Hi Steve and thanks for your input.

Not sure what you mean by the above, could you clarify ?

Current milage is about 84k but engine was replaced under warranty by jaguar in 2002 at 42k. No idea if manifolds etc were changed at the same time.

 

I think he means that whatever fault you've got has caused the ECU to overfuel the engine as much as it can, within it's own programmed limitations.

STFT= Short Term Fuel Trim (changes by the second)
LTFT= Long Term Fuel Trim (changes much more gradually over minutes, hours etc)

 

I think that's what I mean too - whatever's wrong is pouring petrol into bank 2 - hence emission fail + poor economy.
There are three possibilities -
1. Bad plug/plugs bank 2 (which should throw misfire codes)

2. An inlet leak to bank 2 after the MAF
-this is a bit counter-intuitive but what happens is the MAF works out how much air is going into the engine and supplies fuel accordingly.
The upstream O2 sensor then tells the ECM if all the fuel has been burned (low O2 level) and the ECM then trims fuel injection accordingly.
If there's a leak after the MAF then more air is getting in than seen by the MAF. The O2 sensor sees too much air (i.e. not enough fuel) and the ECM winds up the STFT. Once the STFT is as high as allowed it starts winding up the LTFT.

3. The O2 sensor is lying - saying there's more O2 than there really is.
This would normally produce a sensor specific code but, apparently, not always.

You risk damaging the CAT on that side if you don't find the problem.

The important thing is not to go replacing the O2 sensor if it's a leak.

Can you hear any sort of hissing noise from the engine (supercharger doesn't help, I know) ?

How many miles and do you know if the upstream O2 sensors have ever been changed ?

 

Current milage is around 95 k, no idea if o2 sensors have ever been changed.

One question though. would an o2 sensor failure on bank 2 affect both exhausts as the garage tried both sides and both failed.

test results attached.

 

The 'separate' exhaust systems are actually joined near the front of the car, so the exhaust gasses will be the same on each exhaust pipe outlet at the rear.

For V8 engine to perform at it's best, it was discovered way back in the annals of American V8 muscle car tuning that a link pipe was needed between the 2 engine banks at the exhaust near the front of the car.

With truly separate exhaust systems as on classic US muscle cars, you'd find a 'balancer' pipe joining the separate exhaust systems together fairly close to the exhaust manifolds.

On modern cars it's usually done by having the 2 exhaust downpipes feed into a single large front silencer so the exhaust gas pulses from both cylinder banks combine. Then separate outlet pipes are taken from the single big front silencer to give the appearance of separate exhaust systems for each side of the car.

On the Jag V8, each exhaust manifold exits into it's own separate downpipe catalyst-which also houses the O2 sensors. The 2 front downpipe catalysts then exhaust their gasses individually & the 2 pipes are then fed into a single large front silencer, which allows the exhaust gas pulses to intermingle with each other from the separate cylinder banks.

So the quality of the exhaust gasses at the rear tailpipes is the same on both sides, due to the gasses mixing together at the single front combiner silencer.

Therefore, an O2 sensor failure on just 1 bank of the V8 cannot be detected by analysing the exhaust gas from it's associated rear tailpipe. The 'bad' quality exhaust gasses from the suspect cylinder bank will intermingle with the 'good' quality exhaust gasses from the functional cylinder bank in the big front combiner silencer-so the result is equal exhaust gas quality when they reach rear tailpipe.

 

I think there's something to learn from the CO + Lambdas here.
I'll have to think about it and could do with input.
The very high CO suggests the CAT is burning excess fuel so there must be excess air - i.e. there's a leak.
If the O2 sensor was lying there would be excess fuel with no excess air (O2) for the CAT to burn it with so wouldn't one get a high lambda (rich mix) + normal COs ??

I'm no expert on this but stoichiometry is something I do know about.

 

From my past experience with other cars which have failed the MOT emissions test in the same way, with an excess of CO & a Lambda value of less than 1, it's always indicated a rich mixture & the cause was always a failed O2 sensor which gave no output signal.

On the narrow band O2 sensors, a 'low' voltage output signal from a working sensor indicated a weak mixture-so the ECU richened the mixture. As the mixture was enrichened, it would eventually become too rich & cause the O2 sensor output to swing positive & give around 0.8v output. This was the signal for the ECU that the mixture was too rich now & so the ECU would then 'swing' the other way & weaken the mixture until the O2 sensor output voltage fell to around zero-which was the 'trigger' for the ECU to start richening the mixture again.

Thsi process would 'cycle' continuously in closed-loop mode & hold the average mixture strength around the Stochiometric point where the catalyst works at it's most efficient.

