Autosport Labs

Took a look at the OBD2 project- it uses (of course) the popular ELM327 chip that I'm quite familiar with. One goal of mine is to do a comprehensive and open OBD-II reader. The information is out there, just many fragmented and incomplete projects. still, very interesting.

Ok nothing to do with cars or megajolt but wow http://www.youtube.com/watch?v=MvRTALJp8DM

And all those who understand this put your hand up http://www.youtube.com/watch?v=F6zE__J0YIU&feature=fvw Give it a few years and all cars will have one of these.

The Quadrotor stuff is cool to watch (I spent a year or so learning to fly RC helicopters (and wrecked a lot of parts in the process, so that sort of control is impressive), , but I wonder how well it would work in an real-world (unknown) environment. In a pre-mapped closed box loaded with position references and multi angle cameras, all you really doing is making an air-drive servo motor with more elaborate position control loops..

Like a line-following or car-welding robot, they are fine in a controlled environment like a factory, but you wouldnt want to put one in any situation that included unpredictable obstacles or routes.. yet. maybe another 10 or 20 years when (if) reasonably strong A.I. starts to become feasible.

And while I dont like to be a nay-sayer twice in one message, I saw that variable transmission video a while ago and thought "hang on, somethings not right here"..

After pondering it for a few minutes, it occurred to me that the speed-variation input shaft (effectively, the "gear selector") will probably have to resist up-to an equal amount of power that the transmission is "transmitting". Action and Reaction you know..

So if you wanted a 100hp capable transmission, you would need a second 100hp motor to control the variation shaft. Otherwise, the moment the variation shaft was "unlocked" to start varying the drive ratio, it would start spinning uncontrollably.. like a non-limited-slip differential when the load/traction is removed from one wheel.

In fact, unless Ive overlooked something and he's hiding the torque/power somehow (in which case, heat should be appearing somewhere), all the inventor has really done is make a complicated-looking differential.. The exact same input-output speed variation by a third-shaft principle occurs in a "1-wheeler" diff.

Perhaps thats why its still being demonstrated as a hand-cranked scale model with a little drill motor drive.. Try and put some some serious neddies through it and its limitations would quickly become apparent.

Can anyone see any reason why his gearbox wouldnt have that problem ?

The D-Drive system has been thoroughly debunked already long ago. In terms of transmission technologies the most interesting one out there will be the NuVinci, providing that they can scale it up: http://www.fallbrooktech.com/nuvinci.asp

The other interesting one was zero-shift (http://www.zeroshift.com/) which promised instant, clutchless gear changes. Unfortunately they've found it harder to productionise than they had thought.