I wrote it in a short reply to my previous post, but I think it might be interesting to more people, because this problem was reported here in the past, but I never saw a solution to it. So, here is another chapter of my year long story...

I found the knock sensor on a bank 1 was malfunctioning. I replaced it and with a new air/fuel mixture sensor Bosch 15627, the car started a ran smoothly. On the bank 2 is still the original Denso pre-cat sensor234-9016. I was exhilarated. But...after about 12 miles of country roads the check engine light and restricted performance showed their ugly heads again. My generic OBDII reader pulled out a dreaded P1642 code and P0172 and P0174 codes. Right 4 cylinders running rich, left ones lean. Can anybody explain it? Drivability seems to be OK - but I did not drive the car for a year, so I do not have a fresh comparison. After some 5 -10 minutes the restricted performance goes off, check engine light stays, but the engine seems to be taking a full throttle thrust. Only a gas mileage dropped from 24-25 mpg on the highway to 21.5mpg. I went home and measured the short fuel and long fuel trim values at idle, 1500rpm and 2500 rpm in situ. These are the values:
SFT 1 -4.7 tp 5.5%
LFT 1 -20.3%
SFT 2 10.9%
LFT 2 19.5 G /what this G stands for?/

1500rpm
SFT 1 -10.9
LFT 1 - 4.7
SFT 2 10.9
LFT 2 4.7

2500rpm

SFT 1 -7.3
LFT 1 -7.0
SFT 2 12.5
LFT 2 8.6 (G)

can anyone interpret it in the terms what I should be looking for...
thanks. I wish, I will be able to write a final chapter for this story....soon

 

Mixing old and new parts on a balanced system could lead to the Right and Left readings you are seeing. Replace in pairs with parts from the same manufacturer. Replace the remaining old sensors with new ones and it will balance out.

 

I understand that - but why are the fuel trim values so different? Bosch and Denso sensors are said to have identical properties.

 

Edited........

Shot my mouth off when I shouldn't have, see post 6. Won't happen again.

Cheers,

 

when I checked all kind of web sites this Bosch was listed as a direct replacement for the original Denso (only difference was the cost 234-9016 Denso cheapest $399, Bosch $93.
Now explain me how different a sensor can be for 32bit bus, when it is a strictly analog device changing potential/current ratio and not sending bits....

 

Well, see there, I broke my own rule and shot my mouth off in my haste to get up from the lunch table and back to the store. The AJ26 and AJ27's are both 16-bit systems, but the AJ27's do use universal heated oxygen sensors just like the later 32-bit systems. In theory, there should be no issue using a Bosch part over a Denso part. I have never run into any problem as we always use Denso parts, of course; but I have heard of experiences where mixing manufacturers has supposedly been problematic. Remember, there is a heater in there along with the sensor. It could be a variance in the heater.

I'll correct my earlier post and go stand in the corner.........

Cheers,

 

The simple, although onerous, way to check the sensors would be to switch the upstream sensors side-to-side.

Are you prone to making more left-hand turns than right? Centrifugal force could be starving the left bank and making the right run fat! ;-)

 

p.s. you might consider looking at the downstream O2 sensors' voltages to see if they support the trims. Normally not anything to hang your hat on but in your case it might support the difference between the sensors.

 

steve and Beav - thanks for an advice. I hooked my OBDII reader before I swapped the sensor connectors. No 0172 and 0174 codes but those enigmatic 1638 and 1642 popped out. Erased. Swapped terminals and got this readings ?
SFT 1 -14.7
LFT 1 -4.7
SFT 2 25.0
LFT 2 3.1

after a while and rising rpms to 2000

SFT 1 -9.4
LFT 1 -10.4
SFT 2 25.0
LFT 2 19.5

which tells me that both sensors work approximately alongside same amplitude. Now, what is the next step? WIll the post cat sensors, which are very difficult to access, have that much influence?

 

Downstream sensors exist only to check cat efficiency - they have absolutely no bearing whatsoever on the fuel trims.

Do not drive the car with the sensors switched! If you have one side running fat (actual) and the other lean (actual) the trims will adjust the wrong sides, exacerbating the issues and potentially damaging the convertor on one side, maybe to the extent of causing a fire. My previous post was to swap entire sensors side-to-side, not just their leads. I said 'onerous', read: I wouldn't do it, I'm too lazy to waste that amount of effort. There are many tricks a pro utilizes to judge their performance before condemning them, most would be more focused on what is causing the problem they are reporting rather than trying to 'shoot the messenger'.

 

Beav, I would not drive it with switched terminals, but can you explain why the measure values stayed the same on the each bank after I switched the terminals? I drove the car for about 190 miles. After I started the engine, the "restricted performance" came on and stayed for about 3 - 4 minutes and disappeared and the car seemed to perform normally.
STored codes 0172 and 0174 again....

