I’ve observed that many 928 owners are trying to diagnose problems with their
cars. I’m pleased to share some of the facts I’ve learned since I began
providing LH module rebuild service. I must advise the reader that extreme care
must be taken when working near high voltage spark sources. If you DO NOT KNOW
WHAT YOU ARE DOING, DON’T DO IT. You can hurt yourself and damage your car. Have
a qualified Porsche technician do the work, in the long run you’ll save money
(no self-damaged parts) and you won’t get hurt. Odds are your Porsche shop may
not know this detail of information, so feel free to take this information to
them, it could save you money by reducing the time required and improve the
quality of the diagnosis.
There were two different generations of LH fuel injection controllers used during the life of the 928. The 1984 – 1986 Euro/ROW and 1985 – 1986 US/Japan model 928s used an 8039 microprocessor-based design and have a 25-pin connector to the module. For model years 1987 to 1995, the LH module was based on an 8031 microprocessor and the increased functionality required a 35-pin module connector. Spark control also was supported by two different generations of modules. The 1984 – 1986 Euro/ROW and 1985 – 1986 US/Japan 928s use a 25-pin EZF controller and an inductive RPM pick-up and had a vacuum coupling provide engine load information. From 1987 onwards, a 35-pin EZK module was used. The EZK had a three-dimensional RPM, engine load and spark advance map stored within the ROM code, so no vacuum connection was needed.
The EZK system used an inductive pick-up for RPM signals plus a Hall Effect sensor to recognize which ignition coil gets the next spark signal. Also, knock sensors were added to the spark control system. On-going improvements and fine-tuning were achieved by changing ROM codes, so modules were often specified to specific year applications.
When diagnosing no-start situations, ALWAYS clean and tighten connectors, check ground connections and inspect for broken wires. Only after those potential problems are checked-out should a module failure be investigated. The 928 workshop manuals are a necessity for diagnosis, the system is complex and the interconnections are not intuitive. Here are some hints to follow if your LH-equipped 928 isn’t starting or running well.
It is not commonly known that the LH fuel injection module is slave to the EZx spark control. So if you have a no-start condition, always check for spark first. If there is no spark, fix that problem before testing for LH ECU functionality.
The EZx spark unit looks for rotation pulses from the engine RPM sensor as a condition to start the spark system. The EZx then turns on the LH ECU, so a bad EZx ECU may cause the fuel injection not to work. Therefore, always check for RPM pulses from the sensor located at the top of the bell housing for a no-start & no-spark conditions.
Both systems use the inductive sensor (master-sensor). Inductive RPM sensors generate voltage so a VOM can be used for testing. For the EZF, it is connected at pin 19 and 7, where pin 7 is connected internally in the ECU to pin 12 (master-ground). Pin 20 is the shield of the sensor cable and is connected through internally to master-ground pin 12. For the EZK, pins 23 and 6 are used where pin 6 is connected to master-ground pin 18. Pin 24 is the shielding of the sensor cable and is connected internally to pin 18.
Hall Effect sensors can be tested with an LED tester. The EZK generates the Power supply for the Hall Sensor at pin number 5 (12V). The EZK generates a 5V Level at pin number 22, which the Hall sensor can pull down to ground. Pin 22 is the "hot" signal cable and therefore is shielded. The cable shielding is connected to pin 4, and pin 4 is connected internally to master-ground pin 18.
The EZx spark and LH ECUs are under relay switch power control. First verify that voltage is present at both LH and EZx relay pin numbers 30. There should always be voltage at pin 30, regardless of the ignition key position because a direct circuit is provided from the battery. If voltage is present, next jumper pins 30 and 87 and attempt to start the engine. If it starts, there is a relay or control signal to the relay problem. If it doesn't start, you may have an ECU problem.
The LH ECU controls power to the fuel pump. If there is a "no fuel pressure" situation, first verify that the fuel pump fuse is OK. If the fuse is OK, next check for voltage at fuel pump relay pin 30. If voltage is present, jumper the fuel pump relay (pins 30 and 87) and check for a running fuel pump. If the fuel pump runs, there is a relay or control signal problem to the fuel pump relay. If the fuel pump doesn't run when the relay is jumpered, there may be a fuel pump or wiring problem to the fuel pump. One of the symptoms of a failed LH module is no "turn-on signal" to the fuel pump relay.
