LR3 2006 4.4ltr, fuel issues which Land Rover cannot diagnose

[shedlock2000]

Sorry for the long post which follows, but I think you need all this information to help out! I am having a protracted issue with my LR3, 2006, 4.4ltr, LHD, HSE, 120,000miles. She unfortunately became wet in a recent flood event in my town and drew in water through the fuel filler cap (The car was running and being loaded to exit the site when the bank side slipped away. The insurance won't cover any issues and I am a (now) very broke student). The inside of the vehicle remained largely dry, but she was running until she stalled from water in the fuel.

The water contained a soft fine clay particulate which clogged up the fuel pump went though the filter and into the engine. The injectors were clogged but have been replaced. The fuel tank, pump, filler cap, and evap. lines have been changed with spares from a running 2006 (fitted with a recall tank upgrade -- a straight fit, 2010 tank and pump). The battery box has had no water ingress, and I have removed the transfer case module and ECM to inspect for water ingress in the plugs or module (which there was not). The E-box fan shows as 'off' in the code reader, but a false feed runs up the fan fine. The fan fuse is good.

The problems are these:

• After an hour or more of driving, or about 20 minutes of idling in hot weather, the vehicle begins to run progressively leaner and eventually stalls.
• Switching off the key for 10 seconds resets the ECM and the vehicle runs fine for about 10-20 more minutes when the difficulties begin to slowly re-occur.
• Switching off the vehicle for 10 more seconds will again re-set the problem for another 10-20 minutes.
• The symptoms occur more quickly in the heat (especially the 35 degree heat we're having at the moment), but last night, with the ambient air at about 18 degrees, she idled for 2 hours without any difficulty then drove home perfectly.
• Opening the bonnet reduces the time which the issue occurs -- however, she will still begin to fail eventually.
• She seems to perform without issue much longer with a full tank of fuel -- though this may also be just a corollary, rather than a cause.

When the data streams are observed during the lean times, the short term fuel trims max out at 41.4%+ and the O2 sensors run to around 1.16mA (though Bank 1 is usually more extreme than bank 2 -- despite the fact that Bank 2 records more lean codes than bank 1). There are rarely any fault codes thrown -- even if the vehicle stalls, but occasionally a lean code for banks 1&2 are given, or (very rarely) a misfire code. When the vehicle is running normally, fuel trim figures hover around 2 or 3-4% and bounce around as you might expect, and the O2 sensors report 0.01mA and flick around properly. Sensors progress to the former slowly and progressively, and correlate with the progressive lack of power one expects from a shortage of fuel. The O2 sensors have been disconnected and codes reset then a test run has been made under which the vehicle failed again after about 10 minutes of idling and 25km of driving. From this I conclude that it cannot be O2 data which is being improperly managed by the ECM.


Work done has included:

• -- Replaced O2 sensors and injectors,
• -- Blown out fuel lines,
• -- replaced pump and filter (and fuel tank).
• -- Cats. inspected and found to be in excellent condition.
• -- Battery box, Transfer case module, and ECM inspected for water damage.
• -- Inlet manifold checked for leaks; the evap. system has been isolated from the engine to eliminate that.
• -- Fuel tank cap has been removed to check for vacuum issues and replacement fitted.
• -- Engine ECM has been re-mapped to avoid stored data glitches, vehicle taken to Land Rover (who have been unable to diagnose the fault but charged me $1000 for the privilege).
• -- Inside of vehicle checked for water ingress and the passenger (LHD) carpets removed to dry and clean out wiring in the sill.
• -- Battery voltage checked: good and consistent,
• -- Fuel pump electrical-feed plugs under front wheelwell checked for water/dirt ingress.
• -- Tank has been inspected for floating bags/plastic,
• -- Fuel pump removed and bench tested,
• -- O2 sensors disconnected for trial run,
• -- Smoke test carried out,
• -- E-box fan/fuse tested,
• -- Fuel rail pressure checked (80psi)

No existing fault codes show; all historical codes have been eliminated and all systems operate normally. Vehicle ECM has been hard reset prior and after ECM hot-soak.

When the vehicle is just started or cold, it runs perfectly and without issue. However, very, very occasionally (only twice ever), there is some hesitation to start which feels like a lack of fuel too (the key cycle does not start the vehicle but there is a faint try to start. Starting the car is possible on the second cycle).


Further information:
The MAF and MAP sensors record appropriately and within parameters; the coolant temp. is about 95, and the control module temp. between 49-62C; fuel rail temp fluctuates around 50-60C.

Fuel pressure when running well is 80psi (over the 65psi required), but I have not yet run the pressure gauge while the vehicle has been running lean, as this at first only happened on long runs. The new O2 sensors seem to have contributed to a faster onset of symptoms.

