Disco 1 auto Throttle problem

[Derek Martin]

Idle/throttle closed switch??

Not sure how or why - that was just my immediate thought on reading your message

I was thinking that as the voltage never changed when the problem was there, that would rule out the TPS and its immediate wiring, I'd guess it would have reverted back to tickover voltage ?

The post about a rovacom sounds a good idea, when(if) the problem comes back I'll try it then, there must be someone who can do for reasonable price, the last place I tried wanted £50 for just reading the error code from the ecu, it took about 3 mins!.

[Ally V8]

Hi again,I did the resistance check on my throttle pot - 1.963 k ohms at idle falling to 1.565 k flat out.While I was doing this it occured to me that it is really a poor way to test it as the throttle pedal is real "crunchy" in operation,very difficult to get it to move slowly.This is why I scope them,as you can alter the time base to make even the slightest glitch show up.

I'm probably a bit far away to be worth you coming down to me,my workshop is just outside Bath,down in the SW.Good luck with it,just remember it did work properly and they are normally totally reliable - which is why I run one as a works van,the last thing I need is a truck that always needs fixing,I'm meant to be fixing other peoples !

[Derek Martin]

As I expecting, the problem has now returned, it is still happening at around 2000rpm (higher than it was originally), I don't suppose something specific happens to the engine about that rpm that may cause the problem (maybe turbo related?).

I will get the replacement ECU coded up and see how that goes, I'm going to take it to the Bosch specialist and get them to investigate the cause at the same time if the ECU does not fix it.

[sotal]

Unlucky - I was hoping you'd fixed it!

Mine is doing it more often now and I seem to be able to make it do it more often as well, Just about every time I drive down ash bank I can get the revs to the right point and make the light come on.

I'm pretty sure there is 2 points at which the light comes on, one around 1200rpm and one higher somewhere round 1650rpm.

I was talking to a mechanic friend the other day who said they had them quite often doing a similar thing (on cars) and all they did was take the top off and pack them with electrical grease and send them on their way, but if you've ruled out the TPS then that shouldn't do anything.

I can't think how the ECU could die at certain points, it just doesn't seem logical that it would wear out, it should really be something mechanical like the TPS. I can't even see that a dodgy connection would only occur at certain points so it shouldn't be the wiring.

If it's not the TPS the only other part I can think of is the fuel pump as I guess it must interpret the readings from the TPS and turn these back into something mechanical to operate the injection pump, I'm guessing if anything it would be that, that is worn.

New fuel injection pumps aren't cheap though although one a 2nd hand one recently sold on ebay for £41 + £20p&p (item number 170169404769)

I don't know if that part of the fuel pump can be cleaned up and I have no idea how it works - just trying to think through what could be at fault

[Arned]

Unlucky - I was hoping you'd fixed it!

Mine is doing it more often now and I seem to be able to make it do it more often as well, Just about every time I drive down ash bank I can get the revs to the right point and make the light come on.

I'm pretty sure there is 2 points at which the light comes on, one around 1200rpm and one higher somewhere round 1650rpm.

I was talking to a mechanic friend the other day who said they had them quite often doing a similar thing (on cars) and all they did was take the top off and pack them with electrical grease and send them on their way, but if you've ruled out the TPS then that shouldn't do anything.

I can't think how the ECU could die at certain points, it just doesn't seem logical that it would wear out, it should really be something mechanical like the TPS. I can't even see that a dodgy connection would only occur at certain points so it shouldn't be the wiring.

If it's not the TPS the only other part I can think of is the fuel pump as I guess it must interpret the readings from the TPS and turn these back into something mechanical to operate the injection pump, I'm guessing if anything it would be that, that is worn.

New fuel injection pumps aren't cheap though although one a 2nd hand one recently sold on ebay for £41 + £20p&p (item number 170169404769)

I don't know if that part of the fuel pump can be cleaned up and I have no idea how it works - just trying to think through what could be at fault

[Arned]

I am new to the forum but have read the articles with great interest. I have a 97 Auto Disco showing the same problem. My LR garage in France had and still does not have a solution. The readout of the diagnostic program points at the throttle potentiometer, cooling water temperature and bad wiring. I checked the wiring and see no obvious problem. I have also noticed that the problem seems to be temperature ‘independent’ and did not change the temperature sensor. The obvious one to me seemed the throttle potentiometer and thus changed it. The Bosch part, although still very expensive, was half the price of the LR part. This as more of you have already discovered, did NOT solve the problem. I am now investigating all the various sensors connected to the EDC, Mass Air Flow (MAF), Manifold Absolute Pressure (MAP), Exhaust Gas Recirculation (EGR) etc.

Have any of you more news on the problem?

[sotal]

Not found anything more out yet. Not had time to look yet.

I also have found out it doesn't matter if it's warm, wet, cold, sunny etc it can and will do it any time.

I have also found out it is at definite points on the rev range, or pedal position.

