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

[Mandrake]

Density close enough Simon - it varies by the ratio of the absolute temperature (low temp / pressure pretty close to ideal gas) so (2+273)/(17+273) = 0.948 = 5.2% difference.

Ah, good to know.

I too am surprised it makes that much difference, but you can't deny the outcome!

Exactly. The performance of the car last night was phenomenal, almost as I remember it when I first got it. (I had someone who obviously thought I was in an old heap trying to rip me off last night, no chance, I can't wait until its fixed properly. )

It's one thing to look at the scope and see that it's running lean when it shouldn't be and pronounce that that will be the problem - it's certainly A problem, it might not have been THE problem all by itself. It's another thing altogether to apply a bodge that makes the mixture significantly richer and actually prove that nearly all of the performance returns, thus also erasing doubts about other possibilities... (ignition always nagged on my mind)

Didn't this effectively start all at once though, so something must have broken or gone wrong to cause it - unless you just got a cruddy tank of petrol that blocked the injectors - and the ones you are running now are little better....!

That's what I'm still grappling with Richard - if its the injectors, what caused it ? It must have gone from running acceptably to running poorly in less than two months. I do recall that the performance went down hill BEFORE the exhaust broke, so that seems to have been an unfortunately timed red herring.

The fuel filter on the car was very old looking and I didn't replace it until well after the problems had occurred is it possible that there was a tear in the filter paper that was letting crud from the tank through to the injectors ? I don't know if that's even possible but if it did happen I could see crud building up in the injectors very quickly once it tore, especially as I was always running it near empty at the time. Or could there have been corrosion inside the filter that was being pumped into the injectors ? The original filter was a metal type and quite corroded on the outside. In any case it has a nice new plastic Bosch filter on it now - even though the (metal) one I fitted previously has only been on there a year.

Or the other plausible theory is heat soak combined with lots of short trips and poor quality (supermarket) fuel. That's why I posted the link to that injector-rehab article, where it talks about a lack of the right type of cleaning agents in sufficient quantity in cheap fuel that combined with lots of short trips (particularly hard driving short trips) can rapidly accelerate varnish build up in injectors.

From the time I got the car until it started giving me problems it was running almost exclusively on ASDA petrol, it was subject to a LOT of short trips, I did tend to give it a lot of stick on those short trips, (so the injectors would then bake in the heat after it was parked) and to top it all off I usually ran the tank near empty. All an invitation for disaster from what I've now been reading.

Also keep in mind that my original FPR was only measuring 2.7 bars which is approximately a 5% reduction in flow rates even if the injectors were perfect, so we were already off to a bad start. Any injector restriction due to varnish would be additive on top of that, so 5% on its own might have been just tolerable, but two lots of 5% added together from two different causes maybe not ?

It does seem that a relatively small change in mixture towards being lean can have drastic consequences, if you're close to a critical threshold - I've seen it stated that a flow reduction of 10% can be enough to cause a complete lean misfire - that does seem to agree with what I've observed, as an increase of 5% seems to have largely eliminated the symptoms I've been seeing when I thought it would take more than that. Is it possible that the low fuel pressure combined with a 5% restriction of the injectors over a few months of abuse has been enough to cause the issue ?

I honestly don't know until I see the injector cleaning report - if I can see a significant restriction in flow in the before figures versus after I guess that will answer the question. 10% too lean seems to be able to cause severe running issues, timing retard and lean misfires, whereas 10% too rich doesn't cause any performance problems - the only issue would be slightly less MPG and slightly more carbon build up on plugs if you use a heavy foot all the time. (open loop mode) That's why I'm going to the extra step of upping the pressure slightly - better to err slightly too rich than slightly too lean! In closed loop mode it won't make any difference to the mixture, the O2 sensor corrects the change so it won't affect the ability to pass the emissions test, nor the cruising MPG. I can always change the regulator back if I have any doubts anyway, I have one that measures 2.9 bar.

Edit - my only confusion is why the O2 sensor doesn't overule the ECU, by saying "too lean" - it appears that the ECU is taking the IAT as gospel over the oxygen sensor? Why would it do that? Or is it because it is running in open loop at WOT?

Yes - the O2 sensor is completely ignored in open loop mode. The mixture is pre-calculated based on all the other inputs including the IAT and if its too lean because of a fault with low fuel pressure or low injector flow, too bad - you run lean and the ECU doesn't notice and doesn't care. It indirectly notices the engine is knocking and retards the timing but it doesn't see the connection between the two. (At least not this old ECU, modern ones are much smarter!)

In which case why doesn't it advance the timing back once you are off WOT?

I don't know the answer to that to be honest. It may be that it has forced it to switch to a lower octane map - in a lower octane map all the base timings are reduced as well. But I don't know for sure and the ECU isn't giving up its secrets...

