Ah, understood. On other cars that's generally true as well... I keep coming back to the MAF. It would explain a too-rich code (because it's lying to the ECM) and IME it's rare for the engine management to diagnose a failed MAF until it's REALLY failed. Unfortunately, my experience with Jaguars is limited to this very one.

 

Injectors are on the table, but if one injector was bad I'd expect a misfire, and even if one BANK was bad I'd just get one code. But codes for both banks really suggests the issue is in the intake, would you not agree?

As I mentioned in the first post, the MAF has been cleaned and the air filter replaced just a couple weeks ago. The entire throttle assembly replaced less than a year ago - it was spotless when I installed it, and the car has driven less than 1000 miles since.

 

Sorry to have forgotten about the CTS. With all the threads I participate in I can't keep all the details straight, and normally, going back to the first post everytime I revisit a thread isn't practical. However, I just had to go back to your first post to be reminded of the year and model of your Jag. Have you added this info to your signature line? If so, it's not showing up....

Regarding your CTS indicating a stable 190 degrees, that seems low. According to the Engine manual, the thermostat is not fully open until the coolant reaches 205F. I just ran our '03 N/A X308 up to operating temperature, and with the ambient temperature at 49F, the CTS reported engine coolant at 92C/198F at idle and 91C/196F if I revved it for a few seconds. Is it possible that your car has a thermostat with an opening temperature lower than 183F? I think the thermostat part number and opening temp for the S/C and N/A engines is the same, but some parts sellers offer a 170F option. I don't know if 6 or 8 degrees would be enough to make a significant difference in fueling at idle, but if your temp goes down significantly while driving at speed it could.

Regarding the MAP sensor readings, I don't know of a Jaguar reference that states the range, but I just checked our '03 N/A X308 and the MAP signal is 21 psi at idle and doesn't change when I give the engine a short rev. I also tested an '06 X350 XJR I'm working on (4.2L S/C so it should be close to your 4.0L S/C). Its MAP sensor reads 11.1 inHg (5.45 psi) at idle, and when I give the throttle a blip it goes up to 21 inHg (10.3 psi), falls to 6 inHg (3 psi), then resettles at 11.1 inHg (5.45 psi). This makes your 6.8 psi seem plausible. How does yours respond when you blip the throttle?

Cheers,

Don

 

I'd suspect a faulty sensor somewhere. Proper diagnosing of all of them with a power supply and voltmeter sound be adequate to determine something faulty.

 

I'm not sure I'm following - how do you diagnose a faulty IAT or MAF with a power supply? Wouldn't the proper way to do that be with a scope?

 

I took the Jag to work today - nothing has changed since resetting the CEL last week. It hasn't moved since then.

My drive is quite short - about 8 miles incorporating 6 miles of trafficky surface streets and 2 miles of trafficky freeway. Outside temp is about 60 degrees.

I grabbed my Bluetooth dongle on the way out, not realizing in advance it doesn't work very well with the Jag... I'm not sure why, but Torque only reports a handful of sensors. I guess next time I'll bring something better.

I monitored the :

IAT, which stayed pegged at 16C the entire drive. It's not wrong per se - it's just weird it didn't move.

CTS warmed up rapidly to 90C and mostly stayed there as long as I was moving, once I stopped it would creep up to 92-94C. Since it's able to maintain 90C even at 50-60mph on the freeway I'm inclined to believe it's working properly.

(It's an OE thermostat, replaced four years/6k ago. If nothing else, 190C is a totally workable engine temperature... it's not "cold.*)

Unfortunately Torque was unaware of the MAP, so that was off the table.

I could only see STFT on B1 and could not see LTFT at all - no idea what that's about. STFT/B1 was all over the map - anywhere from +3% to -24%. That's clearly not right. STFT was worst right off idle, whether slowly accelerating in traffic or creeping at <10mph. At higher speeds, it would settle into an expected -5 to +5 range.

I'm still very hung up on the MAF, and today didn't help. At idle, Torque reported 8-9g/s which seems high. More interestingly, while coasting to a stop from, say, 30mph, I would see that number plummet to 6-7g/s. I don't understand how an engine at 600rpm can be ingesting more air than an engine at 1200rpm no matter what the throttle is doing. There was unfortunately too much traffic to really get a sense of air flow at higher speeds.

I'm gonna get a different/better OBDII reader and see what it says, but at this moment I really think I have a problem with my MAF.

