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

[CitroJim]

Hmmm, I've seen the discrepancy in coil pack charge/dwell times before Simon when trying to find the reason for high HCs on PavlosV6's car. Despite noticing my own V6 shows similar we swapped coil packs which did seem to help HC a bit but did not cure it.

I'd have thought a zero time here would have caused a cracking misfire given each coil serves two cylinders. It may be a red herring but it will doubtless be worth doing a coil swap if you can. They are expensive and a good source is to find one in a breakers if you can. Remember they were also used on the 406 as well which expands your search. Have you swapped plugs and checked the continuity of the rear bank plug leads? They're not resistive.

The coolant temperature seems rather hot to me. Be warned that the temperature gauge is not an accurate reflection as the two sensors are in wildly different places. The green ECU sensor is effectively reading the top hose temperature downwind of the thermostat and the gauge is reading effectively bottom hose temperature so it will often read lower. A new green sensor might not be a bad idea.

I see the oxygen sensor is reading the lowest it normally goes. Can you observe, with the engine at normal temperature, the reading trekking between that and say 800mV or more over the course of a few seconds? If it's stuck then the mixture will not be optimum as the ECU relies on the oxygen sensor for keeping the mixture stoichiometric.

The MAP sensor looks good. Check it with the engine switched off against prevailing barometric pressure.

Onto the gearbox...

The fault is indicative of the detection of slip. basically the gearbox input speed sensor (not the engine TDC sensor) and the gearbox output speed sensor together help the ECU determine what gear is engaged. Knowing that it knows there must be an exact relationship between input and output speed. If this exact relationship is out by a given amount the fault will be flagged. This may be related to the intermittent engine speed sensor. Normally they're very reliable (thank goodness as they are almost inaccessible) but a clean of it's brown plug might be wise as it can get splashed by coolant leaks.

In this case I think the intermittent engine speed sensor has caused the gearbox ECU to believe there might have been excessive torque converter slip as it again knows what the relationship to input and output speeds should be and can tell if there is excessive slip going on.

That's my take anyway. Clear them and see they don't return... Best I can do with my brain the way it is today

I'm not sure what is meant by the 'Drawing' state. I must do some research...

[Mandrake]

Hmmm, I've seen the discrepancy in coil pack charge/dwell times before Simon when trying to find the reason for high HCs on PavlosV6's car. Despite noticing my own V6 shows similar we swapped coil packs which did seem to help HC a bit but did not cure it.

I'd have thought a zero time here would have caused a cracking misfire given each coil serves two cylinders.

But I do have a bad misfire... Between 1000-2000rpm under load its missing like crazy, stuttering, hesitating etc... while stationary if I lift the revs to around 1200rpm its vibrating like the old 3 cylinder Daihatsu charade I used to have years ago (which have an inherent vibration at that speed due to the timing of the 3 cylinder design...)

It may be a red herring but it will doubtless be worth doing a coil swap if you can. They are expensive and a good source is to find one in a breakers if you can. Remember they were also used on the 406 as well which expands your search. Have you swapped plugs and checked the continuity of the rear bank plug leads? They're not resistive.

I've been thinking about how to troubleshoot this and have an idea. The Lexia seems to have an actuator test that can continuously fire the coils (one at a time) while the engine is not running, if I disconnect the plug leads from the coil pack and substitute a wire positioned near the chassis as a spark gap, I should be able to measure the spark strength from each coil individually - if two coils can throw a decent length spark and the 3rd has a weak or non existent spark, the coil pack must be the culprit.

If all three coils throw good length sparks the next step would be to perform the same test via the cars own spark plug leads to the chassis, again if one spark is weak its a faulty plug lead.

If the plug leads pass the test then it must be one or more fouled or faulty plugs...

The only thing is I'm not sure what the access is like - if I take the top cover off can I get to the coil packs and plug leads fairly easily ? Or is there significant amounts of dismantling before I can test it ? I understand the rear spark plugs are quite a big job to get to ?

