Revolutionary Top Secret Transmission in Racecar Engineering

[evolutionmovement]

They won't give details as its all proprietary, but it's a called the ZeroShift transmission. It's a manual that DOES NOT INTERRUPT POWER BETWEEN CHANGES! It also only needs the clutch for initially moving off. They have test mule TVRs and a WRC sequential box with it. It's supposed to be a very simple system that can be adapted to most preexisting gearboxes and they're claiming may be standard in cars within ten years (heard that before, though).

Somehow the inventor was inspired by working with materials handling and mining equipment and started off by thinking of a bicycling application so that a rider didn't have to back off on a hill. From there he went on to think up and develop an automotive version. Only the gear selection mechanism of a gearbox needs to be changed to adapt this new system. This idea also does not use a double clutch or any kind friction mechanism. I have no clue and my brain hurts trying to figure it out. Vrg3?

Steve

[THAWA]

I believe I read about this or something quite similar in templars supercharger thread on rs25.com Someone was talking about getting manual that didn't need to be clutched or something. I can't think of the name of it off the top of my head, but weather these are the same thing or not, it soundslike they're trying to lessen the gap between auto's and manual's more every day.

[TireIron]

big rigs use transmissions like that. they have to clutch 1st but after that its all pneumatic. they just shift it through all 13 gears like nothin.

[vrg3]

Huh. Interesting.

I don't know so much about how truck transmissions work, but I had assumed that they were kind of hybrids between manual and automatic car transmissions.

The question that pops into my mind is: What absorbs the speed differences when you shift? Cuz a crashbox is also a manual transmission with almost no interruption of power between shifts, but it's hard on the engine and hardware because all the rotating components have to instantly change speed.

From my understanding of bicycle gears, they work kind of like how a crashbox does in principle. The derailleur moves the chain to the correct position and since the linear tooth pitch on the sprocket matches the spacing in the chain, the chain and sprocket find a position where they match up and then they lock together. This happens almost instantaneously. The speed difference is absorbed by the rider.

But this is weirder, because they seem to claim that there is no interruption of power between the gearbox and the road either. If that's true, the shifting collars can't go to neutral between gears.

Huh! Maybe it's a torque-converter-ish device mounted on each shifting collar so that when you shift, you gradually reduce the torque taken from one gear and gradually increase the torque taken from the next gear. Or maybe they're super strong spring-loaded telescoping synchronizers. I don't see what either would have to do with bicycling though.

[vrg3]

Hmm, I just read the article. The thing that caught my eye was the acceleration vs. time curve of a car accelerating from a stop. It appears that shifts aren't instantaneous, but that rather they happen progressively, with acceleration falling linearly down to its new rate (if I'm understanding it right; the figure in the article I have is hard to read because its resolution is low). The article does also explicitly state that momentarily both gears are driving the output shaft.

It also says the ZeroShift magic machinery just replaces the gear selection mechanism.

So it's gotta be just a way to progressively engage and disengage gears, like the guesses I made above. It's simply amazing that they found a way to do it reliably and without overstressing other components.

Oh, and it says he was inspired by trying to think of a better, shaft-based, solution to the bicycle gear problem, to replace the standard chain/sprocket/derailleur.

[ZeroShift]

Hi vrg3

Thought I'd come and have a look at the Forum. The graphs you refer to show two things.

One is longitudinal 'g' (shove in the back) which with a ZeroShift box is uninterrupted by the shift. Put another way, it's 'managed' across the shift. ZeroShift therefore has a straight line longitudinal 'g' trace.

The other trace is engine revs through the shift. Obviously engine revs have to fall when shifting up and rise when shifting down. The 'slope' of this fall/rise (ringed in the RCE feature) is a variable we control to ensure longitudinal 'g' remains constant. Read that another way, the torque in the driveline stays the same through our shifts.

The RCE graphs compare AMT/manual and auto traces against ZeroShift.

Now, that slope. While the revs slope happens over time, the shift is instant: Zero time. Not 'salesman's Zero' or 'Zero-ish' - exactly Zero.

Lots of ideas have come up on other Net threads, some of them may even work. However, the speciality of ZeroShift is the simplicity with which it goes about its business. All the suggested ideas I've read introduce complexity (independent clutches, torque converters etc. etc.). Complexity, in our book, is the enemy. We think synchromesh is too complex.

All the best

P.

[vrg3]

Hi Phil! Thanks for coming by! :)

One is longitudinal 'g' (shove in the back) which with a ZeroShift box is uninterrupted by the shift.

The 'slope' of this fall/rise (ringed in the RCE feature) is a variable we control to ensure longitudinal 'g' remains constant.

Woah, cool. I guess I didn't quite get that from the figures and the article.

