Jaguar XJ-S. Service manual — part 44

171

anyway. He also points out that you can then turn around and sell the EFI ECU that worked with the Marelli, possibly
recouping a considerable portion of the cost of the donor car.

This author is not convinced that retrofitting the Lucas ignition system requires changing the EFI ECU and the harness.
If I didn’t have the replacement ECU and harness handy, I’d try running a shielded wire from the Lucas CEI ignition
amp to the footwell where the Marelli ECU was and connect to the suitable pin in the harness for sending the trigger
signal to the EFI system and see if it works.

If that works, it’ll also be possible to use the early OPUS distributor along with an aftermarket ignition pickup and amp
like those suggested on page -- but you’ll want to recurve the distributor advance mechanisms to the curves used
with the H.E. engine. If the ignition system doesn’t provide a dedicated trigger signal lead, connect a wire to the -
terminal on the coil and run it through a 600Ω resistor for a trigger signal.

Coleman also points out that this conversion will require changing out the tachometer, since a six-cylinder tachometer
is used with the Marelli ignition. Again, if you have the donor car, this is no problem -- unless you have the later style
dash and the earlier tach won’t fit.

REPLACING THE MARELLI IGNITION SYSTEM PART III: As mentioned below, the XJR-S came fitted with a
Zytek engine control system which utilizes a gutted Lucas distributor. The XJR-S is not a common model and the
Zytek system controls both EFI and ignition, so installing this system in a regular XJ-S is likely to be costly. The Zytek
system is supposedly programmable -- you’ll probably need to reprogram it to work with a non-XJR-S engine -- but it’s
reportedly difficult to get the programming software from Zytek. All in all, this retrofit is probably not a reasonable
option from a cost standpoint unless you happen to need a new EFI system as well, and even then it’s not likely to be
the cheapest alternative. Of course, if you happen to have a junked XJR-S laying around...

UPGRADING THE MARELLI IGNITION SYSTEM: Aside from fire avoidance and/or fire repair, some might opt to
replace the Marelli system (or the Lucas systems, for that matter) simply to improve performance. There are several
features that might be worth seeking. Some of the better aftermarket ignition systems have electronically-controlled
advance, making them ostensibly similar to the Marelli -- but they almost invariably are programmable in order to be
adaptable to different cars. This programmability might be of considerable value, especially if the engine is being
modified. Some systems include knock sensors which will allow better optimization of the advance curves as well as
automatic compensation for differing tankfuls of fuel. Some systems use six double-ended coils, thereby eliminating
the need for a distributor.

The ultimate, perhaps, is the “total engine management” system. Such systems replace both the ignition system and the
EFI ECU with an integrated, programmable control system. Some even integrate electronic control of an automatic
transmission -- but obviously you’d need to upgrade the transmission to take advantage of that.

XJR-S Ignition

John Goodman reports that the XJR-S uses a Zytek ignition/fuel injection system -- see pages 306 and 719. “Basically
it is a modified Lucas distributor; even the rotor arm is standard. However the vacuum and mechanical advance
mechanisms have been removed (ECU-controlled vacuum sensor and programmed advance).

“It has a magnetic "Hall effect" engine speed sensor and a similar timing sensor. The 'Hall' effect thing has an inner
and outer wheel. The inner one has one notch and is the timing signal/ speed sensor, similar to the Marelli crank
sensor, the outer star like wheel has twelve notches/spikes to control injectors.

“It appears that to set the timing exactly you really need the dealer to plug in the "JaguarSport magic laptop". In the
set up screen there is a function <reference ignition timing>. This is where you set the distributor to the 10 degrees
BDC at idle for the ECU to get its base line input. After exiting the screen the ignition timing reverts to the control
of the ECU.

172

“So, under the ECU control, at idle you should be seeing 3.5 degrees and at 3000 rpm it should be around 18
degrees if the timing is right. It is a damn inconvenience not being able to check or set the ignition timing without the
software.”

The Zytek system continued to use the altered Lucas distributor even when the baseline XJ-S was fitted with the
Marelli ignition.

Distributorless Ignition

The Jaguar V12 was fitted with a distributorless ignition system for a short time just before it was discontinued, but
didn’t make it into the XJ-S; Roger Bywater of AJ6 Engineering (page 710) says, “It was on the 300 range saloon and
we have a dealer technical guide telling all about it. There were two amplifier modules driving two coil modules each
containing 3 double ended coils for a bank of 6 cylinders. All driven by a Nippondenso Engine Management system.”

Rick Wilder, who owns a ’95 XJ12, says, “I had the impression that the distributorless ignition was only added in the
1996 MY. That erroneous impression was reinforced by alldata.com, which shows the Marelli ignition on my MY. I
just took the valley cover off to see what's up and found the 2 coil packs, each with 6 spark plug wires attached, and no
distributor.”

