2004 XJ8 N/A 88k Miles - High RPM Misfire
 

I'll go ahead and start a thread for the last nagging problem with my 2004 XJ8 "mystery car" that I purchased for 1,400 bucks a few months ago. Hopefully the diagnostic approach will help someone else struggling with the same frustrating issue. I've got everything sorted out, no trouble codes, no idle stumble, the car seems to run fine, engine feels great... except when I put my foot to the mat, at about 5k+ rpm under full load, she trips a cylinder #3 misfire.

Last night I replaced the plug and coil to rule out those issues quickly. No change in behavior, drives great until I dig my foot into it. No overheating issues, no idle stumble, no hesitation below 5k. Transmission seems fine, catches the shifts, I seem to have all six gears. When I pulled the spark plug out last night it looked oddly clean. I'm not really great at "reading" plugs but I'm sort of concerned that it kind of looks like it took a steam bath. Hopefully I'm not uncovering a head gasket issue, I'm hoping it just looks clean because it is running really lean in that cylinder. I think the computer is shutting off fuel when it hits a certain misfire threshold because it flashes the SES and you can feel it fuel cut as revs fall back below ~3k.

I've ordered a new set of fuel injector o-rings and one reman. GB injector. I've also sourced two brand new Jaguar OE knock sensors. When I bought the car, the knock sensors were reversed and disconnected. I repaired the wire harness and got them properly seated on the correct side of the motor but they didn't look to be in great shape.

This weekend (4th of July) I'm going to pull the intake, replace a few gaskets, install a factory valley hose (vs aftermarket), install two factory knock sensors, restore the foam sound insulation, re-seat the intake, install new injector o-rings and replace the injector on #3. I'm also going to install eight NGK IFR5N10. If the aforementioned (which needs to be done anyway) doesn't sort it out I'll try to get it out for a drive on the scantool and investigate further. With a new injector, new plug, new coil, new knock sensors... the best I can hope for is that the ECM is flagging the wrong cylinder.

Any thoughts? I'm sort of stumped with this one.

After replacing the coil pack and spark plug, I also would be in suspecting a problem with the injector on that cylinder.

Yeah fuel injector is sort of where I'm headed. I'm going to be surprised if this is something more serious because the engine feels great otherwise.

If memory serves... in the course of diagnosing other issues with this car I had the intake off and I seem to remember one of the injector o-rings looked cracked. I remember the o-ring issue being on bank 2 though.

I wish cars could talk because mine would have a lot of stories to tell... how a dealer serviced / one owner 04 with 88k original documented miles ended up being totaled in the hands of a crappy indy mechanic I have no idea. I wish the guy I bought it from would just tell me the real story, I'm not upset at all, I just want to know.

I have found over the years so many people just can't tell you the truth.

About what happened to their broken whatever.

Buddy bought over his mower for me to look at couple days ago.

Said he had adjusted the carb, to improve the way it ran.

Then next time he went to use it, said it would not start.

Also it also now backfires, and jerks out of your hand.

Did you hit something, and stop the mower?

NO

Pulled the top off the motor, the shear pin was torn in half, crank was bent.

Asked how the F--- did this happen,if you didn't hit anything?

Well i'm always running over, or hitting something, don't remember if i did.

Reading things you find, can tell you a lot more about the previous owner.

Than his story's as to what happened, or his reason for selling the vehicle.

Two, or four wheels, or anything else, that's for sale.

I bent a crankshaft on an old mower once when I was a teenager, thought it was going to clear the rock :)... you really can't mistake that bone jarring jolt through the handle and the sinking feeling when the motor locks up.

