CVT (Continuously Variable Transmision)

[ddk]

if memory serves me correctly (seldom does anymore) I believe it was the Lasalle (Lesalle lol i don no)( too lazy to google) automobile that used a motor-driven leather edged disk perpendicularly diving a leather covered platter connected to the drive wheel as a cvt transmission with (semi) infinite gearing in both forward and reverse; as the edge of the motor-driven disk could traverse across the whole diameter of the platter. the center of the platter was 'semi-neutral' :lol:
I had an interesting conversation with someone who owned and maintained just such an operational vehicle. It was manufactured around 1914.
Note-and I thought the Honda's cvt was troublesome? ...apparently one has to recover the leather disks after a few 100 miles use.
I believe chain-driven multi-cluster gears are the most efficient method of transferring power from input to output. Efficiency is Kind of Important if one expects to occasionally use an ebike as a pedal-only bike.
...not so much if the pedals are ornamental.

 

[motomech]

my old Honda PA50 used a cvt variable-width pully system
worked ok for a few thousand miles then gave me fits until I finally sold it.
All cvt designs, even the nuvinci, suffer from frictional loss, making them perhaps not so great as a final drive for something human-pedal-operated.
I've thought of implementing something like a dual cage derailleur like the pic.
Should be easy to implement on a recumbent with those stupidly long chain lines.View attachment 1 :lol:

The made in Belgium Honda Hobbit was the worst POS that Honda ever put their name on. I owned and opperated a motor scooter shop for 10 yr.s back in the '80's and I got to the point that I wouldn't take in Hobbits.

And yes, the CVT system is not very efficient. I raced "Mini's"[50cc to 125cc motorcycles] for many years and just for kicks I talked the promoter into letting me try one of my hop'ed-up scooters in the stock Yamaha YSR class. And even though my Aero 50 had more power, I got dusted. But back then[early -90's], there were no CVT aftermarket parts available. But I doubt it would have made any difference. Slipping the clutch out of the corners make all the diffence on sm. displacement race bikes.

The really neet thing about a properly set-up CVT[like stock], is it's ability to keep the engine in exactly the right rpm band all the time. It's uncanny how well they work.

 

[ddk]

The made in Belgium Honda Hobbit was the worst POS that Honda ever put their name on. I owned and opperated a motor scooter shop for 10 yr.s back in the '80's and I got to the point that I wouldn't take in Hobbits...

meh
my japan-made pa50 only served me for ten years thereabouts.
I couldn't get replacement parts for the pulley system for the longest time it seems... had to make my own parts.
Other than clearing the exhaust port of carbon every few months, the engine was really reliable (everything was reliable, excepting the transmission)

 

[motomech]

QUOTE- "Honda PA50
From Wikipedia, ....
....Honda PA 50 Hobbit (American market)/Camino (European market)
Manufacturer Honda
Also called Hobbit (US)
Camino (UK & Europe)
Production 1976-1991 (UK & Europe)
1978-1983 (US)
Assembly Belgium
Class Moped
Engine 49 cc single-cylinder two-stroke
Transmission Automatic clutch single-speed
Variomatic CVT
The Honda PA50 is a moped produced by Honda between 1976 and 1991. It was marketed as the Honda Hobbit in the USA and as the Honda Camino in the UK and Europe. The vehicle itself was manufactured in a factory in Belgium until 1991. "- END OF QUOTE

Now I'm sure that some Hobbits somewhere ran well, but, in general, like all European mo-peds, they are very "finicky". Their sub-systems, crankcase volumetric efficiency, cylinder scavenging and induction are all, each and of themselves, sketchy. They all need to be "spot on" for the things to run well. Otherwise, negative pressure at the carb venturi is poor and the fuel does not atomize well and the result is an engine that runs "ragged".
Not particularly picking on Hobbits, but as a guy who was trying to make a living with a low profit margin, even on the sweet running Japanese machines, taking in mo-peds would have put me in the poorhouse.
Ever try and check spark while pedaling?

