1988 Jaguar XJS

Engine: 5.3 L / 12 cyl / Gas / SOHC
Fuel Delivery: Fuel Injection / P Jetronic EFI
Trans: 3-speed Automatic Transmission
Mileage: 95,000 miles
Emissions: Fed Emissions
VIN: SAJNA5846J

Symptoms: Excessive Emissions, No Start, Poor Idle Quality,
Cuts out, Misfire, Poor Fuel Economy
Occurs: Always

__________________________________________________ __________

Car was towed in. Customer stated wouldn't run and could
smell gas.

We attempted to start the engine, it fired and ran on 6
cylinders (poorly), with a strong smell of gas. Shut car
off and noticed fuel dripping from RH catalytic converter.
Crankcase oil is overfull and stinks of fuel.

Checked fuel injectors, and RH bank (both groups) of
injectors are being held on. Disconnected the power
resistors, still on. Disconnected the ECM, injectors went
off. Obviously the ECM has failed. Looking back through the
vehicle history it has had this same fault about a year
ago. At that time we had ohmed the injectors and, finding
nothing, replaced the ECM - successful repair.

My question, what do you suppose would cause the drivers
for both groups on the A bank to fail open causing this
extreme flooding? If we had a failing injector surely it
would cause just one injector driver to fail on the
affected group, not the two separate groups. Or... could
this just be coincidence.

Your thoughts...


Brian Davies
Technician/Shop Foreman
Steve's Jaguar Service
Canoga Park, California, USA

 

The injectors are switched to ground by the ECU, the drivers are not open if they were the injectors would not fire.

First check the harness for a short to ground, as these harnesses get old they become brittle and the insulation fall off.

If you you have ability open the ECU and idenify the injector drivers. They will most likely be a TO-220 type package transistor bolted to a heatsink and there will be 4 of them.

If they are this type of package looking at them from the front they will have B, C, E in that order left to right. Use a multimeter on the Diode function as follows:

between
B and C one way will be open circuit (it may have some reading from other components on the PCB) and with the meter leads revesed you should have a reading.
B and E will be the same as above
C and E should have no reading either way, this is basically the switch part. if this is short then the switch is on.

B to C and B to E are the switch control parts.

 

Does this car have the Lucas (one big round distributor cap with 12 HT wires round it) or the Marelli 'double looking cap'?

If the Lucas system, and you unplugged the injector loom from the power resistor on the RHS inner wing (fender) AND the injectors stayed on, then the injectors are wedged open. If the injector loom is unplugged then there is no current to the injectors to activate them. Are you absolutely sure the thing you unplugged was the power resistor/injector loom plug? Here is a pic of it, the loom plug is the white one at the bottom of the power resistor:


But the most likely problem (after the injectors have been cleaned and unblocked) is that the injector loom is brittle and shorting to earth, thus holding some of the injectors on. Renew the loom is the best thing to do.

If Marelli ignition, maybe could still be the same problem, but I have no experience of that system. However they have a reputation for distributor cap failure that allows one bank to be sparkless and thus raw fuel gets dumped into the exhaust.

 

BTW the resistors are current limiting if they fail (resistors usually go open circuit) the injectors will not fire.

 

Thanks for the reply Greg. The ECU is grounding the injectors through the "On circuit", I have already established that by unplugging the power resistor (as you pictured it), thus eliminating the "Hold circuit".

Although this car is in pretty good overall condition (we have been maintaining it for years), I have considered harness rub throughs. They could not kill the ECU injector drivers as it would only be grounding the injectors ahead of the ECU (which by testing it is not) and even if it did would not damage the drivers. Now a failing injector could overload the driver in the ECU if it was drawing too much current . But then this would only damage one driver i.e. odds or evens on the A bank. In this situation I have the A bank odds and evens triggered on.

 

Does anyone recall if they have ever had a flooding ECU (injectors stuck on) which bank (or both) was affected?

 

Okay, Ive just opened up an old ECU that we had lying around and have traced the pins to the drivers (transistors). There are four transistors that ground the injectors. Two for the A bank (one for "on" and one for "hold"), and two for the B bank. So one bad injector could take out the whole bank.

I have now taken the 16cu ECU out of the car and taken that apart for inspection. What is immediately obvious is the zener diodes, Z700, Z701, Z703, Z704, are all burned.

 

Warrjon, thanks for the info on transistor testing. I have tested them and it seems the "A bank on" driver has failed.