I'm having a severe problem with the car running far too rich and eventually triggering the CEL and sometimes Restricted Performance.

A while ago I did a full court press and fixed all the vacuum leaks. The intake had some weird crack and leaked from the left front mounting flange. I found a good replacement and swapped the intake. I fixed the place on the right side of the TB elbow and cured the leak there. I properly connected the hose to the left side of the TB elbow. I can find no other vacuum leaks. Unfortunately, that made the rich condition even worse.

I swapped injectors for a rebuilt set - it worked for a while and then the rich conditions returned. I put the old injectors back. No change.

I swapped MAF sensors three times. The original, a new genuine Denso unit, and a new Amazon Brand X. All had similar results, some worse than others, but all run rich.

On average, at idle the LTFT runs anywhere from -10 to -18, depending on MAF. Off idle and at low RPMs, like around town, stoplight to stoplight, LTFT will spike to -13 to -20. At freeway speeds LTFT is pretty consistent at around -8 to -9. A few miles of driving around will eventually result in a CEL.

I have tried everything I can think of except the throttle body itself. I know it's not the original TB. I know I had weird-o TPS errors last year, but they went away after a cleaning of the connectors. Perhaps just a coincidence. It seems that it may have been running rich for a while, but I had offsetting vacuum leaks that masked it. Dunno.

Any advice from the gurus? Can a flaky TB cause these conditions? I consistently get the codes for both banks at the same time, and the LTFT numbers are pretty close to one another across both banks.

Oh, BTW, I just changed cats to what I believe to be a good set of used ones. My downstream O2 voltages look a lot better and a lot more consistent but I can't drive around enough to see if the monitors will set because of this infernal rich condition.

TIA

 

The P0172/5 DTC listing does offer the TPS as a possible cause amongst the sensors:

 

Thanks for the confirmation. I ran across something similar the other night, but for a different chassis type. I've checked the air filter and it's OK. I've checked the fuel pressure and it's OK. The IAT and MAF are integrated, right?

I checked the coolant temp with my cheapie scan tool last night and it seems to be within spec. After sitting a bit it was at 190F, and after driving a bit it showed 200F. So, it's not reading too low.

I guess that leaves the TPS. Should I just replace it, or source an entire throttle body?

 

Yes. The IATS is the part you see looking into the MAFS body.

Sounds like you've covered nearly all bases. If you can lay hands on a TB without sacrificing a kidney then it's worth a try.
Odd that there are no TPS-specific codes being thrown.

I'm not sure if anyone ever found a replacement TPS for the AJ27?

I found this thread relating a similar issue:
https://www.jaguarforums.com/forum/x...28/#post362568

 

Thanks for that link! It helps a lot.

I note that Amazon/FleaBay is full of generic TPS sensors for around $40. Anybody have any experience with them?

I'm kicking myself for not buying a TB when I had the chance a couple months ago. A pick-your-part had a pretty complete XK8 (secondary guides not done) and I got the RH valve cover and the intake off of it, as well as a floormat and the boot cover. I set the TB aside for a possible return trip. I checked yesterday and the entire car went to the crusher. Arrrrgh.

PS - I disconnected the MAF last night for a test run. I got restricted performance, as expected, but the LTFT numbers were essentially unchanged. Does that mean anything?

 

If you are considering replacing the TPS, it might be worth trying to adjust it first. Mark its current position before you start so you can undo any changes you make if it doesn't help. Your OBD scanner should be able to show you the TPS reading (there is a lot more information also available if your scanner can show it).

There is also a service bulletin for the connectors on the TPS - as you don't have a code for the TPS maybe it doesn't apply, but worth checking;-

https://www.jaguarforums.com/forum/a...21-flagged.pdf

You should check and at least clean the connections as a starting point. ETA; re reading your post I see you have already cleaned those connectors, definitely worth revisiting.

For the MAF, given that you have changed it twice and unplugged it and nothing really changed, it's not the MAF.

 

My cheapie scanner does show TPS %. I forgot to mention that, at idle, it fluctuates. It can drop to 1.6%, then jump to 2.7%, then back to 2.4%. That doesn't seem right.