When the vehicle begins to show symptoms, it first presents as a lack of power on hills or accelerating, a general hesitation and lack of responsiveness, and later as a miss; when the vehicle becomes severely affected, there is no acceleration and maintenance of speed is not possible, there is then a lot of coughing and bumbling and the vehicle cuts out. Switching off the key and waiting 10 seconds starts the car as normal and no symptoms present.

I am guessing that it is a temperature-related sensor fail, but I do not know what sensor could cause the difficulties as the sensors involved in fuel regulation have been checked or changed: the MAP and MAF report fine throughout, the O2 sensors have been replaced, and the temperature sensors report properly and accurately. Clearly the ECM is being reset and the learned data is being deleted by switching off the key, but what information is the ECM getting which causes this issue?



Any suggestions, people?
I have been advised to:

• -- check the ECM fan (works fine, but is shown as 'off' by the code reader),
• -- check the fuel pressure during the lean periods which requires the pressure gauge to be continually inline (not possible until my mechanic comes back from holiday and I can borrow his fuel pressure tester). This would only indicate a physical lack of fuel to the engine, not the reason for the lack of fuel (which, I am assuming would also be ECM?)

[Graeme]

I would replace the crankshaft position sensor because some of the symptoms are typical of a failing one and its a sensor that hasn't been replaced. You could try cooling the sensor with an aerosol or water spray to see if the problem is quickly overcome.

[shedlock2000]

Interesting thought, Graeme. I shall just order one as they are not big $. Just out of curiosity, which of the symptoms are the same?

[Silenceisgolden]

I am guessing that it is a temperature-related sensor fail,

When you stop the engine, at first the temperature of the block will rise due to heat soak, so I don't think it is a temperature sensitive sensor at fault. During ten seconds the temperature wouldn't drop.

You seem to have done all the sensible things - hopefully someone here can give you a clue. Very tricky.

[Graeme]

Failing crankshaft position sensors often fail to work properly or work at all when they get hot, whereby even a little rest allows them to cool enough to work again for a while. Your engine's symptoms aren't exactly the same so cooling using a liquid, if access is easy enough, might be a good move. Alternatively if not expensive nor time-consuming to change, it might be simplest to just fit a new one.

[Cambo_oldjaguar]

You said you checked the fuel rail pressure and you got 80psi...

Was that checked with an external gauge or via OBD?

80psi is too high to me, the naturally AJV8 runs at 3.8-5.0 bar normally (55-70psi) and the typical delivery rate of the fuel pump is at 3.8bar or 55psi.

The fuel pressure sensor is a differential sensor that is referenced to manifold vacuum, but if you are looking at it via OBD it should be telling you 55psi +/- a few, but not 80psi, that's too high.

If the sensor is telling the ECU that there's 80psi at the rail then it will back off the pump flow, which could cause a lean condition.

The fuel pressure sensor is not on your list of items that were replaced.

Look into that.

[Graeme]

If the sensor is telling the ECU that there's 80psi at the rail then it will back off the pump flow, which could cause a lean condition.

I went looking for such a sensor in the WSM because I've encountered similar problems with such a sensor on another make, but didn't stumble on the sensor. Where is the sensor?

[Cambo_oldjaguar]

Oh wait, seems that the LR setup is different.

Runs at 65psi normally not 55psi.

And there doesn't seem to be a fuel pressure sensor in the parts list....

Strange...

The Supercharged 4.2L in the RRS & Vogue has the same fuel pressure sensor as the Jags, but seems that 4.4L is a totally different setup...

*now i'm confused*

But here it's mentioned there's a fuel pressure sensor/switch on the rail of the 4.4L http://workshop-manuals.com/landrove...ion/locations/

[Graeme]

I saw a fuel temp sensor there so perhaps its a combination temp and pressure sensor.

Edit: No, likely to only be a temp sensor judging by its low cost. Also a SC'd or turbo'd engine needs a pressure sensor relative to manifold pressure rather than atmosphere, so the pump itself may contain a pressure sensor.

[Cambo_oldjaguar]

The naturally aspirated 4.2L AJV8 in the Jags uses the exact same fuel pressure sensor as the S/C engines, so beats me why the 4.4L LR version doesn't have one.

I suppose there isn't a pressure sensor in the 4.4L it's just the regulator in the tank and the pressure is fixed. That's how it looks in this document ==>> [ame]http://www.disco3.co.uk/gallery/albums/userpics/15405/LR3_Fuel_System_Description_and_Operation.pdf[/ame]

Still wonder why they measured 80psi when the system is meant to run at 65psi though...that doesn't seem right...