The Ratcheting CVT is an IMPOSTER!
HTML-код
- Опубликовано: 4 окт 2024
- Check out my Patreon Page! / geardownforwhat
Buy a ratcheting CVT T Shirt! a.co/dluV19i
or a hoodie! a.co/bLs7kWl
Where you can buy a 3D Printer: goo.gl/gwKB6J
My Favorite Cheap Filament! :goo.gl/FuzECr
For Strong Filament! CF Nylon (NylonX) : goo.gl/a6zeQd
Special Nozzle for NylonX: goo.gl/WTNhxq
Matterhackers: goo.gl/PxNhqg
Follow me on Twitter
/ gdfw17
I post most of my designs on thingiverse
www.thingivers...
Would you help translate my videos? : www.youtube.com...
Links to products in this description may be affiliate links. These links do not add any additional cost to your purchase, but the channel gets a small kickback, which helps us a bunch! Thank you!
“Extremely shelf-worthy,” that describes a lot of the things my 3D printer churns out!
Something to consider. The fact that this CVT has a set number of teeth doesn't exclude it from being a CVT. Because the fluidity in the mechanism comes from the driving arms. One could just as easily argue that it wouldn't be a CVT with Sprag clutch because there's a specific number of sprags.
I've be thinking about this a lot, too. But finally electric motor applications are there everywhere and have a dynamic range from 0 to real quick.
Currently attempting to print a small version and utilize it in an RC application
I feel like this would work extremely well as a bicycle transmission
Despite being "not practical", this contraption would be a great addition to every steam punk themed apparatus.
Don't sprag clutches rely on friction?
The cvt is by no means "continuous" in the real sense of the word. You have a minimum change in the gear ratio (you cant have a 1.0001432634 gear ratio). So as your gear ratio increases, you approach the behavior of a belt driven cvt. (Analogous to a discrete function approaching a linear behavior as delta approaches to 0)
it's perfect for hobbyists though.
How would you go in reverse
Flexi coil used a very similar CVT with Sprag clutches on there air seeder carts to very seed rates. ( grain farm stuff) way a head of its time. It worked great till it didn't. It would
Fail at the worst times
Yeah a week
I have an application . . .
Good
have you seen maker's muse's sprag clutch (3d printed)
Awww! You blew my hopes for a potential bicycle drivetrain.
To know what doesn’t work can be more valuable than knowing what does work, just look at the various lawsuits flying around between electric vehicle companies and the various companies they’ve worked for, one over another claiming intellectual property rights, misuse, etc...
Put it another way; the intelligence services justify massive and secret budgets by often saying, the bad guy only has to be lucky once, they have to be lucky all the time.
It may not be infinitely variable, but doesn't it still fit the requirements that the CVT typically tries to solve? Surely with non-uniform phasing of the ratcheting mechanism, you can get enough ratios to emulate CVT behaviour, and certainly enough to keep within a particular efficiency/power band?
Well, the reason that it is not a CVT might make it even more useful for some applications. I was thinking about what kind of transmission to use in a lathe to translate chuck rotation speed, into
tool holder speed for thread cutting. There is a practical precision limit anyway, and you want have locked in parameters that can be exactly selected in that application.
There is that Chinese mini-lathe I was thinking of try to tinker around where you have to change the gears manually for thread cutting - I found that ok for the price, but wondered if there is not a cheap better solution. Is your design free or affected by the patents you mentioned in the other video?
Good clarifying video do a further explanatory one about the output shaft
This is an important video for one key reason: being able to admit and explain failures during development, and knowing when to abandon a line of thought, is a hugely important aspect of invention. Well done!
Also by the way I once 3d printed a sprag clutch that worked briefly, until it wore a little flat spot and refused to engage properly after that.
Yaaaaa do it
Although shelf worthy, it's still an extremely cool concept. I'd love to see this fabbed up and incorporated into a car as an experiment just to see how it would behave!
Not Shelfworthy at all. Duplicate the input side to replace the output as a variable crankshaft. Time the output ratio inverse to the input ratio mechanically (using intermediate shafts perhaps?). Have fun!
