My problem is that as I drive my car, my AFR's slowly go very lean (across the map). Well into 17 or 18 AFR. It's not directly related to engine temp as the time it takes to slowly go lean is longer that it takes for the engine to come up to temperature. For example: When engine 1st hits operating temp (200*F) the car is still running well or only a little lean. But after another 15-20 mins of driving it gets to the point that it's so lean the car is barely driveable.
The weird part is that the MS seems to be doing this on purpose. Please take a look at the attached datalog and msq. If you look at the data around the 2130second mark (around line number 945) the engine is at idle with the ecu referencing a VE of 43. The MS is using a pulse width of around 2.3ms and the AFR is where I want it (despite lumpy cam).
BUT, if you look at the 2160second mark (around line number 1425) we see that the car is at a light cruise. 2200rpm but VERY lean (17-18afr). What's really weird is that you can see the ecu is referencing a VE of 54 BUT the pulse width is only 1.8ms.
I am baffled as to how my MS can decide to use a pulsewidth of 2.3ms while referencing a VE value of 43 but then uses a pulsewidth of 1.8ms while referencing a VE value of 54. I've looked over all of the other parameters and don't see anything that would have a large effect on the fueling calculation.
Really hoping you guys have some ideas because I am ready to give up
P.S. Ignore the spots where the AFR is reading 10. That is my WBO2 sensor going into warmup mode.
