OpenFlash Tablet v1.55b: BIN files added!

[Luckrider]

Originally Posted by jamesm

Everyone with an 86. @mad_sb did some testing a while back and found his completely stock car to have significant fueling error (~15% in some spots iirc).

If you have an OFT it's free, so there's no reason not to. If you don't have an OFT or ecutek, well, you should .

Do you mind taking a look at some of my logs?

http://www.datazap.me/u/luckrider

I'm getting ready to switch to the 1.55b maps and I might as well start learning to scale the MAF now. I'm genuinely excited for the soon to be added logging parameters.

[troek]

interesting, i will have to do some more reading on this.

[jamesm]

Originally Posted by Luckrider

Do you mind taking a look at some of my logs?

http://www.datazap.me/u/luckrider

I'm getting ready to switch to the 1.55b maps and I might as well start learning to scale the MAF now. I'm genuinely excited for the soon to be added logging parameters.

you can't really discern anything from looking at a log... you have to process and filter the data. if you want.. log the following and email it to me (i believe you already have my email, no?):

short term fuel trim
long term fuel trim
fuel system status
intake air temperature
maf voltage

that way we can filter out all the crap data and plot something useful to look at. if you want to send me your maf scaling (if it isn't just stock) then i can send you back corrections to apply as well.

or you could just watch my screencast here and do it yourself.

[ame="http://youtu.be/NkqY9JVRA4M"]http://youtu.be/NkqY9JVRA4M[/ame]

you can ignore all the bits about process separation and port/di ratios. that doesn't apply to anyone who isn't either completely anal or having aftermarket injectors. if you wanted to dial in di and port independently you could however, though it's pretty difficult without custom maps. you'd have to create two logs, one with di only and the other with port only, and then log the injection mode param to filter on. the great @mad_sb made a thread just recently on the topic for the OFT/open source guys. my method of doing the same requires features not available on those platforms, and as such wouldn't be of use.

[Darkrune]

Originally Posted by jamesm

Everyone with an 86. @mad_sb did some testing a while back and found his completely stock car to have significant fueling error (~15% in some spots iirc).

If you have an OFT it's free, so there's no reason not to. If you don't have an OFT or ecutek, well, you should .

6% beyond safe deviation and about 11% from a good deviation. What the hell was Subaru thinking with this piece of shit tune. I have ran bad tunes in the past in other platforms that were better than this garbage.

[jamesm]

Originally Posted by Darkrune

6% beyond safe deviation and about 11% from a good deviation. What the hell was Subaru thinking with this piece of shit tune. I have ran bad tunes in the past in other platforms that were better than this garbage.

if you ask me... anything more than 3-4% is unacceptable in my tunes. why? because it can be better than that (given a decent maf location). there's no reason for it not to be. granted there's a point of diminishing returns in terms of drivability around 5% or so. with the turbo cars though anything more than that can make for some pretty shitty drivability (particularly if there is a wide discrepancy in error between port and direct injection).

those of you running e85 without flexfuel should keep in mind though that on e85 you'll always see some amount of trimming, because that's the mechanism by which you are accounting for variations in fuel density (e mix ratio). it's not ideal, but it works so long as you give it time to trim out. you just won't have low fueling error, and run the risk of leaning out/going rich if you go into open loop before the high ltft is settled.

[Darkrune]

I typically have experienced ~4% deviation in the tunes that I get. I don't know how to to tune, but experience has led me down the path of knowing what values I need. So 8% would be acceptable for a base or OTS, but a stock tune has no excuses; especially since they engineered it.

[jamesm]

Originally Posted by vgi

i have looked at the shiv's 1.55 rom files to see what tables were changed and it looks like shiv has already re-scaled the maf.

on a side note, i'm not sure that OFT allows yet logging all of those params.

also, didn't you guys say that each (pi/di) afr errors need to be first dialed in separately? that would be pretty tedious process considering amount of time required for closed loop logging (idle - 5 min, slow speed - 15min, and general driving - 30min)? and then my understanding that scaling open loop would also be a good idea? i'm a noob, thus the Qs above.

well maf scaling is a very individual process. general improvements can be made, but it's something that is to be done on a car-to-car basis for best effect.

na guys (or more accurately, those with stock fueling components and maf) really don't need to worry too much about balancing port and di, you just won't have the absolute lowest error possible. the balancing comes into effect when you start changing maf location/size and adding aftermarket injectors. at that point, you can introduce a large discrepancy in the amount of error between port and direct injection, which causes all sort of problems like stalling and idle dipping when coming to a stop, just to name a couple.

