Pulley Question

[jwc4806]

Can a comet type spring loaded driven pulley be driven by changing the belt tension with an idler ? Will the driven pulley change rations as the belt tension increases ?

 

[jwc4806]

Drive Set Up

Please see if layout in attachment would work.

 

[Tpdingo]

I don't see any specific problem with it besides how are you going to keep it in any one ratio. It would take a **** of a lot of force so it would probably have to be driven through a worm drive and a large stepper motor and some electronics...or some complex *** mechanical system for shifting. Remember that the TAV can actively provide lots of force through the engine's power.

And for what? Manual shifting? What are you gaining? The TAV system already does a fine job at selecting the required ratio given the needed circumstances. And it can be yours for 52$, not 520$

 

[jwc4806]

Skid Steering

Just pondering a skid steering mechanism. Slack belts for neutral.

 

[Kartorbust]

I wouldn't want to do that to the CVT belt. You'll probably end up burning that out fast and often. What kind of skid steering do you want exactly? One where applying pressure to the belt and it acts like a brake to turn? Or one where you can turn counter to the other (run backwards) to turn?

What are you trying to build or looking into building? You could look for a Comet 100 differential instead of doing a belt set up. All you would have to do is, add the 2 brakes on each axle half like you would normally, and run 2 separate brake levers or pedals.

 

[itsid]

as Kartorburst said.. that will certainly shred your belt very quickly..

the idler should run on the "outer" the reinforced side to prevent chewing it up for no reason..
then again you run into belt length issues quickly..
or more precisely you are prone to touch the reverse running belt (which is a big nono )

And as Tpdingo said you'd need to lock the lever in place at a set position somehow otherwise it'd be hard to hand on to it for very long..
and worst: you must not set it to a fixed position

A TC pulley works because of two variable pulelys...
and that's NOTHING to do with belt length I'm afraid..

The driver wants to shift up as soon as it's clutch is engaged...
it constantly tries to pull the belt into a larger diameter on itself (and thus a lower diameter on the driven)
The driven is the the "governor" part of this equation..
if it sees little to no counteracting force on the chain it allows the driver to force the belt in some lower orbit (shifting up, spreading it's sheaves)
but as soon as there is a fair resistance on the wheels (and chain) it works against the driver and collapses it's sheaves as hard as it can until it reaches an equilibrium state.
[that's the main reason for your belt wear in a setup like this]

let's asssumey you set your levers to the intended ratio,
now you go downhill, all of a sudden the wheels have much less resistance, the pulley spreads, the belt slacks and in fact is likely to jump off before you can even adjust the lever
to reapply tension.

And the other way around is almost worse...
Nooow you say you could rig up a spring (or rubber band) loaded system on the lever to
balance the idler tension to exactly the needed equilibrium, so the lever shifts up when it can and down when needed..

True but then all you did is build a TC setup with a reduced amount of variance,
by removing one of the variable pulleys.

So, let me suggest some two stage layout instead:

First a 1:1 fixed pulley layout from the engine to a jackshaft.
(lever actuated clutch if you like, but I guess a manual clutch causes less powerloss..
anyways....
ON that jackshaft (spinning at the same rate as the engine now) you have the TC driver (powering the driven just as the fixed pulley you wanted)

Now in order to disengage the TC completely you can disengage the jackshaft,
all variation however is still left to the TC and you can only switch it on or off basically

While I like the idea of the diff (Peerless 100 btw )...
the problem is the overall speed is fixed
and if one wheel is stopped the other one will turn twice as fast as it would at the same engine rpm when both turn.
Viable but awkward to control I bet
Also you'd wear out brake pads quicker than you'd want...

'sid

 

[Tpdingo]

as Kartorburst said.. that will certainly shred your belt very quickly..

the idler should run on the "outer" the reinforced side to prevent chewing it up for no reason..
then again you run into belt length issues quickly..
or more precisely you are prone to touch the reverse running belt (which is a big nono )

And as Tpdingo said you'd need to lock the lever in place at a set position somehow otherwise it'd be hard to hand on to it for very long..
and worst: you must not set it to a fixed position

A TC pulley works because of two variable pulelys...
and that's NOTHING to do with belt length I'm afraid..

The driver wants to shift up as soon as it's clutch is engaged...
it constantly tries to pull the belt into a larger diameter on itself (and thus a lower diameter on the driven)
The driven is the the "governor" part of this equation..
if it sees little to no counteracting force on the chain it allows the driver to force the belt in some lower orbit (shifting up, spreading it's sheaves)
but as soon as there is a fair resistance on the wheels (and chain) it works against the driver and collapses it's sheaves as hard as it can until it reaches an equilibrium state.
[that's the main reason for your belt wear in a setup like this]

let's asssumey you set your levers to the intended ratio,
now you go downhill, all of a sudden the wheels have much less resistance, the pulley spreads, the belt slacks and in fact is likely to jump off before you can even adjust the lever
to reapply tension.