 

[ddk]

maybe I was lucky (except with the transmission crap) :lol:
I traded it straight up for a Honda Express btw
which I hated
...and bought a bicycle to replace it.

sorry to hear about your former business problems
mine died from abrupt technology changes causing my eventual lack of interest.
but we go on living anyway

 

[amberwolf]

The Honda Spree <snip> the smallest and lightest motor scooters

I have most of one of those in the shed; it is indeed light, and might be lighter than my regular ebike, CrazyBike2. :lol:

I dind't know it had even a partial CVT; thought it was just a belt-drive, but I hadn't examined it very closely.

Perhaps sometime I will try working out a conversion of it into a full CVT, or at the least doing something like you described for tensioning. So far I have only used some of the electrical components, lighting, etc., on my ebikes.

 

[johnrobholmes]

My Hobbit is the PAII version, 1982. Out of all the mopeds I have had, it is actually one of the most reliable. Tomos mopeds have clutch problems a lot and the case seals blow out easy if there is ever a backfire. I have a Motobecane, what a nightmare to keep running! Once it runs it runs well, but the timing is always slipping since there is no keyway. I'm getting a Puch soon and they are pretty solid, but no gears or CVT makes them really slow to accelerate.

I have never had a vespa moped, they seem to be relatively solid other than the VERY cantilevered output that doesn't take well to higher power levels.


After playing with classic mopeds for years I have decided I only need one or two in my fleet. The rest of my bikes need to be electric or a modern scooter so I don't have to constantly wrench to keep my ride running.

 

[Hillhater]

But they didn't use a true CVT, instead of a frt. variator, they just had a fixed pulley

I dont understand ?
how can that thing have one variable (diameter) pulley and one fixed pulley with no tensioner to take up the slack ?
Also, if the front ( crankshaft ?) pulley is a fixed unit, why does it have those big flanges like a variable pulley. ?
Just puzzled !

 

[ddk]

QUOTE- "Honda PA50
From Wikipedia, ....
....Honda PA 50 Hobbit (American market)/Camino (European market)

please excuse my dementia but it is likely my so-called PA50 was a Hobbit II purchased mid 1979 with their variomatic (cvt) transmission (the VL series)
...and you're right about it being assembled in Belgium. according to wikipedia
see wut happens when I make an assumption! - i gets to show off my stuupid :lol:

either way cvt's are very useful for maintaining motor rpm but they all exhibit various degrees of inefficiency compared to a simple chain drive
I gave a great deal of consideration with using the nuvinci 171B for my current build, until I discovered they no longer provide warranty for my application. Because I personally have plenty of inefficiencies built-in to my superstructure (ain't so super no mo')

BTW I did try the dual opposing gear clusters, which only required me to flip over the jackshafted one, as I possess multiple freewheel gear clusters with 5/8" axle adapters (three gear clusters -if I can count properly :lol: )
My chain length is WAY to short between clusters to shift through the gears so I never bothered trying to fit a derailleur arm to the jackshafted (sic) gear cluster.
If I were more ambitious I would have mounted the jackshaft in a more suitable location but alas:
-I ain't that ambitious, where I already had the jackshaft bearing plate mounts welded on the trike
-I don't need 'that solution' for my trike, as a single gear is quite suitable for my needs.

 

[motomech]

