Land Rover Fuel Injection System Manual

							
LAND ROVER 
The third is that the cooling system includes a warm water feed to a cha\
mber directly under the 
injection plenum. This assists engine operation by increasing the temper\
ature of the plenum during 
cold conditions – which increases the ability to vaporise fuel. The w\
ater supply is taken from the 
engine intake manifold (front, passenger side close to injector 1) and\
 the return feed is connected 
via a hose directly to a port on the radiator.
 
Idle control system
The notorious weakness of the 14CUX system is its idle control. The syst\
em is intended to operate 
as follows:
 1.  Assume the engine is running and then switched off, at the instant o\
f switch off - the 14CUX  withdraws the idle air valve stepper motor fully, which opens the port o\
n the plenum 
chamber leading to a port on the atmospheric side of the throttle (B) \
via a half inch hose. In 
effect, the full amount of air flow can then pass by this route the next\
 time the engine is 
started and as the air flow is metered (remember it is being drawn from\
 the atmospheric 
side of the throttle and is therefore being metered by the mass air flow\
 sensor) the 14CUX 
will provide an appropriate air/fuel mixture. The temperature sensors ar\
e involved in the 
precise metering of fuel – as when starting a cold engine the injecto\
rs are actively opened 
for a longer duration in order to provide a rich enough mixture. Once th\
e engine catches, the 
injection system runs through a purge fast idle cycle which holds the en\
gine at about 
1800RPM for a few seconds in order to smooth out the engines running (b\
y gaining 
maximum air flow through the intake, and thereby purging any excess fuel\
 which may have 
wetted the intake runners). 
  2.  With the cold engine now running, the coolant temperature sensor sho\
wing a slow rise in  temperature the 14CUX will test the rev speed of the engine and if too h\
igh will close the 
stepper motor by a predetermined amount. It will then pause for a short \
while and retests 
the engine speed to repeat the process. The objective is to slowly reduc\
e the engine idle 
speed to about 720RPM in line with cold operation – a process known a\
s hunt idling. 
 
 
Figure 35 – Idle control system
 
There were multiple problems with the idle speed system on this vehicle \
– resulting in no hunt to 
idle, or a too fast idle speed when warm. An additional complication was\
 that the normal method of 
connecting the plenum port to the idle air valve stepper motor housing (\
ie: (B) to (C)) involved 
using a soft L shaped rubber hose. However, after adding the new rocker \
covers (with PCV oil 
separator at the front of the driver side rocker) and with the plenum o\
rientated as shown, the idle 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (31 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
air rubber hose fouled the PCV oil separator. 
 
The first problem was that the unadjusted base idle speed for the inject\
ion system was much too 
high. The base idle is the speed the engine will run at when the throttl\
e is closed and the fuel 
injection idle circuit is disabled. It is specified by Land Rover as 520\
RPM (+/-25RPM).
 
It is absolutely vital that the base idle speed is checked on a new inst\
allation. 
 
In order to set the base idle, the engine is first run up to full workin\
g temperature. The engine is 
then stopped and the hose connecting (B) to (C) is removed. The two \
exposed ends (at end (B) 
and end (C)) are then blocked and made air tight. The engine is then \
restarted – but this time with 
the throttle initially opened by hand. The throttle is then slowly allow\
ed to close in order to allow 
the engine speed to slowly reduce to its base speed. With the throttle c\
losed, the resulting idle 
speed is adjusted to 520RPM by turning the base idle speed adjustment sc\
rew (via an Allen key) in 
the plenum. This allen key can be protected by a tamper proof plug – \
which must be removed if 
fitted. Once the base idle speed is set, hose (A) can be refitted.
 
As the plenum was delivered, base idle was in the order of 1100RPM – \
which is reasonable given it 
was being used on a larger displacement engine.
 
