Top 100: The most desirable cars of all time

D-Drive redux: about that holy grail thing...


May 17, 2010

Steve Durnin's D-Drive - re-evaluated.

Steve Durnin's D-Drive - re-evaluated.

Every now and again, astute Gizmag readers come to the fore to keep us on our toes - and never has this been better demonstrated than with last Friday's D-Drive Infinitely Variable Transmission article. More than 40 comments and e-mails have flooded in over the weekend questioning the D-Drive's capabilities as a true IVT, and its potential efficiencies. Furthermore, an engineering report was made available on the D-Drive website that flat-out negates some of the key claims that were made in our interview video. So let's take another look at this device in the harsh light of engineering scrutiny.

Steve Durnin's D-Drive gearbox has spurred a lot of interest since it first came to public attention on the Australian ABC's New Inventors show earlier this year. But despite its winning the weekly invention challenge, the device was explained in only the vaguest of terms, giving technically minded folk little from which to draw any proper conclusions.

I contacted Steve Durnin and arranged a meeting, during which we filmed the device from all angles and discussed its potential at length. I then put together a video story, touting the D-Drive's potential as an Infinitely Variable Transmission, talking about its potential efficiency advantages, and explaining how it works from my own (non-engineer's) point of view.

Steve, meanwhile, was working to put together a website for the device - which he made available to me just before the Gizmag article went live - and which contained a link to an engineering report that presented the D-Drive in much more sober terms.

What we got wrong

Firstly, the D-Drive as pictured in our video is not a complete infinitely variable transmission system. At best, according to the engineering report, it is a cheap, innovative and potentially very useful primary component of an IVT.

The key problem here is that the D-Drive's control shaft needs to be driven at variable speeds in order to effect the final ratio - so effectively, you need a variable drive motor attached to the D-Drive before it actually works. e3k's engineering report goes so far as to say the control shaft could foreseeably be driven through an external CVT, using a clutch - which of course introduces not one, but two friction components to the system.

One of the main advantages we spoke of in our original article was that the D-Drive got around the need for friction components and transmitted all power through gear teeth. Effectively, if mated with a CVT, the D-Drive outsources the friction components to the transmission of its control engine.

The next thing we failed to pick up on, but that several commenters have pointed out, is what happens when you run the D-Drive under load. With the control motor running to decide on the final gear ratio, the input motor's power would be transmitted to the wheels, where it would meet resistance under load. That load would then be passed back through the gears to the weakest point - which would be the engine driving the control shaft, if it wasn't up to the job.

So it's not possible to run the control shafts using a small electric motor as we said in the video - in fact, 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."

What you got right

You can certainly rely on Gizmag readers to think through an issue like this one - some of the technical discussion in the comments section of the original article - as well as the discussion threads on Slashdot, Reddit and elsewhere - was fascinating.

The D-Drive does indeed operate as an epicyclic gearset. It does indeed operate in a similar way to the Hybrid Synergy drive on the Toyota Prius - and this is a matter of some pride to Steve Durnin, who designed it with none of Toyota's considerable resources behind him.

It does require an external CVT or some other powerful drive component for the control shaft that will not yield to the torque of the primary input motor.

So where does this leave the D-Drive?

More or less where the engineering report concludes. The D-Drive is an innovative component that could be used in the design and manufacture of a true IVT for use in vehicles or anywhere else a variable drive would be required.

It's not a 'holy grail' and there's no rabbits coming out of hats - and what's more, it can't forseeably operate with no friction components between engine(s) and wheels, or at least something like an electric motor that can be smoothly varied in speed.

The D-Drive should, and will, be evaluated on its abilities inside this scope - and as such it may well still become a very valuable piece of intellectual property for its designer.

Thanks for calling us out

So thank you, astute readers and commenters, for calling bunk on this one. We saw an interesting and remarkable piece of emerging technology that hadn't been covered yet in enough detail to generate discussion. We certainly generated discussion, but we got some key information wrong - for that, we apologize.

