Printable View
I understand that, but in the case of tuning the airflow is the variable not the fuel. We know what the fueling is, that's why injector data is so important.
i do understand what you are saying about the stock MAF transfer, the calibration engineers spend a lot of time calibrating the MAF with expensive equipment. But as soon as you change anything in the intake tract it skews the calibration. I've seen GTOs with a vararam that the MAF has changed 25%.
The injector constant is just the +y in the very famous algebra equation 1x+y=z. While you may not see it as a variable of airflow, the reality of the operating system of the PCM does not really see it as anything but part of a bigger "how much fuel" equation. In this case "air" (there is no "air" in a fuel injection equation in reality, its more just a fictitious number for calibration reasons) is calculated on the left, which in the PCM happens to be defined as fuel injector constant x airflow x commanded AFR. The output as "fuel" on the right as injector pulse width.Quote:
but in the case of tuning the airflow is the variable not the fuel
The real equation is largely just a quadradic one, but you get the idea.
No, the fueling is not a variable. You know the fueling it's the IFR. Everything is based off airflow; the fueling, and the spark. The ends to me doesn't justify the means. There isn't an injector constant, it isn't OBD1. There is an IFR table that is measured and flows a certain amount.
i can appreciate thinking a different way though. It's like how everyone used to scale the tables on lsx stuff but now you can just double the stoich and half the IFR to do the same thing.
A clearer explanation may be offered like this. Ill try to draw a crude example of roughly how I believe the stock "master fuel equation" to work inside of the PCM.. Please do not nitpick or use something like this without caution. This is roughly what the PCM will do when it is putting out about 50 horsepower with 33lb injectors.
calculate 'airflow variable':
Maf freq 5000hrz. PCM references MAF configuration table for mass air calculation resolution... 42.5g/s worth of air is determined to be entering the motor at this time.
Calculate desired lambda:
PCM references PE/Openloop lambda/closed loop stoich for commanded AFR value.. example will use a lambda of 1.0 (14.7 afr)
Calculate "expected fuel"
PCM references injector constant table, and determines that mass injector flow is will output (230mg x 6) 13g of fuel per second
Calculate "required injector duty"
PCM then takes the "required air" and the "available fuel" and cuts the difference based on commanded afr. End result is 30% required injector duty cycle. (there is alot that goes into obtaining the IPW numbers from this, but this is the laymens version)
The point I am getting at, is the fact that the reality is the adjustment of the equation that sits on top of the injector pulsewidths does not have to line up with anything if it works. You can use the system to your advantage, and make ACCURATE FLAT adjustments to the maf curve (no, not just saying x6% to the whole non linear table) much easier in the injector constant side of the equation.
There isnt REALLY anything in a PCM.. to say that "injector constants disappeared in obd2" is crazy.. its the same math, the same tables, the same IPWs, just displayed differently.Quote:
There isn't an injector constant, it isn't OBD1.
Thas a pretty decent map of what it's doing. But there is still other tables that is using the grams of airflow that it calculates and doing it that way means that all of them are off. A serious one is the spark. While you may be getting the correct lambda I still like the computer to know what is going on.Quote:
Originally Posted by darkhorizon
When obd2 hit it because way more complicated than just an injector constant.Quote:
Originally Posted by darkhorizon
Just drove it up to the gas station and for a quick little jog around a strip by my house. I switched over the maf table from when I was close to 5% after a few 30min runs a couple months ago. And it's locking within 7% not sure why all the sudden no problem. I'll be going on a trip for a car show tomorrow I'll scan it to see if it's the same and adjust the maf table more suited to the Fwi I have now but odd I thought nonetheless.
Sent from my SPH-L710 using Tapatalk 2
If you do that while making more serious power, you won't have any ability to adjust for dynair past 0.92 g/cyl.Quote:
Originally Posted by Kyne
In other words, zero adjustment of timing after a medium/low boost setting.
The 3800 PCM is so simple, it likely won't matter either way.
If there were other algorithms running along side that main one, which could directly influence the main algorithm, then it'd be a problem.
Example being that algorithm that estimates the temperature of the air as it reaches the cylinder found in the later LSx PCMs. That would not be one you'd want to fudge.
Here, if you get it to behave, then you win.
I found very quickly that the rate of evaporation is crucial for dialing in the AFR during warm up with very different injectors from stock. The spray pattern of say disc style injectors are more like a fuel hose than a mist. Larger injectors are more stream like as well, even in the pintle style. That correction is non adjustable on your PCM. I simply ran open loop for longer. Also found, the spray pattern should be the factory straight down. Anything else is pretty difficult to tune around. No point in making it more difficult than it has to be.
Goodluck finding accurate data for injectors though. There are many companies out there just making up data or "borrowing" it from other companies.
If you really want the PCM to know what's going on, you're also going to have to know what the stoich of your fuel is. Unless you know how much ethanol is in there, you won't know how to adjust. And if you adjust the stoich value, you have to adjust every value in the PCM that references an AFR.
Well they are magically locking within 7%. I scanned on the way and back adjusted both times. I'm almost to 0%. I have to make a 50mile round trip tomorrow after work so I'll get another two adjustments and I should be gravy to move on. The world may never know why it was so crazy.... It was 100% humidity and raining like hell.. Maybe that's all it was.
Sent from my SPH-L710 using Tapatalk 2
When i was trying to adjust my LTFT i got them all under -.3 and then continued with the tune when it started to rain a little bit and they jumped to -10-15 and couldnt get them back to normal. After the rain i got them to -.3 again.Quote:
Originally Posted by rperry435
By George maybe that's it! lol
Sent from my SPH-L710 using Tapatalk 2
Anyone want to chime in on my maf curve?
Ok... Think Ive got everything just about dialed in..
take a look and comment. let me know if anyone see's something that needs work.
https://www.dropbox.com/s/8qy9m5oeht...ridaynight.hpl
https://www.dropbox.com/s/knsysx69ji...alTimingQ3.hpt
looks pretty good i would lean out the COT just a hair to get you closer to 11AFR at wot. Im thinking its adding a little to much fuel, according to youre wideband. MAF looks great but the VE table could use a tad more work. i use a filter string on mine to help filter some decel and other variable data out of the equation [PID.6310]<12 see if that helps...BTW your trans tables could use some definate work. but thats on you.
I haven't worked on the ve table at all. And I haven't worked on my trans table either so they will be adjusted eventually. I don't know much about VE yet.
I completely forgot about COT
Sent from my SPH-L710 using Tapatalk 2
VE is a waste of time on the 3.8.
Just in my opinion.
I keep forgetting about that, most people just deal with MAF only, and it works, however i rather have that failsafe incase of maf failure and i like that it seems to be a little more crisper running them blended. just my .02Quote:
Originally Posted by Fivefingerdeathpunch
If the MAF ever failed, I sure as hell would not try and drive the piss outta the car. And its very unlikely that it would fail while trying to break a speed record, and if it did the car would just fall on its face and you'd let off anyway. Would be easy to see on the scan what went wrong.
I can understand wanting to use the VE on a different car like a V8/LS motor, but ours just don't need it really. I also never understood why someone would want to spend so much time getting a 3.8 to run good in open loop.
I thought there was no easy way to put 04+ into closed loop.. Maybe I read that wrong somewhere. I'm happy with the difference from stock to what I have now. 19° timing in the .92 and no knock on a 3.6mps
Sent from my SPH-L710 using Tapatalk 2