 

IAT is a thermistor whose resistance changes with temperature. So you can test it by providing a Voltage supply and reading the output with the voltmeter and seeing how the output changes when you pinch it or grab it. You can look at the specs to see if the output seems right. For the MAF, you can supply Voltage and blow through it to see if it changes. (It'd be difficult to determine if this is up to spec).

what kind of scope are you referring to?

I suggested this method because it can be done with household items: voltmeter, 12V (or 8AA batteries in series). There are different ways to go about it but I figured this would be easier.

As for your MAF, they run for a few hundred bucks so I'd test it before I replace it.

 

Yes, this is true, but I can see readings change with an OBDII tool. Grabbing the IAT and blowing through the MAF will tell me "they change," but what I am trying to determine is if they are measuring correctly. Unless I know the temperature of my hand or the strength of my breath, I'm not going to be able to learn that.


An oscilloscope, practically speaking a software based one. By watching the output of sensors as they work, I can observe patterns or lack thereof, indicating flat spots or loss of connection. This is typically the way you diagnose automotive sensors. I was hoping to get to a set of baseline measurements before resorting to this, but it seems nobody has them.

 

I plugged in a real scan tool today and drove around for a bit.


First reaction was a facepalm for realizing my concern about the MAP was just ignorant. The MAP measures ABSOLUTE pressure, so the 6.8psi measured during the freeze frame is 6.8psi above 0psi, which is still some amount of vacuum vs. atmosphere. Derp. I have spent the last month dealing with turbos and my head was into "pressure above atmosphere" not "pressure above 0." So, 6.8psi seems about right, I can probably dismiss the MAP.


My good scan tool was able to display STFT and LTFT on both banks, so that was positive. During my drive, both banks remained with 1% of each other under all circumstances, so I think it's fair to say the in-cylinder stuff is happening in a predictable way... both banks are getting the same result. STFT stayed between -5% and 5%, which is reasonable. LTFT varied between -10% and +5%, which is not unreasonable but does suggest an ongoing rich running issue. I'd like to see that LTFT closer to 0. At no point did it get back up to that -20% number.


CTS remained rock-solid, 190-195 degrees at all times. It's around 75 degrees here, the fan cycled once during a long stop but only that once. I cannot find for sure the specs for the thermostat, but 190 degrees is a very reasonable temp for a boosted aluminum motor, so I'm not too worried about it even if it might be a bit low. It's in the ballpark, not "cold" in an absolute sense.


MAP seemed right. I was mostly in city traffic so generating any real boost wasn't possible, but it did fluctuate between 0 and ~15psi during my driving, with mid-6psi shown at idle. Per above, this seems reasonable.


IAT seems spot on, reflecting the increase in temperature from the other day to today. It seems accurate, I think I can dismiss it.


MAF was different today than the other day, spending most of the time at idle around 6g/s. That's lower than the other day, but it is warmer. I lack the science to say whether 75 degree ambient yields a 1-2g/s loss in air density vs 65 degrees. It seems a bit implausible, so I still suspect the MAF.


The car did not misbehave during Wednesday's testing or today, but it doesn't represent a lot of miles. I did refuel on Wednesday - same station, 15 gallons of Chevron 91 octane.


I think I'm going to try and locate a used MAF to test with. It's not want I want to do, but I need some other readings to compare against and I'd prefer to do that for $50 instead of $250. But, I may cave and just buy a new MAF. It's 15 years old, it certainly wouldn't hurt.

 

The ECU has stored default settings for MAF. I remember running my xjr at idle and when unplugging it while at idle, caused it to shuddered, but did not stall it. I'm not sure if a MAF can cause a car to stall, but from my experience it did not. It could cause some sort of lean/rich issues for sure, which may cause stalling.

 

This statement has been bugging me but I've only now had a chance to confirm my suspicions. Some of the V12s may have had a regulator in each rail, but I can't recall for certain. According to the AJV6/AJ26/AJ27 Engine Managment System manual and parts diagrams, in both the N/A and S/C engines there is just one FPR, in the return pipe at the rear of the A bank fuel rail.

