bike/scooter cvt
 

I don't know if this has been posted before, but currently on one of the scooter boards, this is being discussed. I don't understand why a CVT is generally thought to be inefficient for mpg purposes on bikes and scooters, but if it is, could this possibly be a solution? I'm thinking a manually operated solenoid instead of a lever to facilitate changes. I think something like this could help anyone doing a foot forward design. Any thoughts?

http://www.pattakon.com/pattakonPatBox.htm



PatBox CVT (Continuously Variable Transmission) - Suzuki Burgman Forum

Belt drives such as V belts are slightly less efficient mechanically than cut gears or steel roller chain. This is not to say they are not cost efficient in many applications especially at constant speeds. The purpose of any transmission is to best match the engine output to load demands. Cost of manufacture and reliability are always important considerations. Typical load demands tend to follow an exponential curve and gear ratio spacing usually is specified to match.

The only advantage I see in that design is that you could manually modify the ratio. But the disadvantages are that now you have even more drag resistance. And when you pull on the lever to force the higher ratio it will cause a lot more drag. The CVT transmission is inefficient because it has a lot of drag, and slippage.

If you want to get the most out of a cvt, I highly recommend you get some flat weights, and new springs. I took a Chinese Tao Tao 50CC scooter and changed the cylinder out to 100cc. With the coil, cylinder head, pistion and cylinder, and 27 carb I was producing enough power to get that little scooter to about 68 mph, even with my 225lb butt on it. Of course thanks to the CVT, I ripped through a dozen v belts until I found the right spring, variator, weight combo.

Now you've got me thinking about my little scooter project, I only put about 30 miles on it after my mods, now it sits in the corner of my garage.

Sendler did a fair amount of scooter mods here on an older thread. MotoIQ.com did an extensive project on a Honda Ruckus. Even if they don't slip belts are more elastic than metal.

The one thing a CVT has going for it is the "constant". Trying to force the belt manually isn't going to help efficiency or performance. They are already load sensitive so you can set it up to cruise at low speeds at the torque peak and yet, rev up to the power peak when you pin the throttle for max accel.
.
The problems which reduce efficiency versus a chain and trans are belt losses from bending as it goes around the pulleys. And compression as it goes between the pulleys. Also CVT's don't have quite as much range as a 6 speed trans so the highway speed cruise rpm is topped out in ratio and well above the torque peak unless you give up accel from launch with longer final gearing.
.
Also for hypermilers, on a CVT there is no way to PulseNGlide as you would be able to on a bike with a hand clutch.
.
But there are some CVT scooters doing some good numbers and the CVT takes the driver gear selection out of the equation. Everyone on a Honda PCX can get a really high fuel efficiency of 94 mpgUS depending on how aero you are on the bike and the average speed. A geared bike like the Honda CBR250R has more variation since many traditionally schooled sport bike riders choose to ride around at 45 mph all the time in a gear that is near the power peak so they can have max accel without shifting to "get out of trouble" and end up with 60 mpgUS. Whereas I can take the same bike and putt around at 45 mph/ 4,000 rpm in competition using pulse and glide with engine off coasting and get 135 mpgUS.
.
.
https://scontent-lga3-1.xx.fbcdn.net...da&oe=574451B0
.
.

Apparently the idea is go to heavier rollers or sliders, for maximum top end/ mileage and by varying the preload, be able to have infinite "gearing"(for acceleration) up to the maximum that the heavier rollers would provide. Similar to having say, a six speed with out having the clutch and additional weight (and drag)of the gears. As it is now, scooters are stuck with whatever compromises goes with the weight of the rollers/sliders. They can be changed- lighter weight= quicker acceleration, less speed, more noise, higher rpm, or heavier rollers/sliders= slower acceleration ,higher speed at lower rpm, for better mileage.
It shouldn't increase drag by a significant amount because it isn't powering or driving any thing. It is only applying pressure, at varying degrees, to the drive belt to alter it's position. At least that's the way I see it. Besides, I thought the whole idea of the CVT was to keep the vehicle its most efficient mode- or is it different for scooters. I don't see where a belt driven pulley system is less efficient than one that is gear driven, but I'm new to this.
See post#62 maybe that explains it better than I can.
PatBox CVT (Continuously Variable Transmission) - Page 7 - Suzuki Burgman Forum

