Gernby

iTrader: (1)

I've been working on an air / fuel controller for about 3 years that would tune itself while driving (via WBO2). It functions similarly to an Apexi V-AFC or S-AFC, but on a much finer scale.

Anyway, I used the controller with great success on my '02 S2000 for 2 years, and just recently adapted it for the IS350. To my surprise, it works even better on the IS. The self-tuning process in my S2000 required about 15 WOT pulls on an open road from 2K RPMs to redline before the fuel curve was where I wanted it. However, on the IS, it only takes about 5 pulls! The graph below is my fuel curve before and after those 5 pulls.

BTW, I'm sure an S-AFC would work on a 2IS, but I don't see how the results could be very good. If I remember correctly, the S-AFC adjusts MAF voltage in 1.0% increments, and I'm adjusting at a resolution of 0.1% to a maximum of -1.9% at redline. It seems that our MAF is VERY sensitive.

EDIT: Yellow is stock and Blue is tuned. My target AFR is 13.1:1, but the result is between 13.0 and 13.3.

Gernby

iTrader: (1)

Laptop interface...

morris

iTrader: (2)

so which color is before and which is after?

Gernby

iTrader: (1)

Sorry, Yellow is stock, and Blue is tuned. I set the target AFR to be 13.1-13.2 across the whole range. The result is 13.0-13.3, which is totally acceptable to me. Actually, the MAF's response is not very linear with temperature, so the stock fuel curve changes throughout the day. It would be impossible to tune it perfectly. That is just another big benefit of having a controller that tunes itself. With an S-AFC, you would have to keep going back to the dyno to see how the AFR looks in different climates.

NovaIS350

wow..what do you do for a living? You must be in the automotive industry

Gernby

iTrader: (1)

I'm a software developer with a hankerin' for tinkerin'. This is the only project like this I've ever done.

morris

iTrader: (2)

You dont think they designed it to get richer with higher RPM while under load?

Gernby

iTrader: (1)

Of course they did! That's the point. Car manufacturers deliberately tune it pig rich at high RPMs (open loop) for safety. If they didn't, high mileage, poorly maintained cars with clogged injectors would detonate to death. That's a good plan for the average Joe, but not for an auto performance enthusiast (like myself). I want to tune it for optimum performance with "just enough" margin of safety that it won't detonate. I wouldn't be worried about detonation until it gets above 13.5:1.

phatboyc

By optimizing the air/fuel ratio are you only getting better mpg under load? Or are you also getting better performance out of the engine?

Gernby

iTrader: (1)

It should improve MPG by some small amount, since the ECU is injecting less fuel, but the main purpose is for performance. Here is a dyno plot from my S2000 before tuning (Blue), and after (Red), Peak power was only increased by 5.5 HP, but the area under the curve is significantly higher. Power delivery was also MUCH smoother.

lobuxracer

iTrader: (2)

Are you able to use inputs from both cylinder banks?

carlos

iTrader: (1)

very nice! now how can everyone else do this?

lobuxracer

iTrader: (2)

You get both. It's one of those rare win/wins in tuning.

Gernby

iTrader: (1)

I'm using an aftermarket WBO2 near the mufflers at the moment, but it will monitor both primary O2's in the future. I don't want to talk publicly about why I'm not doing so now. I have to be careful about what I say without starting the clock on my patents.

Gernby

iTrader: (1)

I hope to go into production with a "black box" in the future. However, I'm still quite far from a production unit, as you can see in the picture above. All the components I'm using are "off the shelf", so I would need to spec out custom hardware that has everything needed in one package, and figure out how to finance a first run.