Xantia V6 broken exhaust (update: and rough low rpm running)

[citronut]

the other problem i could see with some sort of gauze placed in the air feed would be how would you secure it to stop it being sucked into the induction

[myglaren]

like these Stevey baby

http://www.3m.com/product/information/H ... -Pads.html" onclick="window.open(this.href);return false;

or

http://www.ebay.co.uk/bhp/scouring-pads" onclick="window.open(this.href);return false;

No, the ones I am thinking of were loosely intertwined plastic strands in an ovoid of around 50x100mm. I can get smaller ones but they are more tightly intertwined and only half the size so they would be sucked into the air filter.
The idea was to restrict fluid intake to the entrance of the snorkel thingy, where it could drip back out, without restricting the airflow.

[Mandrake]

So far mine is running very nicely too. Today was the first time that I've had a chance to drive it any distance since the work done on it last Sunday, and its going like a bat out of hell.

What's interesting is today is very cold (around 2 degrees when I was out in the car) but is also very wet - previously the performance was considerably down in damp/wet conditions with a noticeable under load misfire - today the performance is perfect even in the damp wet conditions.

This suggests my theory of HT leakage on the ignition leads somewhere may have had some merit, even though I couldn't see anything visibly wrong with the leads I took out. (I've kept them so I could potentially do some bench testing on them with one of my spare coil packs)

After posting that last night I was out in the car again well after midnight, by that time it was zero degrees and the road was starting to ice, once again the performance was outstanding, with excellent throttle response and plenty of power - no deterioration at all.

Today I had a look at the air intake issue - unfortunately it may have been a red herring with regard to the stumble off idle. I tried it with the air filter removed, and again with the top half of the air filter box lifted up with the filter removed so that it was basically breathing in through the wide opening in the box, with the air temperature sensor still connected. No real difference in the off idle stumble with any configuration - so it's not the air intake hose being crushed...

I then tried a small adjustment each way to the idle limit stop - again no improvement. What I did notice however is that to achieve a 640rpm hot idle with no accessory load the ICV is almost completely closed - a 25% duty cycle reading on the Lexia. (20% is completely closed)

If I disconnect the ICV bypass hose from the filter box and block it the engine should stall or almost stall - however the idle only drops from 640 rpm to 560 rpm with the ICV path out of action, which means there must be excessive vacuum leakage past the main butterfly, or perhaps from the crankcase re-breather pipe or charcoal canister pipe. (or a small vacuum leak somewhere in the manifold system)

The ICV being right near the limit of its operation to achieve idle may be responsible for the stall/surge cycles it occasionally has, and possibly related to the stumble off idle, but so far I can't find where the excessive air flow is coming from. An intake smoke test looking for vacuum leaks may be the only way to find it...

After not getting anywhere with the stumble I decided it was time to swap my ECU back in. Previously every time I swapped ECU's the performance went up or down. Today I put my ECU in (engine already warmed up) and it ran perfectly from the get go. I went for a longish drive expecting the performance to be flat and disappointing but it was absolutely perfect - if anything even slightly better than the other ECU had been. This was in sunny but wet conditions at around 7 degrees.

Could I have finally solved the intermittent performance problems !?? Or am I jumping the gun. Since replacing the HT leads and TPS last weekend there no longer seems to be any loss of performance in damp conditions (which I attribute to the HT leads) and there also doesn't seem to be any random, intermittent variation in performance unrelated to weather conditions. (which I attribute to the TPS)

Will it stay good ? I don't know, we'll have to wait and see, but it's very promising so far. Lets see if it can go a week with no intermittent issues.

By the way I empirically confirmed the TPS calibration algorithm of the ECU - I remember reading somewhere that the ECU measures the TPS voltage at key on and assumes this to be a closed throttle until proven otherwise. I tested this by looking at the TPS voltage and throttle angle figures on the Lexia.

When I turned the key on with the throttle held slightly open the TPS voltage was higher than normal (as you'd expect) but the throttle angle registered as 0 degrees. Opening the throttle increased the angle with voltage, closing the throttle reduced them together.

If I closed the throttle past the point at which I'd held it when I turned the key on the voltage dropped lower but the angle remained at 0 degrees, now when I open the throttle again it was reading about 2 degrees at the same voltage which it had previously read 0 degrees. If I held the throttle open too much (more than about 5 degrees) during key on this calibration process did not happen, so there must be a limit to the acceptable voltage range before a closed throttle calibration occurs.

So in summary:

a) The voltage reading from the TPS at key on is calibrated as closed throttle/idle provided that it is within a "believable" range, which seems to be approximately 0 to 1 volts. (Normal closed throttle is about 0.6 volts)

b) If while the engine is running a LOWER voltage than the initial closed throttle calibration voltage is encountered, the closed throttle voltage is recalibrated to this new figure.

c) Not tested but I assume full throttle calibration works a similar way - if it sees a voltage higher than the previous maximum, this would be recalibrated as maximum throttle.

