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

[Mandrake]

Loads of theory and huffing/puffing Simon but personally I'd simply buy the correct/genuine part and be done with then you know where you are and aren't introducing secondary faults.

Point taken, but other than paying through the nose to buy through Citroen, how would you suggest being sure of getting a "genuine" part ? Buying through eBay and most online parts sellers you really have no idea which are "genuine" OEM and which are pattern parts of various quality. Paying £65 is no guarantee of getting an OEM piece, and I'd be rather annoyed to pay that and still receive a pattern part.

There is obviously also more than one version of the OEM part as well. Both mine and Addos are stamped Bosch, but both are clearly different, one has a pinhole, one doesn't. So judging what is and isn't "genuine" isn't easy. Chances are there are a number of functionally equivalent variations of the original OEM part. In the meantime I've asked for a replacement or refund, there's no point in me sitting on a faulty part.

[DickieG]

A new genuine Citroen one lists at £72 alternatively contact a local Bosch service agent for part number 0261230057 and obtain a genuine one for less, both of which for me are good value compared with the risk of introducing secondary faults not to mention the time/hassle factor.

[addo]

The one photographed by me is absolutely 100% original to my car; it was serviced by three main dealers who all relied on genuine parts (and when you consider the market slice here, chasing pirated items would've been a WOFTAM) so whatever fluctuations its output may have are clearly either damped by internal circuitry or compensated for in the ECU.

[Mandrake]

A new genuine Citroen one lists at £72 alternatively contact a local Bosch service agent for part number 0261230057 and obtain a genuine one for less, both of which for me are good value compared with the risk of introducing secondary faults not to mention the time/hassle factor.

I'm not sure that number you've given is the correct one. When I search for It I get for example:

http://www.autosensors.co.uk/en/search? ... rch=Search" onclick="window.open(this.href);return false;

The photo appears to be of the later type with the curved connector, not my earlier type.

[addo]

Per your RPO, this one is correct:

http://www.ebay.co.uk/itm/230863382671" onclick="window.open(this.href);return false;

So far as used parts go, it seems to be fitted to all 16V four-potter Series II cars, in addition to the thumper.

[DickieG]

According to the Bosch parts catalogue, Addo's suggestion 0 261 230 012 is for cars up to RP number 8077 and from 8078 it's 0 261 230 034.

[Mandrake]

According to the Bosch parts catalogue, Addo's suggestion 0 261 230 012 is for cars up to RP number 8077 and from 8078 it's 0 261 230 034.

Yep, my RPO is 7905.

I'm contacting the seller again to ask for a refund instead of a replacement. For £58 I might as well go for the genuine one, if it is in fact a genuine Bosch as it claims.

Thanks guys.

[addo]

I'd wager that - unless the ECU changes at that date (and there are no variants noted on the SC pages) - the only change is the connector, with later style plugs being more weatherproof. Simon, be aware those people will soak you sick squid for carriage.

The whole genuine vs fake thing does my head in; I have genuine electrovalves from China labelled "Made in France", have seen Bosch stuff made in half a dozen countries, Ford oil filters with the same part number but different suffix based on country of origin (and thus, destination market), Portuguese wiring harnesses, Polish Valeo radiators, Taiwanese locksets that have to be OEM based on quality and detail. I cannot have a hope in hell of accurately discriminating qualitatively, based simply on country of origin.

[Mandrake]

Simon, be aware those people will soak you sick squid for carriage.

Thanks addo, I did notice that, but I think I need to get a known good one to rule out the MAP sensor as the next port of call is going to be the exhaust system and probably the cat... When I ordered the £20 one I was short of funds, but a combination of starting at a new higher paying job and a sizeable holiday pay from my old employer has left me unusually well off even after christmas expenses. So for the amount of hassle and uncertainty of misdiagnosing the problem I might as well get the genuine item now.

The whole genuine vs fake thing does my head in; I have genuine electrovalves from China labelled "Made in France", have seen Bosch stuff made in half a dozen countries, Ford oil filters with the same part number but different suffix based on country of origin (and thus, destination market), Portuguese wiring harnesses, Polish Valeo radiators, Taiwanese locksets that have to be OEM based on quality and detail. I cannot have a hope in hell of accurately discriminating qualitatively, based simply on country of origin.