This car is insane. If you stand back and look at it on paper or in person it seems like an amazing deal, CarFax looks awesome, private owner, dealer serviced, full service history up to 2015ish. Changed hands in Jan 2015 and only accumulated ~2k miles after that. I purchased from an older gentleman who told me that it went into limp mode and he didn't have time / money to have it serviced. He and his wife only drove it around town on weekends, etc etc. His neighbor friend offered to take a look, he had it towed to the guy's indy shop yard, but decided it wasn't worth fixing. Let it sit with a for sale sign in Timbuktu NY in a storage yard for a year and a half. Body looks perfect, inside and out. Clean CarFax, clean title, symptoms consistent with the story.

In reality I get it home and discover that the engine was removed and / or replaced. Nearly every electrical connector under the hood was destroyed. Whoever attempted the engine removal completely screwed it up, stripped bolts, literally bent the transmission cooler lines into a new route, left half of the intake components off. Screwed up the AC compressor connections. One of the knock sensor connectors was jammed into the harness plug for the fuel rail temp sensor, the other knock sensor pigtail was pinched between the engine and transmission bell housing with the plastic connector ground up by the flywheel. Transmission lines were cut in half by the serpentine belt spewing transmission fluid all over the engine bay. Insanity. Literally everything touched was destroyed. Missing the whole boot on the passenger side rack inner tie rod with about 1/4" play in the exposed joint and fluid gushing out of the rack seal. Someone literally totaled the car with the worst workmanship I've ever seen. You couldn't make this story up if you tried. I'm so glad I bought this car because it would have soaked some unsuspecting buyer.

Right now I'm just hoping that the engine actually was a drop in replacement, that the same crazy mechanical work doesn't extend into the motor. I'm holding out hope because it looks likes the engine was spray painted silver with a rattle can... which seems like the sort of nonsense that a salvage yard might pull when selling a used motor online.

Gentlemen I bought it from claims he bought / financed it at a local defunct used car lot for 10k in 2015 and immediately had problems with it. Essentially parked it because he couldn't afford to repair and pay the loan at the same time. Sounds reasonable except that in the condition it was in when I got it home... it would have NEVER run. No way he could have even got it home from the used car lot five miles down the road. So what did he do, blow up the motor and attempt to replace it himself? I'd love to know, I'm not upset, I actually respect that he tried to bail himself out of a crappy situation. Lien sacked car with a dead motor is everyone's worst nightmare, so maybe he tried to get himself dug out but ended up defeated. Just wish he would tell me because I'm dying to know.

I'm perfectly at peace with the idea that I might need to drop a motor into it. That's why I only paid 1,400. Worst case scenario is it's a rolling chassis for 1,400 bucks... I'm still happy because the body is flawless, interior flawless, it's in great cosmetic shape.

When & if you get to know the seller, you find out just what's wrong with this car.

Fix it, then ask, hey did you have this problem when it was yours?

After they find out you have sorted things out, and are fine about it.

They open up a bit, & often wind up bringing their stuff for you to look at.

An old riding buddy of mine died a while back, bought a motorcycle from him.

Then we rode together for years, didn't talk much just rode the back roads of southern Ohio.

Showed him the modifications made to my bikes, which he then did to his.

Quick update... I pulled the intake last night (for the second time.) Took a little more time to clean it out and noticed a bunch of RTV squished around the front seal that I didn't see the first time. Pulled the front / back covers. Silicone gaskets looked fine maybe even new, someone just squished RTV in there as a "precaution," totally unnecessary. I cleaned everything up, rubbed a little clear silicone grease on the seals and put the manifold back together.

I scrubbed out the valley, replaced the valley hose with an OEM part, replaced both knock sensors with OEM parts, torqued to spec. 15 ft-lb. I also installed some high temp silicone foam to emulate the factory sound insulation foam. I cut and carved the foam around the knock sensors on both sides to make sure they are isolated from outside noise.

Haven't had a chance to drive the car yet. I'm going to get more Mercon SP transmission fluid and do my absolute best to raise the car flat and level to check the fill.

Drove the car last night... no luck, engine running much better overall but I'm still stuck with this stupid high RPM intermittent misfire. I added about 1 quart of fluid to the trans. Car was as level as possible without being on a lift.