Now, that slope. While the revs slope happens over time, the shift is instant: Zero time. Not 'salesman's Zero' or 'Zero-ish' - exactly Zero.

I presume it's necessary for the revs to slope progressively rather than instantly to avoid excessive stresses on the clutch, flywheel, and engine, right?

Now what exactly does it mean for the shift to be instantaneous if the revs don't change instantaneously? Or is "zero shift time" a way to express the fact that there is no time spent in neutral?

Complexity, in our book, is the enemy. We think synchromesh is too complex.

The KISS philosophy is compelling, for sure; simplicity, elegance, and reliability usually go hand in hand. It's a little surprising to hear that you consider traditional synchronizers overly complex, but that must mean the ZeroShift system is really simple. :)

[ZeroShift]

I presume it's necessary for the revs to slope progressively rather than instantly to avoid excessive stresses on the clutch, flywheel, and engine, right?

Right.

Now what exactly does it mean for the shift to be instantaneous if the revs don't change instantaneously? Or is "zero shift time" a way to express the fact that there is no time spent in neutral?

There my man is the heart of the matter. Be assured that Zero IS Zero.

I've said before in another thread that taking a swearbox into meetings might be a good finance source because there always comes a point where the penny drops. It's the "Well, &@$% me!" moment. There's a good case for ZeroShift being how transmissions should have been designed in the first place. This theory falls short because the priorities were different for low-power, low-revving engines.

I have a theory that it's difficult for people to re-examine the assumptions/solutions of descendants more than one generation back. i.e. as a contemporary engineer, you might go back one generation (20 years) and question your father's status quo but people rarely go back two, or in our case, four or more generations. Of course at the key point when synchro may have been questioned, the world was at war. Post-war, gradually rolling out synchro into all transmissions and on all ratios took another generation. Full 'roll-out' of synchro in the 1960s was, by then, 2 to 3 generations on from its patenting in 1928 and therefore - in my opinion - deemed 'beyond question'. What people have tended to do instead (AMT, DCT etc.) is add 'fixes' (aka complexity) to synchro rather than question it. Classic incremental innovation.

Our resident genius Bill Martin, a non-gearbox guy, was free of dogma when he pondered that derailleur a few years back. While we're on the 'social history' thang, Bill's a Kiwi and I've yet to meet a Kiwi who isn't extraordinarily resourceful. Something to do with a culture miles from mainstream manufactured conveniences leads to serious ingenuity I reckon. I know Kiwis who can build houses, boats and cars from a pile of scrap without batting an eyelid.

Anyway, enjoy your debate!

[THAWA]

What articles are you talking about?

[vrg3]

The article from Racecar Engineering that Steve mentioned. Phil sent me a copy which he gave me permission to post as long as I kept the whole PDF intact:

http://www.graphics.cornell.edu/~v/pics ... g_0204.pdf

[evolutionmovement]

I 100% agree with Phil's assesment of the majority of current engineers. As a designer, I look towards innovation rather than development and I'm always amazed that the solutions I come up with for problems that no one else had thought of. I'm usually tentative to mention them as I have no formal engineering background and the solutions seem so obvious to me that I figure everyone had already come to that conclusion and discounted it for some reason. Like my antilag idea that I figured wouldn't work or someone would've come up with it already then I find the Ford and Subaru WRC team had been using the same concept last year and that it will be banned for the 2004 season. I probably came up with the idea around the same time they did - about 2.5 years ago when I just started learning about turbocharged cars.

Its why I mention the Chaparral race cars a lot - those guys thought completely outside the box and revolutionized racing in turn.

Steve

[ciper]

What if this uses sort of a combination geared transmission with a CVT acting as a TC would. A CVT would have a flywheel attached and be connected to the drive wheels (not the engine). While a gear change took place the flywheel and CVT action would keep power to the wheels for a short time.. Once the transmission was in the next gear the CVT and "auto clutch" could reengauge to the engine/transmission. Does that make sense?

[scottzg]

yea, but it seems unlikely, because that would the developing two existing technologies. It would also have a lot of wasted power because of all the spinning bits, 2 flywheels, 2 transmissions, etc. It would be more complex, ergo more expensive to build and more prone to failure (like an automatic ) Also, once you are incorporating a cvt, why shift at all?

[ZeroShift]

Steve:

Fascinated by your Turbo story. There was a guy who raced a seriously modified Morris Minor a few years ago (drag, sprint, hillclimb - it could do just about anything). He had a novel idea for bringing the turbo up to full boost on the startline: he designed and built himself a proper, miniature jet engine to spin up the turbo at tickover. It was awesome, I recall the Minor had everything on it: blower, nitrous, water injection - you name it. It was street legal too.

Have you modded your Legacy then?

