1989 XJS with 89k miles. Purchased with damaged coil and amplifier wires. All of the wiring was repaired and continuity was verified from the ignition ECU to coils, amps, sensors. Rotor, cap and plugs were all replaced. The engine tries to fire but doesn't run. The fuel pump runs but haven't checked the pressure at the fuel rail. Also I haven't checked the compression.

There was no indication that the rotor was arcing to the distributor shaft.

Resistance from ECU to crankshaft sensor is 650. Resistance to flywheel sensor is 695 ohms.

Previous owner thinks one of the CATs overheated but hasn't been replaced yet. I'm trying to get it running before replacing it.

There is spark on at least one spark plug on each bank. It has Marelli ignition system.

Any suggestions for troubleshooting would be appreciated. Thanks Mark

 

I bought a non running Marelli car. 89 model. It had been tampered with unbelievably in a attempt to get it running. I also own a 97 XJR that wouldn't start once. Trust me on this...you can't rely on ohms of resistance when it comes to the crank "sensors". These are actually small magnetos. They produce a AC voltage. Over time, the magnet(s_ break down, and they no longer produce the voltage to feed the Marelli ECU the proper voltage to give it the info it requires to fire right. The first thing I would do, is check the clearance between the face of these front and rear "sensors" to the flywheel and the fingers on the damper. By filing the mounting bosses, get the "sensor" as close to .020 as possible. Then try and start. It may just fire right up as mine did. If not, of all the things to replace, I would replace these. The rear one controls fuel, and the front one controls timing. The only feed to the Lucas ECU is a feed that tells the Lucas ECU engine load (speed). Beyond that, the Marelli ECU and the Lucas ECU operate independantly of each other.

If you need to discuss, feel free to PM me, and I will send you my phone number. I feel very comfortable discussing the Marelli system.

 

Good advice. I would like to talk about this. Tried to send you a PM but appears I can't, maybe because I'm a new member.

First question I have is how do you access the sensors? The second question is how is the gap measured, especially on the flywheel sensor? Maybe the second answer is obvious once I can see the sensors.

Thanks Mark

 

Has anyone looked at the signal between the Marelli ECU and the Lucas injection ECU with an oscilloscope? If so what frequency and amplitude should it be with a running engine?

 

I bought a handheld oscilloscope to try. Two things: first, I had never used a a oscope, and second, I didn't know what I was looking at when I saw it. The single wire going from the marelli ecu to the Lucas ecu is a shielded coax. These "sensors" can more easily be observed with a old anolog volt meter. The needle will sweep showing a voltage, whereas a digital volt meter can't read it because the signal is occurring too fast. From what little I know about it, its something to do with hertz. The 89 cars had coax on the "sensors" only, whereas starting sometime in the 90 model the amps were on coax as well. I built a new harness using coax on both. Again, these things are small generators with a natural iron?magnet core. They break down over time and no longer feed the Marelli ecu the correct voltage resulting in poor running condition and even worse, non running. The car I bought would turn over, act like it was out of time. I corresponded with the lead mechanic at the shop that had it, and he was convinced the car had jumped time. It hadn't. It really was as simple as the rear "sensor"

 

To diagnose this you need to work through this methodically.

You have verified spark (the car will run on less than 6 cylinders, although not smoothly)

Next you need to verify if you are getting fuel. after cranking pull one of the spark plugs and check for fuel, if you find them dry.

Then with ignition on (do not crank) check for 12volts DC on BOTH pins of the injector lead.

If you have 12v on both, then have someone crank the engine and listen for the injectors clicking, use a screwdriver on the injector up against your ear if you need to.

If the injectors are not clicking then I would check continuity of the Tach signal lead from the Marelli to the FI ECU, this is the coax. Check the center conductor, the shield is only connected at the Lucas ECU. Don't worry about scoping this if you don't know how to use a scope it may give you a false reading.

 

Rail, I sent you a PM with my phone number. The front sensor is almost directly under the engine damper. You can put a feeler gauge between the fingers on the pulley and the sensor. On my car, due to casting difference of the timing fingers, there were three different reading between the fingers and the sensor. I filed the mounting boss until I got the tightest clearance of .018 I think, and the other two wound up being .027, and .032 if I remember correctly. The rear one is opposite the starter. You can put a feeler gauge between the sensor and the flywheel. This one I also tried to get .018 by filing the mounting boss. Best I remember, my rear sensor was .047 away from new. Very marginal, and not sure the car ever ran right from new. In all of the tinkering, splicing, and eventual replacement of the wiring harness by the shop that had it, the rear sensor was also wired backward. Positive to negative. Trust me. It mattered. I put at least 40 hours into that whole ordeal, to finally get it running as I felt Marelli intended it to. If you have 12 volts to each coil, that is good. The tach signal has nothing to do with the way the system works. The tach signal is taken off of one of the coil feeds, and only gives the tach something to go by. In other words, its on the output side of the ECU, as is completely secondary to the car running. I had been given advice of my 97 XJR that if the tach signal occurred, then the crank sensor was good. This was not true in my case. The sensor had the correct ohms resistance, the tach signal worked, and from that, the car should have ran. I checked everything. Even cut my fuel line, and installed a fuel gauge, because like the XJS, the XJR has no port to check fuel pressure. It simply wasn't the problem. It was that "sensor". The XJR only has one btw. On our XJSs, the rear sensor should put out somewhere in the range of .6 volts AC during cranking, but can go to as high as 80 volts AC(!) from what I read. The front one is a little harder to nail down. The fingers are spaced unevenly, and I think the ECU uses AC voltage as well as sine? hertz? to calculate timing needs.

 

A generic way, and a bit newish tech to check for spark at each cylinder>


Use an inductance type ignition timing light. Apply one by one and run the engine.
Dark strobe, no spark.


Why comes next????


Carl

 

The 3 teeth on the front reluctor wheel are spaced evenly and correspond to TDC of each piston on the A bank. The VR sensor on the flywheel is engine speed.

 

I didn't take mine off, but it sure appears unevenly spaced as it turns, lol. Either way, I know it isn't exactly the same distance from finger to sensor