My suspicion is the fuel pump

[Guy H]

My suspicion is the fuel pump

This isn't as crazy as it seems at first looks.

You really need to get a datalogger on the engine - or wire up a 'breakout box' onto the ECU and get a passenger watching an oscilloscope. Picoscope do some very good PC based 'scopes and loggers. As you can repeat the problem almost at will it wouldn't be too difficult to track it down with the right equipment.

Anyway onto the pump

The pump receives 2 PWM signals from the ECU, telling it how much to open the 'throttle' and how much to advance the injection timing.

The ECU receives feedback from the pump in the form of an AC signal, varying in amplitude depending on where the throttle is.

It receives feedback on the timing from the injector with the wire, sending a signal telling the ECU when the pin lifts

It is feasible that the pump's sensor is playing up, but to be honest I doubt it - though check the wiring carefully

If the Check Engine light does not come on when it happens then I would still be looking at the TPS and its associated wiring & connectors

Cheers

Guy

[sotal]

Thanks for the useful input there Guy,

The Check Engine Light Comes on every time it occurs, if you keep your foot in the same position the light stays on, movement either way of your foot causes the light to go out and normal service resumes!

Mine does it in to places, which seem to be the two most common places to have your foot, if I'm in a rush and there's not much traffic about so I'm driving around quite quick it never occurs, but pottering around like usual then it often comes on.

One thing which I haven't tried yet is to try it on the drive in Neutral, if I can get the same error then, then that will make test things easy, I can wiggle wires etc to see if it goes off.

I can't see that the ECU would wear out at the common points and the others have shown that the TPS has been proven to be OK, so I think that only leaves the Pump as an item which could have worn at the most used points?

[sotal]

I sat on the drive when I got home, stuck it in Neutral and gently rev'd it trying to be as smooth and slow as possible.

I couldn't get it to do it at it's normal point where it occurs when driving but I continued to Rev it and when it got to approx 2000rpm the engine sounded to struggle a little, I kept my foot still and the revs dropped and the Check Engine Light came on, moving my foot then resumes normal operation.

I repeated this and the same again, I then tried to rev straight through and it revs straight through that point no problem

I got my phone out to film it and did it again, and again it got to the same point and the revs died, so I'll upload a video of the rev counter in the morning.

So any ideas what to measure, I can obviously repeat the fault at will now to demonstrate it to anyone, or to test the output on any cables.

[sotal]

As promised here is the video showing that it's repeatable:

http://www.phpcode.co.uk/CEL.mp4

In Neutral the problem occurs at 2000rpm every time, when driving it does not, which I think goes to prove it's dependant on the pedal position rather than the revs.

So any further ideas anyone? When I get a few minutes to play I'll try setting something up to press the pedal just the right amount, so I can mess with wires to see if that sorts it, might be able to narrow down a faulty connection that way

[Guy H]

So any further ideas anyone? When I get a few minutes to play I'll try setting something up to press the pedal just the right amount, so I can mess with wires to see if that sorts it, might be able to narrow down a faulty connection that way

Looking at the vid it's classic TPS fault symptoms BUT as the Engine Check light comes on that means that a signal is *completely* out of range or disconnected which is good (should be easier to find) the ECU will also have logged what it felt was wrong

Have you got any test gear?

even a good fast multimeter will do

Has anyone with this problem had the ECU error codes read?

As the TPS pot *seems* to be ruled out - my feeling is a dodgy connection or broken wire combined with the vibration of the engine at certain frequencies - will probably be something much simpler and less interesting though

[Guy H]

In Neutral the problem occurs at 2000rpm every time, when driving it does not, which I think goes to prove it's dependant on the pedal position rather than the revs.

Stick it in Drive with the handbrake on (or left foot on brake) and see where it happens then, either the pedal position or revs should be very different

[sotal]

I've got a cheap multimeter, wouldn't say it was brilliant though

I'll try it in drive, just not infront of the garage! We had a neighbour once who had a Jenson Interceptor, he was revving that on the drive when he slipped and knocked it into gear and ran into the garage door, he panicked stuck it into reverse shot up the drive and crashed into his brand new audi, wrote the Audi off and did £20k damage to the Jenson and the garage!

Anyway thanks for the replies, any idea on what I should be checking, obviously I'll start without the test gear and just see if moving any connections between the tps and the pump and the ECU does anything.

[Guy H]

I've got a cheap multimeter, wouldn't say it was brilliant though

Anyway thanks for the replies, any idea on what I should be checking, obviously I'll start without the test gear and just see if moving any connections between the tps and the pump and the ECU does anything.

To be honest if it is one of the cheap 'n' cheerful ones - if I was you I'd go & beg/borrow/steal/buy a reasonable multimeter. A 'scope PC or standalone & someone who can work it would make it a 'walk in the park' lol

Anyway you need your -ve lead or gnd securely on the ECU pin 13 should be a black wire (this is important as that's where the sensors are grounded to)

You need to get readings with the +ve lead off the following pins, before, during & after the problem(progressive pedal movement), should all be pretty stable measurements if the pedal is held.