About the Author
Loz Blain Loz has been one of Gizmag's most versatile contributors since 2007. Joining the team as a motorcycle specialist, he has since covered everything from medical and military technology to aeronautics, music gear and historical artefacts. Since 2010 he's branched out into photography, video and audio production, and he remains the only Gizmag contributor willing to put his name to a sex toy review. A singer by night, he's often on the road with his a cappella band Suade.   All articles by Loz Blain

Im still trying to get my head around this one, and I only had a quick look at the engineering report. But I think Gizmag missed the problem in this update.

As I understand it a brake is all that is needed on the control shaft, not a powerful motor. When the brake is on full the drive is in top gear. When the control is allowed to move freely and there is a load on the output the drive will be in neutral.

The problem that I can see is that the brake on the control creates a source of friction and therefore energy loss.

17th May, 2010 @ 06:03 am PDT

Perhaps a better use of planetary gear sets would be to combine them in series to produce a nine speed transmission and eliminate the need for gear reduction in the drive axle. Combined with a high torque engine - electric or long stroke diesel - you could achieve good performance and a one to one drive ratio with a smaller engine.

17th May, 2010 @ 07:51 am PDT

It can be done with just two moving parts - I did it in 1982.

17th May, 2010 @ 08:22 am PDT

How about instead of using a friction brake, you use a motor operating in reverse as a generator to recover the power? This is starting to sound like a perpetual motion machine, but it's really not (if your "understanding" is correct.)

show me
17th May, 2010 @ 08:32 am PDT

After reading the first article it occurred to me that you'd still need a clutch at least to get started.

If you start a gas motor directly connected to the D Drive, you'll start in gear until some kind of motive force starts turning the other shaft and matches it exactly to the primary shaft's rotation speed. It seemed to me no matter how you arranged it the shaft speeds could not be matched from the initial cranking of the motor to the engine starting.

Hence at least for start up you need a clutch that maybe would only be needed at engine start and engine shut down, but it would be needed.

You could arrange a gear system that drives both shafts equally at start, but now after start up you need a way to disconnect that gear, again you'd need some kind of clutch.

17th May, 2010 @ 09:12 am PDT

If only it had been as stated. Just think, no clutch, no torque convertor, a semi pulling a hill, with just a lever to pull down to keep the r's up on the diesel. Darn it, I watched this video three times before I realized the torque pressure from the load would transmit back to the "control" shaft, (which was mislabled, should have been additional input shaft). My older brother, (80yrs), saw it immediately, and only watched 2/3 of the video to see it. This will work in a hybrid with electric on the "control shaft", and when you pull out on the highway, the infernal combustion is then added.

paublus americanus
17th May, 2010 @ 09:31 am PDT

I think this is a valuable lesson for GIZMAG. And a valuable and unrealized resource. I don't know how many people read this newsletter everyday but it seems to me that you have a very large number of interested and knowledgeable people. Why don't you send out a questionaire every day, in addition to one of your articles at the end? Also, why don't the authors start developing relationships with the best "COMMENTORS"? I mean if one "OF THE CHOSEN" have a suggestion, they could leave a quick voice mail or email or you could even have a video conference!!!

17th May, 2010 @ 09:47 am PDT

HToad's comment bring me another idea. If we need to break the control shaft then we can attach an electric motor to it. Not for driving, just for breaking. We can convert this breaking energy to the main shaft with another electric motor.

We can eliminate most of the energy loss but the system become too complex I guess :(

I hope more ideas will come out. My 2 electric 1gasoline system doesnt look good.

17th May, 2010 @ 10:21 am PDT

These are the very same concerns that came to mind for me upon reading the first article. The conclusion I've come to (upon this article confirming my thinking) is -

With one main drive source/engine two D-drives would be needed, one for the main drive and one for the 'control' drive. The 'control' drive D-drive is then operated by an electric motor. I'm relatively sure that the software is doable.

I sure hope some outfit gets it all going.


Tim Olney
17th May, 2010 @ 10:39 am PDT

The control shafts could be driven by hydraulic motors powered by the same type of pump conventional automatic transmissions use.

Resistance to reverse torque would be simple to control by a variable orifice valve, possibly also with a quick acting one way valve that can be electronically enabled or disabled.