See pdf page 68 for a description and nominal fuel pressures:

Jaguar Service Training Course 881 V6 / V8 Engine Management

According to the Workshop Manual, rich running can be caused by high fuel pressure due to a stuck FPR. Here's the pinpoint test for FPR behavior:



The FPR and return pipe for supercharged engines look like this (photo courtesy of SNG Barratt):


Given that the fuel pressure should increase when the vacuum hose is disconnected from the FPR, it follows that a disconnected or leaking vacuum hose could also lead to high fuel pressure and rich running, so it would be worth checking that out.

Cheers,

Don

 

+1

definitely a suspect

BTW you look to be doing the right kinds of things and making sense so you likely will figure it

 

Appreciated - I will check that out today. If there's a singular FPR, it's definitely suspect, as its vacuum line!

 

There's definitely only one fuel pressure regulator, and it's right where Don said. The SNG Barratt image wouldn't load on my computer, so I don't know what is in the picture, but if you're going to test the FPR, the Shraeder valve to hook up the gauge is at the front of the LH head. You'll need to loosen the wiring harness bracket and remove the part-load breather from the cam cover to get at it. The vacuum line to the FPR is the one you see connected to the top of the SC on the RH side.

 

Been driving the Jag a bit more lately, trying to nail down the problem a bit. I have been driving with an Ultragauge plugged in, displaying STFT & LTFT on both banks, IAT, MAP, MAF, and ECT all at once.

About a week ago it was low on fuel so I refilled, and since that time it's largely behaved itself. Once or twice I noticed an abbreviated example of the issue - just a momentary loss of power that went away before I even had time to look at the scan tool, so that was useless.

Post-refueling, and as often as I checked, all the readings seemed very normal. Fuel trims staying between +5 and -5%, IAT accurately reflecting ambient temps, MAF readings that seemed reasonable (WTF do I know?) at cruise and MAF readings that were appropriate at idle - 6-7g/s. ECT and MAP stable and appropriate. Very discouraging?

Today I had cause to take it on a bit of a high-speed trip - 70-80mph on a 40-60 mile roundtrip. First ten miles were good, but then things started to feel a little off - not tangibly off, but not right. Around mile 40 or so things got weird.

On the way out, with barely any throttle and a 70mph cruise I was seeing MAF readings around 36g/s, which means nothing but is a data point. On the way back, before the real issues set in the same scenario yielded MAF readings between 45-50g/s, which was inexplicable. During this time, I was seeing familiar loss of power but was having a hard time managing the throttle while watching the gauge. But, at one particular point it sagged quite badly, I got my eyes on the gauge and saw 36g/s - I goosed the the throttle (which historically has cleared up the problem) and the MAF reading shot immediately to 122g/s and then everything went back to normal for a bit. I don't know what 122g/s means, but I do question the massive *instant* jump that immediately "dissipated" while my foot did nothing further. That's weird.

As I got back into town and back onto surface streets, I noticed that the MAF reading was creeping ever higher at idle - from a normal 6-7g/s to as high as 9g/s. During these times - with MAF nearing 9g/s - I could watch LTFT creep up to ~ -20%. While doing *nothing* other than sitting there, MAF would sometimes drop back down to a respectable 6-7g/s and I'd see a corresponding righting of LTFT back to ~0%. Bear in mind, this fluctuation in MAF reading (6-9 g/s) was NOT accompanied by any throttle changes or any engine speed changes. That is, sitting there, doing nothing, at dead-even 600rpm the MAF is randomly moving about between 6-9 g/s.

At this point I can only conclude the MAF is not doing well. 20% variance at highway cruise (36g/s to 45g/s at 70mph cruise) and 50% variance at idle (6g/s to 9g/s) can only be explained by a bad reading, or a vacuum leak or some type. Since there is no idle problem, the leak would be pre-throttle-body and there just isn't that much intake there. Plus, one would expect a vacuum leak to *suck* air from outside and cause a low MAF reading, not lose air to the outside and cause a high MAF reading. The fact that it randomly comes and goes makes it seem like an electronics problem... a hole somewhere would leak all the time.

So, I am going to double-check the intake for leaks and I think then get a new MAF. Hopefully that's it.


Edit:


Just as a visual aid, I snapped this photo while driving. This is a 70mph cruise so just *barely* on the throttle. Again, it's really hard to know, but 26g/s does seem off to me. Assuming .8 v/e, that represents around 20hp. I realize it doesn't take power to cruise at speed, but that seems like very little airflow. And why the variance between 26 and 45 g/s (with 36 seeming "normal")?

