AJ16 intermittent misfire

[cjnsharpe124]

My 1995 AJ16 XJS manual has an intermittent misfire that I am struggling to diagnose.

Symptoms - On initial start-up when cold it runs fine at idle with no obvious misfire and drives fine. When it gets up to operating temperature there is a small stumble at idle and on the road it will will occasionally stumble for a fraction of a second as if it it didn't fire on one cylinder.It accelerates fine and I would say the misfire is most evident when cruising along at 50-60mph on light throttle. It is difficult to be precise on how often this occurs when driving. It can occur very occasionally e.g. nothing to 10+ minutes while cruising but then it might happen several times in a single minute (it the road is not 100% smooth it is difficult to distinguish misfire from bump in road. There are no warning lights on the dash and if put ignition switch to position II and leave for 5+ seconds no codes come up on trip computer.

Diagnosis and works to fix so far;
1) Borrowed professional 'all makes' Autel OBDII reader, only code was Faulty Lambda sensor (no engine warning lights on dash) - replaced both lambda sensors with good used items from reputable Jaguar specialist. No change
2) Faulty crankshaft position sensor - replaced CPS for new NGK item. No change
3) Faulty spark plug - Removed spark plugs, all looked to be in good condition with correct colouring and no obvious oil or soot fouling. Replaced all 6 plugs with new Champion RC9YCC items as inexpensive and just in case. No change
4) Faulty plug-on coil - bought new Lucas coil, ran engine up to full temp so idle stumble visible, replaced each existing coil with new item. No change.
Fuel feed problem - On basis that it runs smoother when cold and under acceleration I suspected fuel feed problem so I tried these remedies-
5) Changed fuel filter for new Bosch item. No change
6) Faulty Injector - removed all 6 injectors, sent to injectorcleaning.co.uk for professional clean (very quick turnaround, sent Tuesday afternoon, back Thursday morning before 10.00am with report). Report said all spray pattern were good and flow levels consistent before clean. Flow levels slightly higher after clean. On basis of this I replaced the fuel pressure regulator with a new unit just in case. Still no change.

I would like to get my own OBD reader to see what that says. Does anybody have any recommendations for an OBDII reader as from reading the forums it seems not all new modern readers can read the XJS and I would like to read live data and clear any codes.

Any ideas on what might be causing my misfire, I am stuck.

[ng4xjs]

Sounds possibly a MAF issue. I had something similar. Initially solved by a clean but then unfortunately a new unit (not expensive). My AJ16 (1994) was only OBD1 but by scanner did extract at least the faults displayed on the radio and a tad more.

Nick

[cjnsharpe124]

Still chasing this misfire. Another observation is that the exhaust smells very rich, Fuel consumption is not bad but it has decreased by a few mpg since the misfire started, I tried swapping the O2 sensors connectors just in case I had mixed them up when replacing them but this just made the engine very unhappy.

I now have an iCarsoft LR 3.0 diagnostic reader which after some persuasion I managed to get to talk to the car. The information I am getting with the car up to operating temperature and at idle is as follows:
Short term fuel trim B1: fluctuation from 2.3% -> 10.2%
Long term fuel trim B1: 55.5%
Short term fuel trim B2: fluctuating from -1.6 -> 10.2%
Long term fuel trim B2: 55.5%
Calculated LOAD value: fluctuating 13.7% -> 15.3%
Air Flow rate from Mass Air Flow Sensor: 0.04 g/s steady reading at idle, increases to 0.12 g/s at 2100 rpm
Oxygen Sensor Output Voltage (B1 - S1): fluctuates 0.135v -> 1.235v
Short term fuel trim B1 - S1: fluctuates 1.6% -> 7.0%
Oxygen Sensor Output Voltage (B2 - S1): fluctuates 0.095v -> 1.256v
Short term fuel trim B2 - S1: fluctuates 1.6% -> 11.7%
Throttle opening: 11.7%

Running the car again a few hours later produced similar results with the Air Flow rate from MAF Sensor still at 0.04 g/s.