The coolant temperature seems rather hot to me. Be warned that the temperature gauge is not an accurate reflection as the two sensors are in wildly different places. The green ECU sensor is effectively reading the top hose temperature downwind of the thermostat and the gauge is reading effectively bottom hose temperature so it will often read lower. A new green sensor might not be a bad idea.

97 was the highest reading I saw, it was normally around 80-90 degrees while idling.

I see the oxygen sensor is reading the lowest it normally goes. Can you observe, with the engine at normal temperature, the reading trekking between that and say 800mV or more over the course of a few seconds? If it's stuck then the mixture will not be optimum as the ECU relies on the oxygen sensor for keeping the mixture stoichiometric.

Yes, once warmed up the O2 reading varies from around 180mv and 800mv about once or twice per second, so I take it that means the O2 sensor is ok ?

The MAP sensor looks good. Check it with the engine switched off against prevailing barometric pressure.

The reading with the engine off is 981mbars which without checking the weather report sounds fine to me, so I guess the MAP sensor is fine.

Onto the gearbox...

The fault is indicative of the detection of slip. basically the gearbox input speed sensor (not the engine TDC sensor) and the gearbox output speed sensor together help the ECU determine what gear is engaged. Knowing that it knows there must be an exact relationship between input and output speed. If this exact relationship is out by a given amount the fault will be flagged. This may be related to the intermittent engine speed sensor. Normally they're very reliable (thank goodness as they are almost inaccessible) but a clean of it's brown plug might be wise as it can get splashed by coolant leaks.

In this case I think the intermittent engine speed sensor has caused the gearbox ECU to believe there might have been excessive torque converter slip as it again knows what the relationship to input and output speeds should be and can tell if there is excessive slip going on.

That's my take anyway. Clear them and see they don't return... Best I can do with my brain the way it is today

I'm not sure what is meant by the 'Drawing' state. I must do some research...

Ok what I've done with the gearbox then is to clear the fault codes to see if they occur again.

Perhaps an oil change for the gearbox might be in order in a few months as a preventative measure ?

[CitroJim]

Simon, Brain is tired again

Full reply in the morning. I have much to say

[addo]

There is a coil failed; it's showing a low charge time because of this. I'd buy new off fleabay rather than trust a breaker.

[Mandrake]

There is a coil failed; it's showing a low charge time because of this. I'd buy new off fleabay rather than trust a breaker.

That's what I was thinking addo, although given the cost I think I'd like to manually test the spark coming from each coil first to confirm it before buying...

I can either do that with the Lexia actuator test described above or if necessary take it out and bench test it with a 12v supply. I have enough equipment to do a manual test of it.

So the charge time is simply an indication of how long it takes the coil to charge after each spark, so a near zero reading means that it wasn't discharged (eg no spark) in the previous cycle ?

What is a normal figure that you've seen on your own V6 and how closely do the three coils match when everything is ok ? The two coils that are working seem to stay within about 1 or 2 digits of each other.

I also notice the bad one shows basically zero at higher revs/throttle but at idle it rises a bit, suggesting it might be a very weak spark that fires when idling but is not strong enough to fire under load when there is more charge entering the cylinder...

Even before it got really bad recently there was sometimes a minor but noticeable shudder on heavy first gear acceleration, there may have been a minor misfire under high load even then which was not yet happening at light loads. (Eg again a weak spark)

[addo]

My V6 is a PRV, so it doesn't have coilpacks. In a set on any motor, the dwell times of coils should be awfully close to one another - a couple of milliseconds out, if that.

I think they probably fail because of heat stresses to the "potting" which break the wire coils.

[CitroJim]

I thought I had a bit to say Simon but I think it's been said by Adam.

Somehow I'd missed the bit that you had a very pronounced misfire. Sorry... Anyway, yes. Swap coils. Now, genuine ones are ferociously expensive and I don't rate the pattern part ones you can get for about £120. I've got one here and it's nowhere near as balanced as a good second-hand genuine item...