 

dneu, I think you need to get a little better handle on a diagnostic strategy. Beav's suggestion to switch the two upstream sensors is a good one, but you need to better understand what your trims etc. are telling you. Start with a good read on this thread H20Boy started.

https://www.jaguarforums.com/forum/g...ed-quiz-49317/

He posted a challenge of sorts and I wrote up a paper on fuel trims. That will get you on your way. Then be aware that Long Term Trims are just that, long term. They do not zero out at key off as Short Term Trims do. So when you make a change, like switching upstream sensors, your reference point is lost and new readings can be misleading.

When you make a change, disconnect the battery for several minutes. Now all is back to neutral and you see EXACTLY where the sensors are leading. Whereas with out the battery disconnect, the sensors in their new position have to start working from the skewed number from before. That can be very misleading.

I hope that made sense!

Cheers,

 

XJRGuy thanks for I comprehensive paper.
Actually, it is better than some other I have read.
I have to apologize that i was not thoroughly descriptive when presenting my results. Somehow, I expected that disconnecting battery and discharging the capacitors is so obvious I did not have to mention it. Because this engine has the PCV only on a bank 1, i did not check for any leaks on the left side. Frankly, I cannot think of any unless a lose sparking plug, or something similar weird would occur.
What puzzles me is that even after switching the terminals bank 1 still shows lean and bank 2 rich mixture..I think, my next step is buying a new A/F sensor for a bank 2, although I never experience a "tired" O2 sensor before - but I have to admit, I do not have enough experience with A/F mixture sensors. (Because in newer cars Jaguar returned back O2 sensors, the A/F system might not prove itself as reliable - but it is only my speculation).

 

Small hose going to LH cam cover is your 'PCV'. Instead of a fluttering valve a fixed orifice is employed and very prone to coking. Hose @ RH cam cover is breather hose - supplies air to crankcase at low speed (that the vacuum of LH hose draws), allows larger volumes at high speed to be snorkled in through, well, the fresh air snorkle.

Sorry I was probably more vague than you realized but teaching diagnostics through the use of fuel trims, etc. is way beyond the scope of friendly advice. I rarely look at trims for more than a couple seconds, O2 voltages are where the action is and you need a good multichannel graphing multimeter, 'scope or higher-end scanner to actually 'see' what's going on. STFTs just show the ECU is responding to O2 sensors. LTFTs show history/trends. O2s are the 'Church of What's Happening Now'.

 

Beav, that is OK, I went to trims just for that reason, that I do not have an oscilloscope i could hook to the OBDII socket and my generic OBDII reader gives me just averaged O2 readings, but no oscillation frequencies, nor milliseconds scans...anyway, mu fault that I did not notice the PCV hose and the left bank. First thing tomorrow morning to do. Are the PCV prone to clogging or any other failure?
There is always to learn more...

 

I have checked the PCV line on the left bank and it is intact. but did not check the other end hiding somewhere under the intake manifold...

 

I still get thud P0172 and P0174 codes. Not always and not regularly. But the certain pattern is observable. I usually drive 10 - 15 minutes and "restricted performance" comes up, and the car behave accordingly. It lasts usually 5 - 10 minutes and "restricted performance" disappears, the gauge restores itself to the previous mode and for the rest of drive there is no problem, only engine check light stays on.Have anyone ever experienced anything like that?

 

You've got to be able to look at your left and right O2 voltages (or STFT) in live mode, see which bank it is when this occurs, then start exploring each and every vacuum line and connection... I tell you even though a hose looks good, that norma connector on the end may not be... I found a few on my engine that needed the norma connector o-ring changed - dry rot, and the oil dip stick o-ring was missing, positive crankcase ventilation hose under the throttle body, lots of places to get accumulated air leaks.

 

....after some delays...I had to repair a fender which was hit pretty badly in the parking lot (without a perpetrator's note behind my car's wiper)...a continuing diagnostic of a "restricted performance" culprit. I borrowed a more sophisticated OBDII reader and recorded some live data :
O2B1S2 - 0.975V
O2B2S2 - 0.05 V
O2S11 - 2.051 mA
O2S21 - -0.586mA
it seems to me strange that O2S21 draws 0.586 mA
Some of you suggested to look for air leaks, but I cannot find a vacuum lines diagram with some "testing" points, because it looks like without taking a half of the car apart, I could not reach every little hose which is hiding in the engine bay.
Thanks for any possible advice....

 

a little confusing, isn't it.

What I would look at only (for now) is the 02B1S1 (bank 1 sensor 1) and 02B2S1 (bank 2 sensor 1). The downstream (S2) should always be almost flatlined if the cat is working correctly. when it jumps around alot in voltage, thats a sign the cat is damaged and not doing its job correctly.

Let me take a quick read on my car and tell you what my numbers are doing on those sensors, see if it helps. Give me an hour or so.