And the other way around is almost worse...
Nooow you say you could rig up a spring (or rubber band) loaded system on the lever to
balance the idler tension to exactly the needed equilibrium, so the lever shifts up when it can and down when needed..

True but then all you did is build a TC setup with a reduced amount of variance,
by removing one of the variable pulleys.

So, let me suggest some two stage layout instead:

First a 1:1 fixed pulley layout from the engine to a jackshaft.
(lever actuated clutch if you like, but I guess a manual clutch causes less powerloss..
anyways....
ON that jackshaft (spinning at the same rate as the engine now) you have the TC driver (powering the driven just as the fixed pulley you wanted)

Now in order to disengage the TC completely you can disengage the jackshaft,
all variation however is still left to the TC and you can only switch it on or off basically

While I like the idea of the diff (Peerless 100 btw )...
the problem is the overall speed is fixed
and if one wheel is stopped the other one will turn twice as fast as it would at the same engine rpm when both turn.
Viable but awkward to control I bet
Also you'd wear out brake pads quicker than you'd want...

'sid

Well it does have something to do with belt length. As the belt gets tension, it has no where to go but into the shaves. But still that would probably lead to crazy amounts of uneven wear and what not.

How about a TC and a diff. There, variable speed and skid steer. Also the possibility of using like a 125cc 3 speed motor or something with an integrated shiftable transmission.

 

[Snaker]

A little thin on details but I'm thinking something like:

Spin a jackshaft off the engine.

Put two CVT's on the jackshaft, a left side and a right side.
Toss the drive pulley's rpm sensing flyweights and replace with a lever operated linkage with a throw-out bearing to control the movable sheave. Something similar to a automotive type manual clutch.

Remove the Torque Sensing function in the driven pulley. That would be getting rid of the angle ramps and making so the driven movable sheave movement is linear instead of spiral.

That would achieve a decent skid steering without usual problems of diff braking, the two biggest problems which are the double speeding of the "free" side and quickly wearing out those parts that are't designed for that.

 

[Tpdingo]

A little thin on details but I'm thinking something like:

Spin a jackshaft off the engine.

Put two CVT's on the jackshaft, a left side and a right side.
Toss the drive pulley's rpm sensing flyweights and replace with a lever operated linkage with a throw-out bearing to control the movable sheave. Something similar to a automotive type manual clutch.

Remove the Torque Sensing function in the driven pulley. That would be getting rid of the angle ramps and making so the driven movable sheave movement is linear instead of spiral.

That would achieve a decent skid steering without usual problems of diff braking, the two biggest problems which are the double speeding of the "free" side and quickly wearing out those parts that are't designed for that.

The torque sensing I think its a bit of marketing BS. I mean the plastic ramps mostly just allow for easier shifting and to make the spring's job a bit easier and to get the springs advantage of going in and out and rotating, requiring and less heavy spring. All the "torque sensing" is just the engine speed slowing down when under a real heavy load, the RPMs drop, the weights come in, driver expands, driven closes and bam- it downshifts.

 

[itsid]

Well it does have something to do with belt length. As the belt gets tension, it has no where to go but into the shaves. But still that would probably lead to crazy amounts of uneven wear and what not.

How about a TC and a diff. There, variable speed and skid steer. Also the possibility of using like a 125cc 3 speed motor or something with an integrated shiftable transmission.

misinterpretable wording perhaps..
yes the diameter change request is 'transmitted' due to the invariance of a given belt length for any given setup..
therefor it's of course an important factor.

BUT for a given center to center distance a shorter or longer belt would NOT just mean a differently timed shifting operation.
Both would instead result in a non working setup
therefor belt length and shifting (time and or ratio) are not related
that's wat I meant with the above

'sid

 

[Snaker]

The torque sensing I think its a bit of marketing BS. I mean the plastic ramps mostly just allow for easier shifting and to make the spring's job a bit easier and to get the springs advantage of going in and out and rotating, requiring and less heavy spring. All the "torque sensing" is just the engine speed slowing down when under a real heavy load, the RPMs drop, the weights come in, driver expands, driven closes and bam- it downshifts.

I disagree with that. Early CVT's didn't have Torque Sensing (TS). TS is a add-on feature and is a vast improvement.

Without TS there is no way to manually downshift during heavy loading and the engine gets bogged down. That does end up causing a downshift due to the lost rpm, but that's not a good thing.

When TS detects heavy loading it forces a downshift (backshift) and the engine stays at desired rpm.

Upshifting is also helped by TS resisting or holding back the up shift during heavy loading.

I like to look at it with a bit of generalizing that the basic CVT tuning handles lite or normal loading and TS handles heavy or abnormal loading.

 

[anickode]

Look at the MTD Varidrive system they used on riding mowers. Basically what you're describing, using V belts and tensioners, along with a special pulley.