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

[xantia_v6]

Under most circumstances the "auto-adaptive" adapts correctly to compensate for wear and (presumably) oil condition, but it may be possible that it adapts itself into a state where it misinterprets slippage and gets stuck with wrong parameters. In that case a reset would be a very good idea.

The Mk2 V6 that I had for a while had some difficulty with the auto-adaptive function, and the transmission parameters would vary significantly over the course of a few days, between having very soft changes, with noticable spin-up when taking off from rest, to having rather sharp harsh changes, so it certainly was continually adapting, it was just never happy with the result.

Yes the shift characteristics do vary on this car, sometimes quite good, other times they can be a bit abrupt and erratic, and whether it releases the lockup or not seems to vary.

On your Mk2 did you ever reset the auto adaptives and if so did it make any permanent improvement ? Did you ever do an oil change ?

Could it be that the hydraulics on your Mk2 were intermittently malfunctioning due to sludge build up affecting the operation of the valves, and the auto adaptive was trying its best to compensate for the constantly changing behaviour of the hydraulics ?

The solution for both cars may be a change of oil (or 2 over a period of time) to clean it out followed by an auto adaptive reset a few weeks later after things have stabilised.

I did reset the auto-adaptive several times, and it would take the transmission back to a state where the changes were comparitively rough, but after a while the same pattern would emerge.

My theory at the time was that the symptoms may have been caused by low oil pressure in the transmission due to a blocked filter, but I sold the car before we got around to making a pressure gauge adapter. The car did have some 'transmission history', as it was a 1999 car with a 1997 transmission fitted, and it was filled with dextron when I bought it (but the symptoms above were still present after a couple of fills with the correct fluid.

I think you should pull the transmission fuse, and report how it drives. Take-off is a bit slow, but otherwise it should drive OK.

[Mandrake]

I really hope this transmission hasn't had Dexron or other wrong oil put in it. Many a Citroen gearbox has been ruined by Dexron 3

When I got my 2 litre petrol auto (4HP14) the oil in the gearbox was completely black at only 8 years old and 83,000 miles, I did about 3-4 changes over the first few months (Castrol Transmax M from memory) of ownership to get it back to a nice translucent cherry red.

From what I remember discussing it at that time, not only is Dexron 3 wrong for the 4HP14 and will kill it quickly, mixing Dexron 2 and Dexron 3 often leads to the oil burning and going black fairly quickly, so I was suspicious that it had had a Dexron 3 change, as the practice of people putting Dexron 3 into Dexron 2 gearboxes like the 4HP14 was rife in Australia/NZ at the time...as AlanS had well documented.

Fortunately that gearbox did 4 years and 30,000 miles of smooth trouble free mileage without the slightest hint of trouble after the oil changes, and was still working perfectly when I sold it. Before the oil changes it was working ok but didn't always shift as smoothly as it should have, and it could have potentially been a write off after a few years if I hadn't done those changes.

Regardless of what I find with the engine I'm strongly inclined to do at least one change of gearbox oil, and fairly soon, if only for peace of mind as a blown gearbox would write the car off completely after all the effort I've put into giving it TLC...

Ok, I'll give the fuse a try, I assume it will behave very similarly to snow mode except without 4th, since snow mode starts in 3rd gear and seems to leave the torque converter open a lot of the time. One other thing it will do is prevent the gearbox ECU sending commands to the engine ECU, such as asking it to retard the ignition for torque control while changing gears...(is there anything else the gearbox asks the engine to do apart from torque control via retarding the timing ?)

[CitroJim]

Bit OT Simon but the Chris's recently did an oil change on my XM (4HP18) 'box. The old stuff came out dirty, horrible and black. They refilled it with genuine Dexron 2 and the 'box is now noticeably smoother and happier everywhere. I'm planning for a few swift changes and it'll be interesting to see how the oil looks on the next change.

Luckily on the XM a gearbox oil change is ridiculously easy easy and Dexron 2 is still pretty cheap.

