very cool and interesting invention.
here's article that goes with that video http://www.gizmag.com/steve-durnin-ddrive-d-drive-infinitely-variable-transmission-geared/15088/

*engineering *wheels

Adding video to channels (Engineering, Wheels) - requested by demon_ix.

Sweet sift. I wish I could replace my 020 trans with one of those. =]

During the interview it's mentioned several times that this should be more efficient than a conventional CVT. Conventional CVTs are already more efficient than automatic or even standard transmissions as they IC engine can be run at its most efficient RPM given the load, and even run at a more efficient cycle. I look forward to seeing it implemented in consumer automobiles, some time in the 2030's.

Well, my brain does hurt after watching this video and trying to wrap my head around how this the thing works, but only a little.

Hmmm. Much as I'd like it to be true I'm still not convinced the extra energy needed to control the control shaft is going to make it better than the systems already in use.

I'd like to be surprised though.

Their use of "neutral" for the zero ratio is a little misleading (assuming I'm understanding the mechanics involved correctly). "Neutral" tends to imply that the output is disengaged from the input, but in this case the zero ratio is actually holding the output stationary (ie- you can't freely turn the output arm by hand).

For a true neutral the transmission output would still need to be disengaged from the wheels. It makes me wonder what happens if someone is coasting downhill at highway speeds and then puts the vehicle back into gear without touching the gas pedal. Current CVTs can at least slip if the torque gets too high.

Edit: A ratcheting system would probably solve this issue, for forward motion at least, but would make the next part more difficult.

Also, since I live in the mountains, does anyone know if current CVT vehicles offer any way to engine brake? Using the standard brakes on multi-mile continuous downhills can cause problems, but is the driver generally given a way to "downshift" a CVT?

I'm not sure what I'm seeing, but I like it.

Wait, so if the output encounters any resistance is it not going to transfer any torque to the control shaft? I don't fully understand how it works, but I'd like to see him give that output a little bit of resistance and see what happens.

>> ^robbersdog49:

Hmmm. Much as I'd like it to be true I'm still not convinced the extra energy needed to control the control shaft is going to make it better than the systems already in use.
I'd like to be surprised though.


There isn't extra energy... it isn't like the second motor needs to drive any power... The second motor only has to spin gears. The only work the second motor does is drive the elliptical gears which are in bearings and overcome friction of those teeth and moving the mass of the gears.

Imagine this scenario, a motorized skate board that is powered by a little engine. On top of that you have a deck with a worm tooth gear that moves the deck forward or backwards. So while you are moving forward at a constant speed the second motor moves the deck backwards (so you are moving slightly slower, like walking to the back of the bus while in motion).... It's like that but inside the gears so the second motor spinning will reverse the apparent motion. All the real power is still on the first motor.

The real problem seems to be you would need a mechanical neutral as your secondary motor would have to exactly match the drive rpm (or whatever ratio) in order to not move.. then you slow the secondary motor to increase the output shaft speed.

Another real benefit is that the drive motor can run at any rpm and you just match it with the secondary to stay still.. So you could go 5mph in idle or 5mph at top engine speed. The drive motor rpm - secondary motor rpm = motion and you can just pick whatever rpm delivers the torque you want.

very exciting stuff!

It's a bit late for this. By the time they go through testing, prototyping and get something production worthy, battery technology will improve and we'll all be driving electric vehicles that can go 1000's of miles on a charge, and can be recharged in 5 minutes.

Going to slap a big *downunder on this one.

Invocations (downunder) cannot be called by Asmo because Asmo is not privileged - sorry.

Very interesting.
One potential problem to look at would be if the secondary shaft is run by an electric motor a failure in that motor would result in a Toyota like, uncontrolled "full-speed ahead" mode.

http://dansdata.blogsome.com/2010/05/16/todays-mechanical-conundrum/

^Boise_Lib:
One potential problem to look at would be if the secondary shaft is run by an electric motor a failure in that motor would result in a Toyota like, uncontrolled "full-speed ahead" mode.

All that would do is make it seem like you were stuck in your longest gear, and you could still take your foot off the accelerator. That assumes the sun gear gets "stuck" due to the failure. If it lost all resistance then the vehicle would go into effective neutral.

Having said that, I do wonder how much torque is being applied to the sun gear (the shaft that controls the gear ratio). Something has to keep the resistance on the output gear from turning the sun gear (forcibly lowering the gear ratio). If the torque on the sun gear is somehow negligible then in theory it should also be possible to run that gear in reverse, further lengthening the upper gear ratio beyond what would be possible with a fully stationary sun gear.

>> ^xxovercastxx:

http://dansdata.blogsome.com/2010/05/16/todays-mechanical-conundrum/
wohoo! Dan always does nice analysis of such things.

>> ^Jinx:

Wait, so if the output encounters any resistance is it not going to transfer any torque to the control shaft? I don't fully understand how it works, but I'd like to see him give that output a little bit of resistance and see what happens.


In its current form, he'd likely break the teeth off one of those plastic gears. It would need to be reproduced in a metallic form before that kind of thing could be attempted safely.