Could you solve the unevenness problem by adding more arms with overlapping "power strokes"? It would also help balancing the thing.
And regarding the complexity. Is it more complex than a dual clutch semi-auto? Doesn't look like it to me.
wouldn't it turn into a IVT if you have equal or more shafts than gear teethes on the output shaft?
I would really like to hear the explanation about the output personally, so I hope you cover it in a video.
I love that thing
Hi. Please add animation of variable egsantric.. thanks..
just make another input end to the out put!!!!!
This is what I thought, fair of you admitting all these flaws which render it impractical in any real world use.
Actually there are real world applications which could benefit from non-smooth motion by being forced into a fixed position n times around the shaft... For instance, there is a gear system used for old projectors to advance the frame really quickly... That solution is much simpler though... So I could see this working on a few things, but as he mentioned because of its complexity there are other simpler and less expensive alternatives already available...
It already is, and has been used, for many years. www.zero-max.com/cda-unidirectional-drives. The guts are essentially identical to what he has there. Multiple parallel linkages with a sprag clutch.
You could build a sprag clutch cheaply by opening up a common set of needle bearings and printing a matched inner set of lands. I expect it would wear out over time if made out of plastic, but for the sake or proving a prototype it would work just fine for a short run.
Ratcheting CVT kinda SUS, ngl 👀
A piece missing its puzzle
I would use one in a continuously variable winch
Google zeromax transmission
This is a fun exercise in dynamics of motion.
But we all know electrical 3 phase induction motors or alternators, make a superior transfer from energy to work and vice verse. With contemporary electronics and materials, mechanical systems have place automation, mineral resource extraction, construction, 3R, automotive, aviation, agriculture, shipping, transport, military, etc. We used simple mechanical relays and heavy-duty motors, engines, gears, cams, linkages. Now with servos, VSDs, electronics, hydraulic, pneumatic, systems the flexibility is great. But with good engineering, cams, gears, and relays can give a huge cost advantage and simple maintenance.
Neat machine! What is the gear range of the machine you printed? Are there practical limits on this range? Can it overdrive the output shaft? Would offsetting the arms on the large end increase or decrease torque output? Keep up the good work and thanks in advance!
I was going to say use oneway bearings instead of a ratcheting mechanism. Would smooth it out a bit and quiet it down a bit....
There is a type of gear system which could smooth out the output velocity very much its a gear system that gives a weighted flywheel a kick whenever it's spinning slower as the input gear the down site to this is that it has a rather big amount of friction.
And you could counteract vibration a little bit by installing two CVT's that are mirrored like one has the Ratchets pointing to the left and one pointing to the right and the do the same actions mirrored at the same time.
Have you thought about extending the size of the base and adding a 4th arm to the piece to reduce vibrations?
It is awesome even it is rachating we can use it as cvt
The Ratcheting CVT is actually a clutch.
Nope
someone will find a good use for it :) the inception drive is another attempt of a CVT, but they are looking to make it into robotics, not cars. Those slow robot arms would like an overdrive state when unloaded ;)
I would have to pull out my RC gas car to see the engagement in the pull start sprag bearing. You can find them for cheap if you need a bearing like that again! Keep it up man.
I can imagine it with one way spring clutches which therefore are able to make infinite number of ratios. Possibly. But it will have to be from steel.
The biggest limit in the use of this in automotive use that I can find is that it is freewheeling. The vibrations can be tuned-out, the number of moving parts (and losses) can be reduced by not using the planerary gear system for speed adjustments, the load on the bearings can be reduced by having more, thinner, arms and having them alternating on each side of the output shaft to limit radial load on the output shaft bearings. For the last part, the limiting factor will be that a one-way clutch roller bearing will be quite a bit wider than the needed width of the steel arm but that could partially be solved by having a thick arm and making a lot of cutouts.
Make sure to bring this concept to the attention of your local Formula Student/FSAE team as a CVT as compact as this will be a great performance boost to any Combustion Formula Student car and those teams are always up for engineering their way out of a problem and finding if the freewheeling is acutally a problem.