It does my head in too. I'm not biased based on the country of manufacture though, some of the countries frowned on do make some high quality parts... Taiwan is a world leader in some components like PC motherboards, China despite putting out a lot of cheap crap also produce some very high quality parts for example in the last 10 years they have started making a few really high quality loudspeaker drivers that rival the best in the world. (Aurum Cantus ribbon tweeters for example)

The problem is not country of origin, but confirming the supply chain. A similar problem is counterfeit power transistors. 20 years ago and before the internet it wasn't really an issue but in the last 15 years or so counterfeit power transistors are a huge problem, unless you can confirm the supply chain chances of getting counterfeits in certain transistor types is so high that ordering them online from anywhere except a big name traditional stockist is a complete crap shoot. These days anybody can put something on ebay or throw up a website and start selling stuff, so you really don't know what you're getting in many cases...counterfeits and poor quality copies of components passed off as genuine abound on the internet these days.

[addo]

Unless your MOT is due soon, I'd punch out the cat and spend the moolah on bits for an ML5T swap. This guy is worth chatting to.

[lexi]

It may be the case that the misfiring you discovered from dodgy coil leads and old plugs may have contributed to demise
of Cat innards.

I think the Chinese have the largest ribbon Hf unit facility worldwide now. Mind you, I had Kelly ribbons over 30 years ago

[DHallworth]

Simon, remember I've got a complete V6 here that you could use to test parts from if you want.

David.

[Mandrake]

Hi David,

Thanks, I forgot about that... when I returned the faulty £20 MAP sensor I should have just waited until I visit, swapping a MAP sensor is literally a 2 minute job...I guess I'm just impatient to rule it out and didn't think about it. Mind you if your V6 is more than a few months newer than mine (mine is mid 1998) it will have the later type MAP sensor with a different connector anyway...

[Mandrake]

Ok folks, time for a big update.

The new MAP sensor came, this time a Bosch and whilst it works perfectly (and has the same pinhole design as the original) it unfortunately didn't fix the problem, kind of predictable in hindsight when I couldn't identify any problem with the static MAP readings on the Lexia. At least its ruled it out 100%.

Feeling a little bit despondent that it must be a blocked exhaust after all and with no forum over the weekend (glad its back online again ) I started doing some serious research on the internet into symptoms/diagnosis for blocked/restricted exhausts, which also lead to reading about symptoms/diagnosis of fuel starvation and I came up with some really good resources, including:

http://www.aa1car.com/library/converter.htm" onclick="window.open(this.href);return false;
http://www.brakeandfrontend.com/issue/a ... ntid=38643" onclick="window.open(this.href);return false;
http://www.random-misfire.com/" onclick="window.open(this.href);return false;
http://www.gregsengine.com/using-a-vacuum-gauge.html" onclick="window.open(this.href);return false;

Another resource that was extremely good was the following videos including case studies of diagnosing various related faults, this guy really knows his stuff, I learnt a lot from some of those videos and was impressed by some of the diagnostic techniques:

http://www.youtube.com/feed/UCrf6f8hn5oy4alB2WXJCIqA" onclick="window.open(this.href);return false;

One of addo's comments earlier in the thread came back to my mind that a blocked exhaust can affect the vacuum reading, and since I can read pressure on the Lexia and manually convert it into a vacuum reading I thought it might be worth following that line of investigation.

Various different resources all roughly agreed on what to expect for vacuum readings with a normal exhaust and a blocked exhaust. Using the 4th link above as an example they quote a typical vacuum at idle of around 17 - 21 inches of mercury (Hg) and that on a good exhaust increasing the rpm to 2500-3000 should cause an initial brief drop in vacuum as you open the throttle that will quickly return to a reading that is equal or greater (vacuum) than at idle.

A partially blocked or restricted exhaust should show a reduction in vacuum (eg increase in absolute mb reading on the Lexia) as the rpm is increased to 2500-3000 rpm and higher, the reason being that to maintain a strong vacuum with a wider throttle plate opening requires a higher flow rate, when the exhaust is restricted the flow rate is limited so the vacuum starts ok at idle but drops with increasing rpm.