Amberwolf-
Yeah, when Honda wanted to create a scooter market in 1983, they dumped Sprees for $399 and Az was ground zero. I was working at Az. Honda in Tucson at the time and the feeling at the dealership was, that once they sold a scooter, they really didn't want to see it again. That was the impetus for me open my shop, which I did in late 1984. Soon thereafter, Honda put the "Iowa Spree" on sale for the unbelievably low price of $299. The Iowa Spree was a "de-tuned" model that met the State of Iowa's moped law, ie., not to exceed 25 mph[which, at the time was adopted as an unofficial National standard]. For some reason, Honda must have imported 10's of thousands of the little buggers and students at the U of A and Az. State lapped them up in droves. Honda mandated that the Iowa Sprees be marked with a spot of red paint on the underside of the seat. Many dealers ignored Honda and just sold then as regular Sprees. If you still have the seat, you may see a red spot on the bottom side. Sprees were so popular and cheap, that many students just abandoned them at the end of the school year and if I could get permission from the dorm or frat house administrators, I would recycle them for parts.
So it's not suprising that one ended up in your garage.
One of the marketing ploys that dealerships used, was to promote the Sprees as "mopeds", free from any kind of registration and licensing, which of course, students took to mean they could do anything they wanted to do. The next step for the dealers, was to start selling all 50cc scooters as "mopeds", including the very powerful Aero 50[if you are sensing that I have a very low opinion of motorcycle dealerships, you are right ]. So it wasn't long before there was an army of scooters wizzing around at 40 mph without plates. I left Az in '93, but I think the reason Tucson went from very liberial moped code to one of the most restrictive in the nation, is a backlash to this.
Yes, Sprees were light, when the plastic was removed, it was shocking how little stucture there is. Kind of like giving a poodle a bath.
The area where the single speed trans was most appearent was off the line. Most riders developed what I called the Spree weave. Leaning first to one side and ten the other, cutting sm.partial circles in an effort to build momentum. I doubt that it added to the acceleration, but it was something to do while waiting to get to speed[26 mph for the Iowa, 31 for the standard].

 

[johnrobholmes]

The weaving does actually help going up hills before the power band hits. On the flat I can't see it doing anything though.

 

[gogo]

The "Iowa" model! How ironic, because the Iowa dealership in my city was selling the PA50 Honda Express model that went 30 MPH. This and the low price of the Hondas was a slam dunk on my father's moped shop selling 25 MPH European mopeds.

Luckily for me the 25 MPH European mopeds were detuned 30-35 MPH models and were easily retuned back up. I was king of the hill at age 14 with my 45 MPH 2 speed Tomos.

"The shove" was common practice to get mopeds going back then.

 

[motomech]

But they didn't use a true CVT, instead of a frt. variator, they just had a fixed pulley

I dont understand ?
how can that thing have one variable (diameter) pulley and one fixed pulley with no tensioner to take up the slack ?
Also, if the front ( crankshaft ?) pulley is a fixed unit, why does it have those big flanges like a variable pulley. ?
Just puzzled !

The rear, or driven pulley is spring loaded and the two halves "squeeze" against the belt, forcing it outward, and it that way tensions the belt.
The belt itself, doesn't move up and down the frt. pulley like a true CVT. You can see the mark left on the face of the soft alum. pulley.

One thing to understand about these systems is that it's the sides of the belt that are the "working" surfaces. The belt is a V configuration and the angle is matched to the angle of the pulley faces.

Eventually, CVT or a simple belt drive like the Spree, the belt wears a "step" into the sheave[or pulley half]faces. It doesn't really effect the performance of the system, but can get to a point that it starts to wear the sides of the belt.

CVT or single-speed drive, the clutch can be seen on the outside of the driven pulley. It's a simple centrifuagal affair with three weighted fingers that swing outward and work against a bell housing[not shown in photos].

 

[Hillhater]

The rear, or driven pulley is spring loaded and the two halves "squeeze" against the belt, forcing it outward, and it that way tensions the belt.
The belt itself, doesn't move up and down the frt. pulley like a true CVT. You can see the mark left on the face of the soft alum. pulley..

Motomech, I understand how a belt CV works, but how can that belt be "forced outwards" on the driven pulley without being "forced inwards " on the front pulley. ??
The belt is a fixed length, cannot stretch, and there is no tensioner to compensate for belt length ?

 

[gwhy!1]

can I join in
The belt gets forced to the outer of the front pulley by the weights ( that are on ramps ) that is behind the inner halve of the front pulley these force the inner halve of the front pulley to move outwards this in turn forces the belt to ride up the front pulley as the rpm rises this in turn pulls the belt into the rear pulley which has a big torque spring that normally squeezes the two halves of the rear pulley together. The belt remains the same size so there is no fixed pulley on the picture ( cvt ) shown above if either of the pulley's were fixed then no different gear ratio's will happen then this will not be a cvt

 

