44 replies to this topic

[SteveCanyon]

An interesting new type of CVT, but it has no friction surfaces, cone, etc, to vary the drive output.



It seems to work quite well and not a huge amount of work to get a large enough unit into a car.

[cheapracer]

A good analogy is it comes down to simply being able to mechanically precisely control the slippage speed of a planetary band/drum in an auto transmission.

The bad news is he's spent 20 years to invent the existing system principle that drives the Toyota Prius.

http://www.wind.sann...index_i18n.html

[SteveCanyon]

A bit hard to see from that link, but it does seem to have similarities.

[gruntguru]

A good analogy is it comes down to simply being able to mechanically precisely control the slippage speed of a planetary band/drum in an auto transmission.

The bad news is he's spent 20 years to invent the existing system principle that drives the Toyota Prius.

If this is what it seems to be, it is nothing like the Prius. I haven't quite got my head around it yet, but I think the control shaft can be rotated with essentially zero torque and therefore zero power. The control shaft is acting on two epicyclic gearsets at opposite ends of the gearbox and (I think) the torques applied to the control shaft at each end are equal and opposite. Hence the claim that a very small electric motor driving the control shaft can control much larger torque and power levels flowing through the gearbox.

In contrast, the Priius transmission is just a 3 legged gearbox like a differential, so that input and output can be shared ie any of the three legs (IC engine, electric motor, vehicle drivetrain) can be an input and two of them (electric motor, or vehicle drive) can be used as outputs.

If this CVT is what it appears to be, and it is possible to package it to suit automotive use, it has unearthed something of a holy grail. Pity is it may be too late since it may not be needed in hybrid applications.

Edited by gruntguru, 15 May 2010 - 09:30.

[cheapracer]

No all this guy is doing is controlling the speed of the planetary drum and sun gears albeit that he requires a secondary drive system to do it - thats what the Prius already does in principle using the electric motor as the secondary drive slowing up until the point the petrol motor can go on it's own.

Yes it is clever but will be interesting to see what Toyota says about it.


Anyway I like the very clever but simple NuVinci drive that has been around for a few years now, I guess they are selling so many for bicycles they are to busy for cars....

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

[MatsNorway]

No all this guy is doing is controlling the speed of the planetary drum and sun gears albeit that he requires a secondary drive system to do it - thats what the Prius already does in principle using the electric motor as the secondary drive slowing up until the point the petrol motor can go on it's own.

Yes it is clever but will be interesting to see what Toyota says about it.


Anyway I like the very clever but simple NuVinci drive that has been around for a few years now, I guess they are selling so many for bicycles they are to busy for cars....

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

Cool looking thing that nuVinci system but i believe it would get issues when you give it some propper toques.

Edited by MatsNorway, 15 May 2010 - 11:59.

[gruntguru]

No all this guy is doing is controlling the speed of the planetary drum and sun gears albeit that he requires a secondary drive system to do it - thats what the Prius already does in principle using the electric motor as the secondary drive slowing up until the point the petrol motor can go on it's own.

Yes it is clever but will be interesting to see what Toyota says about it.

If that is all he's doing, its not clever at all and the implementation is very inelegant. I think you are probably underestimating the device.

Edit. If that's all he is doing, there is nothing there Toyota can lay claim to.

Edited by gruntguru, 15 May 2010 - 12:04.

[mariner]

I do not see what is fundementally different between the Fallbrook transmision (variable ball contact point angle) and a Perbury (variable disc angle) CVT. I may be missing something but the Perbury use a disc which is one segment of a theoretical ball between the discs and the Fallbrook seems to use a ball with only one contact track with between the discs.

Am I missing something about the Fallbrook device?

[mariner]

Two posts in a row!

Having re-run the youtube clip of the D-Drive I am struck by the similarities to the Fell railway locomotive system tested way back in the 1950's. Again output speed control by having more than one motor putting power into the gear system but no actual meshing ratio changes.

http://www.paxmanhis....uk/paxfell.htm

Also for those of you who have a copy of Peter Wright's great book "Formula 1 technology " try page 88 which shows the Lotus tank steering differential. There are again some similarities to the D drive system.

