Jaguar XJ-S. Manual — part 64

251

Wrap some foam around the metal filter so it won’t rattle and clang against the floor or battery support. I used the 1/2”
foam tape used for sealing doors. After encircling each end of the filter with foam strips, I added some electrical tape to
hold it in place.

FUEL PICKUP - SALOONS: Saloons -- either XJ12 or XJ6 -- don’t have a surge tank, but they nevertheless have the
same problem with clogged fuel pickups, one per tank. Without having read this book, Chuck Sparks came up with a
remarkably similar fix as that described above, fitting two generic fuel filters in the lines between the tanks and the
selector valve.

FUEL SUPPLY MONITORING: From a fail-safety point of view, the surge tank pickup situation is bad. Regardless
of which pickup screen you have installed or how often you clean it, there is really no way to know if it’s plugged until
you notice the symptoms -- and the first symptoms are likely to be low compression readings in a couple of cylinders!

Of course, a plugged pickup isn’t the only possible cause of lean running under hard throttle. A failing fuel pump, an
obstructed fuel line, or a fuel filter overdue for replacing are other obvious candidates. Whatever the cause, the result
will be the same: if you run it hard, you’re gonna be buying new pistons.

Oh, c’mon, the problem can’t be that serious! How come other cars aren’t burning pistons at full throttle all the time?
Well, obviously because their fuel systems weren’t designed by Jaguar engineers. On most cars, nobody ever even has
to think about the fuel supply; it’s just there, and it works. The concept that an engine could run lean because of a
plugged fuel pickup strikes owners of other cars as laughable.

Of course, you could resolve to never drive the car hard -- but why else would you own a Jaguar? A better solution
would be to have some indication that the fuel supply is holding up before you burn pistons.

Two methods come to mind. The first would be to install a fuel pressure gauge or sensor in the rail. Robert Dingli
says: “I bought a VDO fuel pressure gauge for about Aus$40 and connected it to the fuel rail where the cold start
injector was once supplied. I believe any pressure gauge designed for hydrocarbons would be suitable and other brands
sell for much less. My gauge is mounted in the engine bay as I am paranoid about high pressure fuel entering the cabin.

“There are a couple of things to note about connecting the gauge:

- use high pressure fuel line and fittings.

- don’t mount the gauge on the engine as vibration will kill it.

- use a restrictor in the line as the pressure fluctuations will also kill the gauge.”

If you have the Digital P system, ideally you’d use a gauge that reads the difference between the fuel rail pressure and
the intake manifold pressure. Since the fuel pressure regulators are referencing manifold pressure, this should result in
a steady reading at all operating conditions and any variation is a sign of trouble. If you use a regular guage that
references ambient atmospheric pressure, the fuel rail pressure will go up and down as the intake manifold vacuum
varies, so you will have to monitor both and do a little math to make sure things are operating properly.

For most intents and purposes, a pair of air/fuel meters (see page 36) is a better idea than a fuel pressure gauge.

SURGE TANK DRAINING: In theory, the easy way to drain the surge tank is to pinch off the line from the main tank
to the surge tank and run the pump, which will pump the fuel through the fuel rail and back into the main tank. You
don’t even have to hotwire the fuel pump relay; you can just start the engine and let it run until it stalls.

Good idea in theory, but in fact may not work too well. As described below, the return line from the engine doesn’t
simply go into the main tank, but rather goes through the tank and is directed into the vent connection for the surge
tank. Running the pump will circulate the fuel from the surge tank around and back into the surge tank, and you will
have to run the engine until it burns all the fuel in the surge tank.

252

So, pinch off the vent line too? Bad idea, because then the pump would pull a vacuum on the surge tank and cavitate
long before it empties the surge tank -- unless your surge tank has been fitted with the JLM 2163 vent kit mentioned
below, it which case this might work fine.

Forget that method; there are other perfectly workable methods that are very easy. The intended method of emptying
the surge tank is to clamp the main line from the main tank, and then remove the rubber plug in the floor underneath the
surge tank and open the drain in the bottom of the tank (see page 244). Since the vent line actually opens near the top
inside the main tank (as opposed to where it looks like it goes) you only need to make sure the fuel level is below that
level so it sucks air instead of fuel. Just plan on this job when the tank is less than half full. Of course, you can do it
when it’s full; it’s just that you will first have to drain the main tank down to the level of the vent before the surge tank
will empty.

Another way to drain the surge tank is to pinch off the main line from main tank to surge tank, disconnect a line from
the fuel filter and direct it into a gas can sitting in the trunk, and remove the fuel pump relay and apply a jumper
between terminals 30 and 87 in the socket to run the pump.

Of course, one way to drain the surge tank is to drain both the main and surge tanks; that way you don’t have to pinch
off any hoses. See page 244.

