te... ), which uses a planetary system of swivelling steel balls. Recently, Ratiorez announced their RatioZero CVT ( www.ratiozero...., • This Is The World's Fi... ) that uses one-way bearings (cf ratchets) instead of friction. The use of one-way bearings has the main benefit of being much less tolerances critical than friction-based CVTs. This brings the CVT concept into the realm of makers with 3D-printer.
My CVT uses a piston-like system. A ball chain is just to adjust stroke distance. There are three pistons, with relative phases of 0, 120 and 240 degrees, similar to three-phase electricity. Each piston drives two gears, connected with one-way bearings to the output shaft. One gear is for the forward motion of the piston, and one for the backward motion. So there is a total of six gears, and the system has six phases, one for each gear. The drive shaft is driven with a hexagonal gears. The purpose of the hexagonal gears is to even out the motion of the pistons. This assures that a piston is sliding at a constant speed when it is driving the output axles. This compensation system works exactly for any gearing ratio between the manually-driven input and the output shaft.
Copyright (c) 2025, M. Oskar van Deventer.
Frequently Asked Question: oskarvandevente...
Buy mass-produced Oskar puzzles at www.puzzlemast... (USA, CA) and www.sloyd.fi/b... (EU)
Buy exclusive 3D-printed Oskar puzzles at i.materialise...., www.chewiescus...
Balancing out the speed variations with non-round gears - genius!
Ratiozero does not address the speed variation, citing it as not causing issues in their intended uses. A true stable speed opens up new uses - likely ones with much higher power loads. If the one way bearings can hold at those loads, it might be a good alternative to the existing hydraulic and electrical power unit / drive combinations that are used as a "transmission" in trains and earth moving equipment. However, I don't think the hexagonal gearing can perfectly even out the variation at all gear ratios - my intuition tells me it works ideally at one ratio, or maybe harmonic variations of that ratio.
I find it difficult to get a grasp of this mechanism. Would it be possible to make all those sliding parts from transparent material so we can see those one-way bearings in action?
The files linked to below are quite insightful. It has a bunch of images demonstrating the mechanism.
@@jwrm22 I love you
Even just from an aesthetic perspective I would love to see this
not on his home 3D printer, unfortunately. clear plastic filament ends up looking as transparent as a white t-shirt
@@Vykori that makes perfect sense
The one-way bearing is still using friction to resist the reverse rotation.
Your hexagonal gears idea may indeed expand the application range of the ratcheting CVT a lot.
I would like to see a motor attached to drive the input at variable rates, and the tension on the chain adjusted while the mechanism is in motion. Only because I think it would be cool to see the pistons change amplitude while the output changes speed.
Does this kinetic art have any useful application?
bike,
rotate a crank forward & backward, the bike goes forward,
locked reverse directiion
Variable depth massage gun
fred flintstones would like to have word with you
It's interesting idéal, why Not for an électrique car ?
The title of this video was translated into French. I didn't ask for it, and I cannot remove this feature (I searrched through the settings).
Silly UA-cam. It is virtually impossible to convince the UA-cam algorithm that I am not speaking Dutch, despite have set all settings to English/American ...
It's likely a browser setting. Search for terms like " disable auto translate"
@@ideallyyours No, it is a new UA-cam "Feature" that they have forced out despite (1) no one wanting it, and (2) It clearly not being ready to ship.
Is that because your geolocation is a French region? You could try Google user account settings perhaps
@@DanDart It's just because I'm French. But I speak English, and therefore I don't want the titles of the videos to be translated (considering how bad 50% of the translations are, I really don't want that). I searched through all setting multiple times already, I have set the langues I speak as French and English, but the titles of the videos are still in French... I wrote this comment mainly because I thought that this translation might be wanted by Oskar, it might be a channel setting (but I didn't make this question clear in my original comment, I agree).
Your variable crank-shaft needs work to support higher power, but the funny gear torque ripple compensator is very interesting.
I'd look at the "zero-max" for inspiration. I uses a clever linkage between fixed cranks and ratchets to vary the ratio.
Also, ratchets have FAR better torque density than sprag or roller-ramp clutches. If a ratchet is fabricated with >300 engagement points per turn, then it'll work just as well.
Machined from metal, enclosed and lubricated with a watertight pass through for the chain, I suppose it could work quite well on a bike. Toleranced correctly and well manufactured it would be awesome!
Very clever, bravo, I supposé That behind the first hexagonale gear there is a a came ?
