Those are where I see my sensors.

I have checked the wiring and it seems OK although if I run the car and scanner if I wiggle the connector it might just shift a tiny amount from its static and very low reading.

I guess I'll have to spend the 50 quid on the Titanium sensor. Sensor 1 does bank 1, is the cylinders 123 which are closest to the radiator - right?

Mike

 

Torque App definitely detects 4 sensors and reads figures for them so that doesn't really inspire confidence. I may try another app and see what that says but Torque Pro is meant to be about the best App.

It is a bit puzzling. It's not as if 1 & 3, 2& 4 are identical. Sensor 1 is dead and the other three tend to be very similar.

Mike

 

yep, that is the one!

 

Thank you SparkenZap.

I have been thinking this over trying to work out why I don't have a problem at cold and if I do have the problem of massive lumpy running on the drive I can turn the engine off, restart and it's fine until I race the engine a bit.

So, it can't be the engine temperature as it's still just as hot as it was but the exhaust temperature rises hugely from tickover to 2,000 rpm even on the drive. On the road similar thing it's only when the engine is stressed at 4,000 rpm say that the problem starts and that's probably the first time the exhaust gasses get really, really hot.

I am drawn to the conclusion that it's the high exhaust temperature that triggers the problem and the sensor sits in those gases.

I have no idea why that would seem to limit the revs on the road or make it lumpy on the drive but I'm guessing that the ECU just cannot cope with the nonsense readings it gets from the faulty sensor and makes a mess of things.

I would rather try and understand what the problem is than change parts at random until the problem just goes away.

Having swapped the MAF out I think that can discounted, on the drive the engine temperature, the air intake temp, the coolant temp, the fuel flow would all remain the same. Then revving the engine triggers the fault. On the drive none of those will change but the exhaust will.

Sound logical?

All that said on the road it feels like fuel starvation but I suppose the ECU might be limiting the fuel due to faulty readings.

Thank you all for your patience.

Mike

 

You may be right. The actual closed loop and open loop algorithms for the X-300 are generally not well defined and are apparently not like "other cars" in general, probably since they were early in the ECU game and hey were from Jag-u-wah.

Don;t overthing this one though. If you have flatline sensor readings, and no sensor heater faults and no apparent physical damage in the wiring, there is a damned good chance you have a bad sensor. Normally, I agree wqith your philosophy of having an actual diagnosis before spending money on parts, but in this case, the odds are with the bad sensor. If you really want to be sure, swap the two up stream sensors and see what happens.

BTW, mark the connectors before taking them apart. The ECU can be "orienterd" for either connection, A or B on either of the pair of sensors, so the diagram will not necessarily get you back right.

 

Thanks, being UK I only have downstream sensors and I assume these are Titania type although I thought that where you have 4 sensors the upstream were Titania and the downstream were the cheaper Zirconium.

However, my sensors have 4 wires and therefore I think that makes them Titania.

I'll order a sensor today I have fannied about with this quite long enough.

 

Four wires is no indication of Zirconia vs Titanium. Two wires are for the heater and two to the lambda sensor in either type.

You might be confusing with the later X-308 which had wideband pre cats and "normal" Zirconia sensors post cat.

 

I had the front two sensors replaced several years ago. The new ones failed shortly after installation. Called Landcover/ Jag in Houston where I bought them. The problem was the torque used during installation. It seems they are very, very sensitive to even a slight over torque. As I recall the instructions with the replacements was to only turn the sensors a half turn after initial contact. This made no sense since they are built really stout. But it worked and they have worked for many years now.

 

Thanks for that I shall tell the guy that changes the sensor. I can't get anywhere near it.

I remain unconvinced that the sensor is the problem.

Can you run the engine with the sensors disconnected?

 

Its the crank sensor. It happened to me too. Installed a fuel pressure gauge under the hood, and that was useless. Its the crank sensor. Its a service replacement part. Not hard to change at all.

 

Really? It starts on the button every time never even thinks about it.

I'd be delighted if you are right.

Same symptoms?

Mike

 

Same exact symptoms. Would start right up and idle, but got worse as I tried to figure out what was wrong. I was sitting at a redlight, and when I went to pull off, it acted like a stumble, a misfire. Went where I was going, the car started right up, ran good through a parking lot, but when I got on the road, seems 3500rpm was its limit. I made it to my shop, and began to diagnose. Same thing you've been through it seems.
The thing is, I've had to teach myself about the Marelli ignition, and have come to realize that the same type sensor is used on it as on our 6 cylinder cars. It really isn't a sensor at all. Its a small electromagnetic generator. It sends a AC signal to the ECU as a input. It is made with a magnet. A magnet, when exposed to heat, slowly loses its mag-nativity(?). Any way, on a last ditch effort, I switched the speed sensor, and it literally fired right up, and has been running great ever since. They simply go bad. I wish I could tell you there is a way to test it, but I don't think there is. I was advised that it should read around 700ohms resistance (it did), I was told that if the tach registered rpm when turning over it couldn't be the sensor (it did). But the crank sensor did the trick.