Thought CVT’s were banned from formula?
Not Formula Student. Formula Student is not a racing competition as the usual FIA regulated classes but are instead the worlds biggest engineering competition where the students design and manufacture a racing car to display their skills. There are plenty of videos of Formula Student on youtube so I'd recommend watching a few to feel inspired.
This class has very little restrictions on which technologies that may be used but because the cars are built by students, the tracks are relatively low speed (top speed about 120 km/h) with a lot of sharp corners. This means that a CVT would be perfect for a combustion car so they don't have to change gears all the time. At the same time, these cars have no lower limit for weight so they are constantly being optimized and I believe that this CVT concept may result in a more compact rear end of the car than a conventional gearbox.
Imagine the variable hydraulic pumps like the one featured on ave’s channel pumping into a motor. Bang infinitley variable non friction based cvt
The original version of this drive, using sprag clutches, as patented in the early 1950's is still made by Zero-Max Industries in Minnesota and used all over in industry. They are incredibly cheap in BOM cost, and only mildly expensive in assembly labor while being very useful in particular situations. If powwow reach out to you, point them to the company that literally invented the darn thing!
Zero-max is slightly different in design though.
Sebastian Jørgensen How do you mean? I have the one that runs the power feed to my mill apart on the bench right now, and it seems like the same thing to me.
As far as I can see, Zero-Max changes the gearing by moving linkages while this does it by varying eccentricity. I can't find many good drawings, animations or videos of the Zero-max though so I might have seen a different design than you are using.
Sebastian Jørgensen It has an intermediate arm that adjusts stroke length rather than this internal-to-the-wheel design, but it still is adjusting the length of the linear motion to change the speed ratio.
But that makes it a different technology. As I said before, I haven't seen enough of the zero-max to have a complete overview of how it works but I've found a drawing on their webpage. It seems that the eccentricity is fixed but that the eccentricity rotates inside a connecting rod which has one degree of freedom limited by the adjustable intermediate arm so that the amount that the connecting rod pushes the main link towards the pivot on the output shaft is vaired. This gives a total of 6 joints for each "phase" as far as I can count and also, the connecting rod is accelerated linearly the same amount no matter the speed ratio setting.
In the design proposed by GDFW, has only 3 joints per "phase" and the linear acceleration of the connecting rod is zero at low speeds and rises linearly to the maximum at high speeds. If implemented properly, this design is simpler, which may make it more compact, cheaper and more efficient.
It's still a great contraption though!
I havent sat down and done any design work on this level in a very long time, at least 10+ years. This however has me extremely intrigued. i think i might sit and take some notes on my thoughts currently on how this could potentially be implemented. and in a couple of different applications. To include potential ways to reduce some of the vibrations that will be generated by this.
I would like to thank you in getting this out of practice brain moving again(albiet with some protesting) and giving something to occupy my thoughts with.
Wait, there are uses for incremental output turn...what I can think of, this might be excessively complex...but sometimes that's a good thing...
Another limitation of this system seems to be that the system only allows power to move one way. This may seem like not a problem at first, but this means that when idly rolling the car must use gas to keep the engine moving....
It's still super cool though.
Used to work on mini. Their cvt was always a problem sooner or later.
I'm going to watch the rest of your videos on this machine, but so far, I have to disagree with you on it being "shelfable". I see potential as a bicycle, or more specifically, a recumbent trike, transmission. The varying output shaft speed isn't an issue because a human pedaling a bicycle crank varies the crank speed wildly thru the 360 degree rotation, so any variations in your transmission would be masked completely. The reason for use on a trike is because it can be positioned in such a way to to give much more ground clearance, getting the drive train out of harms way.
What would happen if you made a ratchet setup out of everyones friend, the neodymium magnet?
considering there are already sprag clutches in "traditional" automatic transmissions. while i am not a fan of cvt's i think using sprags in this transmission could still possibly keep it considerable for the automotive industry. as for the vibrations, i wouldnt know how to fix that. good luck!
would it be possible to smooth out the output with a flywheel? is there any reason, other than extra mass, what you wouldn't want to do that?
