I completely agree in keeping this thread on line…although somehow they always trend south! LOL!
My understanding of factory calibrating is that the first thing they do is figure out how the engine moves air through it. Since the engine is nothing more then an air pump. This is what the VE table represents. If the vehicle is equipped with a MAF (like the GEN3 and 4 GM vehicles are) then the MAF also needs to be recalibrated IF the intake side of the MAF has been changed. As noted prior the MAF does not not care what its attached to. Its only job is to report how many grams/sec of air are entering the motor. Because its based upon grams (which is weight) temperature and barometric pressure is “baked” into that value. Hence driving the car on the road is not ultimately needed. The MAF is properly calibrated when at a given MAF Hertz the amount of air is correct. For Ford guys this is AD counts. GM does this by attaching the entire air intake system on to a flow bench and moves a predetermined amount of air through the unit. That airflow is matched against the MAF hertz scale. They then give that data to the calibrations department and move on. Since most shops don’t own a flow bench like this, they use the engine as a flow bench. If you command a specific AFR (or lambda in the real world) and you get something different that specific hertz needs to be adjusted accordingly…long and the short is better done on a steady state device as you do not want any “transient” air changes to effect the readings. Of course you can do enough driving and get close that works for most. Just depends how accurate you want to get.
The VE table (which does not truly exist in any 2006 and up GM PCM), is the second half of the air modeling that is needed. The GEN3/4 motors used blended MAF and SD (speed density) which is why they are more involved then tuning a FORD which is typically MAF based with SD as a default if the MAF fails. The 2006 and up processors use a 15 string polynominal formula that determines the final reported airflow. This is used in liue of a real MAP/RPM VE table which is traditionally used in pre 2006 and aftermarket processors. The reason for this is all the cool features today like variable valve timing (VVT), variable intake runners, boosted engines, cam phasing and multiple cams. Traditional computer technology would require huge bulky computers (to todays standards). Using a formula allows all of these inputs to be factored in for a final sum of air flow. Its an infinite VE table. Flip side is they are very involved to dial in properly unless you understand all the inputs.
Sinces its afr (LAMBDA) is what is being used to make these corrections, the type of meter used and where it is places is also a huge factor.
Much like the MAF calibration the coefficients (the formula) needs to be corrected during steady state. Again since its a based upon grams/sec temp is also baked into it which has no bearing on road driving. I don’t know all the steps in doing it but I’ve seen it setup and its a lot of shit. I also drove a few cars before and after with all these things done and was blown away on how some big sloppy camshafts actually drove when done.
The factory PCM has a ton of “adder” correction tables to deal with real world conditions. Things like air temp vs spark retard, aggressive knock retard tables and some coast down stuff most certain can be fined tuned on the street but all this stuff requires minimal adjustment if all the airflow modeling is done correctly. I have heard Howard say a millions times " the ECM is nothing more then a glorified calculator. Garbage in…garbage out".
I hope this kind of gives some insight on what happens in the thought processor of the ECM and keeps this thread more informational then instigatable!