355 Cam phase sensor malfunction

We aren't thinking that a weak battery will cause a car to only run on one bank of cylinders, are we? Right after start (even on 4 cylinders) the voltage should jump up to 13+ Volts. I have seen EFI cars run on 8 volts before the ECM finally gave up.

 

Image Unavailable, Please Login

 

I had a vague recollection of a faulty crank sensor causing an intermittent problem with something. I couldn't recall if it was a no start or a partial/full loss of power during running.

Was it Beetle? He's had a few problems with these sensors.

https://www.ferrarichat.com/forum/threads/96-us-spec-355-no-spark.560576/#post-145582097

I think there was an incident prior to this.

Logic and overwhelming resistance from the experts convinced me that it couldn't be the crank sensor, but with your ever-changing spark symptoms I think I would have changed it anyway

That link above also shows some grounds. As far as I know, the ECU wiring harness only has one ground that looks like the one in the photograph. But there is also one between the engine and the chassis somewhere.

 

Spiderscott had a rev limiter type effect from his faulty crank sensor:

https://www.ferrarichat.com/forum/threads/my-355-wont-rev-past-3k-rpm.571690/#post-145876909

I also thought of a faulty knock sensor, but then your spark problem changed banks.

I see on page 1, John also experienced the rev-limiter type effect.

Also on page 1:

 

My problem was due to an internally broken wire. Apparently at exactly 4000 RPM vibration or electronic noise due to the break would make bank 1 (my car is a 2.7) misfires like a rev limited. Back off and it ran perfectly.

 

Voltage to the coilpacks is via the ignition lock, footwell fuse/relay panel... and plug 41090 (near the ECU) which you have checked/repaired. The earth for that voltage is provided by the ECU transistors which trigger the pulses to the primary coils. The entire functioning of the ECU and sensors seems to be dependent on that single earth "70047". I don't think the ECU case screws were meant to be relied on.

Grant's Engine Management Wiring Diagram

The only black (earth type) wire I haven't been able to trace is the one on pin 7 on plug 41090 (near the ECU). It may or may not be part of the alarm system.

 

Yes, but I don't think that your problem.

 

Grant, just put your ohmmeter between the engine block and the chassis (bare metal). If it's more than 200milliohms (0.2 ohms), not including the resistance of your ohmmeter leads, your earths need attention. I don't know if I remember correctly, but you may need to take off your underbody fairing panels to see the engine-to-chassis earth. I saw the technician clean the earth during my major (so the panels will already have been removed).

On my 5.2 car, the ECU harness earth is under the cam sensor. I believe that's the one you have already found.

 

Where does the 0.2 ohms come from? I have never seen that quoted. Is that a known industry thing? I'm not challenging. I'm clueless!

 

It was part of a factory engine management system fault finding sheet which my technician had. I believe it is something you are able to access when you have an SD1 tool licence. As mentioned earlier, my new engine harness wouldn't work. One of the faultfinding steps was to check the resistance of the earths. Dispite looking ok and the lug bolts feeling tight, mine were 2 or 3 times the spec, so the earths were cleaned and tightened a little more.

 

Standby, Grant... I'll take a picture... but the earth stud a few inches below the cam sensor is actually the main earth for the whole ECU harness.

 

Image Unavailable, Please Login

This is earth "70047" on the wiring diagram. i.e the only earth. All the black and brown earth wires are spliced to this in one way or another. I've added some extra thick wires on your diagram to show the connections.

 

Wow 45 pages. That's preety cool!!!!

 

It's not the destination, it's the journey...

.... except we haven't moved very far from the garage yet.

 

Hahahaha good come back Ian
I think it's spelt !tushay! I'm sure I will be corrected nowing my spelling prowess hahahaha.

 

+1 The most common by far fix for troubling electrical problems is poor grounds, the other being lack of continuity of a circuit like in John's case with broken conductors in a circuit. One method I do when I'm questioning the continuity of a circuit besides first using a dvom to check the total ohms of the circuit is to "piggyback" the circuit. So lets say its the ECM ground. I remove the terminal from the ECM plug and install a new one on one end of a coil of wire, and then connect the other to the "-" post of the battery. If the problem goes away, your issue lies somewhere in that ground circuit. In GM's case, they used to like to hide splices in their harnesses and not list them in the wiring diagrams and then put them where it was physically impossible to reach with the powertrain installed. So the only other option was to rerun the new wire taped to the harness to the destination. (Ground stud).

 

You can not disconnect the coil ground from the ECU and reconnect it to the battery ground post. The coil on only grounded momentarily to generate the spark. Most of the cycle the coil ground is open. Grant's original scope trace showed that the coil was being correctly grounded when called for. The part of the signal after the ground opens again it what looked bad.

 

Exactly. If the circuit in question was the one (ground signal from the ECU to the coil) then removing the terminal from the ECU connector and temporarily installing a new wire from the ECU to the coil would take the suspicion from that one circuit. What the root of my suggestion is that electrical circuits can be misleading when looking at them on a scope. Poor terminal contact or a corroded splice can cause a small arc across the bad contact and what you see on the scope is that arc, which is a breakdown of that circuit trying to complete itself from one side to the other. A poor ground can show the exact same thing.
Many moons ago at GMTC one instructor attempted to show us a scope pattern on a circuit that was rigged by the students, and no matter where in the circuit we (the students) put the problem, the scope pattern did not change. Only when we fixed the circuit did the pattern change. I'm no electrical scientist for sure, I only try to set up a strategy based diagnosis to keep moving in a forward direction. In a few posts back, I wrote about the human factor because I cannot tell you how many times I was banging my head against the wall only to find in the end it was something I missed because I wasn't thorough enough with my elimination process. I am sometimes my worst enemy.