P.

[evolutionmovement]

ZeroShift - a mini jet engine?! That's great!

I haven't done much to my N/A Legacy yet other than a turbo engine I'm rebuilding in the kitchen. If I had more money I could have tried out my antilag idea when I first thought of it. I was going to take the bleed air off the blow off valve or somehow attach a valve to the manifold that would open at a set boost and store excess compressed air in a tank on the passeger side floor to be released back into the manifold later when the car was off boost. I got the idea by working on the pneumatic system on an SMT pick & place machine. I figured it wouldn't work later on anyway as I didn't think I could move enough air fast enough, but then I read the WRC cars had the same concept. The difference from what I read was the size of the tank and plumbing (all bigger) which would have made it much too impractical for a road car. The concept seemed the same, but they figured out all the details. A much better idea than the burning fuel in the exhaust to keep the turbo spinning. It was in Racecar Engineering Jan 2004.

Steve

[ciper]

Ive seen a car with an antilag system installed before. It had the strangest exhaust noise I ever heard!

[evolutionmovement]

The system is explained in the June 2005 Racecar Engineering. I almost smacked my forehead when I read how this works! Basically between two adjacent gears in a crash box instead of dog rings you have mirrored pairs of 'bullets' that on each end facing a gear have a traditional dog tooth that on the back side (side away from the gear rotation and therefore not under load) is sloped so that it cannot angage from that direction. On the side facing the adjacent gear the design is mirrored. There are six of these bullets; three face in one direction and three face in the other. all six contact the gear in use, but three are unloaded while the other three trasfer torque to the dog tooth on the driven gear. The bullets are connected to the shifter by springs. When the next gear is selected, the unloaded bullets slide to engage the dog teeth of the next gear, while the original gear remains selected momentarily since the springs lack the strength to pull the engaged teeth off against the engine's torque. When the next gear is engaged by the (originally) unloaded bullets, the faster rotational speed relieves the torque from the original gear and the remaining bullets disengage and follow the first bullets to contact the next gear, now leaving it fully engaged.

F'n A!

Steve

[Legacy777]

Know of any pictures??

[evolutionmovement]

Just the Racecar Engineering article. You should be able to pick it up in stores now. I just got it today.

Steve

[xtalman]

Know of any pictures??

There's an animation on the site now

http://www.zeroshift.com/systemanime.htm

[Legacy777]

thanks

Just saw that you're in college station. Another person in texas....woohoo

You get down to houston at all? I'm sure I could make a trip up there if you wanted to hang out some time.

[xtalman]

thanks

Just saw that you're in college station. Another person in texas....woohoo

You get down to houston at all? I'm sure I could make a trip up there if you wanted to hang out some time.

Sounds cool. Would love to see your car in person sometime. I go to Houston monthly, my mom lives there (Spring). I usually take one of the bikes down every few weekends, I've kind of neglected my Subaru . Trying to get it back in decent shape now.

I don't know if you autocross, but if you do, you should come up for one of the summer practices the Texas A&M Sports Car Club holds. I haven't been in a long time, but back when I was active in the club, it was $5 for as many runs as you wanted. Great fun, but chewed up tires and brakes so quickly that I had to stop doing it.

[Legacy777]

I live around the galleria area.

I haven't autocrossed around houston.....mainly because I can't justify driving an hour plus, drive 10 minutes or so, and then spend the rest of the day working the course.

But I've heard some good things about the A&M one. I'd like to do a track day at texas motor speedway....but haven't done that yet.

Drop me a line next time you head down to Houston, and we'll see about meeting up.

[xtalman]

Hey Josh,

Yeah, the TAMSCC events were awesome. When I did the summer practices club they'd give runs until everybody got tired. Usually there weren't any corner workers, and sometimes there weren't any timers. Whenever enough cones got slaughtered, a volunteer would drive his car around and set up the cones again, and everbody would start running again. I was never competitive in my class, but I had fun, and became a better driver.

The really nice thing about the TAMSCC was the area we were allowed to use. It was wide open, and the powerful cars often got up to 80mph or so. The cone setup was different each time. I went to the Houston Police Department autocross one time, and it looked like you couldn't change the track much, or go over about 50mph. This was years ago though, I don't know where the HouSCCA races now, and I've kind of gotten out of autocrossing and into motorcycling now.

Texas World Speedway, I've never done, but spectating is free for most events, so I've attended a few. Some really nice cars out there. I just wish I had the money to prep a car or bike and participate more in these events. I live 15 minutes away, it's a shame I can't take better advantage of it.

Whenever I've got free time in Houston, I'll let you know. I sometimes go down to the Texas Medical Center for work (I'm a grad student).

[Legacy777]

That sounds pretty cool.