 

Is the idle speed also fluctuating or holding steady as those readings change? The computer has full authority over the throttle position and one of its functions is idle speed control.

You can also take the air intake off and look into the throttle body and see if the butterfly is moving like the TPS reading is moving.

 

Interestingly enough, the idle will be stable for a while, at around 650RPM, and then it will drop for a couple of seconds to maybe 500 or a bit less, like the engine will die or something, and then the RPMs will go back to normal. During the process the TPS % usually remains stable. It may dip shortly before, or shortly after, but it doesn't dip at the exact time the RPMs dip.

I will pull of the air intake and watch the throttle butterfly and see what it does when the throttle cable is pulled and report back.

 

The TPS values you see on your scanner on the live data will have a delay compared to reality - you want to see the values are moving roughly in line with the butterfly movements.

Have you checked the fuel pressure at the rail as your idle drops suddenly? It's not something stupid like the aircon compressor kicking in?

 

This idle drop is a fairly recent occurrence. I checked the pressure but not since this started. I happens with or without the A/C being on. A replacement TPS is on order. I guess we'll see how that works, maybe this weekend.

I see a company who rebuilds these TBs, although it's almost $400 for the service. That's a last resort.

 

Have a look around the forum for TPS replacement before you pull the old one off the TB.

There is some dialogue about measuring voltages/resistances, and marking the physical position of the original first.

I wonder if the 'flight recorder' data might contain any clues, if you can get to it? - although if you're not showing a P1582 it may not be relevant

 

Back in October of 2024 I measured the resistance and posted it. However, apparently I only posted for sensor 1 and not sensor 2. I'll back probe 2 and measure it before I start.

I learned a bit more. I theorized that sensor 2 was internal, but both sensors are contained within the external sensor. This seems very similar to the way my BMWs work. As the one increases resistance the other decreases. A sanity check, as it were.

More as I progress...

 

A little OT: there is indeed a sanity check (2 sensors in one), but the implementation is the headache. See this old thread (long) from steveinfrance:

https://www.jaguarforums.com/forum/x...60/#post550812

 

The training guide explains how the TPS works (note that this is for the AJ27 engine):-



The training guide also says the TPS uses adaptive learning, which suggests that the exact positioning of the replacement isn't that vital.

 

I got the new TPS yesterday. I installed it today. I marked the existing butterfly spot with the ignition on, then I disconnected everything and replaced the sensor. The first time was something of a false start. I kept getting TPS fault codes and the % was reading 0.0 on the scan tool. I disconnected the battery, did a hard reset, removed and reinstalled the sensor and things worked much better. After warmup I fiddled and go the idle to very close to 650RPM. I then went for some test drives.

It did make a large difference. When the engine was relatively cold the LTFT numbers were -2 to -3 at idle, off idle to 35mph they were 0.0, and at 50+ they were around -7 to -8. So, I took it on the freeway for about 30 or 40 minutes to get things good and hot. And it started similar behavior as before, but not as bad. On flat terrain, between 60mpg and 70mph the LTFT would be between -2 and -4.5. But on long uphill grades it would jump to -20 with the Brand X MAF and it settled to -18 and -17.2 with the Jag MAF. Long downhill grades, same speeds, would show -14.5 to -15.5. Idle would drop to -7ish, and then to -3ish. After the first run (with the Brand X MAF) i got pending rich codes, both banks. BTW, the numbers with the Brand X uphill were -20, and were jumping between -10 and -14 at idle. After a bit I swapped to the Jag MAF and did another 40-odd minute freeway run. The idle was better, as described earlier. The accelerate and uphill LTFT improved to the -18 numbers above, and when I got back home, after sitting a few seconds the CEL came on. Rich, both banks.

So... the TPS definitely helped, as the idle was more stable, the TPS % at idle was stable (it did fluctuate a tiny bit), and the LTFT was better all around, but not enough to prevent codes.

I've just about had it. Any other suggestions? Could the TB be defective internally? Like the butterfly not actually opening as far as the computer thinks it is? I am running out of ideas as to what else could cause a consistently rich condition on both banks.