I've got three codes related to the transmission that are "sticky" in that they never clear. Even if I leave the car in park, command a reset with the scantool and cycle the ignition key. The following three codes come back immediately and / or never clear in the first place:

P1798 - CAN TCM/IC network malfunction
P0730 - Gear ratio fault - Transmission mechanical failure
P0706 - Gear selector position plausibility fault - TCM/J-Gate module CAN fault

Of the three codes reported, my greatest concern lies with P0730 because I believe that the ratio fault follows a calculation of actual gear ratio tracked by the TCM's comparison of input speed vs. output speed. Theoretically it is possible that my high RPM restricted performance event occurs as a result of the TCM faulting out, misfires might be a symptom of the restricted performance mode triggered by the TCM. Normal messages related to the transmission may not be coming through on the instrument cluster display due to the condition causing P1798. If the CAN link isn't working, I might not be getting the usual "Gearbox Fault" message on the dash.

Hopefully my Autoenginuity scantool will report the calculated "effective" gear ratio that underpins the P0730 fault. If so, I can drive with the scantool connected to see if the value falls out of range immediately before a restricted performance / misfire fault occurs.

I wish there were a way to place the TCM in a diagnostic mode (as with some other makes) to block a fault trigger on a diagnostic test drive. I wouldn't mind letting the clutches slip a little if I could prove that the misfire condition is solely caused by a TCM fault.

Sunroof rain water leaks may have soaked a couple of inline harness connectors carrying the transmission CAN signal. I need to dig through the wiring diagrams some more.

Hello Weekend, Hello Jag... Heading into another fun filled weekend of trying to figure out why she wants to misfire. I suppose it's time to figure out why I've got three trouble codes for the transmission. Will post an update with my findings.

A misfire often upsets a trans so it's often worth fixing misfires or other engine issues first. High RPM and/or misfire - often an air leak. Look at fuel trims and see.

I really want to resolve the P1798 TCM to Instrument Cluster CAN fault. I really don't want to have a CAN network fault lingering in the background when I'm trying to diagnose actual problems. To that end I've virtually gutted the interior because I need to figure out how I'm getting ~1 gallon of rain water in the cabin every time it rains and I want to trace the entire CAN network to figure out why the trans isn't speaking to the instrument cluster. I'm not getting any transmission messages on the cluster display now.

I'm going to pull the front seats and the carpet tonight. Tomorrow after work I'll ask my wife to hose down the car while I sit inside and watch for leaks.

I agree that the high RPM misfire could be causing the trans fault or vice versa. I've replaced virtually everything that could be causing an ignition or fuel misfire. I've lifted and fully sealed the intake. Fuel trims look great, all of the intake duct work is literally brand new. There's no good explanation for the misfire. It jumps all over the place, sometimes bank 1 sometimes bank 2. Just doesn't make any sense.

For the water leak, here are some TSBs which might help:

JagRepair.com - Jaguar Repair Information Resource

Look for XJ501-07 and XJ501-07a - Entry Of Water Into Vehicle –Possible Sources

Dave

Thanks! Yeah XJ501-07 is an interesting read. I'm going to strip down the interior anyway because I need to detail, dry out the carpet and replace the headliner. I'll grab a tube of body panel adhesive and go to work. Surprisingly the trunk seems okay. Sunroof is leaky but the drains seem to flow alright, I'll install a new Jaguar seal on the sunroof glass panel.

On first blush it seems like the issue is concentrated around the passenger side doors. Interestingly I've noted some paint work on that side of the car... maybe minor body damage that never made it onto the CarFAX report.

With so many things wrong with this car I'm trying to make broad sweeps, correcting as many items on the punch list as possible at one time. I was hoping to get her through inspection next weekend but that's going to have to wait. I'm going to tear down the interior tonight, test out the entire CAN circuit and try to narrow down the source of the TCM / J-Gate CAN faults. Then I'll target all areas identified for water intrusion by the TSB and do a full detail / resto on the interior. I'm going to strip down the front passenger door to replace the power mirror, and strip down / replace the headliner, pillar fabric.

