Wow, I can feel your pain. It seems like you've gone over just about all the fuel, spark possibilities. It's a real mystery story. It does seem like the only thing left is a hardware issue. Before throwing in the towel on this engine, you might pull the cam covers and verify that the valve timing is correct. If it were a valve or valve seat problem, it would show up on one cylinder (I would think), but it sounds like the problem is with multiple cylinders. It would be strange if the nikasil decided to give up at this point, but maybe it's possible.

 

I understand that you replaced the ECM was it from a testing and repair company and does it have a test/repair company sticker on it? Now the chance of it having the same problem is slim but it is possible. I would send it in and have it tested.

If I recall correctly you replaced all injectors were they reconditioned or new. When using the sounding check you should hear a ping not a thud sound on all injectors.

The compression check was low and the Restore should have improved it but maybe not enough or it could be a valve seat problem.

It would be interesting to see if the compression increased after using the Restore.

 

I understand the frustration:- if you've reached that point then perhaps put it aside for a while and come back later. I think you have too much sweat equity alone invested in it to throw it away now.

There were times when I could cheerfully have set the MG on fire just so I could have my revenge on the thing. It hasn't been a daily driver since the kids were born so it just went back in the garage waiting for another day. It's all good now and has been for several years.


I'm still agnostic as to whether it's electronic or mechanical: the car thinks it's electronic, but the compression tests seem low. I'm looking at how we can prove/disprove my thinking: do you have a multimeter that can read frequency (either a 'Hz' button or setting)?

It's just a process of elimination, and things are getting ticked off.

 

My worry about getting a replacement engine is that I have other codes that are troubling. Not sure how to troubleshoot them.
U2012, B1352, B1595, U1135, U1041, P20EE. And then I have codes in the seat modules although the seats work fine. B1965, B1957, B1961. My Autel scanner tries to clear these codes but they all come right back. I can't imagine any of these codes pertain to the engine problem. Just a little concerning.

 

Hi,
We (Gus and I doubled) asked motorcarman for his expert opinion on the compression readings you have reported. This was his reply (I didn't expand on the background, so some of the checks you have already done):

"
I haven't done many compression tests in years. You can tell if there is a low or 'dead' cylinder by just the sound of the cranking.
All those look even but unless you get the engine 'up-to-temp', shut down and then check compression, the readings will likely be low like that.

The low tension of the rings in the nikasil engines need oil to seal. A cold engine should start with those readings and THEN once started, things EXPAND and then oil will probably help with the compression and the engine run normally.

As long as it does not smoke or blow oil into the full load breather and air filter box, it is probably a sound engine.

I would swap the coils around or replace them with 'good-used' to see if the fault follows?
Bad connection at the ECM or harness?
Worst case would be the ECM output semiconductors failed and need a replacement ECM?

The AJ26 used external coil drivers but the AJ27 is all contained in the ECM.
"

So it is likely that the motor is good if you don't have the symptoms he has described.

I would treat the other B and U codes as a distraction for the moment.

 

I gave a little info to John in a PM and it is near the same as what you and Bob provided.

Still on standby

Gus
JagRepair.com - Jaguar Repair Information Resource

 

I hope this isn't one of those "unsolved mysteries".

 

Hi everyone. The issue is not solved but seems to have a pretty elusive solution. I haven't had time to do anymore work yet but will. Many of the suggestions have been tried 10 times. I have swapped, replaced, and swapped again, coils on the car. No change. Wiring harness was changed out with a full continuity check, same problem. I will do a resistance check next. ECU has been replaced with no change. I think an oscilloscope check on the coil firing would be good but I don't have that equipment or knowledge. As everyone has told me, replacing the engine will not solve the problem as the engine is probably good. So, I will continue, but I will admit to getting close to admitting defeat on this.

It will be 3 years next month that this problem has keep the car in the garage. At some point my patience will get too short. This is the first time in my life that I feel beat by an automotive problem. I would have already given up on the car but I have done so much work on it over the last 7 years, that I lack the ability to say no mass yet

 

Hi John,
Good to see you're still hanging in there

I have a test that may help, but I need to run it on my car first. I know our motors are different, but it will be valid for either AJ26 or AJ27.

Two quick questions: is the ECM a replacement, or did you send yours out for check/repair, and do you have a multimeter that can measure Hz?

 

It seems that #3, 5 &7 cylinders keep coming up in many threads and the results are often the ECM. I would look to see if that ECM has a repair sticker on it or just send it in to have it checked.

Gus
JagRepair.com - Jaguar Repair Information Resource

 

Aging gas is probably not helping. Maybe replacing this old gas with race/high-octane gas would be a worthy test.

Also, have you thought of more aggressive options, like dropping either water or seafoam in the intake in the hope of doing a deep steaming/cleaning of the combustion chambers? Is there an off chance of helping the piston rings get a better seal in the process?

Best of luck, keep us posted.

 

I got the ECM from Foreign Auto Computer repair. It has this sticker attached. I sent the original one to them and they copied whatever needed to be copied to the new one.

