I'm continuing this on from another thread I started about crank sensors as its now another "TD4 won't start" thread !

Basically I replaced the clutch and flywheel on my TD4 (2003 on a 52, manual, 118k miles) which started perfectly before I touched it, and now it will not start. I dropped the engine and box out to do the clutch change and have been really careful to make sure everything got connected back up, and connected up properly. The engine will now turn over fine, and it will occasionally fire, but it won't catch and run.

I've got a Foxwell NT644 code reader and the live readings are as follows:

Ignition on (not cranking, LP pump silent)
Airflow 3276.7
Manifold 101.4
Low Pressure 100.0
High Pressure Rail 0.0

Ignition on (and cranking)
Airflow 362
Manifold 100.7
Low pressure 334.1
High Pressure Rail 31,729 (I did get a higher reading earlier today - about 44k - but its dropped now)

One thing I noticed is that when I first turn the ignition on, the LP pump at the rear whines (hums ?) for a while (less than a minute) - kind of reminds me of my old Triumph 2.5PI While that is happening I can hear what I can best describe as a "fizzing" sound coming from the area of the HP pump. When the LP pump shuts off, the fizzing at the HP pump stops. More importantly, whilst the LP pump is running, the live reading is at 375. This then drops to 100.0 when it stops, and goes to around 334 when the engine is being cranked.

Other things I have tried:
* I've done a leak back test - all injectors are fine
* I've removed and cleaned the Crank Sensor
* I've replaced the Cam Sensor (with one from, I'm assured, an engine that was running)
* I've replaced the HP Regulator on the back of the pump (with one from the same engine above) Note the fizzing from the HP pump was there before I changed this - its made no difference.
* I've checked the HP Rail sensor connections (all look fine, no corrosion) and cleaned with electrical cleaner
* I've checked the ECU connections (removed when engine taken out)

Whilst the engine was out I removed and cleaned the inlet manifold and rocker cover (decoked with oven cleaner etc), I removed the injectors and cleaned them externally and resealed them with new copper washers and O rings. I've deleted the EGR (stainless tube replacement, blocked the vacuum pipe, blanked off the exhaust port so the whole EGR has been removed). I also replaced the loom from the ECU to the injectors as one connector had originally broken off and been "bodged" to work. When it wouldn't start I tried the old loom back on in case that was the issue, no change though.

Phew !

Is there anything in the above detail that might help diagnose why the bleepin' bleeder won't start ??

Have you changed the O rings in the HP regulator? It's not uncommon for a split O ring to stop it starting. However it would be a real coincidence for the regulator O ring fail during a clutch change.

Nodge68 said:

Have you changed the O rings in the HP regulator? It's not uncommon for a split O ring to stop it starting. However it would be a real coincidence for the regulator O ring fail during a clutch change.

Click to expand...

cheap and easy one to eliminate though

Nodge68 said:

Have you changed the O rings in the HP regulator? It's not uncommon for a split O ring to stop it starting. However it would be a real coincidence for the regulator O ring fail during a clutch change.

Click to expand...

cheap and easy one to eliminate though

I've not changed the o ring, just put the reg on from an apparently running vehicle to see if that helped. I think replacing the o rings and maybe an HP pump rebuild are next on the list, but I was wondering what the HP pump operating parameters are ? No point rebuilding it yet if its within operating spec. Yes, it would be a coincidence but the car had been standing for about 6 months after the clutch failed as the previous owner couldn't get it fixed and lost interest. It started and seemed to run OK when I got it, but I never got to drive it (virtually no clutch). The seal may have been on its last legs. But it started on the button each time before I took it all apart.

I guess as I'm getting readings on the diagnostic tool, that means the HP rail pressure is being measured and transmitted to the ECU by the pressure sensor/wiring - so that should rule that out ?

Its a bit like trying to start a petrol engine with the distributor 180 degrees out, or the timing just being way out. Is it the cam sensor or the crank sensor (or something else) that dictates the timing of the injectors ?

Sounds like you've covered most things. What about swapping the crank sensor? It could have been knocked during flywheel replacement.

Would the crank sensor stop the engine starting ? Does it give a continuous signal to the ECU while the engine is running, or just at start up (as I understand the camshaft sensor does) ?

