Continous variable transmission???

Jasper Janssen writes:

Not a true CVT as they assert since the blades would span integral
quantities of links, but an interesting approach to hyper selection
ratios:

http://www.andersoncvt.com/

It is a true CVT. There is no mechanism that forces the chain links
and the cones into any sort of engagement with the chain. It is, in
fact, pretty close to how automotive CVTs are constructed. The
first DAF CVTs used a belt of some kind on the cones, and torque and
power was limited by the belt. Modern automotive CVTs on the DAF
principle use chains like this one does.

From what i see in the picture is that the cones have slats with a
ridge in their center that engages notches in the chain. These are
shaped to settle into position somewhere in the rotation so that only
one slat is synchronous. As far as I can see, this will not work and
it is not continuous, the chain making digital adjustments to the ribs
on the cone.

The DAF CVT uses friction between the faces of V-pulleys to drive its
chain that does not operate as a chain with engagement features but
rather as a steel V-belt.

Jobst Brandt

On Fri, 21 Oct 2005 12:48:17 +0100, "Tomek Li"
wrote:

Over the years I got used to think that such a transmission would be no more
than another incarnation of perpetuum mobile. But recently I learned that
some quite serious people got on the task, with interesting results.

The links below I copy from uk.rec.cycling

See:

http://www.fallbrooktech.com/NuVinci.asp

http://www.evworld.com/view.cfm?sect...le&storyid=841

http://www.californiabicycleracing.o...interbike.html

So, what do you think?

For a hub as fat as a Rohloff (both in diameter and likely in weight),
they're awfully sparse in their hard information about actual gear
ranges, real torque capacity, and actual cost of manufacture. It's
just a prototype at this point, really. Based on what I see, though,
I wouldn't invest any of my own funds in that company. I don't see
any reason to believe that the product will succeed in this
marketplace; it's not filling a need that isn't already well-served by
mature, inexpensive and reliable technologies.

--
Typoes are a feature, not a bug.
Some gardening required to reply via email.
Words processed in a facility that contains nuts.

On Wed, 26 Oct 2005 22:40:33 GMT, wrote:
Jasper Janssen writes:

Not a true CVT as they assert since the blades would span integral
quantities of links, but an interesting approach to hyper selection
ratios:

http://www.andersoncvt.com/

It is a true CVT. There is no mechanism that forces the chain links
and the cones into any sort of engagement with the chain. It is, in
fact, pretty close to how automotive CVTs are constructed. The
first DAF CVTs used a belt of some kind on the cones, and torque and
power was limited by the belt. Modern automotive CVTs on the DAF
principle use chains like this one does.

From what i see in the picture is that the cones have slats with a
ridge in their center that engages notches in the chain. These are
shaped to settle into position somewhere in the rotation so that only
one slat is synchronous. As far as I can see, this will not work and
it is not continuous, the chain making digital adjustments to the ribs
on the cone.

Yes, looking at it more closely that might be happening. I'm not entirely
convinced it will actually work -- especially at speed and power like that
necessary for automotive application.

It seems to imply, though, that while a bit on each chain link can engage
with those ridges, there is movement in the bit that engages in the
lengthwise direction, and that would seem to me to make engagement analog,
rather than digital -- ie, effective chain pitch can vary. If it can vary
from, say, .45 to .55 and it can then go back to .45 with an extra link
engaged (in other words, the movement is half of the chain pitch in each
direction), you'd get true continuous. Though what the advantage would be
over a simple friction chainbelt eludes me.

The DAF CVT uses friction between the faces of V-pulleys to drive its
chain that does not operate as a chain with engagement features but
rather as a steel V-belt.

Jasper

In article ,
Werehatrack wrote:

http://www.fallbrooktech.com/NuVinci.asp

http://www.evworld.com/view.cfm?sect...le&storyid=841

http://www.californiabicycleracing.o...interbike.html

So, what do you think?

For a hub as fat as a Rohloff (both in diameter and likely in weight),
they're awfully sparse in their hard information about actual gear
ranges, real torque capacity, and actual cost of manufacture. It's
just a prototype at this point, really. Based on what I see, though,
I wouldn't invest any of my own funds in that company. I don't see
any reason to believe that the product will succeed in this
marketplace; it's not filling a need that isn't already well-served by
mature, inexpensive and reliable technologies.