If you examine this switching process-at least for the narrow band O2 sensor operation-you can see that the 'trigger' for the ECU to start enrichening the mixture was when the O2 sensor voltage fell to zero-which it will do with an excess of air in the exhaust gasses indicating a weak mixture.

However, a 'zero' voltage output from the O2 sensor can also happen when it simply fails & gives no output. The ECU sees this constant 'zero' output voltage from the O2 sensor as a signal that the mixture is too weak, so it enriches the mixture. However, because the O2 sensor has failed & is not giving any output voltage, the ECU continues to enrichen the mixture until it hits it's own internal programmed buffers.

It then 'latches up' with a rich mixture & the car fails the MOT test with high CO & a Lambda value of less than 1. This happened to me on several occasions with the older narrow band O2 sensor systems & the cause was always the O2 sensor failing.

The MOT emissions results fail sheet always looked exactly like the results in the OP's thumbnail MOT sheet, with excessive CO & a Lambda value less than 1.

As has been pointed out, any misfire will result in unburned fuel being ejected into the exhaust system. However, as this unburned fuel did not consume the air in that cylinder, then you'll also now get excess air in the exhaust manifold, which the O2 sensor will read as a weak mixture & then try to enrichen it.

However, any misfires on the X308 should throw up a fault code & also cause slightly rougher running. Excess air in the exhaust can also be caused by a weak or failed fuel injector, so these are worth checking.

I've also had new O2 sensors fail a year later if the car has only been used for short journeys, cold starts etc & everything has 'sooted up'. So it may still be worth checking/replacing the Upstream O2 sensors.

There is one thing you can do to check the O2 sensors though-but it involves a bit of work.

Simply swap the O2 sensor from the working cylinder bank with the O2 sensor from the faulty bank-just swap them over & then recheck the fuel trims again. If the O2 sensor is the cause of the problem, then the fault should 'migrate' with the sensor swap to the other cylinder bank.

If the fault stays on the same cylinder bank even after the sensor swap, then you know the O2 sensors are ok & the fault lies somewhere else.

The usual method of tracing individual cylinder faults is to disconnect the ignition coil from that cylinder. Whichever cylinder is faulty will not change the engine performance or exhaust gas fuel trim levels when it's disabled. To disconnect an individual ignition coil, you just remove the bolts & the central thin plastic strip from the cam cover, and then disconnect the multiplug from the coil.

This should theoretically throw up a fault code, so you'll need to need to scan for these & reset the EML with your scanner each time before moving onto the next cylinder.

Because your fault is limited to 1 cylinder bank, it's got to be something that's 'bank specific', like the O2 sensors, fuel injectors, ignition coils, spark plugs or inlet manifold/chargecooler for that bank.

Just out of interest, it may be worth doing a cylinder compression test to make certain there's no mechanical problem. If the cylinder can't attain the required compression pressure, then the air/fuel mixture won't ignite so you'll again get the unused air from that cylinder going to the O2 sensor, which will sense a 'weak' mixture & enrichen it.

 

Intake leaks are very difficult to fine\d visually. There is a tool that some garages have for detecting intake leaks. It is a smoke device that emits a small stream of smoke. You waive this over the motor and the leak will draw the smoke into the leak.
The MAF is very sensitive to intake leaks - one symptom I experienced was surging and failure to idle on cold starts from the lean condition. But as the motor warmed it returned to normal, probably because the leak was so small.
I don't know the protocol for MOT emission testing, but wonder if they insist on testing cold start where leakage is most evident? I know that California will test a warm car and does not insist on a cold start test. When I lived in California I always drove my car fairly hard at least ten miles before emission testing; wanting the motor oil and coolant to be fully warm.

 

In the UK MOT test, any emissions testing has to be done with a hot engine-if it's still cold then the tester will warm it up before checking the emissions.

 

replacement sensor is on order - getting to the old one looks like a pig of a job, does anyone have any tips on that ?

 

Yes, it is a pig of a job unfortunately-the only way to access it is to drain & remove the coolant expansion tank by the bulkhead, then disconnect some of the heater hoses in the area until you can access the small Torx bolts on the exhaust manifold heat shield to remove it & access the sensor.

I drained the coolant expansion tank by inserting a long length of clear windscreen washer tubing into the tank, right down into part of the hose at the bottom & then sucked/syphoned out enough coolant to fill an empty 2L lemonade bottle.

Then I unscrewed the various bolts & carefully disconnected the fragile rigid plastic coolant pipes that go into the expansion tank. Then I used mole grips to clamp & squeeze the main pipe clamps so I could wiggle the expansion tank off the coolant hose stub.

They don't use worm drive clips, but instead use those expanding spring clips which either need a special tool to squeeze together the tangs, or a carefully placed mole grip.

