I have a 1988 EFI 3.5 Range Rover, which has developed the sort of
fault that drives one to drink (well, I can't drive the car so I might
as well)
It starts well from cold, and runs fine. When hot tho' it doesn't
want to know. Leave it 15 minutes and it will start again instantly.
I've changed the ECU for a spare and the symptoms are unchanged. I
suspect the answer is quite simple - once one knows what it is. My
suspicion is a sensor related to temperature is telling fibs to the
electronics. The question is, is this a fault that is familiar to
anyone, and can someone suggest which component(s) to change first
(and where they are)?
Oh, one other point! there is a cold start injector on the plenum
chamber. When I had starting problems a couple of years ago (finally
traced to unclean connector on the ECU) someone on here told me to disconnect it.
I did so, and the thing has started perfectly ever since - until now.
I've also taken the obvious step of cleaning the ECU connector again.
Any ideas, anyone????
Yn amlwg nad ydych yn deallt Saesneg, felly cesio yn y Gymraeg ichi roid manylion y peiriant a'r blwyddyn gwneuthuriad.
Edited by Pugugly on 09/04/2008 at 23:03
|
pmvw
A bit of testing would have been better than replacing parts - particularly the ECU.
When it won't start, is there a spark?
Does the fuel pump run when cranking?
What's the resistance of the ECU's coolant temp sensor when hot?
[That's the 2-wire one in the N/S/F of the inlet manifold.]
|
Ah, but then I'm an old-fashioned Engineer, testing by substitution is a good first step. I have a spare ECU, so it wasn't difficult to swap them.
Inadequacy of information - my fault. There is always a spark. I haven't checked the pump when cranking - but it works when cold, and it works when left for 15 minutes so it didn't strike me as the most likely fault.
On subject of temp sensors, I got hold of a Haynes manual today. There appear to be two. The coolant temp. sensor (front of head, facing forward, one wire) has a temp profile defined in the book. The measured resistances are lower than they should be according to the book. The other is a "thermotime switch" vertically in head just in front of front injector. Testing it according to book it appears to be faulty, it has a resistance of 45 ohms all the time, and should be open cct when hot. However, if this were the problem disconnecting it when engine hot should reproduce its action and the engine start - which it doesn't.
Also, I went to the local RR dealer lunchtime and they had never heard of a "thermotime switch". - but as they also wanted to relieve me of £45 for the coolant sensor (price from my usual source £10) I am not wildly impressed by them
So! Can anyone tell me what a thermotime switch is, and whether a coolant sensor innacurate by about 500% (it measures 60ohms at 80deg C when it should be 330ohms) would be sufficiently in error to stop the engine starting?
|
pmvw
On a 3.5 EFi; the commonest causes of hot starting faults are that coolant sensor, the tank pump and a lack of heat-sink compound under the ignition amplifier on the side of the dizzy. Although, as you always have a spark; then we can rule out the amplifier.
By no means can the fuel pump be eliminated by assumptions like that. They are perfectly capable of refusing to spin when hot. Not hard to check; you can hear the fuel returning through the pressure reg on the end of the fuel-rail.
The ECU's coolant sensor should have a 2-pin square plug with a silly wire clip. As a rough rule of thumb; they should be around 2800 Ohms cold and 280 hot. I've never seen one stuck at 65 before. The thermotime switch runs the cold-start injector for a few seconds at start-up - very rare to see one.
Other than that; the little rev info resistor on the coil needs to be tightly connected and the engine control and fuel-pump relays can't be ruled out either. They are unique types, although they look normal.
|
Suggest check the engine speed reference relay that is located next to ignition coil.Usually mounted in a plastic terminal block.If this fails you will get an ignition spark but the fuel system will not fire up.hth
|
If I had any hair left, I?d be tearing it out!!!
The symptoms are unchanged. When cold, it starts almost instantly. When hot (15 minutes ticking over in drive) if ignition turned off and then immediately key turned it will probably start again. Wait 5 seconds and it won?t start for about 10 minutes. If left for 10 mins or longer starts almost immediately again.
ECU has been changed. Coolant temp sensor changed. Engine speed reference relay checked. There is pressure in the fuel lines when it won't start. When being cranked over there is a spark of at least 1/4?.
Oh, and its probably a red herring ? but the rev counter has just stopped working.
Anyone got any further suggestions ? or a match!!!
|
Does the fuel pump run when it cranks hot?
|
should have been clearer, pump is pumping when engine is being cranked.
It gets worse.........
the rev. counter has started working again(!!!), but now the alternator has packed up. At least, the ign. warning light stays on so I'm assuming its the alternator. I've checked all connectors around it and they seem OK.
|
The good spark that you have is that out of the coil king lead or at a spark plug lead?(possible rotor arm fault).The tacho on these is driven from the alternator;no charge =no tacho.Check wiring to alternator,it may have been disturbed when the coolant temp sensor was replaced.Also check the Efi system earth leads.These are bolted to the rear of engine block ,nearside.hth
|
Spark is at plug lead. I noticed earler today that the rev counter was behaving very erratically while the engine was cranking over. Now that the alternator has apparently packed up the rev. counter is working perfectly.