 

Not that I expected to, but I found no leaks. I was pretty thorough with inspection and replacement when I replaced the throttle body last year... with that big part out of the way it would have been foolish not to! So, I think that's out.

While trying to find the best price on an OE MAF, I ran into this site:

P0172 & P0175 Running Rich - Jaguar XJ8 | Editing Luke

Seems plausible. Rather than bend pins on the MAF I actually dismantled the connector (remove TPA, release terminals) and bent the terminals back out individually. There was definitely a different feeling reconnecting it, so maybe I did some good.

I'll put some more miles on it and see what happens. I'd sure be happy if that was it!

 

It may be nothing more than a poor connection. Did you use electrical cleaner on the plug, and have you cleaned the mafs itself with mafs cleaner? I recall reading that the mafs is not expensive, so you might want to replace it as you mentioned earlier.

 

LTFT that's -ve means too much fuel so correcting for it. You're unlikely to have an air/vacuum leak.

So... now to try to figure what may be happening. You get to imagine and see if you can find something that at least seems plausible (as in fits the known facts in as far as the facts are known).

It's a computer going round a control loop a lot.

Imagine it sees the MAF making it seem there's more air. Computer (PCM) will immediately add fuel to keep the AFR (air/fuel ration) sane.

Then the fuel burns (we hope) and the O2 sensors tell the PCM what really happened some time later (i.e. after a time lag).

Looks like air (i.e. O2) must have been absent so it makes a note to add less fuel (thus -ve trim).

Could be MAF isn't right but maybe there are other scenarios (e.g. part-blocked cat or fuel filter or failing fuel pump can be swines to figure out).

BTW not relevant to this car right now but the "we hope" is important e.g. for misfires - where there's air because it failed to burn. This fools the PCM as it has no "excess fuel" sensor. It adds (yes!) fuel. Misfires are a pain. (As I put, not in this car right now.)

 

Yeah, I mentioned in the OP the MAF was cleaned. I wasn't watching it with a scan tool prior so I don't know if it changed anything, but it's immaterial I guess. I was cleaned within the last several hundred miles.

The MAF is really pricey - aftermarket ones are in the $100-150 range but OE sensors over double that. I have *never* had good luck with 3rd party MAFs on any car, and I'm hesitant to try another. However, in another thread someone mentioned they'd had good luck, so maybe it's worth a go.

Yep, 100% agreed. My working theory is that the MAF is over-indicating the airflow into the engine so X% extra fuel is added, then the O2 sensor finds too little O2 in the exhaust (ie a rich mixture) and trims it out negatively. There could be other components calling for more fuel (like IAT or ECT) but since both readings appear accurate I think they can be ruled out... pointing me back to the MAF.

What would your logic here be? I understand the false lean - eg poor fuel atomization from clogged injectors or low pressure resulting in poor combustion and excess air so you get an "incorrect" positive fuel trim - but I can't wrap my head around a false rich. I can only get to a sensor problem (MAF/ECT/IAT/etc.) from that direction.

YES! Was there not too long ago with my Suburban. But, fortunately for "us" I have a Tech 2 so I could watch live misfire data. Unfortunately, not an option on the Jag. On the Suburban, the issue was a coolant leak that was sucking air into the system and (I guess!) creating an air pocket around the ECT so it showed the engine was colder than it actually was. Once I saw the ECT data it was obvious the sensor was the culprit... but a new one didn't fix the issue. About three weeks later the heater control valve failed spectacularly - fortunately while changing the oil - and all the pieces clicked into place. New HCV, no leak, no misfire. On the Jag... maybe not so easy.

*However*, after reassembling the MAF connector I ended up nearly repeating my Thursday drive late Friday evening, the car behaved perfectly. It was late at night - mid 50s - which historically has seemingly made the problem worse - and there were no issues. OBDII data from the MAF seemed a lot more stable - and *interestingly* ~70mph cruise now shows a steady 25-30g/s. No immediate explanation but STFT is showing generally more positively and LTFT generally more negatively, but it could be temperature-related or maybe adaption related. It'll definitely take some more road time to form a real opinion, but *initially* I'm hopeful. This was the most trouble-free drive I'd have in two months.

 

You may find OE tool(s) can watch misfires. Not that they look to be your problem.

Blocked cats disrupt fuelling. Failing fuel pumps are a pain to diagnose sometimes. They don't fail predictably so a control loop can have problems.

If I had to guess, MAF looks worth a punt first, being easy.