[gruntguru]

Mariner, you can't have watched the video. Perbury and Fallbrook are both friction drive whereas the D-Drive is 100% gear coupled. The Fell system is just another set of differentials (plus some fluid couplings and clutches, both points of energy loss) like the Prius. AFAIK no other transmission can match the fundamantal claims of the D-Drive, which are:
1. True CVT
2. 100% gear drive - no points of slip or potential slip.
3. Single input shaft and single output shaft. The control-shaft input consumes nominal power only - it is not an alternative power input which is being balanced against the main power input as in Prius or Fell.

[Grumbles]

No all this guy is doing is controlling the speed of the planetary drum and sun gears albeit that he requires a secondary drive system to do it - thats what the Prius already does in principle using the electric motor as the secondary drive slowing up until the point the petrol motor can go on it's own.

I'm not sure that it's that simple cheapy, I was under the impression that the control shaft was somehow torque balanced.

If it really does what it he says it will I'd love to see old mate succeed. In an area full of hype and scammers making inflated claims it's refreshing to come across what seems to be a genuine and honest bloke having a go..

[mariner]

Mariner, you can't have watched the video. Perbury and Fallbrook are both friction drive whereas the D-Drive is 100% gear coupled. The Fell system is just another set of differentials (plus some fluid couplings and clutches, both points of energy loss) like the Prius. AFAIK no other transmission can match the fundamantal claims of the D-Drive, which are:
1. True CVT
2. 100% gear drive - no points of slip or potential slip.
3. Single input shaft and single output shaft. The control-shaft input consumes nominal power only - it is not an alternative power input which is being balanced against the main power input as in Prius or Fell.

I think I was trying to make two seperate points. Firstly that the Fallbrook looked rather like the Perbury and secondly that the D Drive had some similarities to the Fell system. I still think there are similarities to the Fell system as an epicyclic gear train is also a differential mechanism until you lock the ring gear. In fact some cars ( like the Olds Toronardo) have used an epicyclic train as the diff.

I think the question is how you can put only a nomimal control force into the D Drive gearset when there is no obvious anti feedback mechanism. It never seems to lock up a gearset so that cannot be a means of applying a torque reaction so logically I would have thought that the torque at the control gear has to vary as the main input torque or the ouput resistance varies.

Maybe the test of the D Drive would be to study what happens when it goes on the overrun as the car slows down. What does the control motor do in such circumstances, with what required torque and is it the exact opposite of being under power?

[arttidesco]

Fascinating stuff but a word to the brave be wary of driving a CVT when in reverse, many years ago in London I got stuck behind a dust cart late on my way to work and decided to back down the street from which I had just come hence, I was estimated to be doing nearly 30 mph in reverse in a CVT Volvo 343 when I lost control and demolished a gate post :-)

Fortunately no one was hurt and only my pride, the gate post along with the tailgate and bumper of the company car needed repairing :-)

Edited by arttidesco, 16 May 2010 - 22:26.

[pugfan]

Cool looking thing that nuVinci system but i believe it would get issues when you give it some propper toques.

If I stand on 175mm cranks with my 90kg then that's 154Nm of torque and that's only passively letting my weight fall.
I've had cars with less torque than that. Admittedly cars tend to have bigger reduction ratios.

[gruntguru]

I still think there are similarities to the Fell system as an epicyclic gear train is also a differential mechanism until you lock the ring gear. In fact some cars ( like the Olds Toronardo) have used an epicyclic train as the diff.

Agreed and My previous posts use the terms "differential" and "epicyclic gearset" interchangeably.

I think the question is how you can put only a nomimal control force into the D Drive gearset when there is no obvious anti feedback mechanism. It never seems to lock up a gearset so that cannot be a means of applying a torque reaction so logically I would have thought that the torque at the control gear has to vary as the main input torque or the ouput resistance varies.

I think the giveaway here is that the D-drive has three (or is it four) epicyclic gearsets when it would only require one to split two inputs in the manner of the Prius or the Fell. The obvious answer is the two reaction torques to balance the control shaft, are generated by the complex arrangement of epicyclic geartrains.

