2 chrome mods that will never happen

[Greg E]

*wideband auto tune and/or closed loop feed back

Get asked a lot if I'm ever going to implement this and the answer is no. I've actully written and tested the code to do this (and it somewhat works) but decided its smarter to not release it.

Main reason has to do with a false promise that it'll work 100% without issue. Most guys want a true plug and play solution for their cars without having to do any tuning. The honest truth is, there is no such thing. Anyone who tells you anything different is trying to sell you something. Each individual car will need its own unique tune. On this platform there are lots of guys who take major short cuts to save $$ and it ends up causing some mechanical issues that make tuning difficult. There is just no good way to write software intelligent enough to compensate that well for all the ghetto/lazy/poor setups out there.

Another reason is liability. The hardware and code can only react so fast and that reaction is based upon feed back from the sensors. This feed back information from your sensors is only so good... If you got any kind of misfire from your motor, the readings from your o2 sensors can be inaccurate thus the code will not compensate correctly. People are quick to blame the "tuner" for mechanical failures and its hard to prove fault one way or the other. I feel its in my best interest to avoid giving some "Internet mechanic" a tool that can be used against me.

Third reason is to force people to actually get their hands dirty. As I said before, each car needs its own unique tune. Doesn't matter how identical your setup is to another car. The general chaos of combustion and voltage feedback isn't as consistent as many want to believe. Tuning a car really isn't difficult at all. Some people are just intimidated by all the features and reading required before getting a good grasp on what's really going on.

I will say that there is nothing wrong with using the 1V narrowband output from your wideband in place of the stock narrowband sensors. Just keep in mind that widebands are finicky devices and have very short life spans so be aware of this if you do decide to eliminate the stock narrowbands. I am working in a plan to have the closed loop code target leaner AFRs for better gas milage but what's difficult to get people to understand is the closed loop subroutines function off feedback and response, not a target AFR (that's why the code cycles the narrowband readings).

[Chris@Rvengeperformance]

I really think you are doing the right thing. Once you get the car cruising and idling well the wot AFR tuning really is pretty trivial if your car is mechanically ok. I would be pretty excited about the ability to alter the cruising afr reliable though, especially when I change over to e85. I think I could lean it out a lot and help lessen the mpg gap between it and 93.

FYI I did use the emanage ultimate's auto tune and if you use it to make a couple pulls until it is right and then shut it off for good it is not a bad feature, but if you leave it on daily it will wreck your tune and make the car very untrustworthy.

[KeithMac]

I really think you are doing the right thing. Once you get the car cruising and idling well the wot AFR tuning really is pretty trivial if your car is mechanically ok. I would be pretty excited about the ability to alter the cruising afr reliable though, especially when I change over to e85. I think I could lean it out a lot and help lessen the mpg gap between it and 93.

FYI I did use the emanage ultimate's auto tune and if you use it to make a couple pulls until it is right and then shut it off for good it is not a bad feature, but if you leave it on daily it will wreck your tune and make the car very untrustworthy.

I've been experimenting with this for a long while now and found some rules, mainly you need a 3 second delay after you've been on overrun before reinstating closed loop feedback to avoid "tainting" the fuel map and causing drivability isssues.

I've rigged a system that looks at tps, when tps is 6% or higher for more than 3 seconds it triggers closed loop operation.

For idle correction the car has to be at a dead stop, 0 vss to avoid overrun wrecking the tune.

V2.32 has just been released which allows correction to be slowed down to 1hz if required.

I battled with PTE1200's and closed loop for a long while until it became reliable and drivable in all situations.

As for O2 correction in boost, its more trouble than its worth..

[stealthII]

Are you planning on doing speed density w/ Chrome? (you've probably already been asked this). As an EMU owner, it's one of the things that keeping me over here instead of going Chrome.

[Greg E]

Are you planning on doing speed density w/ Chrome? (you've probably already been asked this). As an EMU owner, it's one of the things that keeping me over here instead of going Chrome.

I will be working on alternate MAF solutions soon. Still focused on the chrome 2.1 mods and switchable maps.

[J-Groove]

Great post Greg, and right on target.

There is so much liability for an auto tune that even if it works perfectly, someone's build WILL kill it and cause a huge headache for you and put a needless black eye on all of the awesome work you've been doing.

-John

[sergechronos]

Autotune is so overrated, all those artists can't sing without it.

But seriously, even if autotune did work perfectly for everybody and never had any issues, it still would only be worthwhile (imo) for getting you in the approximate ballpark. You'd still need to do an actual tune to really maximize the car for both power and reliability. And if you paid all taht money for a flash ecu and the cables and didn't learn how to do the tuning yourself, then you shouldn't have a problem paying another couple hundred bucks to get it tuned properly.

[Kx1984]

now what about a tune suggestion

my buddy runs his hondata and it has a table where he tells it the desired afr for his wot
and the table fills in with numbers that show how far off that cell/cells are from his current settings
and he repeats it over and over until the numbers are within a certain range if that makes sense

i guess its kinda the same thing as logging and just going back to make changes
to bad our program wasnt hondata its a great program

kx

[Greg E]

I could make an evoscan loggable request that will display the AFR number you need in the fuel table to give the desired AFR based on your wideband scaled readings.

[Kx1984]

if you think we can benefit from that u know better than me