 

Well I find that much more convincing than a dud O2 sensor.

I also find your analysis that the magnet loses its strength convincing. I think I'll be exchanging the O2 sensor for crankshaft position sensor and try that first.

 

So, I guess w ehave given up on diagnosing and starterd throwing parts at the probl;em. Ithink most of us are working on the assumption that you have a scanner reading one lambda with no variation. But then re-reading I realize that you have not really identified which sensors are reading, whether the fuel trims make sense, or whether you are even in closed loop control.

If your scanner is not able to determine that stuff, or you don't know hoiw to use the scanner, then you should fix that. Throwing parts at a 20 year old car can get expensive.

 

It might seem that way but I think I'm right that the car should run OK even if the O2 sensors are disconnected - true?

There are only two O2 sensors. Torque will create meters for 4 but two clearly do not exist - so I am unsure what to make of that.

I believe I mentioned that the car seems fine in open loop, ie cold.

The STFT readings do seem reasonable, I think they should be within +-10% and they look as though they are.

I was advised the LTFT reading at -98% (which it has been for at least a year) should just be ignored.

All the readings apart from the one O2 sensor look sensible so what would you advise, and I am serious as you do offer excellent advice which I for one appreciate?

One reading I do't know about is the advance retard reading which fluctuates a lot but then it would I imagine. Do you know the range one can expect?

Do you believe one faulty O2 sensor could cause all this mayhem?

Supercharged sounds like he had a comparable problem and the crank sensor fixed it. They are known to wear out (as are O2 sensors) and they do create problems.

I feel as though I have nothing to work with in coming to any diagnosis.

I do agree that throwing parts at an old car is a mug's game but I am stumped as is everyone I speak to. I was a little disappointed that the recovery guy who by coincidence was from the local Jaguar dealership was unable to get his diagnostics to work on my car but hey he was in his van and only doing it out of interest.

Mike

 

OK- i understand now, I think! I have not yet gotten access to the computer that has the low cost OBD scanner software (a fellow from work has it.), but I will go see him today and find the name of it. If you have reasonable, and changing, STFT on each bank, then I agree it is unlikely to be the lambdas causing misfire. I thought we had determined one sensor was not changing voltage on the scanner and the other was.
And, yes, the CKPS is well known to fail on these cars, with any different failure modes, including working at one engine speed and not another. And it is also well known to not give good OBDII failure faults.

So, I agree that the CKPS is a reasonable differential diagnosis if no faults are present and the fuel trims are "normal". And the CKPS can be had at a reasonable price and is not too hard to install.

Then, if that isn't it,you are back to fuel pressure.

 

All I know is that I have a ODB scanner. When my crank sensor went bad, there were 4 codes that came up that had absolutely nothing to do with the crank sensor. I'm not for switching parts guys. NOT AT ALL. Yet, when all else failed to identify a problem, its what I did. My experience with this all started with buying a non-running 89 XJS V12 with Marelli off of eBay. I bought it from a shop that had literally ran through roughly 6K in labor hours, and parts INCLUDING having a new ignition harness made. I felt I got a great deal on the car, and began work as soon as I got it, determined to fix what others couldn't. I went through everything I knew. The car sounded like it was timed wrong. Even the shop foreman felt the timing chain had slipped. But I felt 99.9% that wouldn't be the case as I am familiar with V12s. After MANY hours of verifying fuel pressure, timing, tracking wiring, testing the ECU by a fellow lister, I was fully stumped. I then started asking even more questions. I found all info I could on Marelli ignition. Turns out, the rear sensor (the V12 has two), I found that the sensor was .047 away from the flywheel. That was problem number one. After filing the mounting bosses to get the sensor(s) within .020 of the flywheel and the timing fingers on the front, I found that the rear sensor had been wired reverse in the new harness. Thats when I gave up, and built a whole new harness. After that, the car fired right up! A lot of learning. I also found out from some airplane mechanics that magnetos have to be changed on airplane ignition systems for the same reason our "sensors" fail.
So, when my 97 XJR had some of the same symptoms, I quickly went to the crank sensor as a possible cause. Lets face it, these cars are 20 years old. Magnets don't last when heat is nearby.

 

Yeah, I understand what you are saying, but that does not mean that every high speed (or low speed) miss is a CKPS.
My neighbor had sweats and felt faint while he was having a hear attack. That does not mean the everyone feeling faint and sweating is having one too. Not that is should be discounted, either!

 

Well I took advantage of a spare hour and quickly changed the CKPS and it starts OK and runs OK but - and there had to be a but.

It now seems to idle high. I didn't have much time to test it but if say I run it at 2,000 rpm then it will return to a 1100 idle, if rather higher rpm then higher idle. It didn't use to do that.

I assume there's no relearning by the ECU or anything? Is it likely that I just didn't fit it right which would seem to be quite hard to do.

Mike

 

I would not think there was any way a mismounted CKPS could cause high idle. Did you pull the battery lead while changing it? And yes, there are memories in the ECU.