We will have to get some real world data to find out if adding another belt that is sliding along the first belt, and rolling over two more rollers, gains more from rpm control than it loses in friction.
.
If you really want to shift, a manual trans and a chain drive will be more efficient but not as simple or maintenance free as a standard CVT.
.
It looks like Suzuki implements their selectable ratio trim on the factory system for the Burgman 650 by tightening or loosening the tension of the contra spring with motor driven gears. Which eliminates the extra belt and pulleys.
.
GLOBAL COMMUNICATION MAGAZINE | Global Suzuki
.

Suzuki changed its CVT with the 2013 model to be a bit more efficient. So now there's a patbox for the 'fly by wire' set up.
http://www.pattakon.com/pattakonPatBox.htm
I was only thinking of this as it might be easier to package for a recumbent streamliner than dealing with a clutch and forward controls. It looks like it would be a way to "fine tune" a cvt for better economy.

I invented an idea a long time ago on scootdawg.proboards of a center based bearing that will press on one side of the center of the variator, to squeeze it together.

The other side of the bearing should have a spring, connected to a cable-based lever (kind of like a choke).

The advantages are:
- installable on nearly all scooters,
- Only needs small amount of space,
- low friction,
- No excessive modifications. Same belt can be used.
- The choke system can help you to choose between performance or eco riding styles.

Cons are:
- Can't accelerate fast in Eco mode, unless the choke-knob is changed to performance.
- Not easy to adjust while riding. (imagine riding a bike with 2 throttles, one the real throttle, the other, a manually steered variable gearing, that also will affect acceleration and speed).

A chain drive is not more efficient than a belt
at least not modern micro v belts that can be as good as 98%
And some of the big cruiser synchronous belts can handle all the power you can give one tire

But this is not just a cogged belt drive. It is a v belt in a cvt which is much less efficient than a chain.

Yes but you see the constant RPM of the motor and easy power are on the other end of the of the scale
also that is a cogged drive belt in a cvt
but it's not a synchronous belt drive

There are lots of losses in a rubber scooter cvt. Lots of heat. Requires it's own fan. Nothing cogged about it. The v belt is gripped by pressure from the contra spring and weights on the surfaces of it's sides.

It is a cogged belt Google one look at the picture
And it is only griped on one side both the contra spring and the variator float on the belt and provide no traction
like a fat kid sitting on a cape

The belt you are talking about is a synchronous belt
A cogged belt is a v belt with notches or cogs to allow more flexibility and less heat

http://www.jpcycles.com/product/603-...5xgaAsvY8P8HAQ
https://www.vbeltguys.com/products/a...wsQaAkKR8P8HAQ

Whatever. The cuts in the belt of a scooter cvt don't engage anything. They are only there to help reduce the bending losses as it goes around the pulley. The only transfer of power come from the pulley halves squeezing the sides of the belt. There are compression losses in the rubber and a certain percentage of slipping.

That's true but there are things that can be done
The cooler you keep the belt in the first place the less
heat it produces. It slips very little if clean and the contra cams spring and veriator face are in good shape

Chain on the other hand has vibration problems
also has slipping even if it's moved of to the clutch
and does not self adjust for stretch
Also has compression loses to the rubber bumpers in the wheel
It will last only about as long as I can get one of my belts to
Belt start at about $5 and a good quality belt is about $35
A cheap chain for a little bike starts at about $20 if you want a good one
Well... It's not uncommon to spend $150
And it heavy
Pros and cons on either drive