You can replace the power mirror w/o stripping down the door assembly. Also use marine grade clear sealant.

Do each of the ingress areas outlined in the TSB, especially the trunk and rear door suggestions. You'll be surprised how much water collects on the rear carpets due to these areas alone.

Quick update here... In the process of diagnosing the transmission / instrument cluster CAN network fault P1798 I've taken some interesting measurements. What I'm finding is that when I measure resistance between the CAN network pins on the female J-Gate module connector I can see the 120 Ohm terminal resistor in the air suspension module referenced on the right hand side of the diagram below. What is interesting is that when I measure resistance on the two pins representing the left hand side of the circuit diagram... I see nearly 1000 ohms, way more than the 120 Ohm terminal resistor in the ECM. Essentially at any point in the network if I break the "rails" I should see 120 Ohms on both sides. That implies that, at least from the J-Gate's perspective, the 120 Ohm terminal resistor in the ECM is hiding behind some larger inline resistance (likely a bad connection.) None of the terminals are shorted to ground or power. My meter does measure a dead short to ground if the meter polarity is reversed on some of these network pins but that implies that the ICs communicating on the network act like a diode when measuring resistance, not a surprise.

When I unplugged the dynamic stability (DSC / ABS) module connector under the hood last night, it felt loose. I'm starting to suspect that the DSC module pins aren't making good contact. Tonight I'll measure pin resistance there and see if I can find 120 Ohms on the ECM side of that connector. Then I'll put the connector back together and open inline connector IP55, check resistance there. If I'm losing my 120 Ohm terminal resistor between the DSC female connector EC30 and the left half (as diagrammed below) of IP55, I think it is a safe bet that the problem lies in the DSC module pin connection. I remembered a TSB https://f01.justanswer.com/crzydrvr0...J413-02am2.pdf which references problems with the DSC module pin connections specifically for the CAN network. Perhaps I am on to something here. Hopefully this is my problem. Time got away from me last night and I had to get enough sleep to hold it together for my day job today :)


https://cimg1.ibsrv.net/gimg/www.jag...94a96b93ea.jpg

For those who are unfamiliar with the CAN bus architecture. I've pasted an image below from the Wikipedia article showing the "typical" structure of the CAN network. Signal structure dictates that both "rails" or wires float above ground. Rail to rail resistance is 60 ohms with everything connected which is created by closing the rails at each end with a 120 ohm resistor. Two 120 ohm resistors in parallel = 60 ohms total resistance. You can see that our X350 CAN bus looks just like the example on the Wikipedia page.

https://cimg6.ibsrv.net/gimg/www.jag...29475eeba6.png
https://cimg9.ibsrv.net/gimg/www.jag...b79e57637c.png

Sounds like you're on the right track with a bad connection. With all of the water leaking in, and the high resistance, I would suspect corrosion at a terminal somewhere along the bus. Given the low voltage of network communications, it doesn't take much to get in the way of things.

Dave

... and just like that I'm back to square one. It turns out I didn't fully seat connector IP55 when I put it back together the other day. I was looking for corrosion and I opened the connector but apparently forgot that I just left it loose and moved on. Unfortunately that created a red herring when I started metering out the circuit. With IP55 connected properly I now have ~120 ohms on both sides of the CAN circuit as expected, huge downer honestly, I really thought I had it pegged. Now that I understand that the ECM and the ASM are the terminal ends of the CAN network I will run an end-to-end continuity check tonight just to be sure that I don't have some high resistance connection lingering. All of the connectors mentioned appear completely fine. I restored the FP6 connector that was damaged by water under the seat so I know what a bad one looks like. IP55 and TL93 got wet but both are elevated and they never sat in a pool of water. The pins look fine and continuity seems fine through both connectors.

