Ross,

Thanks, correct! I just found the schematics and saw that they were the element terminals rather than the heater terminals. Additionally, I saw that the element inputs and signal ground are all routed to the EMU with the signal at a steady 3.8 volts and and ground at SG (signal ground) at a steady 3.5 volts. My 'current' hypothesis is that this input is a constant voltage variable current input lifted off battery ground for reasons of noise immunity. If so a 'scope won't really show it so I'm back to square one.

What I'll do now is to test the heater element and if it is OK I'll replace the 'A' bank sensor and see what the error codes do.

Jim

 

Thanks, I'll try that!! (Although I'm not quite sure how...) I was actually able to get the code I was looking for today, but while doing it the dashboard lit up like a christmas tree - all the errors I mentioned earlier + restricted performance mode. Yes, live data is there and I am able to record it, but then of course I also get error codes caused by using the scanner. Jima, I'll let you know if I have any sucess.

 

Well, I checked both upstream sensors and both have open circuit heaters. I also checked both downstream heaters and they measure approximately 6Ω so with 100% duty cycle drive they would be pulling about 2 amps which is in the ball park so I know I am measuring correctly. It seems the Jags pulse the heater circuits so it would seem that maintaining O2 sensor temperature is done partly by the exhaust gasses and partly by the heater circuits. I imagine this is especially relevant when the engine is cold. It would seem even with a cold engine the O2 heaters bring the sensors to working temperature in about 8 seconds.

P1646 and P1647 errors can be caused by heater element failure according to the book of DTC's so look like next step is replacement of both upstream O2 sensors.

Look forward to hearing your results too, Bo!

Jim

 

Well, a resolution finally! I took Misty in to the local Jaguar dealer for a 90K service and having explained the mystery of the non-illuminating CEL/MIL they readily agreed to do a full diagnostic read on their equipment at no charge (thanks, Stratstone!). I was relieved to see when I collected it that the only two codes which had turned up were P1646/P1647 which is exactly what my reader displayed.

I just had both upstream O2 sensors replaced today and having collected it and test driven it can report that a) it is much smoother and quieter at idle - in fact I thought I had another engine problem until I realised that what I was hearing was the engine of the Renault in front of me idling and b) the engine feels generally much more responsive and smooth, in fact seems to drive the auto box a bit better as well but that could be my imagination, I guess.

Putting the reader on when I got home I was delighted to see no stored or pending codes, just the usual P1111 'cycle completed' code and the engine was finally in Closed Loop mode - yay!

Displaying the O2 sensor output I have no more flat lines for the upstream sensors but nice, healthy activity on both sets. I have included an image of what this looks like on the EngineCheck screen during a 5-minute cruise to the off-licence (liquor store) to get some juices of the fruit to celebrate that my ECM is not toast for anyone interested.

In summary:

a) still no idea why the CEL/MIL light isn't on and neither does the dealer. I'm probably not going to bother following up on that one.

b) Contrary to some opinions here the EngineCheck reader was (and is) spot on! It doesn't cause any CAN issues, put the car in limp-home mode or bring about a plague of locusts. It doesn't cure cancer either but does seem to do a spot on job of displaying both generic and Jaguar error codes. As a bonus you can see what both sets of O2 sensors are doing and perhaps the Pro edition does even more. Kudos to Gendan and many thanks go to them for the great help Dean there gave me.

c) None of the 'classic' O2 waveforms I believed I was supposed to see after researching the 'net were anything like what actually showed up. The upstream O2 sensors appear to be constant voltage, variable current 'planar' types. Even my trusty Tektronix 7704A oscilloscope was of limited use.

d) Both of the upstream heaters were open circuit. Very odd. Since the ECM monitors the heaters on both sets of sensors it seems to have detected this fault and decided to ignore the pre-cat sensor outputs despite the fact that eventually they would have heated up. Of course, the working temperature would have varied wildly between idle and speed during driving so perhaps it was a design decision to ignore them due to the inaccurate readings caused by the faulty heaters. Of course, this is speculation on my part but the system never went into closed-loop mode with the faulty sensors.

d) Working O2 sensors are a must have - if you have any issues with a lumpy idle, rough running or sluggish throttle response get them checked fast - of course other faults may cause these symptoms but you can at least eliminate the sensors as a cause.

e) Finally - thanks to all who responded here. All ideas and contributions were very welcome and helped greatly.

Jim

There are some other displays of an '01 XJ8 4.0L here http://www.attfield.co.uk/gallery/jaguar/diags.htm

 

Jima's problem is solved, so all is well. I just want people reading up on scanners know that I have not been able to change the configuration of my scanner in the way that Sparkenzap suggested. There's simply no such menu in the device. (Thanks anyway!)

 

Im not gonna read through all the replies but a quick answer, thsoe are both upstream 02 sesnro faults. He's the deal. The light may have been on at one point, but after so many key cycles/starts. If the ecu no longer continues to see the fault it will turn off the light and show the codes in history. Clear them and see if they return

 

Thanks, Brutal, I cleared them quite a few times but they always came back. My real puzzle is that the scanner actually reported that the MIL was on but it wasn't. It passes the lamp test OK so I know the bulb works.

Jim

 

Caution! Caution! Caution! Caution! Caution! Caution!

I can absolutely verify using the Autel MaxiScan MS509 OBD-II/EOBD Scanner initially everything looked Ok but then my 2003 XJ8 went int restricted/performance limp-home mode and reported a transmission/ACS failure and WS509 reported code 1642 (CAN error?). The next day exactly the same thing happened with my 2002 XK8!!!!!!

Seem to have fixed the problem by starting the car with the brake depressed, then clearing the codes with another scanner.

 

Glad you found and fixed your problem.

Relative to the scanners - I think the fundamental issue is that our Jag is an overuser of the CAN bus and was somewhat poorly implemented. I have designed CAN bus systems in the past and have at least some knowledge here - IIRC the XJ has something like 32 processors, all trying share data space on the same bus. Not all scanners can handle this kind of data traffic without choking and not working, or worse, causing CAN bus collisions themselves, leading to spurious readings.

One of the stickies in this forum, again, if I recall, has a list of scanners that forum members have had the most success with.