Because the system comprises fixed, constant mesh gears throughout, the control torque (if greater than zero) would have to be a constant percentage of the input torque. Using the same logic, the control torque would also have to be a constant percentage of the output torque. However as the gear ratio is varied, the ratio of output and input torque will not remain constant so there is an obvious paradox. Does the paradox disappear for the case where net control torque equals zero?

EDIT. I am starting to think this paradox is fatal ie the D-drive may be no more than an over-complicated epicyclic transmission. Apologies to Cheapy and Mariner if that proves to be the case.

Maybe the test of the D Drive would be to study what happens when it goes on the overrun as the car slows down. What does the control motor do in such circumstances, with what required torque and is it the exact opposite of being under power?

I think it would have to be according to the rules of fixed ratio gearsets - the same rules I applied to arrive at the above paradox.

Edited by gruntguru, 17 May 2010 - 02:54.

[Greg Locock]

Turns out that " the engineering report is quite clear on the fact that the 'control' motor needs to be just as powerful as the 'input' motor: "Our designation of 'Input' and 'Control' shafts in this report is arbitrary in that both would conventionally be used to provide power. There is no inherent character of the mechanism that requires the input to be the dominant power-providing element. The torque provided by the control shaft will typically be of the same magnitude as the torque provided by the Input shaft... the Input and Control should be considered as parallel power paths rather than as 'power ' and 'control' elements respectively.""

There is no inherent torque balance that magically reduces the torques. (which given that epicylics are just levers is no surprise)

So, it is very similar to the Prius gearbox, which in itself was based on an architecture proposed in the 70s, which was based on control type gearboxes dating back to WW2, if not earlier.

Note that the control input has to be infinitely variable in its own right.

http://infinitelyvar...sion-Report.pdf

[cheapracer]

Cool looking thing that nuVinci system but i believe it would get issues when you give it some propper toques.

Says the man who will then go out and drive his car to the shops using his FRICTION PLATE CLUTCH.

It has absolutely everything to do with surface area, if it's got enough balls it can take on anything (hmmm, where have I heard that before?).

I'm not sure that it's that simple cheapy, I was under the impression that the control shaft was somehow torque balanced.

If it really does what it he says it will I'd love to see old mate succeed. In an area full of hype and scammers making inflated claims it's refreshing to come across what seems to be a genuine and honest bloke having a go..

Well the inventor himself explains that the shafts and their input speed are the key to the system.

I think its a great idea, don't get me wrong, simplicity is the key to nearly all successful anything and he has simplicity covered - I honestly believe that he could be challenged on it from Toyota if he produces it.

I might ask him nicely if he might make a note here.

[MatsNorway]

Says the man who will then go out and drive his car to the shops using his FRICTION PLATE CLUTCH.

Strange thing we haven`t seen it on the road then.

[cheapracer]

Strange thing we haven`t seen it on the road then.

Hmm now lets see, grab proven reliable clutch kits and transmissions from Asia for $200 a pop

or

invest hundreds of millions of dollars in a complete new drive system that the public may not accept, go through Brand trust damaging early model dramas while paying design royalties along the way.......

Take a public poll about new technology and you will get an overwheming positive result - put it in the showroom and ask people to actually buy it and see what happens......

[MatsNorway]

hmmmm yea i think all the big brands just dropped all development and interest in new technology

Audi had or has a CVT. If that ball transmission was superior, Audi would have bought the company or the patent or leased it, etc.

The NuVinci system was around in 2004 giving the interested companies a lot of time to talk to them etc.


We run something called balldiffs in the rc cars, the diff is 33mm wide the rims are 53. the diff contains 12 carbide balls with a dim of 2.4mm.

it need special grease and routinely maintenance. And it needs to be run in before it works perfect. based upon that I think a ball diff on a street car would be pretty heavy, costly and unreliable. Same goes for a gearbox i believe.

Someone should do some math on the contact patch and friction needed to transfer some 150Nm at say 3500rpm.

Edited by MatsNorway, 18 May 2010 - 18:26.

[gruntguru]

I think its a great idea, don't get me wrong, simplicity is the key to nearly all successful anything and he has simplicity covered - I honestly believe that he could be challenged on it from Toyota if he produces it.