FE wise i owned a honda xr 125 (geared bike) and now own a Chinese gy6 125 scooter and getting the same mileage. Around 90mpg. Scooter is definitely more nippy through traffic but on tje open road the honda was about 15mph faster. That being said i did own a 150cc gy6 that's consumption was about 50 to 60mpg

50 to 60 mpg is a very poor number for a 150 gy6 something is wrong with it. The worst I ever got with a 150 was 75 and that was in an over sized frame zipping around at 55 mph

My current gy6 150 scooter is doing 120 mpg on average and and up to 70mph of course I don't get the same benefits at that speed as I do at 45

Why not a hand-shifter similar to the one which was used in the early Vespa?


Scooters seem to not just get a narrower RPM band, but also not to compensate it with a wider gear spread which would render them more suitable to occasional road commuting.

Dipping into one of my text books, Mechanical Engineering Design 5th Ed. Shigley and Mischke, we find V belt efficiency ranges from 70 to 96%. A notched V belt is different from timing belt system which rivals a roller chain with efficiency of 97-99%.

If that's true, that's really awesome for the notched V-belt. The advantages sure would outweigh the roller chains.

It is likely cheaper to manufacture a sheave system than to cut a bunch of gear teeth.

If you ask me (even though you didn't), the way to go would be to have a CVT for getting up to speed, and then be able to switch it in to a fixed gear once up to speed.

I also assume the reason belt driven CVTs are only 80% efficient is slippage, which is the main reason friction belts are less efficient than chains, and why a timing belt is not. The other losses come from flexing/compressing/decompressing the rubber.

A dual belt system with a V belt over a timing belt and a timing sprocket at the drive and the base of the sheave. Hmmm

i believe the most common problems with the gy6 CVT are
the constant movement of the belt
(the clutch is on the wrong end)
and gearbox final drive
( each gear interaction cost a few percent efficiency)

I would move the clutch to the drive pulley like a comet.
A giant 17 inch serpentine pulley on the back wheel and a 2 inch
pulley on a real swing arm axle jackshaft
it would weigh less handle bumps better
and be more effective

I have a bit larger bike I have tried using a flat cog style belt to compare to a sealed roller chain mind you this is final drive not in the transmission. This set up lost a average of 1.25 to 3 mpg on a 83 550 Yamaha Maxim. I tried thinking the weight reduction would help however it didn't. I put back on my chain after trying several adjustment. I even went as far taking drive pulley off and machining as much mass off as i thought remotely safe to do. http://www.cyclingnews.com/features/...ve-whos-faster found this link when i was trying to figure why.

I had read the study a number of years ago

a chain starts off with fantastic numbers at lower speeds and power levels
but the friction builds fast as power goes up

with a conventional motorcycle belt you start with crummy numbers
from the needed high pre tension
but it never really gets much worse even if you pour on the power

yeah that's why I prefer a microv belt, with as little as 10 pounds it becomes self locking
and it still has the self leveling friction curve but has far better power consumption at low power levels
this is what makes it a favorite with car designers

The advantages I see with belt drive is that it won't require the sprockets to get also replaced like it's done everytime a chain-driven motorcycle has the chain replaced, plus it won't require lube.

One of the things that I don't miss about "real" Motorcycles. :)

Gil

Apart from Harleys and similar models, and also the BMW G 650, I can't remind of any other belt-driven motorcycle.

BTW what are the other things you don't miss about them?

Here are my questions about the CVT:

Why can't a disconnect clutch be installed between the engine and drive pulley to allow slide and glide riding?

Shouldn't a properly designed CVT start applying power at the torque peak and then transition when the engine is at peak HP, constantly keeping engine RPM in that sweet spot?

almost right
you do want the clutch to open at peak torque
but as the engine is not accelerating you want to stay at peak torque
until the CVT is maxed out
at that point the engine will begin accelerating and HP will be the preferred metric

hope that makes sense

That could be more useful while starting the engine.