There's still not much space to work in there with the expansion tank off & you'll probably need to move/remove some of the heater hoses-just the ones that are in the way as you'll need to make enough space to take out the exhaust manifold heat shield.

To unbolt the exhaust manifold heat shield you'll need a small Torx set & some long extension bars with universal joints-the Torx bolts don't face upwards so access is not easy. 1/4 inch drives work best in the limited space there.

Once the heat shield is loose it has to be carefully moved upwards-it's quite rigid & you can't squeeze it through gaps, so you have to watch it doesn't cut into any of the soft heater hoses, so they have to be positioned out of the way.

The sensor itself is behind the heat shield at the top of the catalyst & you'll ideally need an O2 sensor wrench adaptor with the split in it so it goes over the electrical lead. Again, access is not great & it may be tricky to get enough leverage to unscrew the sensor.

The electrical plug is accessed by removing the long centre underbonnet plastic trim panel from the bulkhead-it only has one plastic screw fastener in the top centre. The sensor electrical plugs are in a bank below this trim piece, just above the flywheel bellhousing. You may also need to unscrew & move the Evap valve out of the way to make more space to access the electrical plugs.

It's a fiddly job, so allow a morning or afternoon for it.

After refitting the sensor & refilling the cooling system, the coolant level may drop over the next day or so as air locks vent upwards so don't be suprised if the low coolant warning message comes up-this happened to me although there weren't any coolant leaks.

Oh, and your knuckles may get a bit knarled

The sensor on the other side can be removed very easily in about 15 minutes or less, but Fate usually makes the awkward-to-reach sensor fail first...

 

@Red October ... or anyone else

Can you confirm that the front silencer on a XJR allows the gases from the two banks to feely intermingle?

 

Yes, the front silencer is a single large unit which has 2 inlet pipes-these come from the individual catalysts on each bank of the engine & the exhaust gasses from both banks intermingle in this silencer.

Once the exhaust gasses leave the front silencer, they again follow separate paths to the exhaust tailpipes.

The exhaust gasses from both banks freely intermingle in the single front silencer-this is a design characteristic of V8 engine tuning & can be traced back to the V8 Muscle Car era, where you could find a balancer pipe at the front of the V8 linking the exhaust gas flow of both banks of the engine, before they went into the silencers further down the car.

On modern V8 engines the 'combining' is done via a single front silencer to allow the exhaust gasses of both banks to intermingle. My old BMW 540i V8 had the same arrangement, with individual catalysts for each bank then feeding into a single big silencer in the middle & a further single silencer at the back.

On the Jag they use 2 'separate' exhaust tailpipes, but the gas flows are actually combined at the front silencer after the catalysts-so you can't monitor the individual bank exhaust gasses at the tailpipes as they have 'intermingled' at the front silencer...

It's a characteristic of V8 engine tuning which improves gas flow efficiency-somewhere along the line the individual bank exhaust gasses must intermingle for correct exhaust 'pulse' tuning flow.

 

Don't bother draining the header, unbolt it and tie it to the windscreen wiper.

O2 Socket is 22mm or 7/8th", you want the long slotted version, it makes it easy but a long extension and U.J of 3/8th or 1/2" plus breaker bar, they're in tight. I have a snap-on S6176 but almost any make of the size mentioned will do, Franklin, Teng, Britool etc.

I can do these in under an hour. It isn't as bad as working on Audi/BMW V8's

 

The job will probably take longer than an hour for someone who has never done this particular job before...

The biggest problem was that rigid heatshield-it can't be squeezed to get it through gaps & the sharp edges rub against any coolant hoses that are in the way-mainly the heater hoses.

So I just drained & removed the header tank as it wasn't particularly difficult to do & gave more space to work.

 

Off topic, but the practice of exhaust crossover pipes was developed at one of the Irish universities (not a joke). The effect is to increase torque at low revs.

 

That's interesting & I guess the 'Irish Joke' now is that the annual road tax for anything much over 1.5 litres in Ireland is hugely expensive & absolutely discourages big V8's...

 

Don't suppose you have a link to the technical papers?

 

Hi Guys.

I just got around to changing the O2 sensor and have hit a problem.

When I unbolt the coolent expansion tank, if I tie it off to the windscreen wiper it is even more in the way than if I leave it in place.

I then can't see any torx screws on the heatshield (at least I think its the heat shield) just 2 standard bolt heads. Could it be these ?

I looked at the JTIS disk to see what it said and it says that you have to get to it from under the car ?

Is it possible that we are talking about 2 different engine types ?

Engine was replaced in 2002.

Will try and get a few pics shortly

 

pics