Shame the same does not apply to the HP20 but it has to be a job well worth doing. When you do, give the old oil a good sniff. It's not quite so easy to smell bad things in LT71141 as it smells of dark toast and cat's pee anyway but at least you’ll be able to tell if it's Dexron as that has a more oily smell and the last thing LT smells like is oil

Remember too that LT is not red, it begins as light straw amber and even after a long interval it should be still a little translucent and not dark black. If it is and smells more of burned toast and less of cat and is no longer translucent then you may have grounds to be a tad concerned. The only HP20 I've drained that had black oil in it looked like the Somme inside; it was carnage...

But no worries, just pop the 'box off, pop it in the post to me and I'll do a quick rebuild on it for you...

[Hell Razor5543]

I'd hate to have to pay the postage on that! Time to bring the FCF relay procedure into play, methinks (if i ever think!)

[Mandrake]

Simon, there's something very seriously wrong. Mine doesn’t labour at all at 1500rpm in top gear, let alone third. Gosh, a V6 can go practically everywhere without ever exceeding 2000 rpm and not show the slightest sign of labouring at all, TC open or not. V6 engines don't labour!

Mine cleanly accelerates from 1500 rpm always in any gear and any circumstance. I deliberately tried it today. Whilst I was not watching a Lexia to see when the TC opened I could but I'll need to enlist an assistant. If it will help I will.

I think, in light of what I said before, that the problem is with the engine rather than the gearbox. That groaning is odd though I have to admit. A fault on the engine will make it seem as if the gearbox is about to throw the towel in. Remember, the engine is designed to produce a lot of torque at very low revs and is carefully matched to the gearbox so any deficiencies in the engine will show up more from the gearbox than perhaps they would if it was a manual gearbox.

I've been thinking about what you're saying Jim and there does seem to be a major intermittent loss of engine power - when I first disconnected the oxygen sensor the power seemed almost normal, certainly far more zippy than it has been in a long time, but in the last 2 days since the weather has been very wet with a lot of moisture hanging in the air (coincidence ? I don't know) the power has been quite lacking especially at lower revs.

Yesterday when it was really bad it even had a flat spot at around 2200rpm where if I pressed the throttle a bit it almost seemed to cut out for a fraction of a second - just like a flat spot on a carburettor...

Maybe we need to go back to basics here, what things can cause loss of power ? No (or intermittent) spark, incorrect timing, insufficient fuelling, or low compressions. Anything else ?

Is there any possibility that a slightly dodgy MAP sensor could be causing intermittent loss of power that matches the symptoms ? When I tested it with the Lexia all I did was check that the ambient atmosphere reading was correct with the engine off, and that the reading dropped when it was idling, which it did to about 380mb from memory.

However that's hardly a thorough test - it says nothing about the accuracy of the reading over a wide range, (which I presume is critical for accurate fuelling) nor does it say that its not intermittently reading incorrectly under some conditions of temperature, humidity and so on...whatever is wrong does seem very intermittent, and ever since I've had the car the performance has varied noticeably from day to day...

Is the data from the MAP sensor used to help the gearbox decide when to change gear under load ? If the MAP reading was jumping up and down intermittently could that cause the gearbox to change gear and change back again quickly as it occasionally does ? From what I remember from skimming the 4HP20 manual a complex calculation is done by the engine ECU using various inputs to calculate the torque output of the engine, and that torque figure is sent to the gearbox to help it decide what to do...if that data from the engine ECU is faulty, could it cause the gearbox to misbehave ?

What about an air leak downstream from the MAP sensor ? Could that lead to leaning of the mixture that the ECU wouldn't detect and compensate for ? Is there any piping on the V6 that needs checking for air tightness ?

Another thought - the manifold gasket that needs replacing if I do the rear plugs, what happens if that gasket is already leaking ? Could that upset the mixture and the running too ?

[Mandrake]

When you do, give the old oil a good sniff. It's not quite so easy to smell bad things in LT71141 as it smells of dark toast and cat's pee anyway but at least you’ll be able to tell if it's Dexron as that has a more oily smell and the last thing LT smells like is oil

Remember too that LT is not red, it begins as light straw amber and even after a long interval it should be still a little translucent and not dark black. If it is and smells more of burned toast and less of cat

Sorry Jim, but this left me in stitches.