Another possibility, if the control shafts only need to rotate in one direction, would be one way roller clutches, also similar to those used in conventional automatics. They have next to zero resistance in their forward direction but quickly lock when torque is applied the other direction.

Facebook User
17th May, 2010 @ 03:41 pm PDT

ah well, they say pigs can't fly,,,,,,,,, bloody hell, there goes another one !

17th May, 2010 @ 06:44 pm PDT

@ HToad to Facebook User

It's nice to see the interest and ideas you people are expressing in your comments. But I have to regretfully inform you that the proposed ideas to solve the 'friction brake' or 'ratio control' issues efficiently have already been resolved and have been successfully implemented in the Prius Hybrid Synergy Drive. Please see comments from other astute readers and those from myself explaining this on the original article about the dDrive. If you are interested in the way the torque is controlled and some of the benefits of using the Prius system please find a rather lengthy explanation way down that same comments page.

@ froginapot Gizmag

I wholeheartedly agree with your comments on making Gizmag more interactive with its readers. One thing I suggested was to speed up the comments section, I have posted comments and they have taken over 5 hours to show up on Gizmag. I am not sure what is causing this delay and it tends to make any real discussion or information exchange impossible. Comments made do not and cannot 'self-correct' with such long delays in posting. Is it possible that Gizmags web servers are on Pluto given the time it takes for it to go up? ;) Maybe it would be possible to migrate the comments to a forum type environment and from there the authors can tag some reliable contributors? Just an idea.

@ Grunchy if you read this...

Which version of the Ikona transmissions are you referring to? The non-involute profile gears (aka Harmonic Drive) were used on the Lunar Rover in the sixties, but I have not yet found a variable speed/ratio version. Could you post a link to the actual drive please? Thx

17th May, 2010 @ 08:18 pm PDT

why not place a motor on each output shaft of a common differential? each could play as a power/control motor. The output shafts would then be the input shafts. The speed and power of the main shaft, now the output shaft would depend on the direction and speed of the respective input shafts.

Thom Delahunt
18th May, 2010 @ 08:32 am PDT

Well, since it has two input shafts, could a high torque electric motor be used as the second input? Electric motors have huge, instant torque curves...

Making an electric motor stay in "stop" position for full gear input is not that hard. The Segway does this with its two motors to stop itself. So in this regard I do not think a clutch would be needed for the second input shaft.

In this way the system could work with little energy input into the second shaft.

I am also wondering how much energy is needed to get the transmission into the neutral setting. Granted, it has to spin the shaft, but it is not powering a load.....

At that point the question would be whether the energy expended on the motor is justified by the simplicity and efficiency of the design.

As opposed to a six speed automatic, which are becoming common as auto manufactures seek better MPG....

Or am I still missing it?
18th May, 2010 @ 08:40 am PDT

Poor Steve, it looks like you've gone to huge effort to build a beautifully complicated model of something that has already been invented. I don't really care whether this has been used before in jet engines or Toyota hybrids and the other comments here do enough to knock it down on technical points.

The simple fact is that once you realise that the power of the input and control shaft needs to be balanced, then you could achieve the same result using any old differential or planetary gear set.

Keep trying Steve, but don't pump more money into this one!

Jon Pearce
18th May, 2010 @ 08:51 pm PDT

I'm not sure why commenters are suggesting 'work-arounds' for the 'control' shaft (e.g. CVT & clutch to transition to neutral at start-up) when it works against the input shaft anyway. While an stop-start ICE would reduce the 'control' shaft inefficiency there's no avoiding the energy loss when on the move just to maintain the required ratio. Solving the friction problem by meshing gears surely gives the inventor a clue that the assertion that the energy for the 'control' shaft 'doesn't have to be anywhere near the energy that you're putting into the input drive because it doesn't have to work against that power' (Gizmag presenter, but surely from the inventor) is ridiculous.

What astounds me is that an 'inventor' would spend 20 years of his life working on something without doing basic research - even before the internet & Wikipedia made it simple. Why not just bolt an identical electrical motor to the input shaft of his plastic prototype and watch it freeze up? The use of planet gears (the essence of the 'invention') for power-splitting is very old (well before Toyota used it to combine 2 motors), so why would anyone think that one motor can be almost independent of the other?