Because the load and engine are always directly coupled, the load's inertia itself will act like a flywheel. The reason vehicles have one is because the load can be decoupled from the engine.
What if there were a stochastic element, like some elasticity and a vibrating element, where it would 'dither' the engagement of the cogs? Could it then be continuously variable on average? Each individual stroke would engage a whole number of teeth, but aggregated over time it might be possible to advance a non-whole number of teeth on average.
But I think this CVT would be great for heavy machines that only move slowly but the CVT would have to be pretty big.
i was looking at the output shaft in the last video i was thinking no way that is usable power
In plain English.
eh, I noticed all of those problems (plus others you didn't mention in this video like how it would actually wear down quicker than a fluid based clutch) in the previous video.
I did also come up with similar ideas when I was in primary school (elementary for you americans) while thinking up how an automatic car tranmission functioned (since I had no internet nor anyone to ask around how they worked), I didn't know they were called CVTs or IVTs. Basically my idea came around from using a stator-like disc for one side, and the other side would have more stators but the friction blades are rotatable. Quite simple, but even that design has its flaws, in that the pressure must be contained well somehow (I haven't even looked up how fluid clutches work in automated engines, but I do know there are fluid-tight mechanisms out there otherwise things like hydraulics wouldn't work) and would also experience oscillating vibrations like your product, unless more blades are added (with also, a diminishing return, and having thinner blades also means they're more prone to breakage) and heat dissipation would also be an issue, although adding cooling vanes would be simple enough. If there is a failure in the system, it would be catastrophic like how a single turbine blade in a turbine engine detaching would also be catastrophic.
I never had any machining skills nor equipment so all I really had was lego to try out my ideas. And well, lego isn't exactly suitable for the concept I had.
Just saw this video after posting my comment on your first video moments ago. You explained what I saw in the output shaft. Sprag-clutch sound more expensive than a simple one way roller bearing. I'm a copier technician. Copiers have many one way bearings that last the life of the machine. I know there are one way bearings in car transmissions that last the life of the car. Are those bearings the sprag-clutch type? I don't know. I would like to see your product becoming a success story.
Couldn't you make one with 2 cone shaped cluches? (or try with a rubber cones first) maybe ball bearings in there. Oh maybe 2 cones in an x and the cross point has some sort of connector
could you add a structure like steam train wheels on the output arms? i think the point is not "expensive" but the percentage of energy lost.
Wow! The print looks really good! What printer are you using? Great video!
build a vct variable circumference transmission please
Did you mean a "sprag clutch" or a "trapped roller clutch"? You said the first but showed a clip of the second…
Renold Sprag Clutch: ruclips.net/video/Fsp3fm4KHs0/видео.html
Renold Trapped Roller Clutch: ruclips.net/video/QjR7dimpSJA/видео.html
It's kinda impractical, sure. But it's a neat demonstration of the principles.
It was a long week...
Why cant we go with constant velocity you can add addition set
What if you are going on a crank shaft instead of ratchet shaft?
If you ratchet the on both the forward and backward stroke wouldn't if maintain a constant speed?
Use a SMOOTH output Rod & Rubber blockers to drive the output ! This will make a CVT !
I like that
+Lego Technic Mastery While it is an interesting idea, it introduces slip under normal operating conditions in both directions. The one advantage of the ratcheting CVT is that it has an exact 1:1 angular velocity ratio between the output shaft and the currently engaged ratchet. With a rubber surface you can come close with a proper design but there will always be a non-linear engagement where the two surfaces meet, which means the angular velocity is a function of the output torque.
From a theoretical perspective the plastic ratchet one will actually have an ever so slight angular velocity difference as well, but it's a linear function of the torque and proportional to the Young's modulus of the ratchet material(s). Since rubber deforms in a non-linear way we can't make the same approximation with a rubber output shaft. The same non-linearity would also eliminate the approximation that the output shaft torsion is negligible for small loads. Keep in mind that while there are values floating around for a Young's modulus of rubber, the linear approximation (Hooke's law) used in the model of deformation applies poorly to very elastic materials like rubber. It sort of works for small numbers but that's about it.