Traditionally, friction drives with metal-metal contact have had some
rather touchy lubrication requirements. As in a small amount of water
in the system would cause massive amounts of damage in a big hurry, and
an oil change would be extremely expensive.
A toroid-style CVT I got to see up close required oil costing right
around $45/qt. And it needed a few gallons. It was a big unit running
a conveyor.
So if they did produce this hub in large quantities they'd probably
get a lot of complaints from people who either don't change the lube
when needed, put in the wrong stuff, or got a seal leak they didn't
discover until total failure.

--
B.B. --I am not a goat! thegoat4 at airmail dot net

Tomek Li wrote:
Over the years I got used to think that
such a transmission would be no more
than another incarnation of perpetuum
mobile. But recently I learned that
some quite serious people got on the
task, with interesting results.

The links below I copy from uk.rec.cycling

See:

http://www.fallbrooktech.com/NuVinci.asp

http://www.evworld.com/view.cfm?sect...le&storyid=841

http://www.californiabicycleracing.o...interbike.html

So, what do you think?

I like single speed.

Nikolas Ojala
Tampere, Finland

--
http://www.iki.fi/no

Static truth is dead truth and only
dead truth can be held as a theory.

peterlip wrote:

It'll be interesting to see how efficient a CVT can be. From memory, the
current gear with derailer configuration is around 95-97% efficient, ie
only 3-5% of your power is lost.

The test data published in Wilson's _Bicycling Science_ show derailleur
drivetrain efficiencies from the high 80s to the high 90s, depending on
choice of gears and cleanliness of components. Even the clean, new
example displayed pretty close to 90% efficiency in some gears.

It seems that many folks develop an exaggerated impression of a
derailleur drivetrain's efficiency, simply because that's the most
prevalent design. I think it's prevalence has more to do with the fact
that it's considerably cheaper to manufacture than other kinds of
bicycle transmissions.

Internal gear hubs do not impose an automatic efficiency penalty in all
circumstances, but any kind of CVT probably would.

Chalo Colina

Tomek Li writes:

Over the years I got used to think that such a transmission would be
no more than another incarnation of perpetuo mobile. But recently I
learned that some quite serious people got on the task, with
interesting results.

The links below are copied from uk.rec.cycling

http://www.fallbrooktech.com/NuVinci.asp

http://www.evworld.com/view.cfm?sect...le&storyid=841

http://www.californiabicycleracing.o...interbike.html

So, what do you think?

This is a friction drive that's been around for a long time without
finding a good home. Its spherical idler has a bad mechanical
advantage at both ends of its limited ratio range, besides being
heavier than most chain or planetary transmissions. I guess they
tossed in the towel, so to speak, when they found no positive
engagement gearing to accomplish the impossible. Naming it after the
village in Tuscany where the great Leonardo lived doesn't make it new.

http://www.sheldonbrown.com/brandt/continuous.html

Jobst Brandt

Does your 80-90% figure come from the 3rd edition? Page 300 of the second
edition says "The efficiencies of present transmissions using chains and
derailleurs or hub gears are in the high nineties..."
--
Carl

"Chalo" wrote in message
oups.com...
peterlip wrote:

It'll be interesting to see how efficient a CVT can be. From memory,
the
current gear with derailer configuration is around 95-97% efficient, ie
only 3-5% of your power is lost.

The test data published in Wilson's _Bicycling Science_ show derailleur
drivetrain efficiencies from the high 80s to the high 90s, depending on
choice of gears and cleanliness of components. Even the clean, new
example displayed pretty close to 90% efficiency in some gears.

It seems that many folks develop an exaggerated impression of a
derailleur drivetrain's efficiency, simply because that's the most
prevalent design. I think it's prevalence has more to do with the fact
that it's considerably cheaper to manufacture than other kinds of
bicycle transmissions.

Internal gear hubs do not impose an automatic efficiency penalty in all
circumstances, but any kind of CVT probably would.

Chalo Colina