Google research indicates the flow rate in g/s should be about the same as the engine capacity in litres i.e. 4 g/s so my hypothesis is that ng4xjs is correct and the MAF is faulty, BUT I am not getting any OBDII messages or codes to say the MAF is faulty so before I rush off and try to source a new/replacement MAF is there any way I can test it? I have taken it off and cleaned it using Liqui Moly air flow cleaner and cleaned the contacts but no change, reader still says 0.04 g/s. Can you even get a new MAF? Any advice?

Car is 1995 XJS 4.0 with manual gearbox.

[ptjs1]

cjnsharpe124,

A couple of thoughts,

A stumble, rather than rough idle, feels like a signal-based sensor or cylinder component not operating properly eg Coil, Plug, lambda, CTS etc.

As you've changed coils, plugs, lambdas etc, we probably need to look at other components that start to influence running when the engine is warmed up.

One key one is the Coolant Temperature Sensor. If this is faulty, it tends to manifest itself during hot running rather than cold, often over-fuelling. It's cheap and worth replacing as a consumable item anyway.

Other considerations could be the MAF or TPS, although my limited experience with these is rough running rather than stumbling. I know you've taken readings on the MAF. However, before you condemn it, I recall that I once read of someone who had taken readings that looked wrong, and after a lot of investigation, he established that Jaguar's early OBDII on the late 94/95 AJ16 engines had the MAF readings displayed wrong by a factor of 10 ie .4 was actually 4! Incredible, I know, but I'll see if I can find his article. So, MAYBE don't condemn your MAF just yet.

TPS voltage can be measured, the key readings being at idle and then seeing progressive rises. I can find out the figures if you don't have them. This is a common problem, but usually causes wrong idle rather than stumbling, but could be the cause.

Throttle butterfly at close is crucial and needs to be checked. If wrong, the conflict of air readings can cause the ecu to incorrectly fuel which could cause the stumbling.

The other place you might want to look is vacuum hoses that may leak when hot. There's a typical one under the intake elbow / throttle body.

Hope that give you a few considerations,

Paul

[cjnsharpe124]

Hi Paul, thank you for your advice.

The temperature sensor has been swapped out for a genuine Jaguar supplied part (LHE1600AA) as part of my efforts to chase this down.

I haven't checked the TPS so the voltage readings would be useful if you have them. There seems to be some uncertainty about the correct part number for the 1995MY TPS, I think it is JLM12074 but the Jaguar Classic Parts site doesn't provide XJS VIN numbers for this part.

In my case I think the OBDII reading is out by a factor of 100 but if there is more information on this anomaly it would be useful. I don't want to replace the MAF as I cannot find anybody selling new ones and I am not sure about s/h as mine looks fine but might be faulty. I have found a company in The Netherlands who say they repair it but it is expensive (300 euros plus shipping plus duty and VAT).

Do you know if the MAF sensor voltage readings are available? If so perhaps I could check the MAF output voltage (or current) at various engine speeds using a multi-meter rather just rely on the OBDII readings if they are suspect.

I will remove the throttle body and give it a good clean and check the butterfly at close.

I will also check out the vacuum hoses as suggested.

One other thought is to check the ECU connectors and earth, where will I find these?

Thanks

Chris

[cjnsharpe124]

Still struggling with this. I removed the throttle body and cleaned it thoroughly.

The Throttle Position Sensor is a black Ford device, see picture. I removed the TPS, cleaned it and checked that it did rotate ok. I checked the TPS resistance which measured 120 Ohms at throttle closed and 3925 ohms at throttle fully opened. The resistance seemed to increased in proportion to the throttle opening. I don't know if these readings are good or bad.

Re-assembled everything, started the engine, no change. Engine wants to stall when opening throttle from idle so I did an Adaption of the Throttle Position Sensor using an Icarsoft LR 3.0. No Change, stumble is still there.