Tbere is a lot friction and Energy loss ?
There is already a CVT with a similar mechanism made of steel for a bike. There is also a video about it which is called CVT - This Is The World's First Geared CVT and It Will Blow Your Mind...
Ratcheting CVTs have been around for a long time. But I like how compact this one is, especially compared to the video you reference.
@@brendansimons6811 I like the fact that they reached production level with their CVT and that it does not fall apart under torque.
I like it, and I just went shopping for one way bearings to simplify my own design. I hope you can demonstrate speed regulation soon.
I have no clue if it's useful. But- I wonder how it wears out over time. Some engines continue to work correctly even as they wear out. This one might not.
What is the purpose of the 1-way bearing?
Ratchet
To transmit the power from the sliders to the output shaft.
That's fantastic!
Interesting engineering. Basically a Zeromax . Trying to reduce the speed jitters is made in a nice way...but the problem is not here. The main problem is the "almost" no backlash, "almost" no play and "almost" infinite rigidity... ending in a limited power capacity...at low speed...
Congratulations, certainly you know the "zeromax" CVD
I can see it serving industrial applications, but the full extent I can't firmulte atm.
Oh and your CVT screw thing. Might be soonish i can put my money where my noise bush is. Though i do apologise for the time it took. But it's things out of my hands.
If there might be any useful application, it will probably be a very niche specific obscure device that needs exactly this kind of transmission.
A one-way bearing IS friction! Every continuously-variable transmission REQUIRES friction. This is nothing new
I faul to see the 'Continuous' in this CVT...? Maybe some more explanation next Saturday? 🙂
The idea is, the movement of the chain affects the amplitude of the deflection of the discs. This in turn, affects how much the shaft is turned each stroke.
My comment with the lin got spam filtered, so here's a video code with a better design and a better explanation: G9-N-nIqc4g
Th principle is the same.
@@hxka Link: ua-cam.com/video/G9-N-nIqc4g/v-deo.html
Gotta love the Dutch :)
Very clever. However, it still uses friction inside the one way bearing, so it doesn't actually acheive the goal you set for yourself.
Thanks! To be honest, I do not know how one-way bearings work, or how the compare to ratchets. One could arguably claim that ratchets also use friction inside.
This friction has an infinite coefficient, it is absolute, like a ratchet, so its losses are practically nonexistent.
A belt CVT on the other hand relies on the balance between friction of the belt and its tension, so to achieve a good coefficient of friction (to transmit full power), it has to be heavily tensioned, which in itself causes friction, but this is taken away from the output.
There's another geared CVT I saw designed recently. It may give you ideas on how to improve your current designs: ua-cam.com/video/mWJHI7UHuys/v-deo.html&ab_channel=driving4answers
Thank you. I had already linked that in the description as one of the inspirations.
Genius! :D
How did you generate the gears?
Pretty standard involute gears. I generated them with an Excel sheet, that I developed ages ago.
@@OskarPuzzle The polygonal gears too?
The 'pistons' move in a triangle wave pattern. Without the hexagonal gears they would move in a sinusoidal pattern. Is that correct?
It will create a lot of heat and power lose throufh the friction
yyeeeeeesssss, more CVT
this is just a ratcheting CVT using one way bearings instead of ratchets
Whereas your statement is correct, I wonder why you say "just". The innovation of this design is the hexagonal gears that linearise the motion at any gearing ratio.
Yeah, as everyone in the comments to that """first""" """geared""" cvt was saying this is a ratcheting cvt. These aren't new (check Wikipedia for examples a century old). These never caught on because of nonlinearity of motion. Sure you can try to correct it by changing the shape of the gears but it would only perfectly cancel it for a single ratio. At every other ratio it would still be nonlinear. This makes them unsuitable for pretty much anything.
Your comment about non-linear motion is correct for some other versions, but incorrect for my version. At any amplitude, the piston moves with a perfect sine motion. And my hexagonal gears compensate for this at an amplitude. So my version is linear at any amplitude.
I wonder if it is provable. Like can you show and prove?
Would be interested in that video
@@OskarPuzzleIs the amplitude of the piston associated with the gear ratio?
a company called ratiozero have created a radial geared CVT. I have seen it used on a bicycle.
ua-cam.com/users/shortsDCtGOaCSgK0
Thanks. I had linked to them in the description, as they were one of my inspirations.
I understood zero of this.
Sorry. Bit hard to explain without extensive background knowledge, which is outside the scope of my videos.