90 XJS V-12 High Fuel Consumption & Low Power

I have a 90 XJS Conv. with V-12 that has a Keisler 5 spd Manual Transmission Kit. Car was running fine, but needed a new clutch. Although you can drop the Tremec 5 spd and the Keisler Bell Housing from under the car, I found it hard to get the input shaft to engage the pilot bushing, so I pulled the engine and re-installed the engine and transmission as a unit.

The down side is that you have to disconnect so much to move the wiring harness, etc. to pull the engine that it takes a lot of time to put everything back in place and connected properly.

The problem that I am experiencing is that after connecting everything following reinstallation, I am running rich (trip computer says 4-6 MPG)and have sluggish accelleration with no power. Some gray smoke when trying to accellerate, but Kirby Palm book says that can also be running rich. Some black specks on the back of the trunk (boot) on both sides of the license plate, which is another dead give away to running rich.

The one problem encountered when pulling the engine was that when removing the main electrical harness that includes the ignition amplifiers and connection for the front crankshaft magnetic pickup for the Marelli ignition from the valley of the "V", I broke the connector for the crankshaft sensor. I replaced the AMP connector and also tried replacing the magnetic pickup cable. The problem persists.

I tested injector spray pattern, since it took a couple weeks for the removal and installation and I thought some gas might have gummed up a couple injectors, but the spray pattern was fine. I verified plug wire routing because I had disconnected a couple wires to clear the electrical harness, ran a compression test while the spark plugs were out to see if they were carboned up (no soot on the plugs at all), averaged 200 PSI cold for compression test.

It runs fine at idle. You can smell the gas if you back up in reverse. It is hard to get above 4,000 RPM.

I had to make a slight adjustment to the throttle potentiometer to get .33 vdc at idle. It was .51, but I attributed that to the fact that I had removed the turntable from the tower and there is some play for the four studs when it bolts down. I traced each vacuum line and found it to appear to have correct routing.

I haven't got a clue what to check next.

Any Thoughts???

First thing I would do is double check to make sure that all the sensors, i.e. both temp sensors) were not accidentally disconnected.

Be very careful as you proceed to get this problem solved. The V12s are notorious for engine fires and the low mileage/rich fuel observations sound ominous. Determine if your ignition is firing on both banks. One half ignition system failure is a common problem for the Marelli. If one bank is not receiving ignition spark, you would get similar symptoms.

I swapped out the -712 Bosch inlet fuel PSI regulator to see what effect it had. Problem remains. I did not swap out the -714 on the left side since vacuum for it is regulated by two temperature operated vacuum switches in series.

I went ahead and replaced the female AMP connector for the front crankshaft sensor with a new NAPA 2-17427 Wired Housing. I had replaced the original broken connector with a used one that I had, but was not sure that I had the best contact.

I checked the other connectors and all appear to be properly connected.

Both banks of the Marelli ignition are firing, but I am picking up an occasional miss, but it is not consistent. I checked plug wires again. I will pull the distributor cap and look for carbon tracking next.

I thought about catalytic converter obstruction, but when I experienced that problem before, the tell tale sign was one or both purple hoses for the air injection system located back at the firewall rupturing. That is not happening, so I have ruled out clogged converters.

Is there something to look at with the trip computer interface? I'm asking about the black box mounted on the same bracket as the two fuel system relays in the trunk by the battery. I am wondering how accurate the readings given by the trip computer are.

Should I put a fuel pressure guage in series in front of the regulator, or perhaps at the inlet to the fuel rail? What pressure would be normal?

The other question that I have is about how to troubleshoot the charcoal canister. The Jaguar Repair Operation Manual AKM 3455 lists the canister as a possible cause for running rich. Anyone know how to check it for proper operation? there are three seperate vacuume valves connected to it.

Michael, You are certainly having a lot of trouble with this XJS and I admire your persistence. If it were mine, I am not sure that I would still own it!! My experience says that the trip computer is not accurate as far as MPG. It usually errs on the high side however. If your car is running properly, you should be seeing 18 MPG while cruising at 60 to 70 MPH. If you put a pressure gage on any part of the fuel rail (I usually use the inlet junction to tap the pressure) you should see about 40psi, plus or minus 2.5 as the intake manifod vacuum varies. Check both banks and if you are seeing more than 45 psi it will make it run richer. Less than 35 will make it run lean. I have never had a charcoal cannister that misbehaved so I cant advise you on trouble shooting it. If you suspect that the richness is being caused by cannister being saturated with gasoline, then unhook the intake manifold to cannister hose and plug up all vacuum lines associated with the cannister. Your mixture should return to normal if the cannister is the culprit. Regarding plug wires and spark plugs; I prefer Champion copper plugs and avoid Bosch platinum since they dont feel at home in a large engine displacemet. Your occasional miss might be caused by resistive spark plug wires. In this age of high performance solid state ignition systems, the spark plug wires seem to be the weakest link. I also recommend using the OEM Jaguar wire set on the V 12 even though it is more expensive. Another tip; The fuel pressure regulators can develop leaks in the seal between gasoline and vacuum side in the diaphram. This would certainly be a significant cause of mixture richness. Check for this by putting a vacuum pump on the vacuum hose to the regulator. If it does not hold a vacuum, you have a leaker. Let me know how the problem works out.