Air Flow Meter (MAF)
The Jaguar Workshop manual 4.0 litre section gives a test for the Air Flow Meter but this refers to an air flow meter with 6 connector pins and mine has 3. Pin 1 on the connector is towards the front of the engine and has a black wire (I assume -ve to earth). Pin 2 is Green/Pink, Pin 3 is Red (ish) or Pink not absolutely sure as it is faded. I inserted some rear probes (paper clips) into the MAF connector so i could read the voltages. The MAF wire colours don't match the circuit diagram in the manual adding to uncertainty.
At idle the MAF voltages are;
Pins 1 - 2 Reading is jumping around from 2.25 vdc to 14.2 then seems to settle at about 13.5 vdc., This does not change if I increase engine speed.
Pins 1 - 3 Reading is 14.2 vdc
Pins 2 - 3 Reading is 0 vdc.
Live engine data indicated 0.00 g/s air flow

Research indicated that a Lucas 20AM Air Flow Meter was also used on some V8 Land Rovers or Range Rovers of the same period. The jury was out as to whether they were interchangeable but it did have the same 3 pin connector so I bought a cheap copy that said it was for a Land Rover with a Lucas 20AM air flow meter.
The EMS did not like this unit very much and the stumble became more pronounced. The MAF idle voltages were the same as with my original unit although the live data air flow rate was 0.04 g/s. No progress.

Original Air Flow Meter back on. One change is that the engine now stumbles and can stall when opening the throttle from idle. Live data has Idle speed is 682 - 708 rpm. The Absolute Throttle Position is 12.2% which seems high but I don't have any reference data.

OBDII Fault code is P1178 - Banks 1 and 2 long term fuel trim too lean (air mass flow rate). This is counter intuitive as the exhaust smells rich.

The Fault code is making me think Air Flow Meter but the stumble when opening makes me suspect Throttle Position Sensor. Both are expensive so any help gratefully received.

[ptjs1]

cjnsharpe124,

The wiring colours in the Electrical Manual for the MAF state:

Black with Yellow trace
Green with Pink trace
Pink with Slate trace.

Is that different to what you have?

Paul

[cjnsharpe124]

Hi, yes that is what I have. The printing of the circuit diagram in the workshop manual is very difficult to read and I thought it said W with N for the Black wire but good to know it is correct.

[PaulGover]

This is probably irrelevant, but might be helpful: My 1991 V12's TPS died. It was a 90 degree potentiometer, and after investigation I found a generic part from RS Components (I forget the exact details, but the crucial points are overall resistance (I vaguely recall 120 ohms), 90 degrees rotation, and automotive use, as they have a special track that can handle the constant twitching over a small range of movement that characterises motorway driving). I got it somewhat wrong, as a final crucial point is rotation direction, and I ended up having to make a special bracket to hold it inverted under the throttle mechanism... Whatever, the point is that comparatively cheap substitute TPS's exist.

The other, again probably irrelevant and V12 specific, point is that the ECU isn't actually very interested in said TPS. If I correctly understand the manual, it's only there to tell the ECU the speed with which the throttle is moving, so it can detect when you floor the throttle, and respond with an enriched mixture - like the dashpot on SU carbs. The rest of the time the ECU relies on air flow and the like. When it's actually interested in throttle position - idling - it relies on a separate microswitch.

[cjnsharpe124]

Apologies for the long post but I still have this misfire. To recap the car is a 1995MY XJS 4.0 Manual. The engine starts readily cold or hot. When starting from cold if runs beautifully smoothly for a few minutes than as the temperature increases so the misfire is introduced. When it is warm you can see the revs vary slightly as the engine 'coughs', the misfire is very noticeable at 1400-2000 rpm, it is not consistent but it is persistent. I have checked the spark plugs and they all look good with no signs of fouling or lack of fuel. I have been chasing this for months and have done the following (all parts from reputable Jaguar parts specialists);
1) Fit new Champion spark plugs
2) Fit new Fuel filter
3) Fit new Crankshaft position sensor
4) Fit new engine temperature sensor (the one that goes to the ECM)
5) Put a new coil on each spark plug in turn
6) Fit new Lambda sensors (car has 2)
7) Remove injectors and get professionally cleaned, also fit new fuel pressure regulator at same time
Fit new Lucas DMB411 coils to all cylinders
9) Replace MAF with good S/H unit
10) Check for vacuum leaks using Smoke test. Fit smoke machine to connector between MAF and intake elbow. Smoke escaping from between the intake bellows & throttle body. Tighten bellows tube clamp, no change, replace bellows and use wider tube clamp, re-test no leaks. I though this might be it but misfire persists.
11) Refit original crankshaft position sensor just in case wrong or faulty item supplied (unlikely), no change.
12) Visually check ECM, no sign of water having run down the mounting panel or into the ECM

There is no Check Engine light or fault codes displayed. My iCarsoft LR 3.0 did read the codes and engine data if I told it the car was a 1996 X300 but now it can't read the data for some reason. My little generic Bluetooth ODBII reader and Torque app has more success but data is limited and no codes found. I don't know if it is possible to do reset on the ECM just in case there are stored codes or the new components need to be recognised by the ECU.