Another thing that threw me is usually when a coil falls over it puts the EML on...

Your Oxygen sensor looks fine. Ditto MAP. And ditto temperature

Yep, always a fluid carnage is a good idea on an HP20. Remember LT71141 only and don't you dare re-open the old and frankly rather tedious debate about LT71141 substitutes. If you do I'll never, ever replay to any of your posts ever again

For the record, there is no substitute for LT71141

[Mandrake]

I thought I had a bit to say Simon but I think it's been said by Adam.

Somehow I'd missed the bit that you had a very pronounced misfire. Sorry... Anyway, yes. Swap coils. Now, genuine ones are ferociously expensive and I don't rate the pattern part ones you can get for about £120. I've got one here and it's nowhere near as balanced as a good second-hand genuine item...

Hmm, thanks Jim, now I don't know which way to go.... I'm not keen on a 2nd hand one from a breaker given the age Xantias are now, they're likely to be dodgy already or not last long, on the other hand if the pattern parts are not very good that's a worry too.

What do you mean they're not as well balanced ? Do you mean the dwell readings are not well matched ?

By the way another symptom I haven't mentioned is intermittently a lot of visible "vapour" from the exhaust, only really noticeable on cold mornings, both when idling and revving the engine. I'm assuming (hoping) that its just unburnt fuel condensing in the cold air. It doesn't smell like burning oil. If it is unburnt fuel it must be killing my MPG...its only been visible since the running deteriorated.

Another thing that threw me is usually when a coil falls over it puts the EML on...

I was quite disappointed that no faults were logged too, given how badly the car has been running. Just goes to show that computer diagnostics don't replace common sense and experience!

Your Oxygen sensor looks fine. Ditto MAP. And ditto temperature

Yep, always a fluid carnage is a good idea on an HP20. Remember LT71141 only and don't you dare re-open the old and frankly rather tedious debate about LT71141 substitutes. If you do I'll never, ever replay to any of your posts ever again

For the record, there is no substitute for LT71141

I think you have me confused with someone else Jim! I've always been on the side of using the original spec oil in the dexron 2/3 wars

I did a few oil changes on the 4HP14 in my S1 2 litre, (which fills through the dipstick from memory) is the V6 similar or is it a more challenging job ? No immediate plans to do it, but might as well find out what is involved...

[CitroJim]

What do you mean they're not as well balanced ? Do you mean the dwell readings are not well matched ?

Precisely Simon. It was an unbalanced pattern one, although not causing a noticeable misfire, that helped put the HCs up a bit of PavlosV6's.

By the way another symptom I haven't mentioned is intermittently a lot of visible "vapour" from the exhaust, only really noticeable on cold mornings, both when idling and revving the engine.

No, that's perfectly normal Simon. It's just there's a lot of moist air flowing through the exhaust and it condenses so much it looks like there's a V6 steam engine under the bonnet. It worried me at forst but after owning three of them that all do it, no worries.

I was quite disappointed that no faults were logged too, given how badly the car has been running. Just goes to show that computer diagnostics don't replace common sense and experience!

Quite

I think you have me confused with someone else, I've always been on the side of using the original spec oil in the dexron 3 wars

Thank goodness for that Yes, yours is unlikely to have a dipstick and this does make the procedure a little more fun-filled. I have the official words on the task and if you PM me your email address I'll send them to you, along with a few tips. Good job you have a Lexia as you'll need it as the oil level must be checked at a precise gearbox temperature...

[Mandrake]

Precisely Simon. It was an unbalanced pattern one, although not causing a noticeable misfire, that helped put the HCs up a bit of PavlosV6's.

Hmm, ok. I guess I'll have to consider my options. Is there just one brand pattern part floating around or more than one ?

