Which Temp Sensor

Hi all,

Sorry if this is a stoopid question. . .but: I have very strange happenings with my '97 3.9v8- i have to reset the ECU (usually by unplugging the Air Flow Meter) at the start of every journey, otherwise it is basically undriveable. Once reset it runs sweetly for the rest of the journey, until i turn it off again. I reckon it's to do with temperature as I've already replaced the Air Flow Meter and throttle Potentiometer, but with no avail (old ones for sale if you're interested). So, i'm now thinking Coolant temperature sensor, or Fuel Temperature sensor.
The idle when running on petrol is also pretty wonky (Flucuates a lot), altohugh it's pretty smooth on LPG.
I've bought two replacement coolant sensors, but niether one looks like it fits.

#1: Facet Coolant Temperature Sensor LAND ROVER DISCOVERY 3.9 V8 89-98 | eBay

#2: Facet Coolant Temperature Sensor Land Rover DISCOVERY 3.5 (SALLJG) 89-98 | eBay

I thought it went kind of behind the dizzy, to the left abit, but the sensor there has a different connector to both i've ordered so now i don't know where to fit them (or if i've got the worng one's), and i keep finding contradicting diagrams.

Having some difficulty uploading pics- can anyone tell my where the sensor lives? Sorry for such a basic question. . .

Cheers,
Steve

Hang on. . .I may have found it. :doh:

Does it live behind the dizzy, pointing directly down?

Sorry for being a pleb!

The lower link item is for a 3.5.
For a 3.9 u require either PRC8003 manual or PRC8783 automatic according to my parts book.

director_steve said:

Hang on. . .I may have found it. :doh:

Does it live behind the dizzy, pointing directly down?

Sorry for being a pleb!

Click to expand...

yes.

If you think that it's a coolant sensor then warm the engine up and disconnect the coolant sensor. When the ecu sees an open circuit on the sensor, it will assume a default temperature. The engine will not be running at optimum fueling but it should run fine.

When you say reset the ECU by usually disconnecting the A/F meter, can you achieve the same result by disconnecting another sensor?

Hi,

To be honest, I think the only way I can reset the ECU is via the Air Flow Meter- I've tried the temp sensors, but had no effect. I haven't tried it with the Throttle Pot, as the connector's hidden beneath the Plenum and annoying to get at, but replacing it has had no effect. I do know the ECU is rewired to run without Lambdas, and it was running perfectly until this started to happen.

Since I replaced the coolant temp sensor I still had running problems, but i'm currently running it with the Air Flow Meter unplugged- it's lumpy when it's cold, but when warm it runs really sweetly. So i'm at a bit of a loss. A garage said it might be the Crank position sensor- but I've no idea where that is. The same garage spent ages trying to wire diagnostics into the ECU, but with no luck. . .

At the moment i'm contented with it running how it does- when warm it's pretty much spot on (to me). Then again- I know nothing (full pleb).

Cheers,
S

discool said:

The lower link item is for a 3.5.
For a 3.9 u require either PRC8003 manual or PRC8783 automatic according to my parts book.

Click to expand...

Thank you!!!

One thing that it's not is the crank sensor as the 3.9 did not have one!.
I had the same problem as you with my 3.9 and the engine was fine when the airflow meter was disconnected as the 14CUX ecu then uses the throttle pot to control the fueling. Disconnecting the AFM does not actually reset any thing, but I'm pretty sure what is happening is the ecu sees a fault when you disconnect the AFM and then it ignores the AFM signal for the rest of the ignition cycle and relies the throttle pot to control fueling. When you cycle the ignition the ecu will then start to use the signal from the AFM and presumably it runs badly. I know that you have changed the airflow meter, but if this was a 2nd hand unit then it's faulty! Can you borrow a good one from somewhere?

Disconnecting the vehicles battery will reset the EFI ECU.

Thanks both- i've got a lot to learn about this engine! I haven't disconnected the battery yet, but i'll do that this weekend and see if it makes a difference.

The Air flow meter was a brand new OEM equivalent, so I'. be surprised if it is that, but at this stage anything is possible!

Found this site which might be useful for you.
Lucas 14CUX Fuel Injection System ? Installation and Diagnostic Notes.

Try running the engine and when warm, disconnect the AFM and drive for ~50metres and see if it drives as well as when you do a reset with the AFM. I still suspect that the ecu is running without the AFM info when you "reset" the ecu.

Good idea!

Thanks for the link. Some worthwhile bedtime reading methinks.

S

director_steve said:

Hi all,

i have to reset the ECU (usually by unplugging the Air Flow Meter) at the start of every journey, otherwise it is basically undriveable. Once reset it runs sweetly for the rest of the journey, until i turn it off again. I reckon it's to do with temperature as I've already replaced the Air Flow Meter and throttle Potentiometer, but with no avail (old ones for sale if you're interested). So, i'm now thinking Coolant temperature sensor, or Fuel Temperature sensor.
The idle when running on petrol is also pretty wonky (Flucuates a lot), altohugh it's pretty smooth on LPG.

Click to expand...

You can't reset a 14 CUX ECU by disconnecting the air flow meter. You need to unplug the ecu or disconnect the battery. You don't need to wait, there is no volatile memory in a 14 CUX, the "reset" is immediate as soon as the power is taken off.