The injectors are good and the spark plugs tell me each cylinder is getting fuel and firing so I am wondering if the ECM command to activate each spark plug is inconsistent (hence swapping the crankshaft position sensor again). It has coil on plug so I haven't so far been able to put an indictor light on all of the plugs to see if they are intermittently not sparking. I haven't checked the Camshaft Position Sensor as the JEC magazine article by David Randall (October 2024?) said that the camshaft position sensor was only used during engine start to identify which cylinder was on compression stroke and so start the engine, thereafter the ECM relied on the crankshaft position sensor and as it starts easily.

Any ideas?

[ptjs1]

cjnsharpe124,

I'm sure it's frustrating you, particularly with all the work and parts that you've fitted.
David is correct the engine position sensor does nothing once the car has started, so don't worry about that one.

A few random thoughts:

- The lambda sensors - are they mapped correctly? It is possible in the software for them to be swapped over, ie The one for cylinders 1,2,3 being mapped to the wiring plug for cylinders 4.5.6 and vice versa.

- Lambda sensors - if you monitor their voltage trace using any cheap OBD software app such as Torque, do you see a fluctuation on one of them as the engine stumbles? That might at least help you understand which cylinder group is missing

- Air leaks when warm, emanating on the inlet side, possibly at the inlet manifold connection to the head or between the throttle body and the inlet manifold

- The vacuum pipe underneath the inlet elbow starting to lose it's seal when warm

- The TPS, is it reading exactly the correct voltage? Is it varying smoothly as revs are increased? Has the TPS been reset properly using IDS or equivalent software?

- The exhaust manifolds - prone to cracking as we know, and starting to leak as the crack opens up when hot. Have you taken off the cover and thoroughly checked them for any small cracks?

- Exhaust manifold to head gasket. Is this slightly leaking when the engine warms up?


Good luck

Paul

[cjnsharpe124]

I finally managed to get back to this. I removed the exhaust manifold cover and identified cracks in both front and rear manifolds. This is accompanied by soot staining on the inside of the cover. I can't see any sign of the gaskets leaking.

This could explain the misfire although it is slightly counter intuitive as the misfire only becomes noticeable when the engine is warm and I would have expected the cracks to close as the manifolds got hot and the metal expanded, rather than them open. As others have noted there was no obvious noise from the cracked manifolds when the engine was running. However I will have to replace them as now I know they faulty I need to eliminate them from my investigation.

Reading other forum posts indicates that this is a common problem with no known permanent solution other than fabricated stainless manifolds which are a lot of money even for a rare 4.0 litre manual car. The obvious solution to to fit some good second hand units but I am wary of replacing one thing that failed with something that is exactly the same so is likely to fail as well at some stage. I expect budget will be the deciding factor. I will have a think about it.

[ng4xjs]

I was able to obtain a rear manifold for my car from David Manners a few years ago (not cheap). I still have the original which I removed (somewhere in my shed, I think) which as it turned out only has a quite small crack (repairable).
Nick

[PaulGover]

... I would have expected the cracks to close as the manifolds got hot and the metal expanded, rather than them open ...

Think of a sheet of say rubber with a hole in it. Stretch the sheet; the holes get bigger. Another way to view it: fill the crack with a fillet of cast iron that fits without force; heating the manifold makes both it and the fillet expand, and since the expansion is uniform, the fillet still fits without force; remove the fillet to leave an expanded crack.

[JimMann]

Cast iron can be successfully welded, but it does require a level of skill that is potentially beyond the home mechanic. Try a local Blacksmith, or a welding engineering firm.

Jim Mann