Don't get me wrong, I'd love to see this idea implemented, but I can't see any practical advantage it would have over a traditional belt driven CVT. It's entirely possible that I'm just not seeing it and if so I'd be delighted to hear what that might be.
Sorry, I meant to say the difference in angular velocity due to torque is *inversely* proportional to Young's modulus. What I said would mean that hardened steel would behave worse than plastic under extreme loads which is clearly wrong.
Can you elaborate a bit? I dont understand OP's comment.
How hard would it be to 3D print some (non-ratchet-and-pawl) freewheels for demo purposes?
What I really would like to see you create, is the coolest transmission which *even fewer* people have heard of, the Constantinesco Torque Converter, which is one of the earliest automatics which doesn't use friction or hydraulics.
Interestingly, all the diagrams and prototypes of it which I've seen used a ratchet and pawls on the output end, when sprag clutches would be quieter and more efficient, so there's obviously room for improvement.
Nice explanation. :)
But... Can you make a "true CVT" version of it for... Scientific purposes?
Ti M when you write “for scientific purposes”, it sounds like you want to enjoy yourself with it.
Al Cam Indeed, but I would like to enjoy it in a voyeuristic way, because I just want to watch it working. ;D lol
cvt flip
Wow man, I still think that is pretty well made.
Perhaps you can think up of something that this gear might be useful for and then tweak it so that it does it better then any other.
Great stuff, don't knock yourself.
Highest regards,
Mac
wouldn't a electric motor basically do the same thing? like in tesla's truck.
Yeah that’s what I said
I'm your 50,000th subscriber
Yeah you are! thanks man!
Did I hear intermediate value theorem? Lol
What’s that?
Gear Down For What? It’s what IVT stands for in calculus 2
waste of time
Make it of metal
Man look your output is not finished. You can solve your "engine braking" issue and get reverse.. and smooth the output a great deal. You are so close. Just a few more of the parts you have already. Believe me from just a mech-mind standpoint you want to know this if you haven't figured it out already and you may have. If not I would love to tell you but not here. I could explain it on the phone in 30 seconds.... it would be like a call from Leonardo:)
Has anyone caught the name of that clutch he talked about?
"Sprag clutch". Also known as one-way bearings or one-way clutches. It is essentially a roller bearing which locks up if load is applied it one direction.
Lol, head hung low... been there done that.
Ok so build us a true cvt that we can use. Something for maybe 1 to 5 horse power.
I think using the good old design of scooter cvts, maybe replacing the weights with a lever (so you can vary it yourself), and using carbon fiber filament as a replaceable wear surface would work.
here's what I'm talking about : gallery.autodesk.com/fusion360/projects/scooter-cvt-gearbox
also an "hybrid" could be cool i.e. using a cvt with a lower range, and a standard gearbox, the cvt being enough to jump from one gear to the next with a bit of leeway (75% on the first gear becomes 25% on the second)
Dude, where can i contact you? Ive been thinking about ordering some torque gears for a ecar adaptation
Email is in the about tab, but I don’t sell products directly, sorry.
I'm trying to print one of my own. I used the link and got the design from thingivers, but im not sure how to put it togather, can you help me?
It’s just really hard
Gear Down For What? It would be helpful if you made a video about it, I'm sure other people are interested to
ok 👩🚀 pass it onto a 5yr old as a toy 🤪 he mite make it into a windup toy robot one day :) Oh some engineers tech made a cube that stands on it corners 3 wheels in side NOT fly wheels fook Alien tech is out their 👽 My box of toys walked out the door WTF
oh wow am i first?? awsome video man keep it up
Actually, there's no such thing as "infinite positions" in the physical Universe. There's always the limitations of the atoms and the Planck length.
True!
not in a mathematical point of view ! There's an infinite amount of points (tangents) around a circle !
Sixth!