So yes, the wiper going open circuit momentarily while driving would most likely trigger a closed throttle recalibration and muck up the calibration of the TPS. From what I can see the ECU will recalibrate the TPS voltage range again after a key off key on sequence, which could explain performance loss that occurs while driving which temporarily goes away after the engine is turned off then on again - something which I have observed on a number of occasions.

I'm still at a loss to understand what is causing the stumble just above idle but if the main performance problems are solved its a fairly minor issue that is barely noticed while driving.

[Hell Razor5543]

I wonder if a cruddy/crackly track would cause that problem. I have this issue on my clock radio, and every so often I have to clean it to sort out the sound quality.

[Stempy]

If you have a small air leak it might be worth checking the throttle butterfly pivots. I had this problem with a Volvo a few years back, wear in the pivots had caused enough play to allow air to suck through them.

[Mandrake]

I wonder if a cruddy/crackly track would cause that problem. I have this issue on my clock radio, and every so often I have to clean it to sort out the sound quality.

Cause which problem ? I replaced the TPS with a brand new one so I should hope that the new one is not crackly!

[Mandrake]

If you have a small air leak it might be worth checking the throttle butterfly pivots. I had this problem with a Volvo a few years back, wear in the pivots had caused enough play to allow air to suck through them.

And how did you fix it ? Or did you just put up with it...

[Stempy]

If you have a small air leak it might be worth checking the throttle butterfly pivots. I had this problem with a Volvo a few years back, wear in the pivots had caused enough play to allow air to suck through them.

And how did you fix it ? Or did you just put up with it...

Got rid of the car as it was horrible anyway.

[Mandrake]

If you have a small air leak it might be worth checking the throttle butterfly pivots. I had this problem with a Volvo a few years back, wear in the pivots had caused enough play to allow air to suck through them.

And how did you fix it ? Or did you just put up with it...

Got rid of the car as it was horrible anyway.

Haha ok then, not a viable fix in this situation. I'll just leave it for now, winter is coming and it's not a serious problem...

[Mandrake]

First cold morning start today after swapping back to my ECU again - it takes several revolutions of the engine while cranking before the engine fires, (maybe a half to one second) whereas on the substitute ECU it catches instantly on the first revolution.

This is a 100% consistent and repeatable difference between the two engine ECU's every time I swap between them. Anyone have any thoughts on what the difference could be ?

Different ECU programming ? (For example different mappings for cold start enrichment etc) my ECU is about 8 months older than the substitute one. (June 1998 versus February 1999)

Weaker ignition coil primary drive (from my ECU) combined with normal low voltage during cranking ? One interesting observation is that the substitute ECU reports a battery voltage (via the Lexia) 0.3 volts higher than mine. About 13.6v with the engine running while mine reports 13.3v. (Actual battery voltage is 14.2v while idling)

The difference in starting may not be major but it does show there is definitely some difference in behaviour between the two ECU's, when I would have expected them to be functionally identical.

I'll have to double check next weekend but I THINK they're running the exact same (latest) reported firmware version. I took a screenshot of the firmware versions for mine but forgot to do so on the other one before removing it.

Another thing that crosses my mind is that my ECU is unlocked which means its running modified/hacked firmware, so potentially anything in the firmware could be different without necessarily reporting a different firmware version.

For example the modified firmware could be based upon an OLDER version of the firmware with known performance/driveability related bugs, but with the version number bumped to the current version number to prevent a Lexia from "upgrading" the firmware from the unlocked firmware to a newer locked firmware. (The Lexia will refuse to update the firmware if the reported version number is already the most recent version - thus protecting the modified firmware from accidental overwrite)

If that's the case getting it re-flashed to a standard locked, (but ready to pair with the Lexia) up to date version of the firmware might resolve any issues it has, as well as restoring operation of the immobiliser.. (I have no working immobiliser because the ECU is unlocked, it starts even with the transponder antenna removed)

Can anyone recommend a good ECU "unlocker" that I could email to enquire about the possibility of re-flashing the ECU ? I'd be interested to ask them whether its possible that my ECU is running an unlocked but old version of the firmware that reports itself as the current version - and whether that ever happens...

[Chris570]

Can anyone recommend a good ECU "unlocker" that I could email to enquire about the possibility of re-flashing the ECU ? I'd be interested to ask them whether its possible that my ECU is running an unlocked but old version of the firmware that reports itself as the current version - and whether that ever happens...

Does it matter?
Seriously, the difference is a slight delay in the first firing of the cylinder?

You didn't want to change the oil but you want to investigate flashing an ECU for this?

Am I missing something?

[Mandrake]

Can anyone recommend a good ECU "unlocker" that I could email to enquire about the possibility of re-flashing the ECU ? I'd be interested to ask them whether its possible that my ECU is running an unlocked but old version of the firmware that reports itself as the current version - and whether that ever happens...

Does it matter?
Seriously, the difference is a slight delay in the first firing of the cylinder?

You didn't want to change the oil but you want to investigate flashing an ECU for this?