After all the reading/research/study I've done about engine ECU's (especially scannerdanners channel) I've come to realise that the ECU on this engine is VERY simple minded and crude by modern standards - I'm not even entirely sure that it has a proper adaptive long term fuel trim system as any OBD-II system would. It seems almost as if it has short term mixture control in closed loop mode with the O2 sensor but no ability to lean any corrections in open loop mode at all. It certainly seems that way from the poking I've been giving it...

Or maybe that should just be ignored, and revel in the fact that it WILL run OK

Now all we have to do is make it stop!

If the symptoms go away and the car runs like a charm that's enough for me, even if I've cheated a little bit with slightly higher fuel pressure. After more than a year of hair pulling I'll take any victory!

[Mandrake]

Still running great so far with my resistor bodge. Only drove about 10 minutes last night but it was snappy and responsive right from the get go and during the whole trip. I'd forgotten just how much torque these engines have when they're running properly!

Still a slight hesitation when you punch the throttle from idle in gear or in neutral but much less than before, and I'm not really seeing any of the other array of symptoms anymore. So I think it still needs to be a little richer at least during a throttle snap.

What's really interesting is just how much smoother the gearbox is. I was starting to get a wee bit worried about the gearbox in the last couple of months, enough that I gave it another oil change (although it was due anyway) but after improving the lean condition and doing an auto adaptive reset the gearbox now almost feels like it has nothing wrong with it - even the torque converter lockup transition is quite smooth now and that was pretty bad before and tended to grab suddenly when it engaged - especially when you'd apply some throttle after overrunning.

I'm pretty sure that the hesitation and flat spots in the engine performance upset the gearbox operation - it relies on the torque figure passed from the engine ECU to predictively apply the correct pressure to the clutches. (torque converter in particular to regulate the controlled slip)

So if the driver punches the throttle and if the engine ECU says there is a sudden increase in torque and the gearbox ECU quickly bumps up the pressure to the TC clutch in anticipation but the torque fails to materialise due to a lean stumble or misfire (in fact the torque probably drops to almost nothing momentarily) then higher clutch pressure but a drop in actual torque will cause the TC clutch to grab suddenly and fully lock up - which is exactly what it has been doing sometimes...

Wouldn't that be a lovely surprise if the gearbox starts working normally once the engine performance is solved.

The only other remaining issue is high CO and HC, and those could easily be due to an injector flow balance problem causing the mixtures in different cylinders to be different. (An issue that cheating by bumping up the fuel pressure or tweaking the IAT can't solve - only cleaning or replacing the injectors can address that)

[CitroJim]

Interesting on the difference the 'IAT bypass' has made Simon..

I must try substituting mine with a 5.6K resistor

Excellent!

[Mandrake]

Interesting on the difference the 'IAT bypass' has made Simon..

I must try substituting mine with a 5.6K resistor

Excellent!

Yes, the difference is staggering, I'm annoyed I didn't think to try such an easy test a long long time ago. It took me all of a couple of minutes to do, not counting booting up the Lexia to read the reported temperature!

I think the reason I didn't try it is that although I was aware that the IAT reading contributed to the mixture calculation, I thought the effect of temperature would be quite small, and I had no idea that such a small mixture change could have such a big effect.

It really does seem that once you wander too far in the lean direction from nominal you end up falling over the edge of a precipice very quickly where knocking (and resulting timing retard) and lean misfiring starts occurring, so there is only a few percent mixture change between being just barely ok and major problems. With my tweak it's still probably in the "just barely rich enough" region where there is little safety margin, especially if one or more injectors are running significantly leaner than their neighbours.

The problem with a constant value resistor is the extra mixture richness is not constant with ambient temperature - at 2 degrees it would not make it any richer at all, and the hotter the weather gets the richer it would run. Below 2 degrees it would actually make it leaner.

A much better and more accurate bodge would be to insert a 3.3k resistor in series with the IAT - that would give about the same change at 17 degrees ambient (which was where my testing was done) but still allow the IAT to give some contribution to the resistance reading so that the richer mixture stays somewhat more constant with temperature variations instead of just continuing to get richer the hotter things get.

I would have done that except I can't see a quick and easy way to do so as I don't have a male and female connector to make an in-line adaptor and I don't want to cut any wires... as its probably only going to be a couple of weeks before I get a chance to refit the injectors I might just leave it as it is until then. We're not likely to see any sub zero temperatures or summer scorchers until then.

As for you trying it (I assume you mean on your V6 that isn't quite as zippy as the other one) if yours is not running lean you might find that it runs too rich! Of course you could always connect a 4.7K potentiometer in series with the IAT and adjust it to tweak whatever mixture you like... Higher resistance means lower temperature which means richer mixture.

As a diagnostic tool to help diagnose lean running, an in-line adaptor that connects a 4.7K pot (in series with the IAT) on the end of a long fly lead (into the cabin) would actually be a very useful tool...