COOLING SYSTEM

COOLING SYSTEM DESIGN: In most “water-cooled” cars, the cooling system consists of a circuit wherein a fluid is
pumped into the portion of the block surrounding the cylinders, then up into the heads, then on to a thermostat, and
from there through the radiator and back to the pump. The thermostat and housing are usually designed so that, when
the thermostat is closed, the fluid is not deadheaded but rather bypassed to allow the fluid to circulate around between
the engine and pump without going to the radiator.

The Jaguar V12, of course, is different. It has two thermostat housings, one at the front of each bank. The XJ-S
radiator is a side-flow radiator divided into a top third and a bottom two-thirds; the end tank on the left end of the
radiator either has an internal baffle one third of the way down, or actually consists of two separate end tanks. The
coolant coming from the left bank, via the left side thermostat, enters at the top left and flows left-to-right through the
top third of the core. Then the coolant from the right bank comes in, mixing with this already-cooled fluid. The
mixture then flows right-to-left through the bottom two thirds of the core and on to the pump. Each thermostat also
controls a bypass directly back to the pump inlet via a “cross pipe”; as the thermostats open to allow coolant to the
radiator, these bypasses are closed off by a disk on the bottom of each thermostat.

It is unknown why Jaguar contrived a dual-thermostat system, but it was probably just to reduce plumbing -- not having
to get the fluid from both banks back to a single thermostat housing.

Because the mixed fluid goes through the pump and to both banks, both banks are always seeing the same incoming
coolant temperature. When the thermostats are closed and the fluid is bypassing into the cross pipe, both banks will
also see the same flow rate, and therefore will warm up at the same rate. When the thermostats are fully open and the
bypasses are closed, however, the left (B) bank will always see a lower flow rate than the right (A) bank, and therefore
will run warmer. This is due to the radiator design; the fluid from the left thermostat outlet has to pass through the
upper third of the radiator core to get to the exact same place the fluid from the right thermostat outlet goes to directly.
As a result, the backpressure at the left radiator inlet will always be higher than the backpressure at the right radiator
inlet. Since the flow to both banks is from the same pump and it is not a positive displacement pump but a centrifugal,
most of the flow will naturally take the easier route through the A bank.

The action of the thermostats may accentuate the differential flow at operating temperatures. When the thermostats are
only partially open, the bypass leads directly back to the pump inlet while the route through the radiator has a higher

173

backpressure. As the higher backpressure begins to slow the flow in the left bank, the coolant arriving at the left
thermostat begins to get warmer. As a result, it closes off the bypass some more, further reducing the flow in the left
bank! Some have advocated installing a colder thermostat in the left side to compensate for the warmer running bank,
but this would seem to be the exact wrong thing to do; it could arguably be better to put a warmer thermostat in the left
side so the bypass stays open and flow is maintained through the bank. Flow will be reduced in the radiator, though,
possibly causing the entire engine to run hotter.

As long as the cooling system is clean and operating properly, the differential cooling flow between banks doesn’t seem
to cause any problem. However, there are three implications the XJ-S owner should be aware of:

1) The coolant temperature sensor for the EFI system is on the left thermostat housing. Since this side is running
warmer than the right and the EFI system will be adjusting the fuel mixture for the entire engine accordingly, the right
bank is likely to be running a little leaner that the left. Of course, on cars with Digital P EFI and oxygen sensors, the
system will automatically correct the mixture on individual banks -- as long as you’re in closed-loop mode. When you
put your foot in it, you go into open-loop mode -- and when your foot is in it is precisely when you don’t want to be
running lean! You can only hope that the open-loop scheduling includes enough enrichment to overcome the effects of
the temperature differential between banks. Again, as long as the system is clean and operating properly, there doesn’t
seem to be a problem.

2) The coolant temperature sensor for the gauge on the dash is on the right thermostat housing. This means that the left
bank can have zero flow and be overheating severely, and the driver would have no indication from the dash as long as
the right side still had good flow.

3) While the system works fine when clean and operating properly, the double-pass radiator scheme seems designed to
accentuate any problems that develop. While the coolant flow is always lower through the left bank, the radiator
getting obstructed internally seems to increase the difference in flow -- possibly because the flow from that bank has to
go through twice as many passes of the obstructed radiator, or perhaps just because the thermostats are closing off the
bypass fully. Whatever, the incidence of dropped valve seats or other overheating damage is much higher for the left
bank.

Roger Bywater has a different theory about why the left bank seems to drop more valve seats: “If the V12 overheats
for any reason B bank is likely to suffer more than A bank and I believe this is because A bank is fed from the low side
of the water pump whilst B bank is fed from the top. Therefore any condition which might promote cavitation at the
pump will have a more damaging effect on B bank.” There are even those who suggest that the left bank suffers more
than the right because the big belt-driven fan is blowing air over the right bank, although this is a minor factor at best.
Rather than trying to decide which theory is correct, note that all these theories might have some validity! The B bank
is definitely the one at risk.