Am I missing something?

Does the difference in starting matter ? Not really, if that were the only difference.

Does the fact that performance seems to be much more intermittent on my ECU than the other ECU bother me ? Yes. Does the fact that my ECU is running hacked firmware bother me ? Yes, a little.

Before I did the plug leads and TPS a week ago the performance was always far better and more consistent on David's ECU than mine, proven over multiple back and forth ECU swaps over nearly 2 months.

Now that I think I've sorted the other issues (leads/TPS) that were introducing confounding variables, I'm doing one final test over the next week with my ECU back in to see whether there really is any difference between the two ECU's and whether I should be looking for a replacement ECU or not. The car ran like an absolute dream for the last week with his ECU in - will it with mine ? That's remains to be seen. Past experience suggests it will start misbehaving over the next few days.

The point about the difference in starting is not that the starting is a problem per se, but that it is a concrete, 100% reproducible difference in behaviour between two ECU's which should behave absolutely identically, especially if they claim to be running the same firmware version. (Which from memory they do)

Something is different about them and the way they behave, not just the way they start.

PS not sure what you mean by "didn't want to change the oil" - I did want to change it, and I changed it twice within 6 weeks as I wasn't happy with how clean it was after the first change. It's all a matter of prioritisation of limited resources and time...

[Chris570]

It's all a matter of prioritisation of limited resources and time...

That was my point, the oil changes took a long time to get done (and then changed quickly after that we know)

But so far as i understand it you're running ok again with the original ECU under the same conditions (although not always reproducible) with no problems yet. Yes there is a clear difference in the ECU, it perhaps could be firmware but if it doesn't cause a problem, why bother with it? It's an older citroen, you wouldn't chuck loads of money at it for something else that's so minor.

I bet at least a couple of your spheres are at least a few bars out on pressure, so are mine.

Not trying to be mean but it does look to me like you're trying for an unrealistic goal.

[DHallworth]

Could it be an adaptive value that the old ECU needs to learn after having sensor changed?

I know on the Range Rover ECU's you clear the existing running config which causes it to run like a bag of spanners for a few miles, you take it for a run of mixed driving and it resets itself. According to the workshop manual, this is supposed to be done when replacing any of the engine sensors such as MAF, Lambda's, TPS, etc.

David.

[Mandrake]

It's all a matter of prioritisation of limited resources and time...

That was my point, the oil changes took a long time to get done (and then changed quickly after that we know)

But as far as I can tell, with the benefit of hindsight the oil changes made no real difference to the intermittent performance loss that was dogging it.

It's good that the oil changes were done from a maintenance perspective, they needed to be done as the oil was very dirty - and the engine runs a bit smoother and quieter now but it didn't help me solve this problem. So in that sense I was right to prioritise other possible causes first. If I hadn't been pouring time and money into this fault some of the maintenance items like oil changes would have been done a long time ago.

But so far as i understand it you're running ok again with the original ECU under the same conditions (although not always reproducible) with no problems yet.

Yet. It only went back in on Sunday and there has only been one journey lasting longer than 5 minutes since then. As I'm sure larppaxyz will attest, the issues we were seeing are infuriatingly random and intermittent, disconnecting the battery for any length of time often cured it for hours to a few days. It ran beautifully on Sunday with my ECU on a decent length drive just after an ECU reset but unfortunately that doesn't prove anything until it has run flawlessly for at least a week.

I'm certain I've been dealing with multiple issues at once so isolating and separating them has been challenging to say the least. With HT leads / TPS and ECU all changed I finally reached a point where performance was excellent and more importantly consistent - now I'm taking one step back by reverting to the original ECU to see if any of the myriad of symptoms return.

Yes there is a clear difference in the ECU, it perhaps could be firmware but if it doesn't cause a problem, why bother with it? It's an older citroen, you wouldn't chuck loads of money at it for something else that's so minor.

As I said, all the starting difference proves is that there is a definite difference between the ECU's which is not my imagination or wishful thinking. If in addition performance seems consistent and normal with one ECU (which also happens to start better) but worse and inconsistent with the other one then that tells me the ECU is not healthy.

Maybe its a firmware glitch due to the unlocking, maybe its a 2nd hand ECU which has some "history", maybe the output drivers for the coil pack have been fried by a faulty coil pack and are limiting the current ramps, (perhaps more noticeable during cranking when available voltage is lower) perhaps many things.

Not trying to be mean but it does look to me like you're trying for an unrealistic goal.

I take your point, but I think an engine that runs well every day despite weather, temperature and other random variables, instead of one that loses half its power sometimes for inexplicable reasons is not too much to ask for. We're not talking about a minor barely noticeable loss of power, at its worst it was a gutless wonder that couldn't get up hills without manual down shifting.

It might have been an unrealistic goal if it hadn't been performing flawlessly, reliably for the last week with the replacement ECU. As far as I'm concerned the problems are fixed - the only thing that remains to be seen is whether the ECU was one of the problems.