I'll probably check continuity from the ECM to the ASM (air suspension module) with the TCM disconnected then check continuity from the TCM to the ASM with the ECM disconnected and compare the resistance. Based on the diagram ECM to ASM and TCM to ASM should be virtually identical, I should get just about the same series resistance through all of the interconnected modules.

I connected all of the various components last night and ran the car for a while to keep the battery alive and shake off the spider webs. I also absolutely verified the alignment of the shift cable based on instructions in the shop manual. Just as the shop manual describes I set the cable for reverse by pulling it all the way out and pushing it back one stop with the J-Gate removed, then replaced the J-Gate unit, pulled the shift lever to reverse and verified that the shift cable eye and the J-Gate ball end line up and snap together perfectly. No adjustment needed. As I pull the shift lever though all of the selections, the red light follows the appropriate selection. There really just doesn't seem to be an issue with the mechanical shift link cable adjustment.

On another note I also verified that re-flowing suspect solder joints on the radio main board fixed my radio display back-light, thankfully it seems to be fine now so the tear down effort wasn't for nothing. I re-flowed several solder joints on the J-Gate module circuit board as well, I'll scan codes when I get home from work and just verify that I still have all three transmission codes present.

I'm still getting water front and rear but I have all of the electrical connectors pulled out and away from any areas that are damp. I want to resolve the faults first so that I understand exactly where things went wrong. For no good reason at all (swaptronics) I just went ahead and bought 5 new ignition coils to make a full set of 8 new coils. Instead of trying to target dead ones I'll just have a full brand new set on the motor to rule out any lingering issues with a coil here or there. That means I have 8 new OEM plugs, 8 new SMP UF-519 coils, one refurb. GB injector and all 7 remaining injectors were ultrasonic cleaned and flow checked.

The hunt continues... I replaced all eight coils and installed a used EGR valve that I had purchased when I suspected EGR issues immediately after buying the car. I replaced both EGR gaskets with brand new and applied some copper coat to both sides of each. I also swapped the fuel pressure sensor with a used unit that I had in my parts bin, just to rule out any issues with fuel pressure feedback.

Same exact behavior, under load, high RPM I've got a multi-cylinder misfire. Thankfully I was able to clear transmission code "P0730 - Gear ratio fault - Transmission mechanical failure." I still have the CAN instrument cluster to TCM fault and the J-Gate CAN fault but I don't believe either issue is related to the high RPM misfire.

Has anyone experienced a head gasket failure that behaved this way? I see no problem with engine temp, no issues at all when it is driven enthusiastically. The problem only occurs when I put my foot to the floor and the RPMs come up between ~6300 and red line under load at wide open throttle.

Under load high rpm - may well be a blocked cat.

Thanks, clogged CAT is an interesting suggestion! That's a totally new angle that I haven't considered. What is the best way to inspect for a clogged cat? I've never encountered one before. Should I just open the exhaust and do a visual inspection or is there some way to see evidence in my scantool data?

Additionally I'll mention that I've set up and begun using JLR SDD 130 instead of Autoenginuity. Autoenginuity is fast and simple but SDD gives me access to real-time PIDs from different modules at the same time. I'm planning to take it for a drive tonight to monitor fuel pressure and MAF under load at high RPM. Is there anything else I should monitor while I'm at it? I'll record / snapshot if I see anything interesting.

Depsite setting the multi-cylinder misfire code, I do seem to trip cylinder 3 and cylinder 8 pretty frequently. What is interesting is that both of those cylinders were missing at idle when I first bought the car. Changing the coils eliminated the #3 and #8 idle miss but I'm starting to wonder if I don't have a weak primary signal on those coils. Perhaps the new coils simply respond better to a poor primary ignition signal. Prior to replacing the coils, I swapped them around the engine and there was no change with OE units, the idle miss was still there. I thought that was weird at the time but I assumed I had just lost track of which cylinders' coils got swapped.