No, that's perfectly normal Simon. It's just there's a lot of moist air flowing through the exhaust and it condenses so much it looks like there's a V6 steam engine under the bonnet. It worried me at forst but after owning three of them that all do it, no worries.

Yes, but this seems a bit excessive ?

http://www.youtube.com/watch?v=aYXz79pN-mM

It wasn't like this a couple of months ago, although the weather is a lot colder now to be fair...that was taken on a 4 degree morning...however I didn't see any other cars on the road doing the same thing...

Thank goodness for that Yes, yours is unlikely to have a dipstick and this does make the procedure a little more fun-filled. I have the official words on the task and if you PM me your email address I'll send them to you, along with a few tips. Good job you have a Lexia as you'll need it as the oil level must be checked at a precise gearbox temperature...

Thanks, I'll send you a PM.

[CitroJim]

As far as I know there's only one and it's branded 'Sagem'. It looks OE but isn't... There may be others but I can't say as I've seen them...

Just be careful if you get a breaker one; ensure it's for the ES9J4 and they don't try to give you a coil from the later ES9J4S used in the C5 and late 406s. These are totally different and have a separate coil on each spark plug.

Your steam looks pretty normal to me. You wait until the brass monkeys really put in an appearance, then you'll see some steam! Bear in mind you have six cylinders of three litres moving a heck of a lot of very humid air and water is a by-product of combustion and the action of the cat so a bit of steam is really to be expected. Better than some cars that look like they have a hose-pipe rather than an exhaust.

I wish I understood dew-point a bit better; that'll help explain it I would imagine

[addo]

Dew point is simple; it's intuitive when you think about it.

As to the coils - Valeo 245088 is the other aftermarket you might punt on. About £170 from factors. They look awfully similar to the Alfa Twin Spark coils, but potted in a line. How many pins on the plug - seven?

There are two problems I am finding more frequently on PSA stock from the later '90s onwards.

First is cooling fans drawing excess current; they can cause stumbling and near-failure of the motor by screwing with alternator output briefly. As it's "upstream" of the motor, no fault is logged (sound familiar?). Checking current draw of each fan is the test.

Second issue applies to multiplex cars only; the underbonnet fuse and relay panel is actually controlled by a bus from the BSI/COMM unit and while it may appear perfect can have very poor connectivity. Test is usually a swap with fully working "known good" to see if symptoms change.

[Mandrake]

Having a look to see what's out there, found a supplier of the Valeo one for about £170:

http://www.mister-auto.co.uk/en/coil-ig ... tml?select" onclick="window.open(this.href);return false;

There is a cheaper one on eBay for about £140:

http://www.ebay.co.uk/itm/CITROEN-XANTI ... 1736wt_892" onclick="window.open(this.href);return false;

Searching for that part number from ebay I found this table that appears to list many different equivalent parts from various after market suppliers including the Valeo code:

http://www.partsbase.net/part/sta-iis065" onclick="window.open(this.href);return false;

Quite a few brands listed there but not Sagem, and the £140 eBay listing appears to be a brand called "Standard". Anyone have any thoughts on any of the brands listed on that reference page ?

I would have thought Valeo would be a fairly safe (albeit more expensive) punt as they are an OEM for many parts in Citroens if perhaps not this particular one. Most of those other brands I haven't heard of before.

[addo]

Standard is a parts branding house, just like QH.

It's fair to say the things I worry about, aren't the things Jim worries about - different service environments, different priorities in life. I've had generally good experiences with Asian parts from balljoints to electrovalves, and rank Taiwanese parts only just below Japanese for potential quality.

[Mandrake]

Can anyone tell me what the numbering of the cylinders is when looking from the front ?

I'm planning to take the top service cover off the engine tomorrow to see if I can test the coil pack, so it would be helpful to know which cylinders are 1 and 5...

I presume one of them is in the front row and one is in the back ? I might pull out the front plugs if I can to check if there is any fouling of the front plug of the 1/5 pair relative to the two other front plugs.