[stevieb]

I'm surprised you're so sceptical, given the change in your car when you swapped the injectors. Or are you just sceptical that the cleaning wouldn't restore normal operation if the injectors were blocked ?

I'm just not convinced the injectors could create sufficient problem to cause the problems you've still got. As I said, at least one of mine was ticking loudly, like a noisy tappet, but you've reported no similar noises. And again, mine improved in so many ways with the injector change, but no more than would be expected if you assume one of my old ones had failed. And it wasn't a cure until the fuel pump was dealt with, though maybe you're approaching the same solution from the opposite side.

I still think you're going to find either there's a sensor giving erroneous readings or the slightly lower fuel pressure is having a bigger effect than you realise because the fact the injectors are capable of delivering the extra fuel required when the resistor was added to the IAT connector suggests they ought to be capable of delivering what's required the rest of the time - and that any short-fall in fuelling is caused by something else.

I'd be tempted to change the IAT resistor values to see how much warmer you can kid the ECU it is before the old problem begins to show again. There has to be a point where if there's a single injector giving problems, it'll start to struggle and show at the O2 sensor (if we assume the current extra fuelling is overriding the problem). If you can get very close to ambient temps before the problem shows, then I'd be more inclined to think there's a blanket fuelling problem across all 6 cylinders, and that means fuel supply, rather than delivery.

EDITED

[stevieb]

Yes, the difference is staggering, I'm annoyed I didn't think to try such an easy test a long long time ago. It took me all of a couple of minutes to do, not counting booting up the Lexia to read the reported temperature!

I think the reason I didn't try it is that although I was aware that the IAT reading contributed to the mixture calculation, I thought the effect of temperature would be quite small, and I had no idea that such a small mixture change could have such a big effect.

As a daft thing to try, have you tried a plain resistor across the inlet air temp sensor terminals to kid it that the air temp is constant?

Ahem

[stevieb]

I'm just thinking, I sound very negative about the injector thing. I genuinely hope it's the cure for your problem Simon, but I know that every step of the way we've both made changes that have improved things for a short while and then things have returned to the way they were before - I know I certainly did last year with a lot of work going into detaching that invisible caravan - but it came back again.

More recently, after things had slipped again, both my change of plugs AND change of injectors gave similar benefits, but both only short-term. I think the improvement gained from both those changes merely gave the ECU and me "hope" in that a lot of corrections took place that masked the real problem until the ECU realised the real problem was still there - a lot like you finding problems gradually returning after an ECU reset. I would love to know how different mine would've been if I'd changed the fuel pump first - before all the other changes. I expect that would've masked other problems instead, but I do wonder how and when those other problems would've become apparent. Would I still have felt to need to swap the injectors to stop the ticking (or would it have gone completely, like it did for a short while after Terraclean?). I don't doubt I would've had issues with my MAP sensor (considering the Intermotor one was giving multimeter readings off-the scale - I appreciate you can't measure the effectiveness of the device with the MM, but you can certainly see if the electronic side is goosed ). I also wonder how much further you can get on a gummy throttle body before it needs cleaning if the fuel pump is playing ball, not to mention how much further I would've got on those tired spark plugs.

I think I had a whole host of other problems that were compounding the root cause, which was the fuel pump struggling to supply the juice. But I don't think the injectors were as much an improvement as they appear, considering the problem gradually returned, as it did with new plug, as it did with a new MAP sensor, as it did with the throttle body cleaned, as it did with the ICV cleaned and the weird key/reset thing being done.... The fact your base set of symptoms was there with the old injectors as it is with the new ones (albeit more apparent with the new ones), makes me think the injectors can't be the cause, unless you're very unlucky. That's not to say it isn't them, just that I think cleaned ones will give that temporary boost before things slip again.

Now, a new pressure regulator on the other hand...

[Mandrake]

I'm just thinking, I sound very negative about the injector thing. I genuinely hope it's the cure for your problem Simon, but I know that every step of the way we've both made changes that have improved things for a short while and then things have returned to the way they were before - I know I certainly did last year with a lot of work going into detaching that invisible caravan - but it came back again.

True, but there is a key difference this time - I'm constantly observing the O2 sensor while driving the car - the scope has become a permanent fixture of my dashboard for nearly 2 months... So I'm not relying on subjective impressions alone, I'm looking at cold hard evidence on the scope display. It's also now proven to my satisfaction that the time delay where it seems to work ok for a while is mostly due to the time that it takes the timing to get retarded, as knocking has very specific requirements to occur (including hot cylinder walls) so doesn't happen straight away.

I'm no longer being fooled by the effects of a battery off reset because even when the car temporarily seemed to be running much "better" after a reset or other factor the scope STILL shows a lean mixture and from that I know the problem is not fixed even though the car may not be exhibiting the full array of symptoms yet and may at a casual glance seem to be running fairly normally. I know from looking at the lean mixture on the scope that the symptoms will return - and they did.