I don't think Toyota hold the patent on epicyclic gearing. After reading the engineering report on the D-drive, there doesn't seem to be anything patentable there either. (Just because the patent office issues a patent doesn't mean something is patentable either )

[Greg Locock]

Yes the current Audi A6 has a CVT. I hated it. In small cars I think CVTs are fine, I find them amusing. In performance cars they sound wrong.

[cheapracer]

hmmmm yea i think all the big brands just dropped all development and interest in new technology

Audi had or has a CVT. If that ball transmission was superior, Audi would have bought the company or the patent or leased it, etc.

The NuVinci system was around in 2004 giving the interested companies a lot of time to talk to them etc.

The profitable companies just get on with boring old "what we know". The cost of putting even minor additions into a car are just staggering and risky. They are companies designed to put money into shareholders pockets not to satisfy your quest for technical improvements.

Don't believe that car companies don't investigate every possibility out there but in the end the Accountants have the final say.

Nissan have had CVT's in bigger cars for some time now and it's a realistic production prospect because the tech has been used for 50 years in other forms of mass production and comes down to the use of 4 simple tapered rotors and a belt.

From Wiki...
CVTs should be distinguished from Power Sharing Transmissions (PSTs), as used in newer hybrids, such as the Toyota Prius, Highlander and Camry, the Nissan Altima, and newer-model Ford Escape Hybrid SUVs. CVT technology uses only one input from a prime mover, and delivers variable output speeds and torque; whereas PST technology uses two prime mover inputs, and varies the ratio of their contributions to output speed and power. These transmissions are fundamentally different. However the Honda Insight hybrid, the Nissan Tiida (only the SL model), Nissan Cube, Rogue, Altima, Murano and Maxima use CVT .

I have mentioned before, one minor hobby of mine is to scour through thousands of inventions at freepatentsonline.com - there is some stunningly brilliant ideas out there for cars that will save the whales from cancer, give us 400 mpg running on old cigarette butts and sexually satisfy me while driving but they will never see production.

[cheapracer]

Of interest

No More Gears
A Timeline of CVT Innovation
1490 - da Vinci sketches a stepless continuously variable transmission
1886 - first toroidal CVT patent filed
1935 - Adiel Dodge receives U.S. patent for toroidal CVT
1939 - fully automatic transmission based on planetary gear system introduced
1958 - Daf (of The Netherlands) produces a CVT in a car
1989 - Subaru Justy GL is the first U.S.-sold production automobile to offer a CVT
2002 - Saturn Vue with a CVT debuts; first Saturn to offer CVT technology
2004 - Ford begins offering a CVT

[cheapracer]

Oh and guess what I just found, the NuVinci principle is actually in production for the largest of the Nissans, the Cedric with 400Nm of torque ........

http://www.histomobi...a/tech/90-2.asp

Edited by cheapracer, 19 May 2010 - 01:55.

[Canuck]

Yay - snowmobile transmissions on road cars. I changed my fair share of belts as a kid. They can be dialled in to launch really hard - a thing of beauty with a studded track.

[Grumbles]

What I don't get is this: if the d-drive is really nothing more than an elaborate differential (and the engineering report would seem to confirm this) then what is the purpose of the other two epicyclic gearsets? Smoke and mirrors, or just an honest and naive mistake by the inventor? It's a bit hard to believe they went to the trouble of making the plastic model without making even the most rudimentary torque measurements.

[SteveCanyon]

According to the technical paper that Greg linking to, yes it could be made less complex.

[Greg Locock]

What I don't get is this: if the d-drive is really nothing more than an elaborate differential (and the engineering report would seem to confirm this) then what is the purpose of the other two epicyclic gearsets? ... It's a bit hard to believe they went to the trouble of making the plastic model without making even the most rudimentary torque measurements.

Well, I could be really cynical and suggest that perhaps a plumber didn't think about such things? After all, he let the video interviewer get away with some fairly ludicrous statements, that he must have known were misleading. OK, fair enough, very few journos let you review their material before they release, I've been burned there.

[Tony Matthews]

Well, I could be really cynical and suggest that perhaps a plumber didn't think about such things?

Harsh, but amusing!

[NeilR]

I assume that wasn't me Greg?

Edited by NeilR, 19 May 2010 - 11:22.