Actually, I'm not sure where I saw it, but someone was proposing to put a Sprag clutch on the drive wheel of a scooter. If that were done then P&G would not be a problem, but then one would lose the engine braking effect when decelerating... That would mean you would be much more dependent on the brakes to slow you down. Most likely one would be replacing brake pads and possibly brake discs much more often than they would otherwise.

Let me list them, the lack of a fuel gauge ( low fuel lights don't count), the handlebar vibrations, the uncomfortable seat ( coming from a Yamaha V-star 250), the carburetor, shifting gears ( fun but tedious), also the noise ( they are just too loud), just to list a few. I like that I can jump on my scoot and go. It will start right up whether it's cold or hot. Riding a "real" motorcycle is like driving a race car for the daily commute.

Over the years I have realized that every rider has his own idea of what an ideal motorcycle is. Or they haven't really given scooters a chance, maybe not manly enough.


Gil

It also bothers me a lot.


I used to be more favorable to carburettors. Actually still like them because it's easier to perform some emergency makeshift fix, but I recognise an EFI may be better for basically any other aspect.


I'm more concerned about the clutch than the gear-shifting itself. I got quite surprised when I found out about some American companies that make clutch automation kits for nearly every motorcycle.


The way some folks ride them, I won't even argue with you on that matter. Well, most of the people who get them for commutting back here do so in order to save not just on fuel but also the time they don't waste stuck in traffic congestions...


I'm not even sure about what would be an "ideal" motorcycle for me, but I like the Yamaha Ténéré 250 that is available here in Brazil. I just hope Yamaha fits them with ABS instead of phasing it out in the next year...


I've already considered getting either a scooter or a regional derivative of the Honda Cub, but since they're more optimized for urban commutting it wouldn't be so suitable for me (considering that I would use a motorcycle mostly for medium to long distance trips, plus there is a lot of pot-holes in our streets and it would be quite costly to replace tyres and rims after falling on them).

Looks like we have about the same views on riding. I'm sure that any type of motorcycle is better for the commute in your city. You are right about the possible wheel damage with bad roads. Funny that you mention that, not long ago a member on the PCX forum shared that in Thailand they swap in 13 inch wheels onto their Honda PCX, so that they can fit fatter tires with more side wall, that way it can handle the bad roads. The Honda PCX comes with 14 inch wheels from the factory.

Gil

Possibly.


Something like this would be nearly perfect.
https://scontent-sea1-1.cdninstagram...02647040_n.jpg
http://1.bp.blogspot.com/-fEHQBi-N6O...onda+nx200.jpg


I also usually prefer a smaller rim and tyres with a higher sidewall.

This is not really correct.
A chain starts out with little friction, and does increase in friction the higher your speeds go.
But the same is also true for a CVT belt.
It starts out with higher friction, but then it'll switch to where the belt occupies both the driven and driving pulley on the same diameter.
This is where the belt is performing at it's peak efficiency, because it's bending diameter is the largest on both ends.
When you accelerate further the driving pulley's grip will have a larger radius, and the driven pulley will have a smaller radius, at which most of the heat transfer happens in the driven pulley.
And one of the ways you know a chain is more efficient than a belt, is because a chain will always be cooler than a belt.
A chain also rotates slower than a belt, and is longer, but even if not, it just doesn't have the slippage a belt has. And the higher the speed, the higher the forces, the higher the heat dissipation, and the higher the heat, the more the belt will slip (a quality of rubber, unfortunately).


Anyway, it just comes down to a belt has a lower low speed efficiency than a chain, medium or comparable mid speed efficiency, and a MUCH lower, high speed efficiency than a chain.
And the higher the speed, the more a CVT with rubber belt loses out to a chain.

Modern cvts don't have belts, but also are based on chains, or metal belts that have metal parts as belt (not rubber). They not only are more efficient, but more durable as well.
They're also a lot more costly (about 10-15x more costly).