I tried three things of late, but still am running too rich. Next is to run down the list of items John Farrington suggested.

First, I found out from my teenage son that helped in pulling the engine that the right hand cat fell and jerked the O2 sensor. I swapped out the sensors. I found two things; 1. the left sensor was loose, but was also soot free. 2. the r/h sensor had a heavy soot accumulation.

I swapped out the Fuel Injection ECU to see if that had any effect. It did not. I also swapped out the power resister to see if it had any effect on duration of injection. It did not. I will post again next week after looking into the fuel pressure.

Michael, Your observation that the right lambda sensor had black soot on it is very interesting. It points to the right cylinder bank running rich. Check everything out including the fuel pressure regulator on the right side having a leak in the diaphram. This leak could occur and the fuel pressure regulator still give somewhat accurate regulation of the fuel rail pressure. Hook up a vacuum hand pump to the vacuum inlet of the pressure regulator and see if it can hold a steady vacuum. If it cant, you have a leak and should replace it.

First thing I tried was to pull a vacuum on both fuel PSI regulators as suggested. I applied 20 PSI with a hand vac for 60 seconds with no drop.

I tried running the engine and disconnecting each vacuum line to see if any fuel was leaking. I did not find any.

I placed a fuel pressure guage with a tee at the inlet to the fuel rail. I got the following results:

Idle - 32 psi
2000 rpm - 32 psi
3000 rpm - 36 psi
4000 rpm - 32 psi

Next I drove the car and had someone ride with me to write down pressures at different speeds and rpms using a series of gears to see what the pressures might be:

20 mph @ 2700 rpm - 32 psi
30 mph @ 2400 rpm - 32 psi
40 mph @ 2000 rpm - 34 psi
50 mph @ 1900 rpm - 34 psi
60 mph @ 1500 rpm - 38 psi
60 mph @ 3100 rpm - 36 psi
70 mph @ 1800 rpm - 38 psi
50 mph @ 1200 rpm - 36 psi
40 mph @ 1000 rpm - 31 psi
45 mph @ 2900 rpm - 36 psi

Does this mean that this is not a pressure issue, but one of duration that each injector is open?

Since the left and right banks of injectors are on separate circuits could that explain the different richness per side?

I swapped out the trip computer interface, but the mileage per gallon indication is still very low.

Michael, You did some good work but your fuel rail pressures in general are on the low side which would cause lean mixtures, not rich. The mixture running rich must be a fault either in your fuel injector sensors or your ECU being out of calibration. The sensor that has the most infuence on the mixture calculation are the coolant temperature sensors. (I think one on each cylinder bank) This sensor is a thermistor and has about 3K ohms resistance at ambient room temmperature (20 deg. Centigrade) The resistance will be lower at coolant operating temmperatures. (90-95 C). Your fuel injector literature will specify an entire calibration curve. Thermistor resistances higher than 3Kohms will cause rich mixtures. Your lambda sensor is primarily for fine tuning the mixture to emmission control standards and would not greatly effect fuel mixture. Do you have access to a known good ECU for your V12? Substitution would proove or disproove your ECU. I am sorry that your fuel pressure regulators check out since that was an obvious and easy explanation. Another thing to check out is the red vacuum line that is routed to the boot to your ECU. This vacuum is used in ECU calculations and should be at about manifold pressure. Make sure that the line is hooked into the ECU. About the only way that you can measure the fuel injector pulse dwell time is to use an oscilloscope. This would also assist you in trouble shooting this. You need some information on nominal pulse times for comparison if you resort to this.

I think that you have already checked the injectors for proper, non drooling operation. If you have, ignore this input. If you havent check for injectors that drool, leak or stay on when not pulsed.

Michael, I dont want to cloud the issue, but high fuel consumption and poor power can also be caused by improper ignition timing. Get a timing light on the engine and determine if the ignition is within limits at idle and also if you have ignition timing advance when going from idle to 3000 RPM. I think the advance is calculated by the ECU on the Marelli distributor. The timing advance is a function of engine RPM. Ignition timing advance may be as high as 15 to 20 deg. from idle speed timing. Make sure you have a healthy advance.

Going back to Maynard's suggestion to verify connections to the temp connectors, I wanted to confirm the wiring color code for the sensors to find out if there is a break in the wiring from when the harness was moved prior to the engine removal. The harnesses get brittle with heat and age.