All alternator wiring has been disconnected, checked and returned. All seems OK (ir worked fine after the sensor had been replaced) The ign. lamp is exhibiting what I think are typical fault symptoms - the lamp gets brighter as the revs increase. Substantial earth leads on the back of the block are also OK.
I think I need to give up for the day, time for a visit to the pub.......
|
Any sign of low oil pressure warning light?
I had a similar problem yrs ago on 1990 3.5 Discovery - stalling preceded by clicking sounds from region of passenger footwell. Basically the same car.
Oil pressure sensor was faulty, wrongly indicating low oil pressure. This shuts off power, via a relay (which is in side panel of n/s footwell), to fuel pump, engine stops. If your oil is thinning as it warms, this could be the cause.
Just a thought.
|
|
cuts out when hot? take the oil cap off when running and put your hand over & see if it has a slight suck to it, I suspect it may have an inlet manifold gasket failing causing it to run too lean when hot. When cold the enrichment compensates for a leaking inlet manifold. If it has a slight suck suspect the manifold leaking.
Have you checked fuel pressure? has the filter on the chassis been changed?
I asume you have checked the rubber air intake pipe fron airflow meter too manifold as these do split wide open ?
IDoc
|
I hadn?t gone away, I just had other things that needed repairing more urgently??..
However, I may have a possible solution in the form of the Fuel Cut-Off Valve. It was suggested to me that these can ?stick? when hot. After removing a couple of hoses and some skin I've reached the fuel regulator to which is attached what I think is the valve. It has two electrical terminals, and a vacuum pipe connected to the inlet plenum. I?m thinking it is actually a sensor, not a valve, detecting inlet manifold depression, switching when there is a high vacuum and signalling the ECU to turn off the injectors to save fuel.
The vacuum pipe is also attached to the back of the fuel pressure regulator, presumably to allow for increased(?) fuel pressure when the engine is under load and there is no manifold depression.
Therefore, if this thing is "sticky" when hot as suggested, when the engine is turned off hot it switches off, but then won't switch back on until it has cooled down a bit. So, if I short out the wires that go to it, I am simulating the switch closing and the ECU will energise the injectors.
Can someone confirm if my reasoning is correct ? before I short out the two wires and blow up the ECU????
Thanks in advance
|
pmvw
If you do have that, very short-lived, system; then there will be a relay clipped to the air-flow meter's bracket. [Don't confuse it with the EVAP valve, the air-con idle-up valve, or the heater flap vacuum valve.]
Simply test for a live at any injector plug when it won't go - thorough testing will find this, not guesswork.
|
There is a relay clipped to the air-flow bracket - I've tested it and it works. It would be just my luck to have a unusual system!!!
Its easy to say thorough testing will find it, but I'm rather short of suitable test equipment - I may be able to get my hands on a scope to look for injector pulses next weekend - but at the moment its parked in the road at the end of the drive and all I have is an Avo 8. The only easy means of testing I have is by substitution or experimentation. I'd prefer to describe it as reasoned analysis, rather than guesswork!!! :-)
What I was after was confirmation that my understanding of the function is at least sound, in which case shorting out will confirm whether the switch is at fault or not. If not, then I've lost nothing and and its one other possibility eliminated.
|
Sorry if I appear harsh; but after 35 years of diagnosis, I know that there is no substitute for first acquiring test data upon which to base the reasoned analysis.
Sophisticated equipment is nice, but not vital; I've done 90% of my testing with a simple test-lamp; [although knowing where not to apply it is my advantage.]
When the engine won't fire; the evidence is there for the testing. Compression; spark; fuel. As you are happy that you have a good spark at the plugs and temporary compression loss is rare; then the injector function is the next focus.
A scope makes it easy; but a noid light from a tool supplier will make it visible - albeit with less detailed information available.
As far as the OFSD system goes; it interrupts the engine speed-signal circuit between the coil and the ECU. The ECU would not know that the engine was turning and would not activate the fuel pump and injectors - and the fuel pump is apparently active during cranking?
To disable it; remove that relay and bridge the white/blue wire on terminal 30 to the white/black wire on terminal 87.
On a different tack; have the heads ever been skimmed? I'm wondering about lifter pre-load?
|
|
Just bought a 1988 classic that had this problem. There is a sensor at the back of the engine that has 2 wires going to it. These wires come from a relay that is positioned to the right of the engine bay (sitting on its own). I do not know the tech bit about this sensor, but apparently it was a short lived mod and now is not on later models. To cure the problem I just shorted out the 2 wires as the sensor is normally closed. If this sensor prematurely goes open circuit it cuts off the fuel supply. Have tried to find a replacement sensor but they do not seem to be available .
|
|
|
|
