CVT Tranny?
 

I did some googling, but couldn't find anything about a miniature CVT transmission design. Sounds like a good idea to take advantage of a BL motor's torque, while eliminating that shifting hiccup (and possible loss of control), allow us to use smaller motors, and be able to use the motor for brake and reverse...

IIRC, most CVD designs in 1:1 cars are kinda limited in the amount of torque they can handle. Just wondering if it would be a good idea for R/C use.

I think you mean Continuously variable transmission - I'm not sure it would work as the motor RPM is kept pretty constant over a broad range of vehicle speed. Speed is determined by moving the pulleys which would be hard...

In a brushless motor RPMs are driven by voltage...

http://en.wikipedia.org/wiki/Continu...e_transmission

Yup, that's what I mean.

The motor would be driven to it's sweet rpm spot where it's at its max efficiency. The CVT would just work to keep the motor rpm at that spot (other than initial take-off from a dead stop).

I didn't think that constant RPM is a motors sweet spot IIRC motors didn't like being run unloaded...

No, not unloaded. But each motor does have an rpm where it is the most efficient (see Griffin's motorcalc proggy). The motor would still be loaded. The vehicle would just go faster and faster up to the CVT limit. So, the motor load would increase slowly, but rpms would be kept fairly constant.

Would eliminate setup problems where people gear too high to attain a high speed, while allowing enough rpms at low speed for proper ESC commutation.

You'd have to have a funky mixing radio to increase revs over first 30% say of throttle input and then hold max and then over say 10-100% of throttle movement move the pulleys apart. You also need to shrink the cones to 1/8 scale and have a servo that will take the force of moving the pulleys...

So does this work some what like a torque converter?

I guess I don't follow. The way I see it, the CVT would start out in a "low gear". Once the motor hits the preset rpm (via an rpm sensor), the CVD starts to increase the ratio slowly.

Think snowmobile clutch! That is a cv setup, it uses two seperate clutches connected by a belt. There are different helix, spring, and weight combos for different snowmobiles depending on the weight of the sled and the power being produced as well as what your riding style is and how hard you want it to engage as well as what RPM for it to start engaging. This is pretty much the only way a snowmobile can have a transmission, if you had to let off and shift and hammer on it again it would most likely get you stuck!

There could be potential in designing a system like this for RC but I think it would be better suited to a nitro rather than a BL truck. BL has no problem supplying torque on take off and that is pretty much the whole point of any transmission. The weight that you would save by being able to run a smaller motor would be offset by the weight of the system. Not to mention another tuning aspect, and if you think tuning a nitro motor is hard just try getting one of these set up perfectly! There are all kinds of books on tuning a snowmobile clutch that go way further in d!epth that tuning a nitro motor(if it's hot richen it, if its cold or lacking power lean it out) These clutch setups would be a major PITA!

I think that I will try to design and proto this, but will include an electric cooling fan in the core design, so to prevent Brian from using it. :yes:

I am surised that you are posting this here Brian, when you have access to the super secret area 51 section of the forum. Surely there is more knowledge in one of the allowed users there than all of us cretins combined?

check out the NuVinci continuously Variable planetary transmission, kinda the same idea but a little more high tek. This one is made for a bike, but i would assume that it could be made smaller for rc-applications. would be expensive though.

http://www.youtube.com/watch?v=kVPjhmTThPo

I've thought about designing one of these for a while. It is "CVT", BTW. The problem would still be how to make it handle all the power since CVTs have quite a few moving parts other that the two main pulleys/cogs.

Seen that. Love it. Wish I had thought of it... :lol:

wow ..that is sweet!...cvt?

the concept is pretty cool actually. orrigionally thought up by davinci himself, so you know its legit! ;)

I actually drew this before I hit "Submit Reply" on my first post but I forgot to attach it. :oops:

It's a belt-type CVT. Depending on the size of the belt and what it's constructed of it should handle BL power quite easily.

Going by your pic, I assume that the belt would require a spring loaded tensioner, like a serpentine setup on a car has. Unless you vary the effective diamters of the contact surface of both sets of "cones" exactly to allow the belt to retain the ideal tension at all times.

they do not need a tensioner, you cent the motor shaft and output shaft's Center-to-center in such a way that the belt would allow the motor to "idle"

* i know this because we run a cvt style centrifigal clutch on my racecar*

centrigal force via ramps inside the primary clutch squeze the cones together, puting force onto the belt. At high rpm, then the cones are as close the belt gets tighter- this tension spreads the cones of the secondary unit, making the effective Pulley diameter of the primary larger then that of the secondary, creating overdrive.



Great concept but your would need to do extensive testing with different ramps, weights and springs to find a good mix:party:

Thats cool, I was not aware that a mechanical method was used to vary the pulley size. I figured that it would be electronically controlled via some sort of electro-hydraulic mechanism.

nope its all centrifugal

as it spins, weighted arms pres against ramps that are attached to the moving cone.

heres a pic of mine
http://i152.photobucket.com/albums/s...9081733_02.jpg

I knew about the cone/pulley method, but the NuVinci one was new to me.

There's also a toroidal CVT (from howstuffworks.com):

http://static.howstuffworks.com/gif/cvt-17.gif

Either way, a design for R/C would have to be a small/light package.

And as someone mentioned, the goal of this is not necessarily to use a smaller motor, but to provide a better spread of ratios, especially in the high-speed projects. Motors can only spin so fast, yet gear them up too high and you have cogging at slower speeds.

And linc, I'll just remove/disable any fan you put in there. Don't think a lack of warranty will scare me. :wink:

wow, that would be fun to mmake, i dont know how you would incorberate it into an RC though

We're talking about Vario-matic transmissions right? I saw them used on a build once on Scrapheap challenge (Junkyard wars to you guys):

http://en.wikipedia.org/wiki/Variomatic

The concept was very simple- the two opposing cones changed shape as the rpms increased, causing the drive ratio to alter as the belt shifted shape/posistion between the two cones which spin the belt. Seems the cones were made of a plastic that would deform naturally due to the centrigugal forces at higher rpms. I'd be lying if I said I hadn't though about this concpet for r/c use, just wouldn't really now how to make it a reality...

I would just make it a nitro powed fan and then no one here will want to use it lol. Just think of the huge advantage at the track of a tranny that no one else can/want to use!

Sucks tho if DNF'd a race due to burning out a glow plug in your tranny...

Yeah, right. No one here has ever converted something from nitro to electric! :wink:

Thats not entirely true- remember the nitro powered Gorillamaxx conversion?

have a link?

the CVT has been used for about 8 years in the FIRST robotic competition

Each transmission weighted about 15 lbs, but one engineer explained ot me why they didnt use it. But you just dont see man CVT's out there..

Ill see if i can dig into them engineers brains

My ATV uses a CVT and it has a fan built in so the belt doesn't melt.

Do you do FIRST?!~

Team 87 here:yes: