I decided to create a new thread with a more appropriate title.
I previously created a thread (184110) asking about measuring 2-pin ignition coils.

I'll describe the issue again.
The car concerned is a XK8 convertible 1997. The issue is that the engine is not idling very nice.

I have verified a number of things to have them excluded as cause.

I bought a simple but reliable ODB2 bluetooth interface and also purchased torque pro (thanks to member fmertz for the advise).

With torque pro I could see that the LTFT are negative for both banks, at idle they stay at -15.6. The STFT switch around 0, in graphing mode I can see the voltage of both oxygen sensors swing between 0.1(lean) and 0.8 (richs) volts. The ECM is able to keep the fuel trims in the optimum range for the catalytic convertors.

When driving the car the LTFT go up a little to arround -13%.

I verified the fuel delivery during WOT, and they go richs for both banks, the oxygen sensor voltages keep above 0.8 volts for a long time during the WOT.This is proof I think of sufficient fuel delivery.

I also verified the fuel pressure on the fuel rail with a fuel pressure gauge.
According to a TSB I found on jagrepair.com I tested the pressure values.
http://www.jagrepair.com/images/TSB/...Flowcharts.pdf

During priming the fuel pressure gets to about 3 bar (very fast) and settles at 2.6 bar just as mentioned in the TSB. I also verified the fuel pressure regulator by applying a vacuum of 600 millibar, then the fuel pressure rises to a little over 3 bar (also within spec). Also tested the fuel pressure regulator diafram, the vacuum stays at 600 millibar after 30 minutes, so no leak.

The fuel pump is fine according to the findings above when testing the fuel pressure.

The next thing I tested was the MAF values. I found a picture in another thread (which I cannot find anymore) and saved it.

The values according to that table for a 4.0 NA engine at full operating temp:


The values I found:

Park,idle : 6.0 gm/s
Park, 2500 rpm : 17.0 gm/s
Drive, idle : 6.4 gm/s
Drive, 1500 rpm : 33.0 gm/s

The question I have is: for which engine/MAF are these values valid.

The engine in my car is a AJ26 with a LJA1620AE MAF sensor.

 

You could simply have asked a moderator to change the original thread title for you.

Graham

 

The immediate thought is that your air flow meter is bad because you are showing negative fuel trims. It is as if the air flow meter is measuring high, so the computer adjusts the fuel for too much air, and you end up with a rich mixture. The upstream wideband O2 sensors measure the rich condition and reduce the fuel.

FWIW, the intake air temp is part of the same sensor, and the value is available over OBDII. Check the value. If off, that is another indication this part is bad. You can try and clean the sensor with a dedicated spray, but don't expect too much.

If original, this part is over 20 years old. Check to see if you can locate a cheap replacement on eBay or Amazon. Price for these has come down a lot. I had good luck shopping using the DENSO number, lots of cars use the same part.

Best of luck, keep us posted.

 

That was also my thought as an explanation.
The MAF sensor however is new and original.

 

Can you expand on that? Do you mean sourced from a local Jag dealer, and confirmed suitable for an early XK8? It seems to me that even if new, this is the wrong part. These parts all look the same, and physically fit just fine, but seem to provide different readings. I also ordered one such sensors, supposedly compatible, but with the wrong DENSO part number. It "worked" but my fuel trims were way up from the original bad part, so I returned it. Only a sensor with the right DENSO# worked fine...

 

I bought it at my local jaguar specialist. I believe he got it from SNG-Barratt.
It is an original denso unit.

It looks like this:


A picture of the actual unit in my car,it was taken in the dark so not really sharp:

 

I have some new findings.

I disconnected the MAF, and did a hard reset.
Started the car and monitored what happened.

As the MAF is not connected the ECM will use a default map.
You will get a "restricted performance" message which will disappear after reconnecting the MAF and a hard reset.

During the first minutes the engine runs very lumpy. I let the engine get to operating temperature before looking at anything.

During the period the MAF sensor was disconnected the LTFT stayed at 0 (zero).
I guess the ECM knows something is wrong and it does not make sense to learn and keep settings for the LTFT.

I found something interesting. At operating temperature and idle the oxygen sensors don’t switch, they stay in the lean position.
When revving the engine to about 2000 rpm the oxygen sensors start switching. They go back to lean immediately at idle.
When the oxygen sensors are not activated, the ECM keeps the STFT at +19,5%.
When the oxygen sensors are reporting the STFT go down.
This all makes sense and expected behavior I would think.

I think the fact that the oxygen sensors are not activated at engine operating temperature is suspicious.

I disconnected the pipe from the fender to the airbox, and used a propane torch to feed propane into the intake.
After some fiddling with the amount of propane, the oxygen sensors could be activated and switch lean/rich.
The STFT went down to about +10%, the engine was running quit acceptable after leting it settle for a while.

I think the question now is: should the oxygen sensors get into operating mode at any time when the engine is at operating temperature.

If so I think I should replace the oxygen sensors.
I did mention it in the other thread: the oxygen sensors are almost new but not original (unknown brand).


BTW after reconnecting the MAF and a hard reset the MAF reported 5.2 gm/s at idle (P). Very near the value is should have according to the table in first post.

 

Hmm, not familiar with that air flow meter. It is older than what I have seen. Pricey, too!

Do you still have the old one? If so, try it again and check the fuel trims now that you have the OBD reader.

Have you checked what it looks like? What about the air filter and air box? I am sure you have checked already...

Also, double check the connector and wires. Can't hurt to spray contact cleaner.

How about your intake air temp? On my car, it is part of the same part. The value should be available over OBD. The value should be around ambient. Also, check the manifold pressure sensor value over OBD as well.

Speaking in broad brush, the O2 sensor circuits are heavily monitored, both on the signal and heater sides, so if they go bad, you should have codes.

Best of luck, keep us posted.

 

Bear in mind that O2 sensors have to be very hot and at idle may not be hot enough. (These are non-wideband I believe.)

So if measuring how often they switch you need them very hot...

 

The actions with disconnecting and resetting definitely triggered the ECM to relearn fuel strategy.

I made a nice drive into Germany to do some shopping. I monitored the fuel trims during this journey.
The total fuel trims where well within the normal range. So it did not go to the negative values 15.6%.

On both banks the LTFT still go negative. Bank 1 keeps between -3.1% and -3,9%. During driving bank 2 also keeps in that range.
But when coasting and stopping (idling) the LTFT goes to -6.2%. There is more change in the LTFT in bank 2.

I will have to focus on that (the driver, LHD) side.

Next action will be if the oxygen sensors are connected in the correct way.