Fooling the IAT is the only change I have made in the months that I've had the scope on the car that has actually made the open loop mixture much closer to the rich mixture it should be. It's still not perfect - under some narrow set of open loop conditions the O2 voltage does still dip just slightly from full rich (meaning that the ratio is getting as high as at least 14.2/1 or so, as the sensor range is only 14.2 to 15.2 to 1 approx) and when that happens you can hear and feel a small dip/hesitation in power (very small compared to before, just barely noticeable) which to me is evidence that yes, the mixture is the issue.

So I know that as long as the mixture stays rich in open loop mode the car will continue to run well, and so far it has. Bear in mind that the fixed resistor value means the injector pulse is not being corrected for ambient temperature, so the degree to which the mixture is being made more rich is now dependent on temperature - the colder the intake air is the less boost in mixture richness its getting, so there will be some variation due to this. (Not that I've observed much)

More recently, after things had slipped again, both my change of plugs AND change of injectors gave similar benefits, but both only short-term. I think the improvement gained from both those changes merely gave the ECU and me "hope" in that a lot of corrections took place that masked the real problem until the ECU realised the real problem was still there - a lot like you finding problems gradually returning after an ECU reset. I would love to know how different mine would've been if I'd changed the fuel pump first - before all the other changes. I expect that would've masked other problems instead, but I do wonder how and when those other problems would've become apparent. Would I still have felt to need to swap the injectors to stop the ticking (or would it have gone completely, like it did for a short while after Terraclean?). I don't doubt I would've had issues with my MAP sensor (considering the Intermotor one was giving multimeter readings off-the scale - I appreciate you can't measure the effectiveness of the device with the MM, but you can certainly see if the electronic side is goosed ). I also wonder how much further you can get on a gummy throttle body before it needs cleaning if the fuel pump is playing ball, not to mention how much further I would've got on those tired spark plugs.

I think I had a whole host of other problems that were compounding the root cause, which was the fuel pump struggling to supply the juice. But I don't think the injectors were as much an improvement as they appear, considering the problem gradually returned, as it did with new plug, as it did with a new MAP sensor, as it did with the throttle body cleaned, as it did with the ICV cleaned and the weird key/reset thing being done.... The fact your base set of symptoms was there with the old injectors as it is with the new ones (albeit more apparent with the new ones), makes me think the injectors can't be the cause, unless you're very unlucky. That's not to say it isn't them, just that I think cleaned ones will give that temporary boost before things slip again.

Now, a new pressure regulator on the other hand...

I appreciate your comments and together we have got a lot further than either of us alone, but I think you're making a very important error in one of your conclusions that is clouding the issue a bit - that the physical root cause of the problem is exactly the same in both our cars.

I'm 100% sure that a lean mixture in open loop is the cause of power loss on my car - that is beyond question now. That, and the secondary effect of the timing getting retarded after too much driving on a lean mixture.

I'm reasonably satisfied that your car was also suffering from running lean in open loop mode (at least as satisfied that I can be when I haven't had a chance to put my scope on your O2 sensor) and thus exhibited many of the same symptoms, including the timing getting retarded.

However I'm not at all convinced that the same root causes were responsible, or if they were and there were multiple causes, the percentage contribution from different causes may have been very different. ANYTHING that would cause the engine to run lean under open loop mode would give similar symptoms, and there are several things that could do this. This is a critical point that is being overlooked.

So from a diagnosis perspective the similarity of our cars may well end at the point where we determine there is a lean mixture causing the issue - from there we each have to follow our own diagnostic path to trace the cause or causes, which may not be the same on both cars. Actual objective measurements of the issue (O2 reading) and applying knowledge of possible causes that have and have not been ruled out are the only way to systematically narrow down the issue or issues and eliminate them.

You keep pointing me back to the fuel pump and now the pressure regulator - but I've already replaced both of these - TWICE, and the problem remained. The fuel pump was first replaced with a second hand one - no change, now I have a new fuel pump - no change. Therefore it isn't the fuel pump on my car, even if it might have been on yours... agreed ?

You're now saying fuel pressure regulator - it has been replaced twice, and apart from a slight difference in pressure all three test absolutely fine. I did some careful testing of the one on the car (which came with the 406 rail) just the other day, and apart from it being 2.8 bars (probably due to a weak spring) it worked perfectly in every way.

The first time I changed the fuel pressure regulator I went from 2.7 bars to 2.9 bars - there was a small but noticeable improvement but the problem was NOT fixed! Not even remotely close. Just a small improvement to a very big lack of power.

As far as I'm concerned 2.9 bars is close enough to 3.0 bars unless there are problems elsewhere. So please accept the fact that the fuel pump and pressure regulator are not the major cause of the issue on my car. I've measured the fuel pressure on my car many times - I don't believe you've measured yours at all ?