Do one pin at a time, and take lots of readings as you move the pedal a bit, stop take reading move again etc, then repeat with the next pin etc.

Pin 33 white/pink

Pin 37 white/green

Pin 25 white/slate

I won't tell you what to look for as that can influence what you see but they should all be DC voltages in either 0-5v or 0-12v range

Be careful around the ECU though shorting out pins or grounding them or sticking 12v down them can release the magic smoke very quickly.

[Derek Martin]

Guy, glad to have you on board with the same problem, I put my multimeter readings on an earlier post on this topic, but I did use a digital multimeter, which I have been told average out the readings, so for this problem where the throttle position at which it occurs is so specific, it may not be capable of showing the voltage dip, but there was no change in voltage during repeated reproductions of the problem

My disco is still suffering intermittently from this problem(see original posts), since my last post I have again attempted to check out the wiring, I still have a suspicion that movement of the wiring loom near the ecu can cause it go a way for a period, but this (along with everything else I try) appears to not be consistently repeatable. I have still not got round to sending the car in to get the replacement ecu fitted to rule that out of the problem. (If anyone knows where I can get it recoded for less than £50 in the Stoke-On-Trent area I'd be very grateful if they could pass on the contact details)

I originally thought the problem occurred more often during periods of hot weather, but since it got worse, the temperature appears to have no affect on it.

I have recently had some periods where the engine has run very roughly at the points in the rev range where it generally cuts back to tickover(1600 or 2000) and puts the check engine light on, but on these occasions the light has not come on, and the revs stay around the same rpm. I wonder if this could be a separate problem, or maybe just a change to original problem?

I have had all the same suspicions, i.e. position at which the TPS must spend most of it's time, engine vibration at certain rev ranges showing up wiring problems, engine speed sensor, etc, but I am pretty disheartened about finding the cause of the problem, with so many people suffering from what appears to be a similar problem, somebody out there must have found and cured it.

[sotal]

Hi again Derek, I don't think Guy has the same problem, he's just chipping in to help.

Mine seems to be getting worse, or I notice it more!

I can sometimes make mine run a little rough around the 2k mark on idle, so I guess it's the same, it can easily be revv'd through.

I've not got much time at the moment to sort it but I will have a good look into mine and will let you know if I find anything, still not seen your Disco in Werrington yet though!

[BogMonster]

I've not got much time at the moment to sort it

Best you change that starter motor and use the time saved not having to jump start it every other morning then

[Arned]

The error codes my garage got via their test gear are:

-132 Throttle potentiometer, 1) value outside limits, due to a broken wire or short circuit of sensor wires 2) unreal values of the throttle potentiometer compared to those of the idling switch of the butterfly valve

This is why I changed the throttle potentiometer! Though did not change/repaired the problem.

[Guy H]

No, I've not got the same problem - bet that's gone and blown it now - hope it can't hear me parked outside

I've just done a fair bit of work with ECUs (mainly with Megasquirt) & done lots of weird fault diagnosis on various electronic systems.

Anyway really you need to take those measurements when the Engine Check light is on. At that point something is wrong with the signal on one of those pins (as logged by the ECU) and taking measurements from those pins will tell you what it is. (BTW It's important to use the correct pin for the -ve)

Did you check all the pins I posted?

I *very* much doubt it's the ECU

The best option at the moment is getting some reading off sotal's car as he can reproduce it at will

[sotal]

Best you change that starter motor and use the time saved not having to jump start it every other morning then

At least I've consistently not got much time to do it. What I should do is add up all the time I spend on here and that would probably give me enough to change the starter and investigate the CEL

Guy - I'll try and dig the multimeter out and get those readings asap - thanks for all the help

[sotal]

Stick it in Drive with the handbrake on (or left foot on brake) and see where it happens then, either the pedal position or revs should be very different

I tried that last night, put it in Drive, left foot on foot brake and started to rev, rev'd slowly and gently up to 2500rpm it was fine, tried again and fine again - didn't want to try again incase I burnt anything out! I then put it in Neutral and started to rev it, the CEL light came on at 1600rpm and the revs dropped I pushed a bit harder and it carried on revving up to 2000rpm then the CEL light came on again. So I can replicate it every time in Neutral but not in Drive

[Guy H]

Another one to test - thinking about your starting problems as well - is between the -ve post of the battery and pin 13 on the ECU black wire on the battery (prob via an extension wire) and red lead on the ECU pin.

I think you may have a flakey ground somewhere as well as some issue with the TPS wiring

[sotal]

do you want a voltage reading at that point or resistance.

and just one other thing - where is the ECU located!!

[Pete Attryde]

do you want a voltage reading at that point or resistance.

and just one other thing - where is the ECU located!!

It should be behind the kickpanel on the drivers side. It's tucked in well and getting to the connector is a bit fiddlely.

Pete.