18th May, 2010 @ 09:41 pm PDT

While I haven't looked at this concept I am inspired to see so many people reading, discussing and thinking about an issue.

Does anyone know of any websites set up to collaborate in this way?

Surely if a problem is presented in a forum-type environment with a lot of smart heads we could pretty much solve every problem humans encounter.

It would also prove an invaluable resource for inventors to prevent them from re-inventing the wheel all the time.

Theoretically patents are meant to act as a repository of information in this way. However, i haven't met many people who find reading patents enjoyable. Perhaps we need smart software engineers and linguistic experts to automate scanning of patents to extract a catalogue of information and then tag appropriate articles of Wikipedia with links to relevant patents. !!

What do you think?

19th May, 2010 @ 02:07 pm PDT

@ Ashlin

This sounds like a decent proposal. I can only see one drawback and that is human nature. People making comments about how good or bad, or how to improve a product or idea, will only continue as long as there is no competition between the parties involved. I think the only way this 'good will' can be fully utilized is if like minded and goal driven individuals contribute without regard for their own compensation. Patents tend to always restrict the idea to an individuals benefit, and this benefit is nearly always financial. I strongly believe that any idea should be the property of all, especially those who can make the most of it in their communities (not just to make themselves rich or powerful).

The world we live in 'likes' to protect the rights of the individual, but if every individuals rights are protected who is going to protect the rights of the community? After all we can not survive alone no matter how many rights we have, and many of the rights we have come at the cost of others rights.

But there is hope for us yet, if we can learn to trust each other and put the needs of others before our own wants.

In the meantime though, a interactive technology website/forum would be the go. Maybe Gizmag or its readers could get something like this going. I notice they've done something similar for the Gulf of Mexico oil leak, but it is in no way interactive.

19th May, 2010 @ 09:59 pm PDT

Two electric motors would work just fine. The electronics to control the motors' relative speed should be pretty straightforward to develop.

24th May, 2010 @ 11:25 am PDT

I dont understand all you people saying to use electric motors with variable speed drives. The electric motor itself is an infinitely variable transmission already!!!!

4th June, 2010 @ 03:04 am PDT

You know what? It occurs to me that both shafts are being powered by the same source. Everybody is getting it all wrong because they think the control shaft has to be powered by another source when in fact they are being powered by the same source. Look at the video. They are rotating off the same drive shaft then power is leached to the second shaft, the control shaft, to SLOW the final drive down. Thus it leaches power only to decelerate in to a new ratio, not accelerate.

1st January, 2011 @ 06:52 pm PST

Resurrecting this again - after recently viewing the video for the first time, I thought there has got to be a simpler way. While tossing and turning in bed last night, I "invented" the same thing, only way simpler using planetary gears. I had no knowledge of how they work, it was all thought experiment. It woke me up to the point that I got up and grabbed my iPad and started Googling. Lo and Behold, I had re-invented the Toyota Prius's Power Split Device (PSD)!

As has been pointed out, differentials and planetary gears are nothing new. The Prius proves they can be used in a very elegant way to blend vastly different power inputs in order to achieve a very efficient and smooth CVT and utilize two very different power sources (ICE and EM) in their most efficient operating regimes, respectively. I think it is clear the D-Drive is just an obfuscated and needlessly complicated version of Toyota's PSD wherein the control input shafts are like the Motor Generator 1 & 2 (MG1 and MG2) in the Prius Hybrid Synergy Drive system.

16th April, 2011 @ 10:34 am PDT

I'm amazed that people fell for this. The thing is just an obfuscated differential, as was obvious from the video.

The transmission of the Model T Ford worked on similar principles. It used a planetary differential with a brake band on the outside ring gear. Applying the brake to the ring gear made it transmit power, thus combining a clutch and a gearbox. That's essentially what the D-Drive is doing, but with about four times as many parts.

John Nagle
17th November, 2011 @ 10:16 am PST

rxcited is exactly correct. The Prius planetary arrangement is quite brilliant, someone deserves an award for that. A differential is equivalent in that it has 3 input/output shafts. This allows you to have two power sources in and one output. The D-Drive is just a very complicated example of this principle.

