I use dynamic airflow corrected by lambda when I'm tuning gen3's. On gen4 I just turn off fuel trims and DFCO and calculate both airflow modes directly. Paste and smooth as I get closer to complete I'll use paste>average.Originally Posted by LSxpwrdZ
No, there is not, it's just another way of calculating the conversion factor from the fuel mass to air mass. I simply hadn't stumbled on the CommandedLambda PID until I switched to the 3.0 version.
The product of the CommandedAFR and CommandedEQRatio is the stoich ratio for the fuel blend needed to convert the observed fuel mass to the estimated air mass.
Since what ever pulse widths were logged, they are off (of the correct fuel mass leading to stoich) by the amount included in the feedback from the WBo2.
Example
- the goal is to have 0.865 lambda, so EQRatio will be 1.156 in the PE table (as a side note, I tuned my MAF in the dyno to achieve what ever ratio is in the PE table, ie. no MAF scaling in the PE table)
- there are two fuel blends, one having a stoich ratio of 14.4 and another 10.5
- the commandedAFR's would be 12.46 and 9.08 respectively
=> by multiplying the CommAFR and CommEQR we get the "normalized" stoich ratios for both, namely 14.4 and 10.5
Now, it doesn't matter what the actual WBo2 reading is, it's the feedback by how much the pulse widths (assuming the injector data is correct, if it's not, then the error in flow rate is included as well) are off compared to stoich regardless of the fuel blend in use.
If, for example, the WBo2 reading is 0.75, it means you would be commanding 33.3% too much fuel (and vice versa if WBo2 shows lean) to achieve the stoich ratio which is the "confirmed" multiplier from fuel mass to air mass in order to calculate the GMVE, hence the multiplication by WBo2.
So basically if you know the actual measured lambda (wideband), how much fuel was injected to get that lambda (inj PW and inj flow rate), and stoich of that particular fuel then you can calculate how much air there must have been to get that lambda reading, and from there a little wham bam math with IAT and MAP and out comes GMVE. Yes?
Now, that's what I would call an executive summary!
So, yes that's reasoning, except I would now use MAT. Of course, with higher airflow the difference is negligible.
The other approach is to know the air entering the engine (through well calibrated MAF and with the help of WBo2).
In the end there's a limit for what can be achieved, as calculated GMVE changes slightly from day to day. +/- 2% or thereabouts is good enough for me.
Last edited by barum; 09-04-2016 at 10:27 AM.
You're right. My apologies. When I was checking my math on it I was forgetting to change my commanded AFR to coincide with the commanded EQ.
Whats the deal with eq ratio commanded being in units of PHI? I read about AFR and FAR but its not quite sinking in. Not sure if someone has a simple-ish answer for this. It seems like everything is good for running with an alcohol sensor and variable AFR ratios but I just want to be sure.
That may be a screw up.. I haven't had a chance to really try it anymore, but when I put it in phi it worked out way better than when it was in lambda based on some old logs I was trying it on
Post a log and tune if you want help
VCM Suite V3+ GETTING STARTED THREADS / HOW TO's
Tuner by night
CPX Tuning
2005 Corvette, M6
ECS 1500 Supercharger
AlkyControl Meth, Monster LT1-S Twin, NT05R's
ID1000's, 220/240, .598/.598, 118 from Cam Motion
2007 Escalade, A6
Stock
Yeah thats the one I read sadly enough. I guess I better close the laptop and get something to eat, pretty sure Ive reached my saturation point for the day.
Thank you again
Yeah I made a phi and lambda graph and neither seemed very close to the other airflow models like others have mentioned about what bank the wbo2 was in.
Wow, this thread is awesome...I will have to reread it about 50 times to absorb it.
Thanks to all of you who have researched this info...just when I think I am catching on, I read a thread like this.
GHuggins- I have read your Injection timing stuff...and Im trying to figure out how this applies to 2.4VVT(unknown cam specs), so I can implement into my tune. Thanks.
Posts like this are what makes HPT forum great. I have read countless hours of all the the above mentioned authors, I,m wondering "how do i fit all this info into my tune"?
Thanks to all of you for your hard work...I paid attention in Math and Im still lost LOL
Last edited by JMsquared; 09-16-2016 at 10:03 AM.
2006 Cobalt SS 2.4 SC 5spd
GMPP Stage 2 w\2 pass plate
Bst ref FP Reg at tank, E67 ecm
ZZP mid-length header\catted dwnpipe
253 Whp so far
Like, for all the contributors!
this thread is awsome now i just need to figure out how to add the custom maths lol
2013 GEN F GTS (the stormtrooper)
mods- BTR stg3 cam, pac valve springs, harrop intake, ID850s, forced induction interchiller, 2.55 upper/9.17 lower pullys, snow performance water/meth injection, defillipo 3" catback, cat delete pipes, built and tuned by me
Why does it need to be done in open loop? To me it seems that it would average out the injector pulse widths between the two while the narrow bands are dialing things in. When I forced open loop and ran maf only my lambda error was bad and it was showing knock. It was disconcerting to see the car that unhappy to be running open loop in the part throttle area. Im glad to see some more people found this thread it went a little cold for a while.
is anyone willing to post a scanner layout with these custom histograms setup as i can't seem to figure it out
2013 GEN F GTS (the stormtrooper)
mods- BTR stg3 cam, pac valve springs, harrop intake, ID850s, forced induction interchiller, 2.55 upper/9.17 lower pullys, snow performance water/meth injection, defillipo 3" catback, cat delete pipes, built and tuned by me
Ill post some up before the end of the night I just want to make sure I haven't messed with them too much I was trying to simplify them to some of slowgoats suggestions. Im learning this as well so I don't suggest anything I say on this matter be taken as scripture.
here is a parameter and graph that I use like I said I "simplified" is some based on slowgoats suggestions just make sure your channels support everything in the maths. All of my graphs are the same so copy them over and add the maths.
Cheers I'll try and get these done when I get home next weekend
good deal man safe travels man.
You need to be in open loop because you're using a wideband lambda sensor and not the stock o2 sensors. If you stay closed loop then your calibration isn't controlling the fueling, your o2 sensors are. The o2 sensor corrections are not included the pulse width, just like the offset isn't included. What's displayed is the calculated pulsewidth based on the calculated airmass. If you're can ran for crap in open loop then your tune in crap. I've only got my phone for internet for awhile so I can't look at the scanner files you're posting.
Reply With Quote