[Hillhater]

gwhy,.. which pulley are you refering to as "front" ? .. the smaller one ?
your description is correct for a normal CV/variator drive, ..... but motomech is saying that the smaller (left hand) pulley is fixed size and the belt does not move up or down this pulley ???
Like you , i dont believe that could be a variable ratio drive. That smaller pulley looks very much like a variable ratio pulley ( with parts removed )

 

[Dauntless]

If you count Leonardo da Vinci's sketches you could say the Italians came up with the CVT first. But the first patents and the original seagoing uses were all German. The British got around to putting it in motorcycles The U.S. was a late comer to them. There are SO MANY types of transmissions lumped into a main group called Continously Variable Transmissions, the cone drive and hydrastatic drive that makes some riding mowers turn on a dime are CVT's. So if you wanted to explore the possibilities, you'd be busy. They've been tried on so many things, at times they are a success. Such limited potential overall, but I just think an electric motor in some varieties of CVT's could be great.

I knew I should have posted the video I saw last night of the GY6 conversion, putting the electric motor right where the gasser used to hook up. Can't find it now, but it sure looked like it turned out well.

 

[gwhy!1]

gwhy,.. which pulley are you refering to as "front" ? .. the smaller one ?
your description is correct for a normal CV/variator drive, ..... but motomech is saying that the smaller (left hand) pulley is fixed size and the belt does not move up or down this pulley ???
Like you , i dont believe that could be a variable ratio drive. That smaller pulley looks very much like a variable ratio pulley ( with parts removed )

Yes the smaller one ( front ), I agree with you in the picture this is a variator with the outer halve of it removed.

 

[motomech]

If you count Leonardo da Vinci's sketches you could say the Italians came up with the CVT first. But the first patents and the original seagoing uses were all German. The British got around to putting it in motorcycles The U.S. was a late comer to them. There are SO MANY types of transmissions lumped into a main group called Continously Variable Transmissions, the cone drive and hydrastatic drive that makes some riding mowers turn on a dime are CVT's. So if you wanted to explore the possibilities, you'd be busy. They've been tried on so many things, at times they are a success. Such limited potential overall, but I just think an electric motor in some varieties of CVT's could be great.

I knew I should have posted the video I saw last night of the GY6 conversion, putting the electric motor right where the gasser used to hook up. Can't find it now, but it sure looked like it turned out well.

From this site,

http://www.yamaha-motor.eu/designcafe/en/about-bikes/scooters/index.aspx?view=article&id=441402&segment=&archive=false&page=2

QUOTE- "Among the most interesting large capacity scooters ever built, there is the American Salsbury from 1947. This 320 cc announcing 6 HP at 2680 rpm runs very smoothly (slightly) above 100 kmh. Before invading Italy, scooters had indeed a tremendous success in USA in the late 30's and we owe to Salsbury the invention of the automatic belt pulley drive which equips these M85 models as well as all modern scooters."- END OF QUOTE

The type of CVT being discussed here, belt driven with movable sheaves, is a direct decendent of the Salsibury scooter/

 

[motomech]

double post, internet in Nicaragua

 

[Dauntless]

Yeah, from that site I'm trying to make out if we can SEE anything valuable in the pic. (Looks more like a chainsaw.) The problem is all this great old technology that we're more likely to be able to replicate at home than we are the more complicated modern stuff is just so hard to learn anything about. Can you make out anything from this picture on the 1910 Zenith GraduaGear?

Easy to find pictures of the whole bike, what's inside can remain a mystery. The 1912 Rudge Multigear pulley system is another that might be great if you can decipher it. Rudge Wentworth has been high on my list to replicate if I built a boardtrack racer, as it had pedals, as well as the slogan "Rudge it, do not trudge it."

Since there's been so few 4 wheelers with CVT, but SO MANY 2 wheelers that worked well, I think there's something to be assumed from that. Just another interesting page, though.

http://what-when-how.com/automobile/continuously-variable-transmission-cvt-automobile/

 

[Dauntless]

Ah, WN.com has a playlist on CVT's. Without having sat and watched all the videos, I'll fly on faith and say there's going to be some great info there or good ole' WN woudn't have bothered.

http://wn.com/continuously-variable_transmission