I did some more investigation and found the high res be totally dead. I hook up a led on the high res line while rotating the engine and it is constant 5volts, I hook it to the low res signal and it goes low to high while engine is rotated. I haven`t figure yet, when the line is high=5v, does the beam goes through the slots or at 0volts the beam goes through the slots on the opti wheel.

It looks like the optical sensor is done. I serviced it soon and I hope I didn`t went too much on the sealing.
Changing the optical sensor will be POA, since the wheel is dot welded. Do you have any experience sourcing reliable parts. I found some chinese crap sensors and some so called japan sensors.

that I was ready to throw in the towel yesterday.
It might be because the approach is not headed in the right direction.


All pre lt1 pcms were time based spark, which is also inferior due to spark scatter at startup and heavy throttle change. It wasn`t used until ls1 pcm which are much faster and accurate for the time based spark.

I traced some pcm code to see how it handles the loss of high res and it looks it is using some really simple conversion from degrees to ms.
We don`t need good running engine on low res. We need limp home running engine.
I have an idea how to sequence the coils if high res is bad. Measuring the off and on time through 1 low res event and comparing the off and on time to some percentage. Since the different slot widths, a pattern can be made for identifying the cyl id.
Your only hope is to utilize the system timer. I will get you some ideas from the lt1 code, which also uses 16bit timer. It is incremented each cpu clock cycle.

Do you have some info on the avr registers to look at.