Unfortunately I don't own a portable oscilloscope but I can hook up my bench scope at idle and look at the primary on 3 and 8 vs. known good. Cylinder #4 has been well-behaved since I bought the car so I sort of use it as my benchmark.

Cylinder 3 and 8 are consistently misfiring at high RPM. Compression in both seems normal, no evidence of oil or water contamination on the plugs, if anything they look like they are running lean. I swapped the coils around yesterday and cleaned up the fuel pump inline harness connector FP2 which got wet under the seat along with FP6. Fuel rail pressure rises ~15psi to 65psi right around the time that the misfire occurs. The problem is its hard to tell if that happens due to computer intervention to shut down injectors or if it is normal for WOT under load.

Regardless, I am still getting the same exact behavior. Manifold absolute pressure looks fine, fuel trims look fine, all of the transmission codes have cleared out. The only thing I can think is that I'm not getting a good primary ignition signal at high RPM from the ECM. LT and ST fuel trims seem awfully low at high RPM for a car that consistently misses on two cylinders. Regardless of whether I have a fuel or ignition misfire I would expect the trims to indicate a lean condition resulting from incomplete combustion. Hopefully I can disconnect the O2 sensors without sending it into limp mode. I'd like to take it on an "open loop" test drive to see if the behavior changes.

Cats may be damaged due to this lingering misfire condition and all of my test driving but I don't think the cats are clogged enough to cause a misfire. There is no sign of a clogged cat in the MAP data and occasionally it will pull hard without misfiring. Injectors and coils have all been swapped around, 3 and 8 are missing consistently. On a rare occasion I pick up a misfire on cylinder 7 but SDD shows intermittent and it might happen once every 4 or 5 test drives.

All brand new SMP UF519 coils and coils have been swapped around six ways from Sunday, misfire does not follow the coils... replaced injector 3 to rule out an injector issue, swapped injector around, doesn't follow injectors. Short of any further ideas I am going to start digging into the primary ignition and injector driver circuits.

For my diagnostic test drive tonight I am going to plug ALL of the vacuum accessories. No PCV, no EVAP and no Power Brake Booster. If I'm still missing I will pull the fuel rail pressure sensor vacuum line (plug the port) which will trick the computer into running rich. That should improve the problem drastically if I have some unspecified air leak via manifold, duct, etc.

I had misfires at high RPM before when my catalytic converters were shot. The clog allowed enough exhaust flow to get through at low RPM to where it wouldn't cause misfires, once it got higher though, the exhaust gas would get backed up and superheat the downpipes and cause random misfires.

Have you checked that cats?

I fabricated a threaded plug to go in place of the upstream O2 sensor that gave me a hose barb for testing back pressure. I installed a very sensitive analog gauge that ranged -15 to 15psi. From Idle to rev limit at ~4200 rpm I have absolutely no back pressure in the exhaust, in fact it pulls a slight vacuum at high RPM as expected (exhaust rushing past the gauge port.) Pressure pulses at idle are ~1/4 psi max and appear very consistent on both banks. No obvious dead spots to indicate a sticking exhaust valve or anything unusual happening with the engine. No difference in back pressure behavior from bank 1 to bank 2 and 99% of the time my misfires originate on bank 1. If it were a clogged catalytic converter I would have expected to see a major difference between the two banks. What other method would you use to check for a catalytic converter problem? MAP pressure / vacuum behavior looks picture perfect on the scan tool. No vacuum pulsation to indicate an intake timing problem or intake valve issue. Vacuum at idle is ~19 inHG, I'm not sure what a normal baseline idle vacuum would look like for this engine because I didn't conduct this extensive testing on my previous 2004.

The only thing I noticed that seems unusual to me is that my MAP sensor reports 13.92 psi with the key on / engine off. I would have expected to see something closer to atmospheric 14.7. My altitude is ~950 ft above sea level so I should be really close to 14.7. Granted the MAP sensor calibration may be poor but why would that only cause a problem on bank 1?

