Just spitballing here:- after dibbit mentioned the air assist, I investigated what it actually does. According to the description, extra air is allowed into the injectors until the engine reaches around 70° Centigrade, beyond which it is shut off. I doubt it's the problem, but that may explain why the trims are better with a cold motor.

A bit left-field - does the ECM part # match what your VCATS sticker shows?

 

The regulator is new, I replaced it right after I bought the car. Earlier I had the same suspicion that fuel pressure might be too high and I put a gauge on it. The pressure was slightly less than 40psi.

The EVAP system works and the monitors set. I've replaced the canister control valve (the one at the back) and the purge valve (the one at the front), smoke tested the system, replaced one flat o-ring on one canister and replaced the other which had a large crack. I now get no more gross leak, small leak, or valve malfunction codes.

I will try that throttle recalibration.

The paper air filter is about 2 years, maybe 10K miles, old. I pulled it out a couple days ago and it looks pretty clean still.

I presume it knows it's running rich only because the LTFT numbers are way negative. The upstream sensors must be sensing too much fuel. Currently, one of the upstream sensors came with the new cats, one was from an old cat, both are Denso and both appeared to be fairly new as they were in pretty good condition both at the tip and the exterior of the sensor. This was happening with the old cats, using the existing Denso and a new Bosch that I'd installed when these rich errors started and I only got rich codes on bank 1. Didn't help, problem got progressively worse.

 

I would change the air filter - an old dirty air filter (even if it looks clean) can make the MAF misread - it's cheap enough anyway to not really matter if it makes no difference.

Yes, that is what is happening (too little oxygen in the exhaust). My suggestion to get the exhaust gas analysed would confirm if your upstream sensors are giving sensible values - do the lambda values match on your scanner versus the workshop's tailpipe analyser.

You are rapidly running out of things to change - you probably need to be looking out for a replacement throttle body to try next unfortunately.

One other random thought - have you tried loosening the MAF to deliberately introduce an air leak? Can you get the trims moving towards zero?

 

the upstream AF sensors never skew, if the pump cell fails they throw a DTC. they are stupid reliable

TPS is almost a complete non-factor for fueling calculations unless it has to run alpha-n when the MAF is unplugged. check base fuel pressure and if the fpr vacuum line is not hooked up

 

I can try the loose MAF thing. I sourced another TB from a recently deceased 2001 Van den Plas. Unfortunately, I don't know the mileage. I pulled it day before yesterday, and they didn't note the mileage anywhere. Interesting, it has had its TPS replaced, as it's not original.

 

I had already checked the FP and it was correct. I don't have a mechanism to test it while driving. Any suggestions? The vacuum hose is connected.

 

check your gauge

see if removing intake vacuum increases pressure

 

I'm leaving for vacation early in the AM tomorrow. So... I may not get anything accomplished until I return. At that point, first order of biz will be to replace the TB with the used one and see if the numbers change.

 

That does seem like the next logical step. Keep us updated when you are back from holiday.

 

Before I went out of town I sourced a TB from a 2001 Van Den Plas that had recently arrived at a pick-a-part yard not too far from here. I figured, for the princely sum of $49, plus an extra $15 for the money-back warranty, it was worth a shot. I installed it yesterday and did a hard reset. I left the original Jag MAF installed.

The car fired up immediately, did the high idle thing. and settled into an idle just slightly above the 650 RPM of the old TB. I connected the scanner and the mixtures were leaner than the old TB. I let it idle as I shuffled cars in order to take it for a test drive. While sitting I heard the revs go up and down and up and down and then stabilized. When I got back to the car it had gone into failsafe mode, with a code of P0121 Throttle/Pedal position sensor A Circuit Range Performance. I cleared codes and things went back to normal.

I did a couple of brief test drives. The mixtures were noticeably better, but as the car got hotter they were still too rich. As a last resort I unhooked the top of the airbox so that there was a small gap between the lid and the air filter. I then checked mixtures and they were substantially better. I then figured that if a little is good, more is better. I removed the air filter entirely, buttoned up the box and started the car. It did NOT like that at all. It started, but would barely idle and wanted to die. It wouldn't rev properly either. Strange. I don't recall ever having a car that wouldn't run without an air filter. So, I put the filter back in, left the top of the box loose, and went for a drive.

The LTFT numbers dropped like a rock. They seemed to get better the longer I drove. At idle they started around -3 on both banks, eventually dropping down to around -0.8. Driving on the freeway, between 60 and 70, resulted in trims between 0.0 and -0.8. Stop and go, from 0 to 40-ish MPH showed similar, sometimes actually going slightly positive, like 0.6. Downhill runs at speed maybe -3 at the highest (or maybe that should be lowest ) but usually lower. I drove freeways speeds for about 40 minutes, in two parts. By the time I got back home all of the monitors EXCEPT the freaking CAT monitor had set.