Looking at the wiring diagrams for a 90 conv., it appears that the white temp sensor on the l/h coolant rail is for the climate control?

That leaves the green sensor on the r/h coolant rail as Fuel ECU? and the air temp. sensor on the l/h air filter inlet?

I'm going to check the vacuum line in the trunk for integrity. It is connected, but is also 16 years old. I did swap the Fuel Inj. ECU, but do not know for certain that either is free of problems. I will also check the green sensor for resistance to see what the value is.

More to come.

With regard to the question of ignition timing, this car has the Marelli ignition which means that the ignition ECU in the passenger footwell sets the timing regardless of distributor position. However, there is a timing mark that can be viewed with a timing light and I will check that to make sure that this is not an ignition fault. I had to replace the ignition ECU with a salvage unit once before when the vacuum sensor for the ignition failed. The only way to run reasonably well was with the vacuum disconnected. I will check that out as well.

IIRC, the temp sensor on the left bank is for the engine management system. The temp sensor on the right bank is for the gauge.

There are two temp sensors for the climate control. One, and I think it is attached to one of the pipes to the heater core, prevents the blower from operating when the coolant is below a certain temp. The other is for control of the air being blended.

The white coolant temperature sensor is for fuel injection ECU. It is the most critical sensor for fuel/air mixture. The black air temperature sensor has very little influence on the FIUECU. Also the oxygen lambda sensor is not a primary measurement but only for fine tuning.

I checked the vacuum line for the Fuel Injection ECU for leaks as was suggested.

First, I found that the red plastic vaccum line is connected to the Marelli Ignition ECU in the R/H passenger footwell.

Next I found that one of the black - white stripe vacuum lines with a T on the end connects to the Blue vacuum Max Throttle Switch wired in series with the micro switch that rides the throttle turntable ramp. The other end of the tee connects to a vacuum port.

I disconnected and plugged the vacuum line where it connects to the Fuel Injection ECU in the trunk and disconnected and plugged the line at the vacuum port and pulled a 20 PSI vacuum on the line. There is a small 1 1/2 inch by about 3 inch vacuum accumulator just under the ECU. The line would not hold the vacuum.

I disconnected the line from the blue max throttle switch, plugged it, and the line still would not hold a vacuum. I found a one-way vacuum switch about 1" in diameter that is black and white with arrows indicating which end is to connect to the "Motor". The switch was under the cowl about 6" away from where the cables and hoses pass through the fire wall.

I found a smaller than 1/8" male connector on each end and the leaks seemed to be here. I replaced the fitting with a 1/8" male connector which was tight in the hose on both ends of the check valve.

Now the line holds vacuum for at least 2 minutes.

Problem has not resolved.

I have not replaced the white thermoswitch on the l/h coolant rail yet. I will try that next.

Replacing the white sensor doubled the mph to 12 - 14 at cruise from the 6 mph I was getting.

I swapped the Fuel Inj ECU with the spare that I have and could only get 12. So there does appear to be something wrong with one of the 16CU's that I have.

Since the rate of engine accelleration is slow, what about the ignition module signal to the fuel inj ECU?

How do you check the signal being sent? Is it by voltage?

The other thing I will try is swapping the Marrelli ignition ECU with the one that had the bad MAP sensor to see what happens.

I sent an email and left a voice mail for Cosaro electronics in Houston. I want to send one of the Fuel Inj. ECUs in to be tested to see if that is the problem.

You can still smell the raw gas in the exhaust, but there is not a lot of smoke. What smoke is present appears gray.

When your catalytic converters have been subjected to as much torture as your fuel consumption habit, they injest a lot of impurities that must be burnt up/expelled. Give the gray smoke a little time before you call it a fault.

I tried disconnecting and capping the vacuum line to the Marelli ignition ECU, but it did not seem to have much of an impact. I reconnected it. I plan to swap out the Marelli ECU tomorrow to see if the car accellerates any faster and if the fuel consumption improves.

I will also drive the car for a day or two to see if the fuel odor goes away.

I never heard back from Cosaro Electronics about sending in one of the ECU's to get it checked out. I had left a phone message and sent an email. The website was still up. After a week, I gave up on them.

I located a rebuilt Fuel Injection ECU from Coventry West. It arrived yesterday. It did not have any effect. When backing up you can still smell the gas and you can smell the gas under the hood as well.

I will disconnect the charcoal canister to see what happens and plan to test the injectors one more time, mostly because I am out of ideas.

I checked the timing. With the Marelli ignition you don't have the degree scale, so it is hard to tell what you have. There is a scribe mark on the pulley and a stationary pointer. The pointer lines up with the scribe mark at idle. As the engine accellerates, the scribe rotates counter clockwise as the engine speed increases.

I am running Champion Spark Plugs. The wire set came from one of the large Jaguar parts suppliers.