The second problem was that both the throttle potentiometer and idle ste\
pper motor were faulty. 
The throttle potentiometer fault meant that the system always thought it\
 was off idle (it was stuck 
set to about 80% open throttle) – which rather nicely demonstrates t\
he value of an air flow 
metering system given that even with this fault, the system was safe and\
 driveable. The idle valve 
fault was caused by a stuck stepper shaft – which is an extremely com\
mon fault with this design 
(given that the stepper motor shaft is exposed to engine blowby gasses \
and oil). 
 
The third problem (as stated above) was that the standard Land Rover h\
ose linking (B) to (C) 
fouled the new position of the PCV oil separator. The hose was therefore\
 discarded, and replaced 
with two short rubber hose joiners, and a neatly bent 10mm copper pipe w\
ith soldered 10mm to 
15mm reducers on both ends (all painted red). These reducers were a sn\
ug fit in the rubber hose 
pipe. 
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (32 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
Figure 36 – Idle air supply pipe (red) – no longer used
 
While this solution did resolve the problem of the original hose fouling\
 the PCV oil separator, it 
came with the disadvantage that the amount of air entering the idle circ\
uit was proportionally 
reduced due to the smaller diameter of the copper pipe. Testing revealed\
 a low idle speed – 
typically 200RPM lower than expected by the 14CUX, and so the ECU routin\
ely reported fault code 
17.
 
 
The idle air flow pipe was therefore reconstructed using half inch coppe\
r painted red (but with no 
restrictions) as shown below. The rubber hose ends coupling to the plen\
um and the idle air valve 
body were, again, ends cut from the standard hose. 
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (33 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
Figure 37 – Idle air supply pipe (red) – finalised
 
The figure above shows the two rubber joining hoses coupling a red 15mm \
copper pipe formed to 
avoid fouling the PCV separator. 
 
There are two more general observations that follow
 1.  The first is that the engine management switch to hunt mode is direc\
tly linked to the output  from the transmission road speed transducer. The 14CUX repeatedly waits \
until the vehicle 
comes to a stop (with a slight pause) before implementing the relative\
ly slow hunt to idle. 
This means that the switch to the lower idle speed is relatively slow –\
 and arguably slower 
than it should be. 
  2.  The second is that the ability of the injection system to properly c\
ontrol idle speed is directly  linked to the cleanliness of the idle air bypass stepper motor shaft. It\
 is possible to clean this 
part (as shown in Ginetta notes at 
http://www.g33.co.uk/fuel_injection.htm) and it is 
suggested that this step be taken on a fairly regular basis. 
 
 
Exhaust and Lambda sensors
The 14CUX engine management system used in 1992 range rover vehicles has\
 a tune resistor 
option fitted to the wiring loom permitting the selection of different m\
odes of operation based on 
the emission legislative requirements of the geographical location of th\
e vehicle when sold. The 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (34 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
loom simply contains a half watt resistor in a plastic case. The value o\
f the resistor defines the 
mode of operation – with the following options …
 
l     180 ohms – Australia rest of world
l     470 ohms – UK and Europe non catalyst
l     910 ohms – Saudi non catalyst
l     3900 ohms – USA and Europe catalyst
 
Non catalytic operation is much simpler in the sense that no extra exhau\
st sensors are required – 
but this mode of operation has the significant disadvantage that the sys\
tem is unable to self 
compensate for any changes in air flow. In effect, the 14CUX will operat\
e using a static fueling 
map, and any engine changes that tend to fall outside of the assumptions\
 built into that map will 
result in substandard performance. 
 
By contrast, catalytic operation requires the use of two oxygen sensors \
(lambda probes) one fitted 
into each exhaust down pipe, but this mode of operation does provide the\
 14CUX with a 
mechanism to monitor the combustion effectiveness of the engine by asses\
sing its exhaust oxygen 
content levels. In effect, catalytic mode of operation allows the 14CUX \
to operate as a closed loop 
feedback system. 
 
The vehicle employs catalytic operation requiring lambda probes in the e\
xhaust system and 
therefore uses a 3.9K tune resistor. 
 