We've both played injector roulette on a second hand set of injectors and you might have lucked out but I've possibly been unlucky. (So far I've been very unlucky with S/H parts in general, getting a number of duds)

Remember that 3 of my S/H ones were well stuck and took a lot of persuading to start working at all, I honestly thought the last one would never un-stick, I had practically given up on it after more than a half an hour of constantly tapping and knocking it about before it finally burst into life. The only thing that would have caused them to be so stuck would be the pintle resting in hardened sticky varnish deposits, deposits that may still be there, and may only be removable by ultrasonic cleaning.

For all we know that 406 went to the scrapyard in the first place because it kept suffering from low power and an intermittent misfire from a sticky injector that nobody could fix! Nobody is going to advise fitting a full set of brand new injectors on a 15 year old car so even if the tech that worked on it made the correct diagnosis the advice was probably "scrap it - a new set of injectors costs more than the car is worth". This was after they had probably spent money on the car fitting a new coil pack, leads and plugs and found the misfire persisted. Just speculating, but we have to consider all possibilities when considering the provenance of second hand parts.

Until I see the cleaning report in my hand we can only speculate as to the condition or performance of the original injectors. I'm pretty sure that my original set did have at least one injector that was intermittently sticking or sufficiently down in flow to cause a single cylinder misfire - that symptom at least seems to have gone with this set of injectors - but look how much cleaning I subjected them to to even get them working at all - I gave them a 20 times normal strength injector cleaner run through them on the test bench multiple times, but even that is nowhere near as effective as ultrasonic cleaning for baked on deposits.

Lets just wait and see what the report says, and then if the report looks favourable (sufficient performance issues to explain the problems in the "before" testing) lets wait and see how the car runs and save the speculation until afterwards if it's still not 100%...

[CitroJim]

Another most interesting and absorbing set of instalments to this long-running saga I may not comment much or very constructively but I am reading and learning with every post.

Simon, I know this is a bit like the pot calling the kettle black but are you always up as early as your last post suggests on a Sunday? I know I am but then I always am 7 days a week and I live in a time-shifted world where mine starts at 4am and ends to all intents and purposes at 8pm!

I'm wondering if you're the same as me and do all your best work in the early morning?

I've been up since 4 and baked a lovely loaf and cooked the Sunday dinner already as well as doing a load of laundry and other housework...

Off for two hours of karate at 9:30... Or early this afternoon in my world

[Mandrake]

Have you adjusted your board preferences for daylight savings yet Jim ? I noticed that mine hadn't been so it claimed I had posted at 5:51 am. I think not!

But yes I was up before 6:30 this morning. When I was younger I was a night owl - weekends I could easily sleep in until 10am and I'd be up until 1am even with work the next day. Not any more! Now if I'm up past about 11pm I'm shattered the next morning, and on work days I struggle to get out of bed.

The annoying thing is that on the weekends (unlike a week day) I'm usually wide awake feeling fresh and wanting to get up for breakfast and a cup of tea by about 6:30 if not 6:00! So for me a weekend sleep in nowadays is waking up really early, having breakfast, reading the morning internet at the breakfast table (including this forum) and then back to bed again for a bit!

[CitroJim]

Good point on the board settings Simon. I must check. I'm a shocker for forgetting them when the clocks change...

[stevieb]

I appreciate your comments and together we have got a lot further than either of us alone, but I think you're making a very important error in one of your conclusions that is clouding the issue a bit - that the physical root cause of the problem is exactly the same in both our cars.

But is that not the case? We've both suffered identical symptoms, except in my case last summer they were extreme and after a couple of sensor swaps they settled down to match the level you've been experiencing. We've each replaced different parts in a different order to reach different conclusions at different times, but ultimately if the symptoms are identical, then the chance of the root causes (and I have little doubt there have been a number of root causes, not just one - some of which have been cured along the way) being different are very slim.

I just can't help thinking that at this point something is different between the two cars, and I'm trying to get to the bottom of what it might be. Obviously the injectors are a complete variable as we've both fitted used parts, but you had the same symptoms before you swapped them over - admittedly with two weak pumps as did I with one weak pump. So are both your sets of injectors duff, and causing symptoms that in my case were caused by a weak pump, or is there another fault that's remained constant and has merely been exaggerated by the change of injectors? You say the problem is worse with the replacement injectors, but the pressure regulator was changed at the same time (to another one that is theoretically working below spec), introducing another change who's effect can't be quantified (I kept my original fuel rail/regulator), so suddenly there's a doubt (at least in my mind) about your pressure regulator as well as the injectors. But again, my mind looks for the constant - if we both had ten problems causing an identical set of symptoms, then which one of those ten have you yet to cure, and how many of my ten were actually irrelevant in the context of causing leanness and knocking?

I'm 100% sure that a lean mixture in open loop is the cause of power loss on my car - that is beyond question now. That, and the secondary effect of the timing getting retarded after too much driving on a lean mixture.