For 1984 – 86 model year 928s, the fuel pump relay is the source for voltage to the fuel injectors and the LH module provides the fuel injection current path to ground. Therefore, a bad fuel pump relay can also result in no voltage to the fuel injectors. The fuel injector voltage source was changed to the LH relay for model year 1987 and newer 928s. In 1989 an ignition circuit control was added as a safety feature to open the fuel injection circuit (stop injection) if no heat is found in the exhaust stream. Look for this module in the passenger compartment near the EZK and LH modules.
A simple way to determine if the “hot wire” Mass Air Flow sensor is working is to disconnect it and start the motor. If it runs better when the MAF sensor disconnected (in limp-home mode), then the MAF sensor is likely failing.
As crazy as it may seem, cars are different. A friend of mine had an LH module that would work in other cars, but not in his, so he spent many more weeks looking for the problem in other areas. The “suspect” LH module ultimately failed, even in other “test cars,” when only a few weeks prior it was running fine.
If you have some specific questions diagnostic questions, please feel free to send me a note at firstname.lastname@example.org.
The L and LH fuel injection modules provide the switch between the "-" side
of the fuel injectors and chassis ground. Even though L-Jet schematics
look like it might be possible for different fuel injectors to open at the
different times, that isn't the way that the circuits work. The analog L-Jet
system has all internal signals derived from of the input signal provided by the
transistor spark module.
LH systems have all fuel injectors ground ganged via the same input, thus it is impossible for injectors to open at different times. LH systems also derive fuel injection timing from the signal provided by the EZF or EZK spark modules.
More advanced fuel injection systems provide for sequential fuel injection, where each fuel injector may open at the optimum time for each cylinder intake - but those were not installed on the Porsche 928.
The electronic systems provide lower fuel consumption due to finer tuning and "off throttle" fuel cut-off provided by L and LH systems.
The LH box on 928's fires in batches of 4 injectors. Bosch just took a 4 cylinder LH box and basically doubled the amount of injectors it can run. Hence the ability of the '87+ to shut down 4 injectors at a time with LH 2.3 when there is an ignition problem on one side.
The LH system fires all injectors at the same time by grounding pin 10 on the controller. They have a mutual +12v line attached to the other pin. I checked my s4(1988) and some schematics of other models. I'm pretty sure the injector firing has not changed during the lifecycle of the LH controller used in the s4/gt/gts. So it is not a "LH controller" thing to manage. Fuel is injected into the intake room and sucked in when the valve opens. There's several injection bursts (1/8 per time) waiting before the valve actually opens and sucks in the fuel vapor.
The limp mode relay does come into play at the later models, cutting off the +12v to a part of the system (4 cyls) and therefore making it shut down
+the fuel supply. You want that because no ignition will make the engine drown in fuel, oil shear disappear rapidly causing wear, and a catalytic converter that will give a new meaning to "glow in the dark" (really!)
In further checking, it appears from all diagrams for all years that all
injectors are fired at once. Four injectors are tied to one point and four to
another, but all are tied together into the LH controller. On the 89 and newer
the two junction points of the four injectors are routed separately to the
outputs of the cut off relay. When one or both of the exhaust sensors detects a
problem with too much fuel in the exhaust (extreme temperatures) it cuts off
power to the injectors making firing signals from the LH moot.
Fuel pressure at the injectors is kept as constant as possible. When you hit the throttle in an LH car, the fuel pressure rises to keep the pressure differential across each injector constant as the manifold pressure rises (less vacuum). This is done by the fuel pressure regulators.
When I wrote about ramping the fuel up I was referring to the amount of time it takes to equalize the opening time. If all the injectors open at once to meet the fuel demand then the engine would flood. Remember the injectors open once for each cylinder that fires but inject only 1/8 of the fuel needed. The fuel then accumulates in the intake manifold until it is ready to be used by the cylinder. Yes the pressure can vary and thus change the amount of fuel injected but this is not going to buy much. As far as the injection responding faster than one can press the accelerator, there is always some latency in solenoids. They are mechanical devices and can only respond as quick as the spring can close the value and ready it to open again. If you listen to the click, click at idle you can get a feel for the number of times the injectors are opening.
(compilation from posts: Dan, Rich, Jon, and Theo)