Bywater relates “...an observation from when I first started doing Japanese Heat Damage Tests, which involved an
extremely arduous simulated incline run. The test was carried out on a chassis dynamometer in a temperature controlled
environment with airflow coupled to road speed. The punishing part was a simulated gradient lasting (going from
memory) 20 minutes at high load and moderate speed in which the airflow through the radiator is barely adequate to
cool the engine.

“Before we added supplementary fans to live with the test most cars would overheat badly towards the end and I
remember that a V12 dropped a few valve seats on one occasion. The dropped seats were in the left hand head. This
was a test fleet car in sound condition so all those theories like radiator or heater blocking are not relevant. I do not
accept that airflow around the engine is a significant factor.

“I think just about any car would have overheated in the circumstances. A pair of electric fans mounted in front of the
radiator kept the problem under control (I cannot be sure if that was done on the V12 but it certainly was on the 4.2) so
maybe that should be suggested to anyone whose driving pattern gets into similar territory. Personally I was dubious
about the fans making much difference but the engineer in charge of cooling systems assured me it would work and he
was right. The setup was used in production on XJ6s for the Japanese market only (it is shown on the parts slide) but I
don't think the V12s ever had it.”

174

RECOMMENDATIONS: Despite the proven shortcomings of the cooling system design, many experienced XJ-S
owners insist you can avoid trouble by scrupulously maintaining your cooling system. This author does not agree. If
you put any stock in what I say, you will make four changes to your cooling system immediately:

1) Install coolant filters in the upper radiator hoses. I recommend the Tefba’s, although the Ganos also work fine.

See page 190. Note that you might also want to install a convenient coolant drain to aid in servicing these filters.

2) Modify the banjo bolt in the air bleed port at the top right corner of the radiator so that it actually allows air

through. See page 43.

3) Install a “flushing tee” in the highest point in the heater hose, and use it whenever you’re changing coolant. See

page 179.

4) Install foam weatherstripping in every nook and cranny where air might get around the radiator rather than going

through it. See page 226.

Those are all easy. They won’t cost you $100, and they won’t take more than a couple of hours. In addition to those
four items, there are two things that should be done when the opportunity presents itself:

1) If the radiator ever must come out for any reason, convert the system to single-pass. If that radiator is going back

in, have the radiator shop modify it first. If a new radiator is going in, make sure it’s a single-pass radiator. See
page 198.

2) If you ever have any trouble with that belt-driven fan, chuck the whole thing and install an electric fan in its place.

See page 219.

HOW HOT IS TOO HOT? Just about the first question everyone asks is just how high the needle can get before it’s
time to start worrying. Unfortunately, all too often owners ask people who should know, including supposedly certified
Jaguar mechanics, about their gauges reading halfway between N and H, and are told, “That’s OK, they read there all
the time, don’t worry about it.” These people are liars, and you can tell ‘em Kirby Palm said so. Many, if not most, of
the maintenance headaches occurring on Jaguar V12’s can be traced to overheating -- not just one-time overheating, but
long, persistent, chronic overheating, the type that can only happen when the owner is ignoring the gauge -- usually
because someone told him it was OK. If anyone tells you your temp readings are OK, ask him if he will guarantee that
and agree to pay for engine repairs when the valve seats drop.

Trying to get a handle on just what constitutes a reason to panic on the XJ-S temperature gauge has long been difficult,
since reports from owners seemed inconsistent or contradictory. The gauge being an example of British electrics has
only added to the uncertainty. However, by noting many such reports on the internet discussion list and following
several of them while the owners checked things and made corrections, this author has been able to derive the following
guidelines. Note that these descriptions apply to the vertical or “barrel” coolant temperature gauge found in the pre-’91
XJ-S; those with later cars with round gauges will most certainly see similar behavior in general, but I can only offer a
few reports on where the needle sits when the car is running properly -- see below.

Some vertical gauges reportedly have temperature scales in degrees, while others have merely a C at the bottom, an N
dead center of the scale, and an H at the top. This description will refer to the C-N-H scale but both type gauges are
apparently identical except for the paint.

There are basically four situations that the vertical gauge will indicate:

1)

If your car has a properly-operating cooling system fitted with 190°F (88°C) thermostats,
the needle will always be sitting on the N when the car is warmed up. Perhaps just a hair to
the high side of the centerline of the scale, but always within the width of the letter itself,
never above it.

2)

If your car has a properly-operating cooling system fitted with 180°F (82°C) thermostats,
the needle will always be sitting about 1/4” below the N when warmed up.