Well, the actual readings I got were 18.3"HG at 680rpm idle and 22.4" HG at 2500rpm. So not only is the vacuum within the normal range at idle (suggesting no significant intake leaks) but the vacuum strength increases with increasing rpm, which means the exhaust is NOT blocked. I only checked it up to about 4000rpm but it continued to increase with rpm to that point.

One other sign of a good or blocked exhaust with vacuum pressure is how low and how quickly the vacuum drops when the throttle is blipped, as shown in this video around 3 minutes in:

" onclick="window.open(this.href);return false;

Now its a bit hard to tell on the Lexia just how quickly and low the pressure is dropping since is a sampled numeric reading, but it seems to be dropping very low (mb almost rising to atmosphere) more or less instantly, also a good sign, however I think its worth me testing the manifold vacuum with an analogue vacuum gauge to confirm the diagnosis as you can judge actual response time, peak readings and so on much more easily. They're not expensive at all so I'll get one shortly.

Other signs that its not a blocked exhaust is that it just doesn't fit the symptoms - although there is an overall loss of power now at all rpm as it has progressively got worse, the initial loss of power was wide open throttle at low rpm, and performance was more or less normal at higher rpm with lesser throttle. Everything I've read on the internet from people that have had blocked exhausts (melted cats etc) say that it limits the high rpm performance - the car might idle ok and drive slowly ok but it can't exceed a certain rpm or top end power is severely down, almost the reverse of what I've been seeing.

So I'm now reasonably sure that its NOT a blocked exhaust or cat, and whilst I don't know for sure whether the cat is still ok in terms of emissions reduction (since its been running with a misfire for a period of time) I at least think its not blocking the flow, thus not limiting power.

A more concrete way to know for certain is to measure the back pressure directly - again lots of resources on the internet for that and figures on how much back pressure is typical etc, for example:

" onclick="window.open(this.href);return false;

Most vacuum gauges can also read positive pressure up to about 15psi, which is enough to diagnose back pressure using a direct reading (good below 2psi at 3000 rpm, bad up to 10psi) - so if I can get an adaptor I should be able to unscrew the oxygen sensor to take a back pressure reading there, however I don't think I need to go that far because of what I've discovered below.

So what could the problem be then ? In watching some of those videos I realised that I'd been overlooking fuelling. Could there be something wrong with the fuel pump, a blocked fuel filter, faulty fuel pressure regulator, faulty injectors etc. There could be a low fluctuating fuel rail pressure, there could be a normal pressure on light delivery sagging with higher delivery etc, I realised I'd completely overlooked fuelling.

After watching those diagnostic videos I'm convinced that my problem is in fact a fuelling problem, and that my next step is to get fuel pressure gauge to run some tests including static pressure, delivery, injector pressure drop tests etc... all these tests can be run with the Lexia as it can pulse the injectors with the engine off, manually operate the fuel pump etc. All I need is a 50psi pressure gauge with a schraeder valve and I should be able to do a full array of tests on the injection system.

Do I have any hints that there is actually a fuelling problem ? Yes in fact. The oxygen sensor has been reinstated recently and I realised that it might help confirm or deny my fuelling hypothesis. In normal idle, cruising or light acceleration the ECU should operate in closed loop mode to regulate the fuel mixture to 14.7/1, in which case the oxygen sensor will show multiple transitions per second, which it does.

However wide open throttle or heavy acceleration will put the ECU into open loop mode and it should run rich. From all the reading I've done the ECU always errs on the side of running rich if it can't use the oxygen sensor reading, and that includes heavy acceleration, particularly to help prevent detonation. So the test was obvious - drive up a hill around 25mph locked in 3rd (snow mode with the lever in 3rd) at around 1200-1600rpm where the engine has no pickup and open the throttle wide. If it was running properly it SHOULD be rich with wide open throttle.

