Ok, so last question... ya right. I would like to know if there is a way to reduce the amount of toggling of the O2 sensor. I understand the purpose of the toggling is to feed the catalytic converter. Since I'm not running cats, I'm wondering if there is a way to reduce the range of toggling. I assume it has to toggle some but I'd like to know if there is a way to reduce the range that it toggles. I've looked through the hac $O2 (1619427 / BJKW) but couldn't find anything that seem to point to a way to control the range of toggling rich/lean.

 

The other purpose of the toggling is due to the nature of the O2 sensor itself. The sensor has a very sharp response as it goes lean or rich. This means that the O2 PID does not know the actual air fuel ratio, just whether the engine is running richer than stoich, or leaner. Due to this, the only way that the closed loop fueling can work properly is to have the gains set up in such a way that the fueling always ocillates around 14.7:1. The actual changes in air fuel ratio aren't as large as they appear from the O2 voltages. With everything working properly, the actual AFR stay relatively close to stoich for the fuel in use if you hook up a wideband.

If you want to reduce the magnitude at which the switching is taking place, the proportional gains and lean/rich thresholds are what you need to look at. By reducing the gains and rich/lean thresholds, the fueling can be held closer to stoich. In some calibrations, though, the rich/lean and mean rich/lean thresholds are already set to .450V, leaving no dead band. In this case, adjusting the proportional gains can usually get the desired result.

Like I hinted at above, dont get real hung up on the voltages you see in your datalog. The main purpose of adjusting the gains is to get them down far enough that the engine doesn't surge from the AFRs going too lean or too rich.

 

As I watch the scan, it toggles between 50m volts and 800m volts very rapidly, which to me seemed excessive. The datalog shows it about twice a second, but the cross counts is much higher than that. Voltage threshold VS air flow seems pretty tight with a dead band of only ~25m volts. Makes me wonder if the gain is too high and giving a lot of over shoot. I'm assuming the proportion gain VS RPM VS MAP would be where to play with that.