The lambda probes were purchased online – complete with plugs for the\
 range rover donor vehicle 
(it is possible to purchase cheaper variants of the probes without conn\
ectors – and this needs to be 
checked at the time of ordering). Price was roughly £70 each. The th\
reads of the standard lambda 
probes are M12 with a 1.25mm pitch thread. The two down pipes were marke\
d up for suitable 
lambda sensor positions and then fully removed from the engine, separati\
ng them from the back 
pipe at the split point under the gearbox. It was fortunate indeed that \
the exhaust system was 
relatively new and constructed from stainless steel. Penetrating oil and\
 some sharp blows from a 
hammer at the joint was all that was required to separate the unions. On\
ce on the bench the 
construction work for each down pipe began as follows
  1.  The down pipe was cleaned and punch marked at the hole location
2.  A 13mm hole was drilled into the exhaust pipe at the punched locatio\
n. The hole was  dressed and cleaned and then roughened in preparation for welding. 
3.  An M12-1.25mm thread stainless steel nut was ground down to lose rou\
ghly a quarter of its  height. The grind was curved to match the curve of the stainless steel e\
xhaust pipe.
4.  A MIG welder was fitted with 0.8mm stainless steel welding wire, and\
 set to a high carbon  argon gas feed in preparation for 1mm welding. 
5.  The nut was G clamped to the stainless steel exhaust pipe, and then \
tack welded into  position.
6.  Final checks were made that the position was correct and that the la\
mbda sensor could  thread and appear in the exhaust stream correctly. 
7.  The nut was then seam welded round the lower edge directly sitting o\
n the pipe. The weld  was allowed to cool and then the area above the weld (leading up to the\
 high point of the 
nut) was welded.
8.  The excess weld was ground down, smoothed and dressed to provide a g\
as tight seal.
9.  Copper grease was used in the nut thread and the lambda sensors were\
 checked for  clearance.
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (35 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
 
Figure 38 – Welding lambda sensor threads into the exhaust downpipes
 
 
 
Bench testing the electrical wiring loom for the ECU
The wiring loom taken from the donor vehicle was in generally good condi\
tion. Nonetheless, a 
number of wires close to the 14CUX harness plug had been cut, and some s\
ignal lines were broken 
perhaps due to the flexing incurred in removing the loom as one complete\
 piece. Both main and 
fuel relays had been included in the loom along with the primary connect\
or used to link the 
injection system to the vehicle (normally located behind the dash).
 
Before any of the loom was fitted to the vehicle it was assessed on the \
bench primarily to 
electrically test the loom for continuity and to repair any damage. Two \
other reasons made this 
step essential. Firstly it was necessary to understand precisely how eac\
h connector was wired (so 
that faults could be diagnosed when fitted in the vehicle). Secondly it\
 was necessary to re-route 
some of the loomed cabling due to the reversal of plenum orientation (c\
ertain plugs had to be 
moved from the passenger side engine bank to the drivers side and from t\
he rear of the engine to 
the front). 
During bench checking a full set of hand drawings were used to record th\
e wiring connections and 
layout. These are copied in the following set of figures. 
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (36 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
 
Figure 39 – Original loom connector layout in donor vehicle
 
 
The original layout reflected the needs of the donor vehicle. Due to the\
 changes in orientation of 
the plenum, these changed when fitted to the target vehicle.
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (37 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
 
 
 
 
 
Figure 40 – New loom layout suited to the target engine bay
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (38 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
 
The new layout was ideally suited to the target vehicle. Exposed loom se\
ctions were rewrapped 
using self amalgamated tape. 
 
 
Figure 41 – ECU Connector layout and numbering (and tune R)
 
 
 
Figure 42 – Injector electrical connectors
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (39 of\
 55)16/10/2010 9:03:18 a.m. 
						

							
LAND ROVER 
Injector wiring is banked as two banks of four (one bank on the passeng\
er side of the engine, and 
one on the driver side). 
 
 
 
 
 
 
Figure 43 – Exhaust Lambda electrical connectors
 
 
 
 
http://www.conehead.org/Projects/Landrover/EFi/efi%20-%20web.htm (40 of\
 55)16/10/2010 9:03:18 a.m.