I can imagine something like this jamming up easily at high speeds though. Because of its reliance on gears and shafts, there's no real way to compensate for materials warping and shifting under temperature variance. This could be a very nasty repair bill when your transmission explodes at highway speeds.

>> ^Jinx:

Wait, so if the output encounters any resistance is it not going to transfer any torque to the control shaft? I don't fully understand how it works, but I'd like to see him give that output a little bit of resistance and see what happens.


This is my point. Yes it would transfer torque to the control shaft. The more power you put through it, the more power you'll need for the control shaft. There will be a power cost to run at a speed where the control shaft isnt locked. This is why I think joedirt is wrong.

A quick thought experiment joedirt: Imagine this system hooked up to a 2000hp diesel unit. If there was no extra power input needed then the little electric motor he has here would work fine. Do you really think it would?

There will be friction to overcome, and the gearing need for such a large power plant would generate a lot of friction. There would also be, I believe, a torque transfer to the control shaft that would be far, far from negligible.

It will certainly need a good chunk of power to control it. This doesn't mean it won't be better than the current systems, but it certainly isn't the case that there will be no power drain from this transmission.

^ That was my thought. The "control shaft" would have to handle as much torque as the drive shaft to prevent the drive shaft from reversing the control shaft. But that's just at first glance. I'd love to see the results of the powered tests when they get to that stage.

Inventor Steve Durnin is from *dowunder

http://www.gizmag.com/d-drive-redux/15120/

^ they go through an engineering report on the device. key points being the "control" motor would have to be as powerful as the "drive" motor (in terms of torque), and control motor would also need it's own CVT, which introduces a friction component into the system.

The whole idea to have gears in the first place is to convert between torque and speed. The second motor in this gizmo not only takes away the speed of the output, it removes the torque as well.

This discussion does bring up a few good points..
http://dansdata.blogsome.com/2010/05/16/todays-mechanical-conundrum/

He is correct about the fact that the output shaft can not spin faster rpm then the input shaft. But hypothetically you fix gear some ratio that makes highway driving efficient, but that could be a major problem.

Also, I realized that even though the output shaft can be reversed, it occurs to me that the action of reversing the output is driven by rotating the sun gears in reverse, which I would think means the secondary motor is then applying the torque, and probably means in wouldn't be good in a car.

Of course that doesn't seem right because if you are working to 90% of the input speed (secondard motor is moving at 10% of inputs rpm) does that mean that you lose some torque..



Also I think the comment about tank tracked system is interesting. You could then put the full motor output on both shafts like a tank does on the tracks.. But you shift the power from primary to secondary and eliminates the electric motor issue. ie. neutral is equal power (at any rpm) to both inputs. And full reverse is all on A and full forward is all on B.

>> ^Psychologic:

^Boise_Lib:
One potential problem to look at would be if the secondary shaft is run by an electric motor a failure in that motor would result in a Toyota like, uncontrolled "full-speed ahead" mode.
All that would do is make it seem like you were stuck in your longest gear, and you could still take your foot off the accelerator. That assumes the sun gear gets "stuck" due to the failure. If it lost all resistance then the vehicle would go into effective neutral.
Having said that, I do wonder how much torque is being applied to the sun gear (the shaft that controls the gear ratio). Something has to keep the resistance on the output gear from turning the sun gear (forcibly lowering the gear ratio). If the torque on the sun gear is somehow negligible then in theory it should also be possible to run that gear in reverse, further lengthening the upper gear ratio beyond what would be possible with a fully stationary sun gear.


You're right, of course. Thanks Psych

>> ^Payback:

The whole idea to have gears in the first place is to convert between torque and speed. The second motor in this gizmo not only takes away the speed of the output, it removes the torque as well.


I was noticing that too. It doesn't seem like a system that would be useful for heavy vehicles or fast accelerations. The shortest gear would have the least amount of power.

It would also seem to reduce efficiency in the lower gear ratios, unless they're somehow harvesting the power lost through the control gear.

Im not yet convinced that the planet gear doesnt cause lots of imbalance and thus make this transmission useless on high-rpm applications.
However, if it really works, I want one made for my car! And the funny thing is, those made custom for older cars wouldnt even cost that much, since its an extremely simple construction, unlike conventional transmissions.
Custom made sequential gearboxes for $75,000? Forget it. This one would wipe the floor with em and would only cost a fraction!

I was thinking about this last night. This is a crappy transmission. Output is only ever 1:1 or less. It is, however, a pretty decent clutch. You can disengage the engine before shifting a regular manual gearbox thus saving wear or torque loss on friction plates.

I upvote because I now understand a practical use for it. As a full transmission, it isn't practical.

*doublepromote... not your everyday clutchless variable transmission...

Double-Promoting this video back to the front page; last published Saturday, May 15th, 2010 1:10pm PDT - doublepromote requested by MycroftHomlz.

Tags for this video have been changed from 'steve, durnin, gizmag, gear, Transmission' to 'steve durnin, gizmag, gear, Transmission, d drive' - edited by kronosposeidon

*downunder

Adding video to channels (Downunder) - requested by kronosposeidon.

Awfully similar to how a Prius's transmission works.

I suspect he won't be able to patent it once Toyota hears about it.