aidandj

iTrader: (5)

Rev matching without DBW is hard. And getting it perfect is even harder. This is the closest ive seen.

Nvm cant find it. It was on a yellow car. Something mid engined.

Joe Perez

iTrader: (8)

Because unless the entire rotating assembly of the engine is made out of vacuum-forged formulaoneium, it takes quite a great while longer to increase or decrease the speed of the engine for the purpose of rev-matching than it does to move the selector form between two gears.

In a modern F1 car, the driver is there simply to act as ballast and talk on the radio. And yet even with drivetrains controlled by Skynet and costing more than the GDP of Bolivia, they still have clutches.

patsmx5

iTrader: (16)

Because I don't think the advantages of directly controlling each gear individually as suggested are worth the work. I don't think it would shift much faster (only tiny gains IMO), and it would increase the amount of work/engineering a lot. Now everything designed has to package inside of a transmission, for example vs controlling the already available/reliable shifter that's external. Yes it moves in two axis not one, but that's about the only downside. There are several upsides.

As I mentioned in post 1, if this happens, I'm not spending thousands to get it done. Might as well buy a sequential, or different transmission. So that also rules out 100's of hours of R&D too, just not worth it at all.

It would have to be affordable, and buildable withing a reasonable amount of time. Otherwise it's not worth messing with.

Mathew, the car already has a hydraulically actuated clutch, no need for a different version of the same thing. If I want to control the clutch hydraulically, that's as easy as installing a tee fitting and electric solenoid upstream of the tee to lock out the existing master cylinder. That's the easy part, hard part is how do you do the rest while remembering affordable/buildable.

So far nobody has any suggestion on how to improve upon the two electric solenoids/two linear pots method of controlling the existing shifter. What about the clutch? I know a pnuematic cylinder/tank/pump will make that work with great speed, but that's a heavy system. Is there any way to make that lighter and still as fast or at least comparably fast?

patsmx5

iTrader: (16)

Thanks for the link. I don't read German, but from looking at the pictures it appears their control system was all hydraulic. I saw a linear pot on the clutch, and it was hydraulically controlled as well.

I know automating the clutch would make the shifts faster. I know a pnuematic cylinder can push in the ACT Extreme faster than I can. That is the most time consuming part of the shift is pushing the clutch in. I can change the gear relatively fast once the clutch is in.

I suppose I need to focus on how to control the clutch at this point, since that is the low hanging fruit regarding faster shift times. I think my original idea would shift the gears, though it wouldn't be as fast as directly controlling the individual gears as suggested. I think that difference would be like 20 ms though, not 100.

jpreston

iTrader: (4)

Maybe try searching the formula SAE forums? A lot of them use pneumatic shifters on their bike transmissions. I know it's a completely different style of transmission and they don't need clutch control, but you can probably find some good discussion on pneumatic vs. electric.

There has to be some reason Xtrac and Mclaren chose pneumatic over electric.

Sneek Peek, KPAX Flying Lizard McLaren 650S GT3

patsmx5

iTrader: (16)

I built a pneumatic system for our FSAE car years back when I was on the team. They are common there, but they have sequential transmissions, so it's really easy to do. Up/down on one lever, and no clutch. One push/pull actuator and you're done!

The reason for using pneumatic over electric is it's faster. Looking at the link you posted, that car looks awesome.

Joe Perez

iTrader: (8)

While continuing to maintain that this whole idea is Bad and Wrong™, I offer the following insight:


Most Miatas already have a belt-driven hydraulic pump installed as standard equipment from the factory. And it's an easy retrofit for those which didn't come so equipped, using all stock parts. Just add a balloon-type pressure accumulator and you've got a system capable of driving a hydraulic TOB.

Of course, you'll never manage to get the car moving from a standstill without some fairly exotic electronics.


No good can come from any of this. And that's not Luddite thinking...

patsmx5

iTrader: (16)

Good point Joe. My car actually has that hydraulic pump too. I had not thought of that...

patsmx5

iTrader: (16)

Also, I don't need it to take off from a stop, or rev match. Nothing fancy, all I want it to do is shift faster and more consistently than I can. That's it.

The whole point of this thread is to see if this can be accomplished cheaply and simply enough to be worth attempting.

matthewdesigns

iTrader: (8)

Sorry for the delay in responding. Yep, they are both twin clutch, dual input shaft transmissions. There is slightly less load on the synchros for this reason, but the difference between the loaded and unloaded input shafts in at least the GR6 is less than 1000rpm, and the engagement of the selector ring takes about 0.2sec, so there's not really much difference between a traditional manual and the GR6 with respect to how much work the synchro is doing.

Once a clutch is disengaged and the gearset unloaded in a traditional manual, with respect to synchro load it is pretty much behaving in the same way as the unloaded side of a twin clutch trans.

Faeflora

iTrader: (2)

faeflora does not approve :[

Zissou

By a strange coincidence I may have info on exactly what you're looking for. A mechanical engineer friend sent me this link the other day.

Very last photo of the article, Ikeya Formula sequential shifter, converts forward and back movement into a h pattern shift.

Look Closer... Do You See It? - Speedhunters

Leafy

iTrader: (1)

There's been other systems like that. Still basically useless without a dog box. The real answer is to just call up qualife, write them a check for 15k and wait for your box of air shifted sequential to ship over from england.

Savington

iTrader: (31)

Rotational inertia and synchros are the limiting factor, not the human. Since you're keeping the stock flywheel, you'll never solve one of those. The other one might be solved if you replace every gear and selector in the transmission ($$$$$$$$$$$$$$), but if you do that, it's easy enough to buy a trans with external shift rods and a Z-gate shifter.

Shortpersonbk

Why not swap in a DCT trans from a generation agos m3, or isf or something? Some guys got them to work on supras and they hold alot more then the six speed from a miata. The trans them selves are like a grand the rest is just getting your standalone to control it (if yours is good enough) and making an adapter plate/driveshaft. The problem with automating the miata six speed is setting it up to actually shift properly will take along time and most likely nuke a couple transmissions before its correct so your already invest alot of money into it. Another and less automated option would be to move onto a twin disk clutch most multiple disk clutches are alot more user friendly/lighter unless it is a mini disk setup.

loveSONG769

kinda like BMW M5 semi-automatic