But it was LEAN!! In those conditions, any throttle from about half throttle to full throttle was giving a steady 0.1v reading from the oxygen sensor - although I couldn't find a long enough hill nearby I was able to confirm that it ran lean continuously for at least 5 seconds with no transitions with as little as half throttle at 1500rpm, and certainly with higher throttle. Drop down to under half throttle in the same conditions and the oxygen sensor swung up to 0.8v and stayed there for a few seconds, (rich) or alternated lean/rich. (Back into closed loop mode with an ideal mixture)

So I'm almost certain that there is a fuel delivery problem that is causing the engine to run very lean under heavy load / wide throttle - exactly as I thought a few months ago but never really followed up on properly because I got distracted by the faulty ignition leads and gearbox groaning.

It fits the intermittent nature of the problem too, because while I find it hard to believe that an exhaust blockage could vary from day to day on a regular basis, I can readily believe that a low fuel pressure/delivery fault could be intermittent, there are many possible faults that would affect fuel pressure and delivery that could be intermittent including electrical faults with the fuel pump supply.

One other thing I learnt from the videos and research I did (although it is mentioned in the ES9J4 documentation as well) is that if the mixture is lean and the ECU is unable to richen it enough under load it will retard the ignition to prevent/minimize detonation and misfire as much as possible. So not only does it run lean but it runs retarded as well - thus no power. This ties in with what I noticed where the ignition timing was retarding unusually under load Jim. It's the ECU trying to protect the lean running engine from detonation.

One final confirmation - I've noticed a bit of intermittent rough running again in the last few weeks that seems exactly like a misfire - particularly just after starting cold, just as the engine rpm falls below 1200rpm it vibrates noticeably like its misfiring, but it does settle down as the rpm drops and after a minute or two its gone. If its running lean due to fuelling problems and the ECU is not richening the mixture that would explain it, remembering that the car must run in open loop mode for the first 2 minutes while the oxygen sensor heats up, so it has to "guess" on the correct fuelling based on past fuel trim calculation - but if the fuel delivery is intermittent it will guess wrong sometimes. Once it warms up and switches to closed loop it can more accurately compensate the mixture in response to varying fuel pressure by adjusting the injection time, thus hiding the symptoms.

So there you have it, I think I've finally ruled out exhaust restriction and narrowed down the problem to a fuel delivery problem, with a bit of luck some testing with a fuel pressure gauge should allow me to narrow it right down. Given that they're cheap and its a good chance its the cause I think it would be prudent for me to order and fit a new fuel filter right away as well as getting the gauges.

One last thing, in my research I've learnt all about things like long term fuel trim (LTFT) and short term fuel trim (STFT) but I've realised that I can't read these figures with the Lexia ? Is the MP 7.0 in the V6 too old to report LTFT and STFT, although I'm sure it uses them internally for its calculations ? Does anyone know whether a generic ODB reader can read LTFT and STFT from an ES9J4, or indeed whether any ODB readers work on the Xantia, or is it only the Lexia/ELIT etc that can talk to them ?

Also does anyone know how to get a Lexia to graph parameter measurements on a Xantia ? I know for certain that a Lexia can do it on some later models of Citroen as another forum member has posted screenshots, but I can't find a way to get it to graph ANY parameters on the Xantia. A real shame as being able to graph things like the oxygen sensor transitions would be extremely helpful.

[DickieG]

Does anyone know whether a generic ODB reader can read LTFT and STFT from an ES9J4, or indeed whether any ODB readers work on the Xantia, or is it only the Lexia/ELIT etc that can talk to them ?

From my experience generic OBD readers don't have access to any more info than a Lexia.

Also does anyone know how to get a Lexia to graph parameter measurements on a Xantia ? I know for certain that a Lexia can do it on some later models of Citroen as another forum member has posted screenshots, but I can't find a way to get it to graph ANY parameters on the Xantia. A real shame as being able to graph things like the oxygen sensor transitions would be extremely helpful.

I've seen access to graphs on my Lexia 2, not sure about the 3 laptop version, I'll have a look at mine tomorrow but as with all things on Lexia what/how much you can access largely depends upon what particular vehicle you are testing as the available screens vary wildly.


As a thought have you conducted a compression test? Worth eliminating as it's very easy/cheap to do and really should be one of the first places to start when carrying out diagnosis on issues such as the one you describe.