 

I come late to the party but feel your pain. My XKR is driving me crazy right now with a no-crank situation, so I don't have all the answers, but here goes nothing...
When I checked your location, Florida, I question whether you got stuck with a car drowned in a flood. I bought a couple of Jags from Texas and was worried about that.
The issue of low compression is thought provoking. If nikasil cylinders were disintegrating, rings fracturing or valves burning, one would expect to see some at or near normal with some very low readings. Low compression across the board, after seeing the photo of disconnected cam chains, makes me wonder if cam timing is off. Been there, done that. As an alternate theory of cam timing, you wouldn't think the intake cams were inadvertantly advanced right at idle without the tattletale Check Engine Light going on with that code.
With a bombardment of multiple Codes, the first thing that comes to mind is poor grounding. Sensors misreading low voltages throw codes. It won't cost anything but free time cleaning grounds, bumper to bumper. I understand the harness and ECU have been replaced, but electricity doesn't care about anything but going from high potential to ground and extra resistance has a way of throwing a wrench in the works. Up here in Michigan, we give salt baths to vehicles annually. I've been known to add extra engine-chassis grounds and even lacing braided copped ground strap directly from negative post to numerous grounds on chassis, engine and trans. I recall a Honda that refused to start until I moved the ground from a 10mm screw in sheet metal to the block. It fired right up, but nothing else on the car worked! Bad engine-chassis ground. The air-fuel ratio meter on wife's MG TD quit, the one-wire O2 sensor quit grounding in the stainless steel bung on the exhaust pipe. Try adding a small ground wire on some clean, shiny screws from the engine to chassis (but don't use the valve covers as they likely won't ground).
Just for kicks, up the voltage before startup. The Jags came with a very large ECM battery with slightly higher than normal voltage. Try adding a heavy duty battery charger or jumper cables from running vehicle to give sensors an advantage. The alternator provides significantly higher voltage once running, so if it spits out codes after it is running, then the issue isn't a weak battery. Use a meter to check voltage is above 14V when running.

 

I can't stop but think they may have overlooked something in THE ECM or the problem still existed when the ECM was installed causing it to fail. Just a thought!

Gus.
JagRepair.com - Jaguar Repair Information Resource

 

So it could possibly be the same unit you sent them, repaired and re-stickered?

I have documented some tests that will help to further pin things down. They should be straightforward, but I need to go through them on my car tomorrow before I post them. Theory and practice & all that...

 

No, definitely not the same unit. There was a mark on mine that was not on the new unit.

 

I keep pondering your mix of symptoms and am getting a vibe I'll share.
From what I gather, you experienced instantaneous rough running; start up followed by immediate shut down; combinations of codes including misfires.
The fact that it starts indicates the ECM is happy with what feeds back and allows cranking. My problem is I can't get cranking, so somewhere along the line, I can't get over a preliminary hurdle.to get as far as cranking. Nevertheless, you can eliminate spurious things like park/nuetral switch issues, inertia switch, fuel problems. One point is, I'm guessing the XK8 ECM reads the crank position to establish the cam relationship as far as power strokes vs intake strokes, whereas some engines get this from the cam position sensor.
The idea the engine fires and then shut right down precludes closed loop issues like O2 sensors, etc. It seems the ECM is missing essential feedback and quits.
The terms "crank sensors" and "camshaft sensors" opens up cans of worms as they can vary widely in functions. How does the XK8 ECM come the the conclusion there's a misfire if you can plainly see steady spark? It may be reading the crank position sensor and receiving poor feedback, misinterpreting the crank velocity as stumbling between cylinders. The fact that the crank sensor was swapped out, one would assume it isn't the sensor. I replaced a crank sensor on a V6 and it ran just fine, on 4 cylinders- it just couldn't pick up one notch in the flexplate. I had a 4 banger drive me to drink with symptoms like yours until I ran a new wire from the ECM to the crank sensor as overheating the engine had corrupted the wiring harness. Track down the ground from the crank sensor and clean it. I had another engine with all sorts of gremlins until I tracked down the screw where the engine sensors grounded and realized the problem- Honda used thermostat housings to ground engine electronics. The gasket insulates the housing from the block and the corroded 10mm screws didn't conduct worth a darn! I've also had issues where the sensor was too short and other situations where it was a hair too far away from the ring,
On the other hand, the XK8 ECM might start off looking at the crank and then switch to cams and that signal is out of spec. I'd have to go back and look.

One other question I had, has the engine ever overheated badly? I've worked on some that had overheated and lost compression. Throw in a new set of rings and it was back to normal. I've tried to chase down Jag compression specs but found it a dead end. Someone mentioned the Nikasil engines had rings with less spring to reduce wear. That may or may not be true. I seem to recall readings 180-190 range on one of mine. With 10.75:1 compression ratio, I'd expect to see that, but an engine with a tight grouping like yours should still run fine. My Desoto flathead ran on 40psi compression readings.

 

These days, I've forgotten more than I know about my Jaguars. After reviewing the Tech Guide, I now recall the "lump" on a cam for the ECM to pick up cam position and the flexplate is a large tone ring for precise crank speed feedback.
The engine cranks holding off firing until the single lobe passes informing the ECM where cams are and then unleashes "lightning and thunder" switching to the crank tone ring & speed sensor to control idle speed. "The ECM produces steady state running at the target idle.speed within 2 seconds of firing, " A poor signal from the crank could be the reason the engine shut down after a few seconds firing. An inconsistent signal could explain why the ECM reports misfiring on so many different cylinders.

 

Ignition
The ECM supplies two ignition amplifier modules, which each supply four spark plug mounted ignition coils. The ECM varies the ignition timing to optimise power, emissions and
driveability at all operating conditions.
A diagnostic signal from each amplifier module enables the ECM to monitor the supply to each ignition coil. If the ECM detects a failure it disables fuel injection to the affected cylinder

Another area to examine, at the top of the firewall behind the engine. I would still bet the ECM monitors engine speed very accurately from the crank speed sensor to identify very slight dip in rpm between cylinders to label as misfire.

 

the ECM may indeed detect a misfire, but there won’t be any variation of rpm between cylinders,

……unless of course the crankshaft is broken and is no longer a solid connection between every cylinder….



Z