I'm reasonably satisfied that your car was also suffering from running lean in open loop mode (at least as satisfied that I can be when I haven't had a chance to put my scope on your O2 sensor) and thus exhibited many of the same symptoms, including the timing getting retarded.

However I'm not at all convinced that the same root causes were responsible, or if they were and there were multiple causes, the percentage contribution from different causes may have been very different. ANYTHING that would cause the engine to run lean under open loop mode would give similar symptoms, and there are several things that could do this. This is a critical point that is being overlooked.

So from a diagnosis perspective the similarity of our cars may well end at the point where we determine there is a lean mixture causing the issue - from there we each have to follow our own diagnostic path to trace the cause or causes, which may not be the same on both cars. Actual objective measurements of the issue (O2 reading) and applying knowledge of possible causes that have and have not been ruled out are the only way to systematically narrow down the issue or issues and eliminate them.

You keep pointing me back to the fuel pump and now the pressure regulator - but I've already replaced both of these - TWICE, and the problem remained. The fuel pump was first replaced with a second hand one - no change, now I have a new fuel pump - no change. Therefore it isn't the fuel pump on my car, even if it might have been on yours... agreed ?

You're now saying fuel pressure regulator - it has been replaced twice, and apart from a slight difference in pressure all three test absolutely fine. I did some careful testing of the one on the car (which came with the 406 rail) just the other day, and apart from it being 2.8 bars (probably due to a weak spring) it worked perfectly in every way.

The first time I changed the fuel pressure regulator I went from 2.7 bars to 2.9 bars - there was a small but noticeable improvement but the problem was NOT fixed! Not even remotely close. Just a small improvement to a very big lack of power.

As far as I'm concerned 2.9 bars is close enough to 3.0 bars unless there are problems elsewhere. So please accept the fact that the fuel pump and pressure regulator are not the major cause of the issue on my car. I've measured the fuel pressure on my car many times - I don't believe you've measured yours at all ?

We've both played injector roulette on a second hand set of injectors and you might have lucked out but I've possibly been unlucky. (So far I've been very unlucky with S/H parts in general, getting a number of duds)

Remember that 3 of my S/H ones were well stuck and took a lot of persuading to start working at all, I honestly thought the last one would never un-stick, I had practically given up on it after more than a half an hour of constantly tapping and knocking it about before it finally burst into life. The only thing that would have caused them to be so stuck would be the pintle resting in hardened sticky varnish deposits, deposits that may still be there, and may only be removable by ultrasonic cleaning.

For all we know that 406 went to the scrapyard in the first place because it kept suffering from low power and an intermittent misfire from a sticky injector that nobody could fix! Nobody is going to advise fitting a full set of brand new injectors on a 15 year old car so even if the tech that worked on it made the correct diagnosis the advice was probably "scrap it - a new set of injectors costs more than the car is worth". This was after they had probably spent money on the car fitting a new coil pack, leads and plugs and found the misfire persisted. Just speculating, but we have to consider all possibilities when considering the provenance of second hand parts.

Until I see the cleaning report in my hand we can only speculate as to the condition or performance of the original injectors. I'm pretty sure that my original set did have at least one injector that was intermittently sticking or sufficiently down in flow to cause a single cylinder misfire - that symptom at least seems to have gone with this set of injectors - but look how much cleaning I subjected them to to even get them working at all - I gave them a 20 times normal strength injector cleaner run through them on the test bench multiple times, but even that is nowhere near as effective as ultrasonic cleaning for baked on deposits.

Lets just wait and see what the report says, and then if the report looks favourable (sufficient performance issues to explain the problems in the "before" testing) lets wait and see how the car runs and save the speculation until afterwards if it's still not 100%...

My simplistic process of elimination says that your problem has been there regardless of the parts changed, which points to the root cause (if we take it to be a singular cause) being a constant, and therefore it's not one of the parts that have been changed. I'd say the only exception is the pressure regulators, which all fall below spec and could be considered a constant for that reason.

The only time I veer away from the argument that it's a constant and towards the injectors is because you noticed a deterioration when you swapped them last, but the point I keep making is that we've both found swapping of parts providing a temporary cure - masking the uncured problem with or without an ECU reset - before the root problem causes the symptoms to return, so it's entirely possible a change of parts could also exaggerate the problem too. Despite the temporary improvement suggesting there was a change, it's possible I needn't have swapped my injectors when in fact my fuel pump was the greatest cause of my problems. It's possible you may not have needed to swap your fuel pump when the injectors are the cause of your problems - but I think you'd be very unlucky to have the same symptoms on both sets of injectors, despite the symptoms being worse on the second set. Which brings me back to the idea that it's a constant and not a variable...

Anyway, that's just me struggling to put my thoughts into words. I'm going to take a back seat now and see what transpires, rather than risk muddying the waters further.

