I read above that you solved it by changing the crankshaft position sensor, That's great. I must have got the posts out of order, but anyway. we have good news.

if you like comparing things you can do a resistance check through the sensor and compare that to the new one you get and your old one.

You can also hook the leads to a meter, put it on volts and maybe with the analog feature if you have it or use an analog meter.
If you wag a screwdriver in front of the sensor and you should see it pulsing without being connected to anything else.

the flex plate has something close to 50 or 60 holes in it , then there is a space where there are no holes. the ICU might count the holes to see if it reads the right number to know if it's reading correctly. I've worked on machines that do similar with encoders, they count the pulses and there is normally an index pulse. if the computer counts the wrong number the machine goes into a fault mode. I'm not sure if a volvo computer does this, maybe.


with mine the gear on the starter has 5 teeth, it went up following the flex plate and bashed the sensor, then I had two problems , one was a bad sensor and the other was a bent up flex plate, to make things more confusing I also had that bad coil connection. all the gear bits were scattered about the flex plate area or had fallen out. I had a starter so that was quickly solved. Im starting to work on my 85 porsche 944 , it has two of these crankshaft position sensors. I dont know why it would need more than one but it seems to have 2

in a volvo I figure it really could not sense the position at all until the crank makes a revolution to see the blank section and yet they seem to start firing without needing to complete a revolution. I think maybe there is some way it pulses the injectors and sparks without having that reference during initial cranking otherwise how could it fire that soon? I did have a situation where i had solved the TDC sensing issue and the flex plate issue but it would fire at the beginning of my cranking but would not continue to run. this seemed to confirm the above theory. maybe the computer recognizing it's cranking and gives some random pulses even not knowing the position of the crank then when it sees the pattern it latches in and continues reading the sensor each revolution.

when I had a disconnected coil lead due to rust on the terminal I had this symptom. It would fire on first rev but would not start or run. i figured since it fired it must have spark but I think it only got a spark for the first revolution.

i remembered being a kid and working on cars with points style distributors, and seeing the same symptom but in that case it was something called a ballast resistor. If it blew it would fire when cranking but not run.. I dont know if earlier volvos had this ballast resistor or not. it might have been there to protect the points from sparking, or something similar. I think in those, the crank position , in reference to the key, had a bit different wiring than the run position.