[threep]

I watched the video and instantly thought "Toyota Prius" too. I hope he hasn't sunk all his savings into this device, I'm not sure his patent would stand up to scrutiny.

[mariner]

One racing connection with CVT is that DAF ran an F3 Brabham back in the 60's fitted with one of their "rubber belt" CVT's.

I also seem to recall that Williams tried a CVT on an F1 car at some time.

This is only from vague memory but I think there is a basic difference between the older "rubber belt" CVT's ( and the snowmoble ones ) and the newer steel belt versions in that the rubber ones pull the belt to transmit torque whereas the steel ones can also push the links to do the transmitting.

[Greg Locock]

I assume that wasn't me Greg?

Correct. idiot motoring journo from The Age, and idiot of unknown 'profession' from Geelong Addy. Both agreed to let me check their articles before publication. Neither did. Those sound fellows from The Australian (John Mellor) did run their article past us and it became our standard technical publicity handout. Win win.

[McGuire]

I NEVER allow sources to review my copy in advance. ESPECIALLY engineers. It's a nightmare, like inviting a bystander to be in your movie: Five minutes later they want to direct. Ask a source to check your copy and suddenly he is a grammar and style authority. In the case of engineers, they can't understand that the story is not necessarily geared toward readers in the autistic spectrum.

The practice is also unethical -- imagine a politician asking a reporter if he can proof a story in advance. Don't even ask a reputable, professional journalist. He will take it as an insult. If an aspect of a story is beyond my technical expertise, I'll go to a knowledgeable and objective third party, not to the propeller-head who developed the gadget. That's just common sense.

[McGuire]

Correct. idiot motoring journo from The Age, and idiot of unknown 'profession' from Geelong Addy. Both agreed to let me check their articles before publication. Neither did. Those sound fellows from The Australian (John Mellor) did run their article past us and it became our standard technical publicity handout. Win win.

No matter how good the product or how much I like it, I would be embarrassed to have my material recycled verbatim as a company PR handout. I'm a journalist, not a shill. I don't want my stories to read like something the company's communications dept. could have written, thus sparing them the trouble. Every day of my life people at auto companies, INCLUDING engineers, tell me that is up is down, east is west, and water flows uphill. It's my job to listen to that stuff and nod my head at regular intervals. Goes with the territory. However, it is not my job to run to my laptop and regurgitate it.

[Greg Locock]

Then they shouldn't have agreed to do let me check their articles. They did, therefore they broke their word.

Since you don't have the faintest idea what the Mellor article was about, or what publication it was for you also haven't got the faintest clue whether it was appropriate or innapropriate to recycle it as a technical handout. FWIW it was a full page on our solar car with lots of cutaway drawings and specs. Because we checked it the numbers and specs were right.

[NeilR]

McGuire I might agree with your view on journalistic independence to some extent, but it does depend on the context and situation to a degree. I have let engineers look at the article I was to publish, for their own satisfaction. If corrections were suggested then I would consider each on it's merit as to whether it was improving the story or correcting the facts within the article. I have had the situation where demands were made for certain material to be included and my response was to dump the whole article on the spot.
However I have had contact with engineers in five countries now and it is admittedly a small sample. I have found that the UK/Aust/German engineers I have dealt with so far are more concerned that the facts are true and correct i.e. there are no factual errors. The engineers from the US and Canada are much more promotion oriented and have a view of providing information with a specific slant towards marketing.

[Canuck]

Ha. HA!! No offense to any of the journos (nor am I commenting on any of their specific work) here but I don't think I've ever read an article that called a turd a turd. While you may not sing the praises of an inferior product, I've not yet seen any journo anywhere (withthe exception of some online writers and of course Top Gear) trash a trashable product. Instead it's a warm but positive review followed by full page and sometimes 2-page adds by the reviewed outfit.

Honesty in periodicals isn't.

[SteveCanyon]

Ha. HA!! No offense to any of the journos (nor am I commenting on any of their specific work) here but I don't think I've ever read an article that called a turd a turd. While you may not sing the praises of an inferior product, I've not yet seen any journo anywhere (withthe exception of some online writers and of course Top Gear) trash a trashable product. Instead it's a warm but positive review followed by full page and sometimes 2-page adds by the reviewed outfit.

