More news! Now the engine is running poorly all the time. I drove a 400 kilometer test, and the engine seemed to use only 5 cylinders the whole time.

Here is a video from the injector test. From 00:39, i try it with 2 bar pressure. Doens't it spray perfectly? Aren't the injectors ok? They all acted the same with that 9 V battery I used. Although this test is months old.



I checked the cables from ECU to coil and saw no corrosion. I also tried to shake the connectors (both heads) while the engine ran perfectly. If there was a problem with the contact, shouldn't the shaking cause some misfiring?




lambda, o2, sensor, jaguar, curve, test, xj6, x300



Above is a random results I got from a jaguars-specialist. Its one of the lambda sensors. It's a bit strange and pointy, isnt't it? However, disconnecting the sensors doesn't affect to the engine run whatsoever.

 

It seems that this diagnostic procedure has run off into the weeds. You have identified a weak spark on cylinder #3. If that is a valid determination, then you need to consider what can cause THAT! I do not know how O2 sensors, injectors, fuel pressure or any of the noise could cause that symptom. So, I would say you need to re-run your spark test to be SURE. Once you are sure you have a weak spark, figure that out. You should have the same 12 volts on one side of each coil plug and the ECU should be switching the other side to ground, then open circuit- which causes the plugs fire.

Without a scope, you can still get an idea of what might be happening by reading both DC and AC voltage, to ground, on the switched side of the suspect coil vs. the other coils. Then move to the ECU connectors to assure the same results.

IF
1) spark #3 is really weaker,
2) the wiring and connectors are good,
3) the coil #3 controlled side voltages are different for #3 vs the others,
and
4) The coils have been swapped with the same result
THEN
It is almost sure that the ECU ignition circuit for coil 3 is the problem.

At that point you either swap out the ECU or try and get someone to change the output transistor for circuit 3 and see if that fixes it.

 

You are right. It is just that I made pretty similar tests as you described more than 10 times and faced a dead end. Guess I have to do it again.

But how should I measure the AC/DC voltage from ECU side of the connector to assure the same results? Do I have to break the connector? Or should I use some needle etc?

 

Yep, you can either pinhole the wire or probe into the backside of the connectors. You should also be looking for a friend with an o'scope. Most any scope will do for looking at the spark primary waveforms.