SURGE TANK DRAIN UPGRADE: Somewhere along the line, the drain plug in the surge tank changed from
C41803 to CBC4518. The newer drain plug is magnetic, while the earlier item was a simple plug. If you are working
in the area and find no magnetism in your drain plug, you definitely should opt for a replacement; any particles that
stick to that magnet are particles prevented from plugging the pickup screen.

The 1988 Jaguar Parts Catalogue lists only the first part number, but later editions apparently confused things a bit.
Wally Magathan says, “According to my parts manual, RTC9900CA, Jan 1989: on page 2D 05, there is a diagram of
the sump tank, with the drain plug part annotated as having been replaced by a magnetic plug. One puzzlement,
however: the diagram is indicated to pertain to cars equipped with 3.6 engines only. There is no diagram labeled for
use with 5.3 engines.” Kevin Darling responds: “This puzzled my local Jag dealer as well. But the RTC9900CA parts
book is wrong, since both my 1985 and 1990 V-12's have the sump tank.”

Supposedly the CBC4518 plug will fit all the earlier cars.

If you’re cheap, of course, you could always simply drop a magnet into the surge tank while working on the pickup. It
might be more challenging to clean, but it should do just as good a job of keeping ferrous particles off the pickup
screen.

SURGE TANK VENTING: The vent line from the surge tank doesn’t go to the vapor separator, as one might expect.
Instead, it connects back into the main tank about 2” above the bottom. However, this is not a simple opening into the
tank. Inside the tank, this fitting continues as a tube that makes a couple of bends and terminates with an open end
pointing upward a couple of inches from the top of the tank.

Meanwhile, the fitting for the return line from the engine compartment is at the bottom left rear corner of the tank.
Inside the tank, this fitting continues as a tube that runs across the rear of the tank -- there are actually a couple of clips
brazed to the inner surface of the tank to hold the tube in place -- and then bends upward and terminates with a 180º
hook that pokes its end down into the opening of the pipe from the surge tank fitting described above. The connection
of the two pipes is not liquid-tight; if you tip the tank over and pour a liquid into one of the fittings, it will leak out of
the connection within the tank. But it doesn’t appear to be a very loose fit; the tube from the return line appears to
nearly fill the opening in the end of the tube from the surge tank vent.

In other words: the fuel returning from the engine compartment passes through the main tank but really goes directly
back to the surge tank, with a leaky connection inside the main tank to allow air to bleed out or something.

Mike Morrin provides a possible explanation for this Rube Goldberg: “Let's say you are nearly out of fuel, with a little
fuel still sloshing around the bottom of the main tank. You park, with the engine running on the side of a road with a

253

lot of camber so that all the fuel runs to one side of the tank. The main outlet from the tank is now above the fuel level.
The engine can keep running (or restart) for as long as there is fuel in the surge tank.

“With the arrangement you describe, the engine can keep running until all of the fuel in the surge tank is burned in the
engine. If the return pipe just went to the main tank, the fuel circulated by the pump would all end up in the main tank,
and the surge tank would be empty in a minute or so.”

Apparently something about this system didn’t work as intended, though; in Issue No. 2 of the XJS Bulletin, Ray
Ingman reports that Jaguar issued a kit to provide an additional vent for the surge tank, JLM 2163. The kit includes a
replacement pickup tube that features a vent connection. A line from this connection is simply teed into another vent
line up near the vapor separator.

Ingman describes the incentive for this fix as a “problem of fuel vaporization within this tank under adverse, e.g. hot,
climatic conditions. This results in flat spots and hesitation in engine performance.”

SURGE TANK VENT CONNECTION: There are three threaded fittings on the fuel tank in the ’83 coupe. The main
drain line from the tank to the surge tank requires a 13/16” wrench and the return line from the front of the car requires
a 19/32” wrench -- both readily available at most hardware stores. However, the fitting where the surge tank vent
connects to the main tank requires a 1/4” Whitworth/5/16”BSF wrench.

Fuel Pump

FUEL PUMP CONTROL: On both the Digital P system and the earlier D Jetronic, the fuel pump is controlled by the
EFI ECU via a relay in the trunk. The ECU contains a circuit that determines whether or not the engine is actually
running and shuts off the pump if it has been motionless for more than a coupla seconds. You can observe this circuit
working by turning on the ignition without starting the car, and listening to the pump run; it should shut off in a couple
of seconds.

Note: there have been some reports that owners have gotten the idea that they are supposed to wait until the pump stops
running before starting the car. This is not the case; the fuel pump control circuit is only provided as a safety device.
When starting, the key should be turned from off all the way around to start promptly.

The fuel pump control circuit is unreliable in the 6CU ECU, and sometimes will fail to keep the pump running more
than a coupla seconds regardless of whether the engine is running or not. Typically, the starter is engaged, the engine
fires, the starter is released and the engine quickly dies -- misleading the mechanic to believe that the problem is related
to coming off the starter circuit. Of course, like most electronic failures, it can be intermittent -- runs perfectly for a
while, then dies, then inexplicably runs perfectly later.