I've just put the cars history together, and one thing I noticed is pushing me down the "rebuild the HP pump and regulator" route. Just over 2 years and 10k miles ago someone obviously filled it with petrol by mistake. Theres a bill for having the fuel drained, then a week later a bill for a new LP fuel pump and filter (looks like a local garage, not LR or specialist). Don't know how long it was driven for or any other details. I wonder if that would have damaged the HP pump ?

According to RAVE, the crank sensor is essential for injection timing, and is used during running, while the cam sensor is just used during starting.

For proper running, the sensor must be placed at the correct (non-adjustable) distance from a reluctor ring on the crank. The ring has teeth all the way round except for a gap for synchronisation. Does the rev counter move during cranking? I've heard this can be a crude way of confirming crank sensor function.

Again, from RAVE:

Supply pressure below 0.5 bar - HP pump shuts down.
... Below 0.8 bar - limp mode

You appear to be getting 334 kPa, or 3.34 bar, so this should be fine.



Rail pressures: minimum - 200 bar (2900 psi) for start-up and 300 bar (4350 psi) at idle. Max pressure 1300 bar (18850 psi).

Annoyingly, the units are all different. In kPa, the figures above are 20,000, 30,000 and 130,000 kPa.

You seem to have over 30,000 kPa at crank so that should be fine.



How sure are you that the crank sensor setup or its harness aren't damaged? Any codes related to it?

Thats the type of info I was looking for ! That's brilliant, thanks guineafowl21 !

What is RAVE and can I access it or do I need to subscribe ?

It does look more and more like its the crank sensor - I didn't notice the rev counter, I'll check it.

The rev counter does not move at all whilst cranking. I will investigate further, fingers crossed this is it

Whistler said:

Thats the type of info I was looking for ! That's brilliant, thanks guineafowl21 !

What is RAVE and can I access it or do I need to subscribe ?

It does look more and more like its the crank sensor - I didn't notice the rev counter, I'll check it.

Click to expand...

I think @GrumpyGel has a decent link to RAVE - it's the proper LR workshop manual. I can't remember where I got mine, but it's a downloadable CD image that runs as a pdf.

As I understand it, if the rev counter moves you have an engine speed signal (from crank sensor via CANbus) but if it doesn't, it may mean the rev counter is disabled until the engine is running. Systems vary, so don't get too excited.

If swapping the crank sensor, remember that the harness or connector may be damaged. The harness is a special shielded affair - if damaged, a good signal may be lost in noise and crap by the time it reaches the ECU. You may need to scope the crank signal at the ECU to find this out. Good luck.

Ah, OK - thanks Is there any way to test the sensor now its off the car ? It only has two pins, should there be continuity between them ? I'm getting a resistance of 0.56 using a multimeter.

How would I go about scoping the crank signal at the ECU ? This sounds like the best idea but I'm assuming its beyond my multimeter ! I've had a look at the harness - it looks the same as the other harnesses in that area (the only shielding is a corrugated plastic outer jacket which all the other cables have). No obvious signs of damage, and I didn't disconnect this when I dropped the engine, I just unplugged everything at the ECU.

I do feel this is the right track though, like I said its like everything is there but the timing is out - which seems to point to the crank sensor.

Whistler said:

Ah, OK - thanks Is there any way to test the sensor now its off the car ? It only has two pins, should there be continuity between them ? I'm getting a resistance of 0.56 using a multimeter.

How would I go about scoping the crank signal at the ECU ? This sounds like the best idea but I'm assuming its beyond my multimeter ! I've had a look at the harness - it looks the same as the other harnesses in that area (the only shielding is a corrugated plastic outer jacket which all the other cables have). No obvious signs of damage, and I didn't disconnect this when I dropped the engine, I just unplugged everything at the ECU.

I do feel this is the right track though, like I said its like everything is there but the timing is out - which seems to point to the crank sensor.

Click to expand...

The sensor is a two-wire variable reluctance type - this means it's a coil around a magnet, whose magnetic field extends towards the tone ring on the crankshaft. As the tone ring rotates, the teeth and gaps have different resistances to magnetic fields ('reluctance'), so the magnetic field near the coil changes. This generates a sinusoidal voltage between the two pins, like a little generator. The sensor is therefore not externally powered.

If you have a code reader, does this not give an engine speed PID? I suppose some reset during cranking, which isn't much help.

Your meter will be measuring the coil, so 0.56 ohm sounds a bit low, bearing in mind the resistance of the probe leads. The only way to be sure is to measure a known-good sensor as RAVE doesn't give details here. A resistance of about 500 ohm would be more likely, but I can't be sure. You may have a shorted crank sensor.