So, I guess we're on to something. The air filter doesn't look bad, but I ordered another cheapie off of Amazon for testing purposes. If it were restricting things, and apparently it was, why didn't the MAF compensate for reduced airflow? I still have suspicions about that MAF but I didn't want to change too many variables at once.

I'm going to drive a bit more today and see if anything changes.

 

A new clean air filter is essential really if you want the MAF to work properly. The MAF doesn't measure the air flow going to the engine, it takes a tiny sample of air in that flow and the computer works out, based on that tiny sample, what the total airflow is. An old dirty, or even non-OEM filter, is going to disrupt that air flow which might cause the computer to come up with the wrong total - perhaps that is what is happening in your case.

By the way, you need to add an air leak after the MAF, not before it, if you want to try and get your rich trims down.

 

One would think so. And I was looking around for a good way to create a downstream "leak" on a hot engine, and not finding one, when I took a flyer on opening up the airbox. And all of the fuel trims suddenly became pretty normal. They are very close to 0.0 in most situations. Which is a far cry from what there were prior. Which seems odd, but there it is.

The current filter is a Mann, IIRC. It's slightly dirty, but I've had far worse. I'll replace it with a temp one and also dig around and see if the "snorkel" thing that goes through the fender isn't somehow blocked. No idea how that could happen, but it's one thing I haven't checked.

BTW I drove around for about 45 minutes yesterday, all freeway, and the trims were all quite low. The CAT monitor still didn't set. Cataclean is on order.

 

Maybe 0 if running OL (open loop)?

Dipstick slightly out might be an easy air leak.

 

I installed a new air filter that I got from Amazon. The old one was either Mann or Mahle, don't recall. It's going on 3 years old and has perhaps 10K miles on it.

The car started immediately and, after a small warm up settled to a very steady LTFT of +6.3/+5.5. First time ever seeing a positive figure. Revving would cause them to drop to near 0.

I took it on a test drive of about 22 miles, with a stop in the middle for ice cream. Hard acceleration resulted in LTFT of 0.0 both banks. Steady 60-70 mph on level ground gave +0.8 bank 1 and -0.8 bank 2. Uphill grades would usually show like +1.2/+2.3. Long downhill grades would be almost a mirror image of the idle numbers, that is -6/-5. At the end of the second leg the idle numbers had dropped to +6.3/+4.7. Stoplight to stoplight, the numbers dropped from +6.3/+5.5 at idle to +0.8 to +2.1 accelerating to 40-ish.

Frankly, I don't see how these numbers could improve much. I have no idea what they were when I first bought the car.

 

Amazing the difference a $10 air filter makes and so easy to overlook it. I'm guessing it was so blocked it was either significantly reducing the air flow or significantly altering the MAF reading. Just goes to show as ever, we should always check the basics first before deciding it's the complicated stuff.

Maybe ask the admins to mark this as resolved as it's a good thread.

 

Hard accel is probably WOT (wide open throttle) and commonly doesn't report trims other than as 0.

Trims to check for air leaks are at idle, hot engine.

 

I put a quart of Cataclean into 1/4 of a tank of gas last night and went for a ~50 mile trip. The only noticeable difference was a slightly higher positive LTFT on Bank 1 at idle. I drove at varying speeds from 55 to 80. At the end, no dice. CAT monitor still shows INC.

This is infuriating. I guess I'll see about getting a temp "moving" pass from DMV, via AAA, and take a road trip to San Diego. If that doesn't fix it, well, at least it will make an almost-finished project for somebody who lives outside of the PRC (People's Republic of California).

 

Just drive it, the monitor will eventually set.

 

That's the next step. It's just that the People's Republic of Kalifornia makes it difficult. The registration expired in April, and I had no idea at the time what it was going to take to get it running. You can't register until it passes smog and it won't pass if the CAT monitor isn't set. So, I put the car on non-operational status. The upshot is I have to get a 1-day permit to drive the car. And I have to get said permit each day I want to drive it. Major PITA.

After all this hoopla we're finally back to where we started... trying to get the CAT monitor to set. Sigh.

 

It's a real Catch 22 - you have to drive the car to get the monitor to set, but you aren't allowed to drive the car until the monitor has set. How likely is it you will get caught driving it with the expired registration? Where I live, I have been pulled over by the police two or three times in the past 17 years to check documents, so I would probably just risk it.