I've never heard of an oil being described as smelling of burnt toast and cat pee before. As it happens I have a very poor sense of smell so I'm not sure that I would be a good judge of whether it smelt more like burnt toast or more like cat pee.

[CitroJim]

I'm delighted I gave you a laugh with the oil Simon

Now, yes, there's a problem in the engine no doubt about it.

The MAP sensor is a possibility and cheap to replace with an ebay one. Ditto with the coolant temperature sensor. I'd replace both and then replace the Oxygen Sensor after seeing how it goes with the first two replaced. In conjunction with this, get the plugs replaced and check the rear bank plug wells for oil ingress.

Air leaks south of the throttle body are possible and if the manifold has been off in the past and the gasket not replaced than it's very possibly it's letting in air and upsetting things. Also, there's a small pipe by the injector plug over on the LH side of the rear bank that takes a vacuum signal to the fuel pressure regulator. A leak in this pipe would both allow false air in and upset fuel pressure regulation under vacuum at the same time and might well give rise to the fault you have.

Early V6 inlet manifolds had a blanked off spigot at the back of the manifold covered with a rubber cap. I'm almost sure yours won't have one but it's worth a check to see.

Have you checked there's no air leaks on the carbon canister purge pipe and canister? You could disconnect the pipe at the purge valve and plug it off as a test.

If all this fails to give happiness then I'd be keen to measure the fuel rail pressure to check it's up to spec. and to ensure it remains constant under all vacuum conditions.

So...

Replace MAP and CT sensors
Replace Plugs and manifold gasket
Check for inlet air leaks
Check carbon canister
Replace Oxygen Sensor
Check fuel pressure

We'll get there

[lexi]

Simon: As is my wont, with all these boxes I changed the oil first weeks I had it.

I think I did just the one change of 3 to 4 litre. I got 2x 2ltre from Arnold Clark of the correct LT Esso stuff.
I remember the oil was dark as they all are but quite clean compared to my mk 1 V6. Did not notice any red in it and remember being quite happy with what I found.

I don't think I did 1K miles in that one so maybe oil aint bad. Miles are still low for year? Dunno really but another change will do good.

[Xantidote]

still got a broken exhaust - my patch up job is hanging in there but I think its leaking a bit now

In the distant past, some engines (the Rover 3.5 litre for one), were very reliant on back pressure in the exhaust. Couldn't be the case here could it?

[Mandrake]

The MAP sensor is a possibility and cheap to replace with an ebay one.

Just had a look on ebay and they all seem to be around £65 I'm a bit baffled as I'm sure I saw some for around £25-30 only a few weeks ago. I'll have to look a bit harder. What price have you seen them at ?

I notice it has an o-ring on the stalk, is it worth me popping the sensor off and checking the state of the o-ring in case it has a leak ? If there was a leak there it would read ambient pressure ok but would be inaccurate under vacuum, and the inaccuracy may vary intermitently...

Air leaks south of the throttle body are possible and if the manifold has been off in the past and the gasket not replaced than it's very possibly it's letting in air and upsetting things.

Could be a double benefit for doing the rear plugs and fitting a new gasket then.

Also, there's a small pipe by the injector plug over on the LH side of the rear bank that takes a vacuum signal to the fuel pressure regulator. A leak in this pipe would both allow false air in and upset fuel pressure regulation under vacuum at the same time and might well give rise to the fault you have.

Ah, this could be well worth investigating, I wasn't aware of it being there or its purpose. Is it relatively easy to identify or does someone have a picture of it ?

Early V6 inlet manifolds had a blanked off spigot at the back of the manifold covered with a rubber cap. I'm almost sure yours won't have one but it's worth a check to see.

Actually I think it does have it - its on the right hand side (when looking from the front of the engine) on the rear manifold facing towards the firewall, right ? The rubber is a bit squishy feeling although I don't think its leaking. I will check it again though.

Have you checked there's no air leaks on the carbon canister purge pipe and canister? You could disconnect the pipe at the purge valve and plug it off as a test.

