Problem fixed, but car is not starting... I just pulled one wire from connector under the dash that was supplying 12v to ECU pin as per correct V2 car schematic. In V4 car this specific wire was not present. ECU now works fine and responds to sensors (like rad fans kicked in when I forgot to plug in temp sensor).
So, what do we have today...
To run it needs compression, fuel and spark.
Compression seems to be there - I can feel it from cranking engine manually and by the sound starter makes.
For fuel and spark we need ECU... Here comes the fun part.
ECU is stock 3-plug 1S ECU for 440cc injectors, and some VF turbo, I believe. Right what I have in car now.
In any case, my other car being stock V4 STi works on this ECU flawlessly. In fact, it really helps that I can switch parts back and forth between the cars to test anything.
All the chassis harness is stock V2 except for ignitor deleted because I converted car back in the days to run newage coils.
All the changes to adopt wiring harness to another ECU are done in separate patch harness.
Engine gas 6-tooth crank pulley, which I believe is what it needs - I couldn't remove crank sensor from my other car to confirm... Cam pulleys are also correct, (black plastic with 7 magnets) - I can't imagine that I messed up that bad and somehow put wrong ones or did that incorrectly.
I tested crank sensor wiring tapping signal wire with test light connected to 12v and fuel pump starts working with boost solenoid clicking after I stop. Similar things happen when I just crank the car, so I assume that crank sensor itself, it's wiring and ECU are working as intended.
Car has no maf sensor and map sensor is incorrect for this ECU (V2 and V4 are different, I planned to fix this later in ESL ECU or replace it with GM sensor) but I believe that it shouldn't affect startup because I stared black car on this ECU with both of them fully disconnected. Car just dies after 5 sec of working, I assume because cranking and starting routine is over. But starts perfectly.
Ignition is a bit customized, this ECU is meant to control coil pack with two channels, but car has newage COP ignition using all old COP wiring.
This mod is well documented and quite common - two coils are connected in parallel and I will check the wiring once again, but two of four pins of ignition control in original COP wiring are not changed at all, so I assume that in worst case if there's a problem, at least two cylinders should sort of fire up but it's not happening, there's no sign of any combustion events...
I also checked +12v at coils feed and it's ok, power is there (At least at ign on, probably need to check that at cranking)
I haven't done any checks for injectors yet. But wiring in general was checked several times, which doesn't mean that I couldn't miss something.
I have custom fuel lines with aeromotive FPR and pressure gauge showing around 3 bar, so some pressure is there.
I spend last week trying to sort it out and found couple of issues and oversights, but now is sort of "should be working", but it doesn't.
Maybe you can advise me something?
I need to confirm cam sensor signal and timing sync at ECU somehow. What is the best way to do it?
Also I plan to re-check wiring of ignition and injectors circuits but I don't feel any of them firing now. Will probably also take one coil out and put old spark plug in to see it better, maybe it's firing after all...
But something tells me issue may be no ECU sync.
I swapped cam sensors between cars and my other car starts every time without any problems, still need to try to start it with cam sensor disconnected just to check...
Also worth noting that fuel in the car is from 2015, but I tried adding 5 liters of this fuel to my other cars and I didn't noticed any changes. In one case other car had really low fuel lever so whatever I filled in definitely burns.. after that I considered that it's not worth draining fuel tank completely and I decided to try to start it as is.. Maybe it's worth adding up some fresh fuel? But again, considering that there are no slight signs of combustion events at all, I believe that problem is elsewhere...