Chris Tacklind
20th May, 2013 @ 12:54 pm PDT

Seriously this was done ages ago with tank drives and military vehicles.

Just go to youtube and watch the training video. . about 9 mins in they talk about using planetary gears and two inputs to deal with variable gear reduction. It's called "Planetary Gears, Principles of Operation, Multiple Sets 1953 US Army"

It's one of the basic principles of planetary gear systems. . .the rotational speed (tooth count) of the two inputs on the set maintains a speed between the two. With the lost speed being converted to torque. . . as all speed reduction does. . . . .the power is split between he two inputs and applied to the single drive output. . . .

why does this keep getting reinvented? That's the actual issue. . .I think it torques peoples head so much that they get confused and grant the patent again. .

Monica Austin
29th September, 2013 @ 03:25 am PDT

Use a torsen differential to split the power to the two shafts (input and control). Brake the control shaft to control its speed and the overall gear reduction. Of course effeciency will be lost through the braking action- but how much would depend upon the type of brake.

Douglas Havlir
28th November, 2013 @ 07:20 pm PST

multiple patents on the same principle, now if it can be shown to offer a big percentage on efficiency over other similar designs, then maybe a patent on that part of the design. Test ,test and retest, we need efficiency,not a repackaging of the same. I hope it is a step up

Thomas Lewis
11th December, 2013 @ 10:02 am PST

I will give the short version: this is basically IMHO a differential gear setup. Once you accept that and stop concerning yourself with all the technical principles, and instead think about what you know of car differentials IN PRACTICE, you will figure what's wrong here.

In short, it's able to vary SPEED but NOT TORQUE.

And you all know why, because you've all been stuck in a patch of mud in a conventional, simple differential-operated car.

The wheel that is NOT turning when you are in the patch of mud - does it supply loads of torque, or is all the energy going into spinning the wheel that is already free?

That is correct - the wheel with NO GRIP spins its nuts off, pointlessly. The wheel with the grip, does not bother, because a differential gear arrangement is specifically designed to let the wheels run at different speeds, otherwise it would be like a solid axle, and judder as you drive around corners and the outer wheel needs to go faster to 'keep up' with the inner one, 'cos it has a smaller distance to cover.

Of course, all it needs is a brake attached to the freely-spinning wheel (shaft in Durnin's case) and then that would FORCE the other wheel to move! Hang on - that's friction. Oh dear.

To recap - it's a differential gear unit; it does NOT increase torque which is the whole point of gearboxes (making the engine put more revolutions in for every revolution of the wheel = mechanical advantage so more torque); his other motor is effectively producing a braking input to one shaft (by 'fighting against' whatever the natural free-spinning desire is of this er, invention).

Look, it came up in 2010 on youtube, and in 4 years apparently the guy never got some torque measurements.

That's it - a simple differential gear unit and like that device, it does NOT multiply torque on the SLOWER moving wheel (read 'shaft' here) and yet NEITHER does the FASTER-moving shaft have more torque.

Think differential, think free-spinning one of a pair of wheels (read 'shaft' here) and realise the reason it is difficult to get your head around is because this thing goes round and round in circles without doing the thing it is implied to do - it does NOT produce more torque as the output shaft is slowed down, nor does that happen at the other shaft that is spinning faster.

You can work out all sorts of additions to make it do something useful, but in practice, it'll need a brake on one shaft and some way of INCREASING the torque to the output shaft, to do anything other than bamboozle people. I am slightly surprised it made it to a plastic model, or that he did not simply modify a differential gear unit off an old car and try it on a go-kart or something, to verify that it could give a useful variation to torque on the output shaft.

Sorry, this is not what it's been billed as, or any new tech, or any real use. Took me 45 minutes to realise this, I am disappointed in myself!

Ian Margetts
7th December, 2014 @ 05:11 pm PST
Post a Comment

Login with your gizmag account:

Or Login with Facebook:

Related Articles
Looking for something? Search our 29,894 articles
Recent popular articles in Automotive
Product Comparisons