No idea Jim, any suggestions on what it looks like ? Half the pipes and small objects under the engine bay are a mystery to me. That wouldn't be a cylinder thats vertical near the back of the butterfly housing would it ? What does it do ?

If all this fails to give happiness then I'd be keen to measure the fuel rail pressure to check it's up to spec. and to ensure it remains constant under all vacuum conditions.

How does one go about measuring the fuel rail pressure ? Can the Lexia measure it or do I need a special gauge spliced into the fuel rail somewhere ?

So...

Replace MAP and CT sensors
Replace Plugs and manifold gasket
Check for inlet air leaks
Check carbon canister
Replace Oxygen Sensor
Check fuel pressure

We'll get there

I really hope so, it worries me when I don't know whats wrong with winter fast setting in, and I miss having it running smoothly...

[Mandrake]

still got a broken exhaust - my patch up job is hanging in there but I think its leaking a bit now

In the distant past, some engines (the Rover 3.5 litre for one), were very reliant on back pressure in the exhaust. Couldn't be the case here could it?

You could have a point, when the exhaust between the cat and the middle silencer came apart it killed the performance dead. I fixed the joint temporarily a few weeks ago and at first it was air tight, but I've noticed that there is a small amount of leakage at the joint now, I've seen water dripping from it when the engine is running too indicating that it must be leaking...I really need to get the exhaust fixed before my bodge job falls apart, but I've been unable to find anywhere yet that is willing to repair the cat tail instead of just replace the whole thing.

[Mandrake]

Simon: As is my wont, with all these boxes I changed the oil first weeks I had it.

I think I did just the one change of 3 to 4 litre. I got 2x 2ltre from Arnold Clark of the correct LT Esso stuff.
I remember the oil was dark as they all are but quite clean compared to my mk 1 V6. Did not notice any red in it and remember being quite happy with what I found.

I don't think I did 1K miles in that one so maybe oil aint bad. Miles are still low for year? Dunno really but another change will do good.

Thanks lexi, I remember you saying you had done a change of oil, although if the stuff in there was dirty or the wrong stuff it would take more than one change to be rid of it. Maybe I'll focus my efforts back onto the engine for the moment.

I'm not sure the exact mileage I've done but I've probably done over 2000 miles in it now, twice as much as you did then. It's also only since the MOT in August that things seem to have gone really wrong (broken exhaust, poor running etc) so I guess I'm lucky this didn't happen before the MOT preventing it from passing...

[Hell Razor5543]

Silly question, but is there an oil flush that could be used safely on the gearbox? I know nothing about working on gearboxes, but I have seen flushes for engine oil.

[Mandrake]

Silly question, but is there an oil flush that could be used safely on the gearbox? I know nothing about working on gearboxes, but I have seen flushes for engine oil.

I don't think there is anything like that, at least not for gearboxes like the 4HP20 which require "special" oil that is not used in any other gearbox.

The problem with auto boxes is that you can't drain the torque converter, so less than half the total oil comes out with each change which means progressive dilution of old dirty oil is the only way to clean them out, the dirtier the oil is the more cycles of dilution it requires. (up to 3-4 if its really bad)

[Chris570]

Honestly, looking at this you've got a LOT of work to do. Make it easy on yourself (and us to diagnose)

Can i just make one suggestion? (ignore me if you want to)

You know the exhaust is a potential problem, fix that.
The inlet should come off and at the same time you can check/change the rear bank plugs and check the little vaccuum hose jim is talking about (it's really obvious) the cannister you're talking about is the idle control valve and shouldnt have any bearing on this but when you take the inlet off (not a hard job at all) you'll be able to see the state of play. To get the inlet off should take no more than 15 mins, i'm not joking.

Once you've done those two things you've ruled out an awful lot of potential issues and you're really narrowing it down. Otherwise you might as well paw through the service manual and say "well it could be that" *next page* "or that".

My point is simply that you need to rule more stuff out before you get the next theories in

Ignore me if you like but that is exactly how i would go about fixing the issue.