My '03 Discovery has begun a sporadic fault. Symptom being the engine management fault light comes on and the accelerator pedal has no effect although the engine continues to run at tick over. This has happened twice now at about two weeks apart. First time just before Christmas on a busy dual carriageway during peak period. I managed to coast up onto the grass right out of the way of the traffic fortunately or some loon would no doubt have come to a sudden stop in my rear end. I called the RAC out and on arrival the fault had disappeared. the engine started and ran ok so we travelled to a nearby car park where the RAC guy plugged his "universal" comms laptop into the link but the Land Rover system would not talk to his set. He theorised that the fault was transient but would be in the engine management memory and a Land Rover agency could interrogate it. Having tried and failed to make an appointment at a reasonable time before Christmas the problem recurred on boxing day but I found simply turning off then on again was sufficient to reset the unit again. I am now due to take the vehicle into an agency next week but they were not confident the memory would still hold the fault. Have we anyone in our numbers who have knowledge of the engine management system on a Discovery 2.5 Diesel and is it feasible to get hold of the commes interrogation software to run on my own laptop?
the place I get mine ('02 Td5 ES) looked after (Rogers of Bedford -
formatting link
- excellent people, horrid web site) had trouble getting hold of one until a couple of years ago (presumably as a result of HP stopping development, as described on the web site).
A quick Google yields interesting results;
formatting link
there;
"Meanwhile, Ken reports that you can purchase Rover Test Book software (that will go on an ordinary PC) for US$250 at
formatting link
look under "Manuals - OBD". Bruno at
formatting link
says that they are working on a DYI kit for the VCSI and hope to have it out in June 2004. Jeff Noxon has designed a very simple RS-232 / ISO9141-2 optoisolator interface - the details are available free online. Ross-Tech is another good source of information on OBD".
A reasonable starting point would be the throttle potentiometer, assuming the disco uses such a thing as its throttle position sensor.
Might be an intermittent wiring fault or a dodgey pot (or a million other things, but you have to start somewhere). Try reseating the connectors. If you can ascertain that it is a simple pot, disconnect it and use a multimeter to see if it's giving sane and stable readings as you move the throttle over the operating range.
As you mention, the ECU *should* record its last error at least and Land Rover should be able to read this out. However, they might try to charge a full hour's fee (like 80 pounds or whatever) so another approach is to try a small garage or tuning specialist - such folk have offered to do a diag dump for me for as little as 15 pounds, until they found out their kit wasn't compatible with my mouldy old '98 Daewoo.
They have those? MM. I suppose they would have. Diesels don't vary air, just the amount of fuel injected. Used to be a mechanical cam on the injection. I guess a pot makes more sense.
I'd hazard a quess at corroded connections to the throttle potentiometer.
That, going OC would generally give you a warning light. Its about equivalent to a snapped throttle cable..
The ONLY major faults I have had with my defender - broadly the same guts as a disco - have been due to wire corrosion type things.
- water in fuel light stayed on. Replaced sensor
- fuel gauge packed up. replace sensor
- sometimes doesn't start. Wiggle leads on starter motor.
- lights inside sometimes don;t come on. Wiggle door switches.
I would be strongly tempted to see if you can find the pot and simply remove its wires, clean the connections and put some contact spray and silicone grease on them and plug it all back together.
I doubt the pot itself is defective..you would get irregular throttle response, not a complete cut.
I don't think you need anything more than common sense and a bit of logic and understanding.
UNLESS there is a wiring loom fault or a fault in the central EMS, only one thing fits the symptoms. An open circuit (or POSSIBLY a short) in the throttle pot circuit.
The fact that its intermittent as well fits that scenario 100%.
Simply find what the accelerator connects to, - should be some kind of electrical lump with two or three wires going to it - and pull off clean and replace the connectors.
It MIGHT be the other end of the wires that go from their to the ECU. Or in the loom between, but I doubt it.
Agreed. It *has* to use such..since the ECU fully controls the injectors. And therefire the throttle pedal must talk to the ECU and not anything in the engine direct.
I think not a million other things.
Consider. The engine went to idle (default safety setting) but otherwise ran perfectly. A light came on. Now that pretty much tells you that the ECU said 'I have an out of spec condition, I don't know what to do, and this is my best response'.
Since the engine didn't misfire, go smokey or in fact do anything untoward, one may conclude that by and large the fault wasn't with any other sensors like temperature and so in, or with an injector or the injector system. I.e. it wasn't with the feedback part of the engine control at all, it was with the primary input to the ECU.
Couple that with my experience that the worts feature of that era and type of landrover is crappy electrical connections, and I feel there is very strong evidence that its a corroded connector to the pot.
Now it gets speculative, because I am not intimately aware of the software in that ECU.
It is pretty standard practice to monitor all the resistances of the sensors (and they are all by and large variable resistors/thermistors etc) and to generate a fault condition when they go outside range. However whether this will result in a single instance of the fault, that then goes away causing the system to go into failsafe, or whether it will note the fault, light the lamp and *carry on* is a moot point.
I know that my Jaguar used to go into 'limp home' mode (50mph max and no bloody power) on a variety of faults - high gearbox temp was one. But stopping and restarting cleared the fault.
It may be that ANY instance of dodgy throttle pot on the Disco results in 'go to idle and stay there' it being deemed that to drive with a throttle sensor problem is potentially too dangerous.
