Several months ago my 2004 XJ8 with 155k miles began giving me a restricted performance message on the dash with an occasional hard start or unstable idle right after startup. Temperature seemed to have no affect, it was very erratic and I couldn't correlate any pattern of behavior. Ultimately I plugged in my diagnostic scantool software and found two history codes indicating a lean condition on both banks. Short term and long term fuel trims observed were running 12-18% on both banks at idle. Off idle fuel trims mellowed out and the motor seemed enthusiastic as usual when accelerating. Idle fuel pressure was healthy. Suspecting a hefty vacuum leak I went searching around the engine bay but found nothing untoward. I picked up some rubber vacuum caps from a local parts store and with the scantool connected I started eliminating vacuum accessories. I opened the breathers and capped the ports, eliminating the crankcase altogether... no change. I opened the fuel vapor line and capped the port... no change. Booster line... no change. Finally I came to the fuel rail pressure sensor, pulled the little vacuum line and plugged the port and suddenly the fuel trims returned to normal. I've been driving the car for 4 months without a single hiccup with the vacuum line eliminated between the TB and the fuel rail pressure sensor. Passed emissions in New Jersey without a problem and I haven't noticed any major difference in performance. What is wrong with this picture? I've picked up a used pressure sensor for 40 bucks and I plan to swap it out this week but there isn't anything obviously wrong with the old one. I figured it might have a hairline crack creating a vacuum leak but it tests solid with a vacuum hand pump. So does the hose. What should I look at next assuming the sensor isn't the problem?

 

Might want to check everything going from the MAF sensor to the throttle body.

 

Last night I replaced the vacuum hose on the fuel rail pressure sensor with the scantool connected. Long term and short term fuel trims immediately spiked and remained high, just as before. Long term reached 12% and was still being pushed upwards by the short term value. Essentially after 4 months of driving without a fault, the fault was still present with the vacuum line connected

Air intake duct downstream of the MAF seems fine, no noticeable leaks, no cracked plastic or rubber. At one point I tried replacing the MAF to eliminate this condition.

After verifying the condition still existed, I shut the car down and replaced the fuel filter. The fuel filter seemed restricted (got a mouth full of gasoline testing that theory) compared to the new one and after replacing it the fuel trims smoothed out. Replacing the pressure sensor dropped the trims a little more. Everything seems stable now and I took it for an hour drive last night with no restricted performance message.

I'm hoping that the restricted fuel filter was just reducing flow without noticeably affecting the fuel pressure at idle. That would explain why the fuel pressure looks good but the engine is registering a lean condition. In any event I'll report back if the condition returns. Otherwise, fingers crossed, this one was an easy fix.

 

Interesting. Thanks for the report. Hopefully that takes care of it.

 

So interesting and curious that idle trims were high. One would think that fuel flow would be lowest at idle and would have minimal effect.
I'll be doing mine as mine, at 185k miles, has +12% trims at just idle, 600rpm, at 700, or cruising it is 3.9/+4.7%. both banks are the same, no other issues.

 

If you measure (hot engine) at idle and then just rev (2000-2500rpm, not critical) what do the LTFTs do?

 

I chased EVERYTHING else because I thought it was impossible for a clogged filter to cause the issue but so far so good. Still running strong and trims are still correct after a couple hundred miles of driving. I commute about 55 miles each way with this car every day. I think that removing the vacuum line (pressure calculation compensation) from the rail pressure sensor just cancelled out the effect of the clogged filter and prompted the computer to artificially increase the pump PWM signal which overcame the filter restriction. You can try an easy experiment... just unplug the vacuum line at the throttle body and push a rubber cap over the open port. See what your trims do with the pressure sensor vacuum line eliminated (just make sure the port is firmly plugged on the engine to kill the vacuum leak you just created.) Chain auto stores sell a little package of vacuum caps which makes it an easy test. You can also use those caps to eliminate vacuum accessories one at a time until you find something that improves FT behavior. That will help narrow down the location of any vacuum leaks.

Changing the filter was easy after releasing fuel pressure from the schrader valve on the rail. Keep a fire extinguisher handy in case things go sideways on you. I highly recommend that you loosen the fuel line fittings before removing the filter bracket as it provides support to the lines while you torque the flare nuts. Reverse procedure to tighten as well, cinch up the bracket before you put final torque on the line fittings. You'll also want to replace the small o-rings on the flared ends of the fuel pipes at each end of the filter. If the o-rings are missing try looking inside the old filter to see if they stuck in there. Either way each pipe should have an o-ring on it. Viton and HNBR o-rings hold up pretty well to gasoline. HNBR (usually green or blue o-rings) are readily available in parts stores for AC refrigerant lines. I often use HNBR on fuel lines because you can pick them up in assorted packs in the AC recharge aisle at most stores.

I bought a smoke machine on eBay and I plan to smoke the exhaust and intake system soon just to see if I find anything else.

 

BTW I think the purpose of the vacuum line on the pressure sensor is to match the differential pressure calculation to what the fuel injector sees at the intake port. One end of the injector is exposed to fuel pressure while the open end (when it opens to spray into the engine) is at manifold pressure (vacuum.) Fuel flow will depend on the difference between those two pressures and the computer adjusts fuel pump power based on that value. The rail sensor has a vacuum line because the engineers wanted to make the pressure reading as close to what the fuel injector experiences as possible.

When you eliminate the vacuum connection to the pressure sensor, one side of the sensor still sees fuel pressure but the other side sees atmospheric pressure... by definition atm is higher than manifold vacuum. The difference in pressure is reported to the computer. Since atmospheric pressure is higher than manifold (vacuum) the difference will always be a lower calculated value and therefore as long as the vacuum line is disconnected the computer will always be running the fuel pump a little harder than it should. That's why it just happened to correct for the clogged filter. The computer will tweak the pump PWM up to show the same rail pressure but that rail pressure reported on the scantool is actually a lot higher as seen by the fuel injectors when the vacuum line is gone. To us it looks like nothing changed in the fuel pressure signal with removal of the vacuum hose because the computer adjusts the pump to yield the same pressure but the engine is getting more fuel.

After diagnosing this issue I had to think for a little to understand why eliminating the vacuum line to the pressure sensor would compensate for a clogged filter.

Confirming this behavior would be easy as you could apply maybe 5 psi of air pressure to the sensor vacuum line with the port plugged on the engine and it should cause a rich condition with FT values moving low as the software artificially increases pump power... I may not get around to trying it

 

Actually what I may try is pulling the vacuum line off again. This time with the fuel filter fixed that should produce a rich condition and I should see the FT values go low... Wish I thought to try that the other night when I did the job originally.

 

Just posting an update... The fuel filter wasn't the problem. The issue has returned with a vengeance now and the car won't run at all if the outside temperature drops below 25F. I've started a new thread because I'm chasing new leads.

 

A restricted performance should have set a DTC?

 

Yes, it did. I've continued the drama on another post because the behavior changed quite a bit and I am still not 100% sure the two problems are related.

https://www.jaguarforums.com/forum/x...eather-194673/