Sensor Questions?

[Edward Hernandez R]

"Oh man, now I'm confused.  MAF isn't good as the basis for fuel injection alone, since the amount of air past the MAF sensor isn't necessarily the amount of air into the engine.  Remember the plenum stores air, so a sudden increase in the amount of required air will draw a vacuum on the plenum before the MAF actually sees the increase  in airflow."

You are absolutely right. That's whys carburetors had pump shots. EFI systems(even SD based) have acceleration enrichment strategies that are a function of throttle transients. That's one of the main reasons for using a TPS(throttle position sensor) on hot wire MAF engines. SD systems are NOT "perfectly capable" of learning about engine mods which affect volumetric efficiency and never will be because of the way they work. You have to teach them. Part throttle learning from O2 sensors doesn't work "perfectly" because engine pumping losses change dramatically with throttle angle. Vol eff curves at part throttle bear no resemblance to WOT vol eff(In fact, there is no such thing as "part throttle vol eff"). Properly designed MAF systems learn by themselves. You just can't beat measuring air consumption directly as opposed to guessing and inferring it.
   I'm not discarding SD based systems outright. They work great if you have the time to map a system out, and they are less expensive. It just seemed to me that they have recently been touted as being capable of doing things things they simply cannot do as well as a MAF system. If you have the time and facilities for mapping and engine's air consumption, good for you. Most of us do not.