 

The problem I have with that theory is that the MAF reads the actual air flow and the computer uses that as the basis for fuelling. If the MAF was disconnected or over reading and the throttle was not open as much as the computer had commanded, then you would get a rich condition - so you are really back at wondering if the MAF is faulty. btw if the MAF reading was too different from what the computer expects based on the throttle opening, I think you would get a code.

Other thoughts - blocked cat or faulty O2 sensors. Is it worth taking the car to a smog station and see what the actual tailpipe readings are?

I've no idea what effect a faulty Air Assist valve on the throttle body would have - maybe worth looking into?

​​​​​​​See what everyone else thinks.

 

So it's possible that this issue was already baked in when you got the car as someone had already changed the TB?

What is the part # on the original TPS? 198500-3250?

 

That is the P/N on the original TPS and the replacement. The original is stamped Denso, the replacement is not.

Yes, it's entirely possible that the problem was baked in, and was somewhat masked by the several small vacuum leaks. It also sat for 8 years in the PO's garage. Some joker had put an extremely crappy rebuilt trans into it, and it failed, hence the 8 year hiatus. The replacement was probably well used. The TB has markings showing it came from a salvage yard. Given the quality of the trans replacement, maybe the TB was also junk to begin with. Who knows? The quality of the work done by the so-called European car specialist was generally sh*tty.

To recap, the car always took a fair amount of driving in order to get the cat monitors set. It got progressively worse until it wouldn't set at all. During this time I had some lean conditions that I partially fixed. Then I began to see an occasional rich error but then the trans failed. After trans replacement, and still not getting the cat monitor to set, I fixed all remaining vacuum leaks that I could find, which really made the rich condition worse.

At this point I've tried a new set of injectors - no change. I tried a new MAF - no change. I replaced the cats with a presumably good pair, the downstream numbers look better, but no change. I tried a 3rd MAF - no change. I have replaced the TPS and the rich problem got better, but it still runs rich at acceleration, high loads and relatively low throttle opening (uphill on a long steady grade), and downhill grades (low load, low throttle opening). It's better at idle and at 65mph on level freeway.

The one thing I noticed is that when relatively cold, the numbers look pretty good. Once the engine gets heat soaked the problems get worse. I swapped the MAF last night when the engine was hot, and that made little difference. The Jag MAF seems to be better than the Brand X Amazon one, which isn't a big surprise.

So... maybe the TB is experiencing some kind of thermal failure?

 

Throwing in these inputs:

• Can you put more texture on the analysis of the fuel pressure? I am more familiar with the 4.2l engine, but I assume the fuel pressure to the injectors has to be a constant. As the injectors are in the intake, it means the absolute pressure has to be adjusted by the regulator, based on a vacuum line to the intake. When the pressure is read with a gauge, that pressure should vary as per throttle position/intake vacuum. I was hoping you could confirm if this was in any way observed during your test. The idea here is to check the regulator. If the injectors are seeing too much pressure (regulator faulty or vacuum line clogged/split/plugged), it seems the engine would run rich.
• I believe there are several separate codes for this (and you have not listed them), but what about the evap system? If somehow the valve is stuck open, fumes from the tank would keep feeding the engine with extra fuel and make it run rich. There is a purge valve behind the fender liner. There are several part# over the years, some of then NLA. I'd say open it up and double check if there is any sign of a hack job around that. As a temporary measure, maybe figure out a way to make sure none of the evap fumes can go to the engine and test again.
• It would not hurt to "recalibrate" the throttle cable sensor: key on, engine off, slowly exercise the entire travel of the gas pedal, including kick down switch if you have one. The slowly release. Do this several times because why not. This should "teach" the ECU about the entire set of values for the reading of the gas pedal sensor.
• Are you running a normal, dry, paper air filter for now?

Best of luck, keep us posted.

PS: All you have is an O2 sensor, there is no fuel sensor to read off of. When the engine runs lean, you can read excess O2 from the sensor. When it runs stoichiometric (perfect air/fuel ratio), there is no oxygen left. How does the engine even "know" you are running rich? Does it all point back to the AFM?