and yes it is very, very tedious. but also very very important for some applications. by the time you've done a few of them, you can nail down a turbo car in a few hours. custom maps make it much easier, but that doesn't apply here. it's the biggest thing people overlook that actually effects the day-to-day drivability of your car (again, much more so once you've gone big with a turbo or something like that). dyno charts sell tunes where a video of someone creeping through a mcdonald's drive-through without drama doesn't, so unfortunately it doesn't get much love. that's a big part of why na e-tune customers (with stock fueling components) tend to be a lot happier than those with fi. it's also a big part of why the OTS maps are so successful.. there's just a lot less to worry about than there is when you start changing maf and injectors (especially if you do them at the same time, which is common with fi installs).

[Luckrider]

Originally Posted by jamesm

you can't really discern anything from looking at a log... you have to process and filter the data. if you want.. log the following and email it to me (i believe you already have my email, no?):

short term fuel trim
long term fuel trim
fuel system status
intake air temperature
maf voltage

that way we can filter out all the crap data and plot something useful to look at. if you want to send me your maf scaling (if it isn't just stock) then i can send you back corrections to apply as well.

or you could just watch my screencast here and do it yourself.

you can ignore all the bits about process separation and port/di ratios. that doesn't apply to anyone who isn't either completely anal or having aftermarket injectors. if you wanted to dial in di and port independently you could however, though it's pretty difficult without custom maps. you'd have to create two logs, one with di only and the other with port only, and then log the injection mode param to filter on. the great @mad_sb made a thread just recently on the topic for the OFT/open source guys. my method of doing the same requires features not available on those platforms, and as such wouldn't be of use.

This is something I can definitely work on as soon as we get more logging parameters from the OFT. I currently don't have access to intake temp or fuel system status (is that just a 3 state parameter logging DI, PI, and DI/PI combo?).

[steve99]

Originally Posted by vgi

i have looked at the shiv's 1.55 rom files to see what tables were changed and it looks like shiv has already re-scaled the maf.

.

I would agree changes to maf scaling evident

[jamesm]

Originally Posted by Luckrider

This is something I can definitely work on as soon as we get more logging parameters from the OFT. I currently don't have access to intake temp or fuel system status (is that just a 3 state parameter logging DI, PI, and DI/PI combo?).

It's one logged param, injection mode. 1 is port only, 2 is direct only, 3 is combined.

[Luckrider]

Originally Posted by jamesm

It's one logged param, injection mode. 1 is port only, 2 is direct only, 3 is combined.

Exactly what I figured it was... and... you already deciphered what the raw data means. @shiv@vishnu. Lets get those new loggable parameters rolled out so I can feel like a tuner

[Hanni_0176]

Originally Posted by jamesm

you can't really discern anything from looking at a log... you have to process and filter the data. if you want.. log the following and email it to me (i believe you already have my email, no?):

short term fuel trim
long term fuel trim
fuel system status
intake air temperature
maf voltage

that way we can filter out all the crap data and plot something useful to look at. if you want to send me your maf scaling (if it isn't just stock) then i can send you back corrections to apply as well.

or you could just watch my screencast here and do it yourself.

you can ignore all the bits about process separation and port/di ratios. that doesn't apply to anyone who isn't either completely anal or having aftermarket injectors. if you wanted to dial in di and port independently you could however, though it's pretty difficult without custom maps. you'd have to create two logs, one with di only and the other with port only, and then log the injection mode param to filter on. the great @mad_sb made a thread just recently on the topic for the OFT/open source guys. my method of doing the same requires features not available on those platforms, and as such wouldn't be of use.

Awesome video dude. Just subscribed to you on youtube. I plan to watch all of your other tuning videos as time permits. You, sir, are a huge asset to our community. Thank you for all the time you've spent so far, and keep on doing what you do.

[brn12345]

Just came back from testing out the new 1.55 map. I ran the stage 2 EL and it's a big step up from the 1.34 map I was running before. Mid range is much more responsive and the soft limiter is much better. Highly recommended! Thanks Shiv@vishnu

[Shiv@Openflash]

Originally Posted by brn12345

Just came back from testing out the new 1.55 map. I ran the stage 2 EL and it's a big step up from the 1.34 map I was running before. Mid range is much more responsive and the soft limiter is much better. Highly recommended! Thanks Shiv@vishnu

Glad to hear it!