If you set the meter to AC volts you might detect the signal (up to 2 V, I think), but a normal multimeter will only average the output and not give an idea of noise. An actual scope would be better. The screen would be a woven mesh inside the actual cable, rather like coaxial cable for a TV, that is earthed in the ECU. The easiest thing would be to get a good sensor, test it for resistance and try it in the car.

If you find someone with a scope, the crank sensor input is between pins 6 and 31 of the middle ECU connector.

Thanks for all that guineafowl12 !

Sadly I had a serious sense of humour failure with the Freelander yesterday so took a night off and only ventured back into the garage this evening. I took the crank sensor out and put a meter over it and measured the voltage when I waved a large spanner across the end (you could feel it in the magnetic field). I couldn't read the voltage, but it was definitely making current. I then tested the continuity from the socket terminals back to the plug that goes into the ECU. Perfect, no breaks. So I am ruling out a crank sensor issue.

So, with a fully charged battery I plugged in my diagnostic and span the engine. The engine RPM was shown (around 255). Again, I think this rules out a crank sensor issue.

Then I looked at the fault codes. I thought I'd already checked this, but there were 3 showing:

0100 Air Flow Meter
1E25 No Description Available
1620 Security

Thinking these might have been stored from some time ago (even though I thought I'd checked and cleared everything down previously), I cleared them down and span the engine up again, then rechecked the codes. The same codes were back.

Any ideas where I should be looking to resolve these ? The security one makes me think the immobiliser is stopping it starting, but then I thought I read the immobiliser stops the starter motor turning ?

From some research, code 1E25 appears to be a cam sensor code.

From memory, the air flow sensor and cam sensor harnesses join and run together. Could they have been damaged? The MAF won't stop the engine starting, but the cam sensor will. Might it be worth checking continuity on the cam sensor harness?

Pins 4 and 17 of the middle ECU connector. I think it should also have battery voltage when key is on II.

Another evening spent in the garage

First thing I did was change the battery in the alarm remote. I locked and armed it, then unlocked it. I plugged in the code reader, cleared the codes and then attempted to start the car. Still no joy, but after a few minutes of cranking I then read the codes again and only the 1E25 had returned. Not sure why the MAF one has gone, but I think the security one was because the battery was almost dead in the remote.

Anyway, next I traced the wires from the MAF and Cam sensor and they do run in one bit of cable. No damage anywhere, so I tested the continuity on the yellow and brown wires from the cam sensor to the ECU plug - perfect, no issues (even waggled the loom about to check). Also tested the voltage from the red wire to ground - 12.4v with the ignition on. So I don't think its a harness issue.

Tried to start it again - nothing. So I put the original Cam sensor in (I'd swapped it out for one from an alleged "running " TD4 in the scrap yard). Tried it again and it was firing a lot more, but still failing to catch and start. So I reckon my scrap yard cam sensor is faulty ! I'm going to buy a new one and try that - hopefully I 'll be able to report a running TD4 shortly.

Oh, and there was a puddle of coolant under the car - the rad is leaking !

Someone called - but I was away in my Freelander for a couple of days and didn't see.

There's a thread in the Freelander Technical Archives for Rave...

https://www.landyzone.co.uk/land-rover/rave-disk-maintenance-manual.260227/

Thanks for the link, I will have a look

I bought a new, genuine LR cam sensor which arrived this morning. I went straight out and fitted it, held my breath and turned the key. It spun up and tried to fire - didn't catch
Silently cursing I tried again, mentally running through how much money I'd lose if I just scrapped the Freelander - and it started So much for the assurance the TD4 in the scrap yard was running when it came in, the cam sensor I bought from there was obviously duff.
Anyway, I let it run for about 10 mins, then checked the error codes again and there were loads - just about everything you could think of. I cleared them down and started the car again and it ran a lot better. I let it run right up to temp and then read the codes again - no codes recorded . I've spent the rest of the day doing other jobs on it, engine running a lot of the time, and no more codes recorded. I just need to check the coolant again tomorrow to see if its leaked again.
I'd just like to say a heartfelt thanks for the help and information from the site and contributors (especially guineafowl21 ). I don't know where I'd be now without the encouragement.
Its not over yet, I've got a list of other jobs to sort still, but somehow it feels doable now the engine is running.

Excellent stuff.
The two most annoying things in mechanics:
1. Intermittent faults
2. 'Known good' parts that are duff.