I guessing you have cats in which case the air flow meter is not likely to be the problem because for the most part the ecu ignores it and trims the fuel mixture using the lambda sensors so you might want to check/test the lambda sensors. I would also suggest checking the ignition timing.

If you have been disconnecting sensors (MAF) you might find your ecu is in limp mode so will just be dumping fuel in so it runs. You mention LPG - I don't know anything about that but perhaps get it serviced.

You didn't say what the running problem actually is (or did I miss it?) some more info might be useful. HTH

Hi Kev,

No cats or Lambdas, but ECU rewired to accept that. Running problem was that it would run fine for about 10mins, then become extremely lumpy, misfiring and stalling. I would then unplug the AFM for a couple of seconds, then replug and it would run smoothly for the rest of the journey, until I turned the engine off again. I have found that if I leave the AFM unplugged it runs really nicely, and my fuel economy has improved, so i'm at a loose end, but not an urgent one.
LPG system is new, and properly set up- I've checked- and it runs beautifully anyway.

Any ideas much appreciated.

S

director_steve said:

Hi Kev,

No cats or Lambdas, but ECU rewired to accept that. Running problem was that it would run fine for about 10mins, then become extremely lumpy, misfiring and stalling. I would then unplug the AFM for a couple of seconds, then replug and it would run smoothly for the rest of the journey, until I turned the engine off again. I have found that if I leave the AFM unplugged it runs really nicely, and my fuel economy has improved, so i'm at a loose end, but not an urgent one.
LPG system is new, and properly set up- I've checked- and it runs beautifully anyway.

Any ideas much appreciated.

S

Click to expand...

You need to check your air flow meter. If you have a decent DVM you can check the air flow signal and CO trim. Do a search I've posted on testing AFM before as have others IIRC. The other thing is that it could actually be the ECU. I had a similar problem some years back, the ECU wasn't reading the air flow signal, it took me ages to find this out, even had the ecu and AFM tested and was told both were fine (ECU wasn't), still on original AFM but swapped out ECU and not looked back. If you can borrow another ECU it may be worth doing a substitution test but first check/test AFM it isn't difficult and only takes a few minutes.

Of course there are other potential causes but no harm in starting with air flow meter and ECU then go from there.

Thanks Kev- much appreciated!

No problem - I've been there. It took me 12 months to get my Rangie to run properly when I first got it.

One other thought, and you might have seen I say this a lot - have you checked for air leaks? I would strongly recommend taking off the plenum and ram housing and giving both a thorough clean then re-assemble making sure the plenum is properly sealed. While you're doing that, check/set throttle disc, clean all breathers and vac lines/ports.

Once you get it back together, fire it up then check the ignition timing, to start with I think I would set it for petrol then set the base idle. Provided it then runs correctly on petrol you can alter the ignition to suit the LPG speaking of which it might be a plan to have the lpg system serviced.

If problem goes away when you leave the AFM connector off and also when you disconnect the plug and plug it back in, then I believe you issue is caused by a fault on the AFM “circuit”. I’m more familiar with other injection ecus and these will lock out the AFM for the duration of the ignition cycle if they see a serious problem with the AFM. I suspect it’s the same with the 14CUX.
Now ECUs will detect shorts to ground and open circuits on their inputs, but also do a plausibility check to see if the signal looks correct compared with other signals such as engine speed and throttle angle. When you disconnect the AFM, the ecu is detecting the o/c and therefore ignoring the AFM for the rest of the ignition cycle. It then uses the throttle pot signal. When you start your vehicle up then allow it to warm up, it runs badly but the ecu still sees a AFM signal that it considers to be in the right ball park so it does not ignore the AFM and assumes it to be correct, but it sounds as though it actually running very weak.
If you have replaced the AFM then, we have to assume that is ok, that leaves the wiring and the ecu. One common problem is water getting into engine bay connectors and its then drawn through the harness with capillary action into the ecu. If this is happening the copper wires will be green. If the wires are green open up the ecu and check to see if there are signs of corrosion or dendritic growth. These can lead to small currents flowing between adjacent pins on the pcb and this could lead to the ecu seeing an incorrect AFM signal.

Update

Ran it with the AFM plugged again today (started the engine with it unplugged, then plugged it in) and it made a massive difference! Much smoother, much quicker to rev and generally a nicer drive. I've inspected the wiring and it all seems to be in good condition- i can't find any green wires and the connections all seem good. I haven't inspected the Ecu yet as it means taking the seats out (they're comfy 1/2 leather rover items) to get to the battery compartment where the ECU is.

So i'm back to square one.

I understand you can test an AFM with a multimeter. Could someone enlighten me as to how and what readings i can expect?

Cheers,
Steve

Another update-

She was driving really nicely again tonight (covered quite a few miles), even after switching her off (which normally knackers everything) until i hit a humpback bridge a little enthusiastically and the air filter fell off (it's a rubbish after-market thing that falls off frequently). Once it had fallen off it was driving like sh*t again, until i put the filter back on AND unplugged the AFM for a few seconds (i tried unplugging it for just a second, but it made no difference- it has to be a bot longer).

Also, i noticed there was a heap of play in the throttle cable, which i sorted out.

Could it be, after all this time, that my problems come down to a crap air-filter that keeps falling off and sending the ECU into Panic mode, compounded by the fact the thtottle cable was really loose?