If you’d like to test this circuit to determine if this is your problem, simply remove the fuel pump relay in the trunk (the
one without a red mark on it) and put a jumper into the socket connecting terminals 86 and both 87’s. This will allow
the pump to run whenever the ignition is on. If the car now runs perfectly, the fuel pump circuit in the ECU (or the
relay itself) was the problem. AJ6 Engineering, Corsaro, Beckman, or any other place that repairs ECU’s should be
able to fix this problem for far less than the cost of a new ECU.

If you’re willing to risk the odds that you will be in an accident where you are knocked unconscious, a fuel line is
severed, and something ignites the leaking fuel, you can leave this jumper in place permanently -- but a better idea
would be to cut the orange wire from the 85 terminal in the socket and connect it directly to ground and reinstall the
relay. That way, the relay will continue to handle the heavy current to the fuel pump instead of pulling it through the
ignition switch, as is done with the 86-to-87 jumper. Note also that if you happen to have a leaky injector, turning the
ignition on without starting the engine can fill a cylinder with liquid fuel, causing considerable damage the next time
the starter is engaged.

You might also consider upgrading to the later 16CU ECU. See page 294.

254

FUEL PUMP POWER SUPPLY: Elsewhere in this book there are lengthy discussions of problems with voltage losses
in the original Lucas wiring. The electric windows, the windshield wipers, and the headlights all commonly suffer from
low voltage due to high-resistance connections or simply undersized wiring. Why should the fuel pump be any
different.

The voltage to the fuel pump, unfortunately, is a far more serious concern. The fuel pump provides a constant flow of
fuel. The engine takes what it needs, and the excess is returned to the tank. As long as the engine is at low power, most
is being returned to the tank, and a reduction of pump capacity would go unnoticed. When the throttle is floored,
however, a fuel pump working at low voltage won’t provide enough fuel to keep up, and the engine will run lean --
precisely when the EFI system is in “open-loop” mode and desperately in need of enrichment. The inevitable result
will be burnt pistons, in short order. What’s worse, it’s quite possible you’ll spend thousands of dollars rebuilding the
engine without ever figuring out why you burnt pistons -- and hence may do it again the next time you floor it.

Now, today, before the next time you hammer that gas pedal, go out there and measure the voltage across the terminals
on the fuel pump itself. You’ll have to remove the spare tire, the trunk carpet, and a couple of panels to get at it, but it’s
not difficult.

At idle, the voltage should be at least 11V. Better yet, have someone hold the engine at 2000 rpm. At that speed, the
voltage should be at least 13V.

If it’s not, fix it. For analysis, you can use the same VOM while you’re there to determine where the voltage losses are
-- in the ground connection, in the fuel pump relay itself, in the wiring from the battery to the fuel pump relay, in the
wiring from the fuel pump relay to the pump. In the XJ-S, all these items are within a couple feet of each other in the
corner of the trunk. In the saloons, the battery is up at the front of the car, so testing each part of the circuit is a bit more
involved.

Better idea: just forget the testing and replace everything. Run a new 12-gauge wire directly from the battery post to
the fuel pump relay. Run a new 12-gauge wire from the fuel pump relay to the fuel pump. Run a new 12-gauge wire
from the fuel pump to ground. Make sure you’re providing good connections to the fuel pump relay socket, not just
splicing into the existing wires near the terminals -- the existing connections where the terminals are crimped onto the
wires may be the problem! If the voltage at the pump terminals still isn’t what it should be, replace the fuel pump relay
itself -- that’s the only thing left it could be. All of this could be done in perhaps an hour for ten bucks worth of wire,
connectors, and a relay. Money well spent.

FUEL PUMP NOISE: A noisy fuel pump is a very common complaint. The two most common causes, though, are not
the pump’s fault. The first is a problem in the mounting. The pump is supposed to be mounted in foam rubber and
connected at both ends with flexible hose so that it can vibrate without transmitting sound into other parts of the car.
However, its near-vertical mounting often allows the entire pump to slide downward through its mounting until the
outlet barb, within the hose, is jammed hard against the trunk floor. This allows the vibrations to be conducted directly
into the structure of the car. To prevent this, the author wrapped a piece of soft rubber around the pump and held it
securely to the pump body with a hose clamp. The pump could then slide downward only until this rubber contacted
the mount, and could not slip any farther. Problem solved.

Ed Sowell has a related problem: “Mine was really noisy for a while. Then I found some jerk had put a floor jack
under the boot floor pan, pushing it so that it ran into the pump. Got quieter when I pounded the floor back down.”

The other likely cause of pump noise is a plugged pickup or other restrictions on the suction line to the pump. See
FUEL PICKUP on page 249.

FUEL PUMP OVERHEATING: Joe Bialy warns against running the pump with an empty tank, or otherwise with no
fuel in it: “The fuel is the only means of cooling the pump windings.”