It doesn't look like I could visually see the catalyst honeycomb on these cats if I pulled the pipe down because there is a bend just upstream of the converter. I would need to buy an inspection scope to look down in the tube and I could only justify that random expense if I observed some back pressure during testing. Again, the problem is concentrated on bank 1 and mostly Cylinder 3, therefore it is hard to justify buying brand new catalytic converters without some way to verify.

This problem is exhausting (no pun intended)... I have absolutely no other trouble codes on the system now, I've cleared all of the electrical issues. Transmission is clear, all network faults are resolved, absolutely nothing else wrong. When it misbehaves, cylinder 3 misfire is the most common but I occasionally pick up cylinders 1 and 7 as well so it is 99% related to bank 1. Maybe two or three times in all of my extensive testing I have pulled an intermittent code for cylinder 8. I would say 90% of the time it is cylinder 3, 99% it is on bank 1.

What am I missing here? This is the most challenging car problem I've ever faced. The fault is NOT intermittent. It bombs out and misfires every single time I romp on the gas, usually it will hit about 6k RPM and drop into limp mode instantly. The problem is that I can't reproduce the failure in my driveway, it has to be under load. I beat the daylights out of it the other night, just pulling hard, triggering the misfire, and repeating, over and over again, hoping to catch some coolant loss or something to give me direction. Absolutely nothing wrong. I watched every sensor I could think of, nothing unusual. Coolant level is totally stable, no odor or residual pressure in the jar. Running compression is consistent across all 8 cylinders. No other codes triggered. Vacuum looks great, exhaust back pressure is normal. All 8 coils are replaced. I'm 99% certain that the problem doesn't move with the injectors but I'll verify one more time tonight. Crank and Cam signals look consistent. Fuel trims look normal through the whole RPM band up to where it fails.

Tonight I'm going to take it out and drive with the O2 sensors unplugged which should eliminate the trim system and rule out any issues related to the O2. I'll then try unplugging the MAP sensor for another drive, not sure if that puts in limp mode or not.

:icon_bricks: HELP

Okay interesting result last night... unplugging O2 sensors made no difference in the misfire behavior. I also confirmed that cylinder 3 is still showing a miss despite swapping the injector to #4. I had a misfire code for cylinder 8 as well, which happens occasionally.

Grabbing at straws, I connected my analog vacuum gauge to the fuel pressure sensor vacuum reference and found ~17.2 inHG at idle, in agreement with the MAP sensor, but seemed a bit low to me given that I had previously noted 19. I was getting irritated because the fan was running constantly while I was trying to poke around so I reached inside and shut the AC off. When I came back to look at my gauge I had over 20 inHG! AC switched on drops my engine vacuum ~3 inHG at idle. Various electrical loads have a big impact as well. This seems unusual to me... not sure if it means anything but I'm going to investigate further. Tonight I'll pull both O2 sensors off and see how the vacuum compares to idle with sensors installed. If I see a big change with sensors out then that gives me justification to tear down the exhaust and investigate.

I'm also going to watch battery voltage during my high speed runs to see if it could be an alternator problem. I'll probably order a PicoScope today so I can try to catch the coil primary and fuel injector PWM in the act while it is actually misfiring on the road.

After months of struggling with this problem I've finally had a breakthrough! The high RPM misfire became consistent enough to trigger restricted performance mode with a quick snap throttle to 4k RPM while in park in the driveway. This gave me an opportunity to try a bunch of things quickly to see if anything changed the behavior. Lo and behold, unplugging the fuel rail pressure sensor electrical connector eliminates the problem! Not going to lie, I jumped up and down a little and had a celebration in the driveway. Enthusiastic test drive confirmed that the high RPM misfire is totally gone when the fuel pressure electrical connector is disconnected!!!