I'm looking forward to the conclusion, whatever it may be

[Mandrake]

But is that not the case? We've both suffered identical symptoms, except in my case last summer they were extreme and after a couple of sensor swaps they settled down to match the level you've been experiencing. We've each replaced different parts in a different order to reach different conclusions at different times, but ultimately if the symptoms are identical, then the chance of the root causes (and I have little doubt there have been a number of root causes, not just one - some of which have been cured along the way) being different are very slim.

I'm not sure that I agree that the cause must be identical at a component level. As I said, I'm pretty sure a lean mixture under load is the key common factor between our cars and possibly other ES9J4's with similar symptoms - if it runs too lean under load it will have certain knock-on consequences that provide most of the symptoms. (I now think even the surging idle is related as that has got far less noticeable since my IAT bodge)

But just because the end symptoms are the same doesn't automatically mean the physical cause of the lean running MUST be the same, nor that there is necessarily down to a single cause on its own, on an old, high mileage car.

I just can't help thinking that at this point something is different between the two cars, and I'm trying to get to the bottom of what it might be. Obviously the injectors are a complete variable as we've both fitted used parts, but you had the same symptoms before you swapped them over - admittedly with two weak pumps as did I with one weak pump. So are both your sets of injectors duff, and causing symptoms that in my case were caused by a weak pump, or is there another fault that's remained constant and has merely been exaggerated by the change of injectors? You say the problem is worse with the replacement injectors, but the pressure regulator was changed at the same time (to another one that is theoretically working below spec), introducing another change who's effect can't be quantified (I kept my original fuel rail/regulator), so suddenly there's a doubt (at least in my mind) about your pressure regulator as well as the injectors. But again, my mind looks for the constant - if we both had ten problems causing an identical set of symptoms, then which one of those ten have you yet to cure, and how many of my ten were actually irrelevant in the context of causing leanness and knocking?

Yes but the pressure regulator that's in there now is still a higher pressure than the original. I very much doubt my original pressure regulator was any higher than 2.7 bars when the car was running fine for the first 6 months. Low pressure will be a fatigued spring which will have very gradually got that way over years. My fuel pressure gauge testing says that apart from small differences in regulator pressure, all three regulators are absolutely fine.

The only way these type of regulators typically fail is a ruptured diaphragm, and none of them have that. Sometimes an old regulator might not seal perfectly on the seat which will cause static fuel pressure to be lost with the key off but will not have any effect on running when the regulator is open anyway - and the regulator in the car currently does not have any leakage either.

Anyway, that's just me struggling to put my thoughts into words. I'm going to take a back seat now and see what transpires, rather than risk muddying the waters further.

I'm looking forward to the conclusion, whatever it may be

Yes, I guess we'll know one way or the other soon enough...

By the way I've just realised today that I may have made a critical error in my testing/cleaning of the second hand injectors - and I'm surprised nobody pulled me up on it when I showed photos of what I was doing...

What was missing from my test rig of fuel pump in a bucket, fuel rail, injectors, injector pulser etc ? A fuel filter!! Yes, I started off with a brand new clean bucket for the fuel pump to sit in, but I was using my old original worn/dirty fuel pump with no filter in line.

What if the bearings were failing on my original pump (it was quite noisy before I replaced it) to the point where the pump had started shedding microscopic metallic particles (remember the fuel flows right through the centre of the motor, through the commutator etc on these type of pumps) and with no filter to catch them I could have easily been contaminating the the S/H injectors with pump debris during testing... Not to mention the inside of the pump may have just been dirty, again no filter to catch anything that came out. Very silly of me not to think of this.

I also wonder whether my original pump starting to fail could have forced enough crud through the original fuel filter to start clogging up the injectors ? Could that have been the genesis of trouble all that time ago ? Could it all have been avoided by a new fuel filter the moment I bought the car ? Just a thought...

[Mandrake]

Well my injectors have come back complete with report and I have to say that at first glance I'm a bit worried - they weren't nearly as bad as I was expecting (or hoping ?) in terms of flow balance and overall flow rate, meaning that I'm not at all sure that the injectors are the cause of the lean mixture now...

Here's the report:

https://dl.dropboxusercontent.com/u/782 ... report.jpg" onclick="window.open(this.href);return false;

I had scratched the cylinder numbers onto the yellow tag at the top of each injector but apparently the guy at injectortune didn't notice as he scratched his own ID numbers in the plastic near the connector, so I've written in pen on the report the actual cylinder numbers of each injector so that the report can be correlated back to specific cylinders on the car.

From what I can see the "before" report is far better than I was expecting - I don't think a 1.1% imbalance between worst and best injectors is that bad for a 15 year old car, and the biggest difference before and after cleaning was only 3.3% - again far less than I expected. I'm happy that they're now all identical and that there was a 3.3% improvement and together with the increase in fuel pressure it might be enough - but it's not what I was expecting.