Again, I am not sure whether the pot is wired AS a pot, or just as a variable resistor, but I suspect, motor manufacturers being what they are, that its as a resistor. Saves a wire dunnit?. Inspection would show whether its two wire or three, anyway. I would ALSO ssdupect that 'high resitsance=low throttle' is the way any sane person would do it, so that high resistance in contacts results in low throttle..we don't want the thing to kick into full chat if a wire falls off - even momentarily. Shorts are far less likely..
So, logic suggests its a two wire system, that is monitored for resistance that is too low, or too high. Mostly the latter.
And it went too high,. That MIGHT be a faulty/worn pot track, but 03 is a young car, and it seems fairly unlikely to have worn in that time. Alasos a worn track is FAR more repeatble..go to that throttle posuitin and it WILL show a problem in a large number ofcases. Its not doing that.
OTOH MY experience of the Defender 2.5 tdi, which shares most of the engine parts with the Disco is that dodgy connections are mandatory on these vehicles, It's their weakest link frankly.
I suspect that ALL that plugging in a diagnostic will do, is tell you that a high out of range throttle pot resistance was detected. That takes you nowhere. It just means that the garage can do what I have done and intuit a problem in that circuit. And charge you £80 for the privilege. All they will then do is simply replace the throttle pot anyway, because that is a guarantee that it's not that as a problem, clean up the connectors and stuff it all back together and hope that the £400 quid cheque doesn't bounce.
If it is an intermittent connection, and if as I suspect it takes but one instance to blue screen the ECU till its rebooted, as it were, then my pragmatic advice is to find the connections, do a thorough job of cleaning and putting water repellent stuff on them, and then run the car a bit .. it will always get you home, if you simply switch off and restart, until the time it doesn't, at which point the fault will be a lot easier to find.
Now, anyone know why my defender suddenly has got a long soft brake pedal? Just been serviced (always a mistake) by the very expensive dealers..who have in the long past under previous management managed to introduce expensive faults..its got plenty of fluid in it..so can't see how air got in..
Thanks for the pointers from everyone. I'm goiing to take a look for the throttle pot this afternoon whilst its still fine weather. I had suspicions that it might be pot as I "think" it is developing a slightly non linear response to my foot position although once you get an idea in your head its all to easy to convince yourself of something :-( The logic about high resistance is good thinking - if the vehicle designers are logical people?
It must have some form of TPS (throttle position sensor) even although a diesel. It might be a pot or a rotary encoder. My first checks would be to make sure the connections between it and the ECU are clean and secure. And to look at its output with a scope.
The trouble with third party software and hardware to read *any* codes is the makers deliberately change these at regular intervals simply to make things difficult for independents - unless it is one of the OBC codes which the US have insisted are universal. Or at least they have to be in the US. And of course they don't really have diesel cars there yet.
Yep - what I really meant was as opposed to some sort of other fancy position sensor (eg Hall senor and moving magnet or optical wotsit rather than a pot) - I'm out of date on these things, so just making allowances :)
I agree with you that the pot is most likely, like 99% likely, but given it's a computer based system, always worth keeping an open mind - never underestimate the ability of a computer to fail in some weird-arsed way based on an unlikely combination of edge cases - I've been caught out before (not often mind) where the obvious cause of the fault wasn't actually the cause.
Do they still use "Lucas, Prince or Darkness" components?
Its the overall production engineers who will insist on replacing e.g. gold plated contacts or proper screw terminals with the thinnest and nastiest of silver plating, or worse still brass, and expecting that salt laden winter spray won't affect anything at wall.
MY dodgy starter motor connection SEEMS to be the the line to the contactor. Push fit spade.
An intermittent high resistance is what you would get from a dodgy variable resistor..likely to give you the odd bout of low throttle when you least expect it, or ultimately not enough throttle at all.
Lucas ceased trading years ago. They were bought out - by of all things - a US company. Don't think they are now in the automotive field. The spares supply is taken care of by IIRC CAV.
The latest defender STILL has switches intstantly recognisable to those of you who grew up with Morris 1100's and Marinas..
I think the disco is a bit better, and the freelander certainly is..the one we have is almost all 'BMW' including the engine..a very different beast altogether.
The 2.5TDI though is a really great engine. Slogs on for ever.
Well I took a look but the promised rain came on almost immediately so I've called it off for the day. I think I found the throttle potentiometer - a shiny round metal component on the pedal axis. Problem now is going to be actually getting to it without taking the drivers seat out. Its tucked up tight into the corner of the drivers footwell behind various obstructions. There isn't any obvious signs of dampness or corrosion from what very restricted view I did have so I might have a search round for a local small firm diagnosis service since the agency were talking =A380 plus VAT. At that price I could justify a connection lead and some software if its anything like reasonably priced.
I had some dealings with a company that does third party software and diagnostic help for ECU's and so on.. the manuals are thick and heavy, and so is the kit, and they owed their existence to the fact that very very few garages actually knew how to use it.
In the one case where I had an injection issue, they and I were correct in our diagnosis. The garage I took it to were not.
The info is closely guarded by the manufacturers, and they tend to release it (they have to under EEC rulings) in as terse and unreadable a form as possible.
I am 99.99% certain that diagnostics will not tell you what you don't know already. It's that shiny round component on the pedal axis..or rather the wires attaching to it in all probability.
You won't see any obvious signs of corrosion. That will all be inside the spade terminals or whatever it uses to attach to the wiring.
A good blast of contact cleaner and a wiggle may fix it even if you cant get the wires off. A good wiggle with the engine running may well induce the fault too.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.