Here is why this is a weird result (and totally unexpected):
- Early in my diagnostic process I swapped the fuel rail pressure sensor with a used spare part that I had in my parts box from my previous '04. High RPM misfire didn't change at all. The used sensor was always presumed good, I just got it with a used fuel rail assembly.
- Early on, I tried unplugging the vacuum feedback hose and plugging the throttle port which would have commanded excess fuel under all circumstances (rich condition) but it had no impact on the high RPM misfire.
- Even though the retention tab is broken on the harness connector, the female pin sockets feel tight (tested with a male terminal) and don't show any sign of corrosion, seal is intact. I zip tied the connector down to hold it firmly in place due to the broken tab. No change in behavior.
- Pressure feedback appears to be normal and stable on the scantool when the sensor is plugged in, ~55 - 60psi throughout the RPM band. Fuel pump PWM driver ramps up as expected to maintain pressure.
- At first start, fuel pump primes the rail with no issue and the engine starts enthusiastically. Idle to 4k RPM is buttery smooth with no hesitation with the sensor plugged in.
- Throughout the RPM band up to 4k RPM the fuel trims remain low and steady. Immediately before I get the restricted performance message, trims jump to ~10% but it happens so fast it's hard to tell if it's a cause or a symptom.
- With all of my intake sealing work we are at 0%/0% LT/ST at idle on both banks! Totally sealed, no air leaks at all.
- O2 sensors respond quickly and seem to be accurate. Unplugging the O2 sensors (front and rear, either bank) has no effect on the misfire condition.

Action Plan:
- I am going to install a brand new, Jaguar OEM fuel rail pressure sensor (ordered one on eBay)
- Ford sells a replacement electrical connector under Motorcraft Part #WPT118, I'm going to order one, extract the terminal pigtails and crimp new terminals and wire seals onto the harness wires. I always hit each terminal crimp with a tiny bit of solder. Will perform a zero splice repair.
- Replaced the fuel filter Saturday but no change in the behavior.
- I'm going to trace all of the fuel pump wiring to see if there is any chance I have a bad electrical connection to the fuel pump.

At least I have a solid diagnostic path now!

Update: I ordered the AMP connector kit from Ballenger Motorsports https://www.bmotorsports.com/shop/pr...oducts_id/3418
My preference is to crimp (and solder) new terminals on the old harness wires whenever possible and install a new plastic connector body. This approach avoids a splice repair using pigtails. $8 for the BM connector kit vs. $50 for the Motorcraft pigtail kit.

*** RESOLVED FINALLY ***

After months of battling the intermittent high RPM misfire I finally resolved it by replacing the fuel rail pressure sensor with a brand new Jaguar OEM part and replacing the associated electrical connector with a kit from Ballenger Motorsports https://www.bmotorsports.com/shop/pr...oducts_id/3418. I crimped and soldered all three female terminals and applied a little bit of silicone grease to each of the wire seals.

All in all, proof positive that sometimes it is something really simple and cheap but extremely difficult to identify. The breakthrough came when the problem became consistent enough to be reproduced in Park in the driveway by simply revving the engine. In hindsight my only clue should have been a brief spike in short term fuel trim values just a fraction of a second before reaching 4000 rpm under load. I'd get a blip of high trim values and then the ECU would go into reduced performance mode.

If you are experiencing a misfire under load, don't automatically assume that you need to replace the catalytic converters. I was becoming really convinced that the cats might be to blame but I'm glad I held out a while longer because it would have been gut wrenching to sink 1k into converters only to start it up and have the same issue. Visual inspection of the cats gave me confidence to rule them out. I've seen damaged catalytic converters before and it is usually really obvious that they are melted, clogged, or broken. Mine looked completely normal. An overnight soak in lacquer thinner revealed no dissolved carbon buildup or loose chunks of catalyst. If you think it might be cats, probably worth pulling the old ones off to have a look first.

Congratulations on a job well done, keeping at it will eventually solve the problem.

Not an easy task, but necessary, unless you have the money to pay to have it done.

Phew!

That's a very rare issue so no wonder you've had trouble finding it.