Hi All,
Have any of You tried to use a maf sensor for load, instead of the Map sensor?,
I ask , becourse i run a Toyota 4age with ITBÃ,´s and turbo, and i have some trooble running it like that,
I runs great on WOT(all the way to 8500rpm), and light load op to around 5k rpm, thn itÃ,´s running really rich, and yes i just lean it out,
but the next time i use the same load zone, then it can be different, if the road is not straight.
SO i also got a really strange ve table, as all other then WOT need almost none fuel, but i need some of the higher map areaÃ,´s to stay rich,as i had it on dyno, and the boost dropped o,1bar more, then on the road, and it got as lean as lambda 0.93.
/Skassa
Interesting - I wonder if this is a manifestation of the same problem I had with the Griffith?
I found there was a MAT error, such that the MAT sensor reading changed non-linearly with manifold temperature, resulting in the air mass calculation being "differently incorrect" depending on temperature.
I found this was usually a small effect on the Griffith (NA car, usually running with a tightly controlled MAT) but sometimes the EGO correction couldn't cope in VE table cells where it was usually good.
I am writing some Perl to analyse this, but it's not fit for distribution yet. I suspect that you could filter a good size log file to extract all the MAT readings and Ego corrections for a particular VE cell and plot one against the other. If there's a correlation with a non-zero gradient then you have found the problem.
MAF would negate all of these problems as a matter of course. You would just set the airflow correction table for MAT globally to 100% in the firmware and use MAT purely as an indicator. Of course, the VE map would be a little weird when based on MAF, but it'll work fine - just connect it as you would an external MAP sensor.
Not too sure how boost control would work then, as it is expecting a MAP sensor...