Honesty in periodicals isn't.

You need to read some Aussie articles then. We have more than enough journo's that'll tell things as they are.

[McGuire]

Ha. HA!! No offense to any of the journos (nor am I commenting on any of their specific work) here but I don't think I've ever read an article that called a turd a turd. While you may not sing the praises of an inferior product, I've not yet seen any journo anywhere (withthe exception of some online writers and of course Top Gear) trash a trashable product. Instead it's a warm but positive review followed by full page and sometimes 2-page adds by the reviewed outfit.

Honesty in periodicals isn't.

Here in my part of the wrench-turning end of the media where we deal with parts for the most part (as opposed to the "clean-hands" mags that primarily review cars) we mainly ignore shitty and/or useless products. They get no mention either way. As a practical matter, that also tends to piss them off so they don't advertise in the magazine either. Works for me.

It's rather like something I did many years ago, reviewing books for a literary journal. We mostly reviewed books worth reading. Books not worth reading were generally not worth a review, unless there was something especially noteworthy about their awfulness. There are more than enough good books to talk about, and people read literary journals to find good books. That worked for me, too, as I would much rather be reading and writing about good books than bad ones, which would be pretty depressing for both me and the journal readers.

Now, much like automotive performance parts, whether a book is any good depends a great deal on the end user. I approached every book this way: who is its reader? If the book has readers who will derive genuine benefit or entertainment from it, that's a book worth talking about. And in the review, that's who you are talking to. There is no point in slagging Stephen King week after week to people who read Camus and David Foster Wallace. They're not his target audience. King is not fine literature but while it's not my genre, he is a good writer and the best *at what he does.* Likewise, lots of products are not of F1 quality but do have a legitimate use and true value for amateur enthusiasts. We speak to that. Others are just junk for any purpose so why give them the ink. Our readers don't buy or care about that stuff anyway. Same kinda thing.

You can generate some cheap buzz and fake controversy via the easy expose route but at the end of the day, that's not why people read the magazine. We trust our readers to have a basic degree of automotive knowledge. Top Gear doesn't grab my interest all that much because they don't. They'll take the easy shot for the cheap laugh every single time, without fail. Gets kinda boring after awhile. Some episodes that's all it is. It's interesting how in general, the more people tend to know about cars, the less they tend to like the show.

Edited by McGuire, 22 May 2010 - 02:10.

[cheapracer]

. It's interesting how in general, the more people tend to know about cars, the less they tend to like the show.

Totally depends if you take it on as a motoring show or as simple entertainment.

[Tony Matthews]

Totally depends if you take it on as a motoring show or as simple entertainment.

We've been here before, on another thread. Top Gear is 'bloke' entertainment involving cars, not a car program. James May seems to know a bit about engineering in general, how much the automobile features in that I don't know. The other two probably only know 0-62 times. I don't care, it has given me a few laughs between the yawns. It looks better than it is (to some!) because the competition is generally dire. 'Britain/America's Got Talent'? Please...

My experience is that car salespeople know very little about cars, but why should they?

[McGuire]

Totally depends if you take it on as a motoring show or as simple entertainment.

I'm sure you're right. I'm probably not a very good judge of any television.

[Canuck]

In all fairness McGuire, your mag is just about the only one I read these days and I'm
afraid it's sporadic at best. I have managed a couple of issues of RET and while interesting I can't help
but be skeptical of any pub where the articles are about (or by) the advertisers. That would be my inherently distrustful nature in large part I suspect. My commercial experience with magazines however coincides with my earlier statement. 2 Canadian pubs and a two in the Middle East both operated in the "we'd love to write about your business...hey, have you seen our advertising rates?" mode.

I've always been impressed that the TG crew were unafraid to out a car's (or manufacturer's) performance even when it was a car that one of them owned (Clarkson's Ford GT for example). I don't think North American viewers get that kind of blunt information very often. James Mays' drive in the Astin "gentleman's racecar", sweating to death in his underwear while Hammond and Clarkson snickered in air-conditioned comfort - exaggerated for comedic effect I'm sure but drove home the point that a lack of air con had significant drawbacks in the real world. Of course in the real world I'm not lined up to purchase any of the more desirable cars they drive either.