The spray pattern is worthy of mention though - a bit more subjective as it's gauged visually by the operator during the testing but the two that gave the worst spray pattern result ("fair") turn out to be cylinders 2 and 5 - the middle cylinder on the front and rear banks. The reading I've previously done suggests that the middle two injectors on in-line fours and the middle injector on each bank of a V6 are usually the worst affected with varnish build up simply because they run a bit hotter than the outer injectors and are more subject to heat soak after the engine turned off due to their central locations - so its interesting to see that those two did indeed fare the worst as far as spray pattern went.

It is still possible that "lean" running symptoms could be a problem with the spray pattern rather than absolute flow volumes - again, reading that I've done previously says that if the spray pattern is poor, eg just a jet of fuel instead of an aerosol like misty spray, that the large fuel droplets that are not sufficiently atomised won't burn, effectively causing a lean burn (and left over fuel that causes high HC readings) even though the total weight of fuel injected might be correct. Under these conditions you need an overly rich mixture (in terms of grams of fuel for grams of air) to actually get a balanced burn as you have to compensate for the portion of fuel that didn't burn to let the engine run properly - but even then the HC in the exhaust will be high.

Anyway, I'm a bit sceptical about the potential for change after seeing the report, but all I can do is fit them and see what happens - weather and other commitments mean I probably won't be doing it this weekend but am hoping to get the job done the following weekend.

The 3.5 bar fuel pressure regulator looks like it might or might not fit - I'm going to try it in the spare fuel rail first to see if it will actually go in and let me put the retaining clip over it - you can see in the pictures below that the length of the 25mm shank is a bit different - but depending on where the step in diameter is in the housing it presses into it may or may not be a problem:

(original 3 bar regulator on the left, 3.5 bar one on the right)





Hopefully it will fit, now that I've seen how little problem with flow rate there is with the injectors a small bump in pressure might be a necessary evil to try to get this thing running properly, even if I never figure out why it might be needed....

[Mandrake]

I wasn't planning to do this job this weekend but when the weather was (intermittently) good today I started to get paranoid that I was letting a good day go to waste and that it would rain next weekend so I started on it on the basis that I might not get it finished today but that we don't really need the car this week if I was delayed.

Most of the job went well - while I had everything out I took the rear spark plug cassette out and checked for oil - there is still a small leak but nothing important so I soaked up what oil there was, I also removed the rear spark plugs to check their condition - two of them I wasn't 100% happy about, one had an odd carbon patch half way down the insulator and the other had a suspicious little mark like some carbon tracking part way down the insulator.

Nothing major but since I have a spare set of barely used (less than 1000 miles) spark plugs of the exact same type from a previous spark plug swap where I had bought new plugs in advance but didn't really need them I decided to swap those two over. I haven't checked the front ones yet.

The cleaned injectors were installed in the rail along with the 3.5 bar regulator, I also decided to swap in my original pulsation dampener rather than the one from the 406 rail as I had noticed the pressure gauge flickering quite a bit at idle when I last checked making me wonder if the pulsation damper that came with the 406 rail is faulty - as it's its job to smooth pressure fluctuations when the injectors open and close. (It's basically a diaphragm with a spring behind it like a tiny little accumulator sphere... )

I put the rail back in and because I wasn't 100% convinced that the pressure regulator is a good fit I thought I better check the pressure and check for leaks BEFORE reassembling the inlet manifold etc... so I put the pressure gauge on and used the Lexia to operate the fuel pump without running the engine.

Good thing I did! No external leaks but the pressure regulator is not working properly... Instead of 3.5 bars I was only measuring at most 2.5 bars, and the moment the pump stopped the pressure plummeted to zero in about 2 seconds. To rule out leaky injectors, fuel pump or some other issue other than the regulator (even though I knew nothing else could be wrong) I tried pinching off the return pipe after the regulator - the pressure was rising to 6.4 bars and would stay steady at about 6.2 bars after the pump shut off without falling at all, proving that the only source of leakage (high leakage!) was through the regulator.

So either the internal seat of the regulator is not closing/sealing properly or the small o-ring in the picture above (which seals between inlet pressure which is applied between the upper and lower o-ring, and outlet overflow which is at the bottom) is leaking due to a bad fit.

Correct me if I'm wrong but looking at those pictures I don't think the shaft or small o-ring are in any way smaller than the original making the o-ring a very unlikely source of such high leakage - leaving a faulty regulator that isn't closing as the cause.

So I've contacted the guy from injectortune regarding the regulator and I'm now off the road for a few days meanwhile as whilst I could fit my original 2.9 bar regulator and get the car back together I really want to increase the pressure and I really DON'T want to be having all the fuel rail etc out again a second time. So I'll wait and see what he comes back with first.

The truth is I wouldn't have got the job finished today anyway - I was playing hide and seek with the rain all day - I was out of the rain and then back out into the sun again about four times today and I had only just discovered the pressure regulator wasn't working when it started raining AGAIN and it didn't stop raining again until it was getting dark...

Ah well, I will get there eventually!