Thank you for this. So, since when I was testing the bank B sensor cable (car running, wiring connected to ECU, but unplugged from sensor) I could see 0.057V DC. Maybe this is normal. Other wire seemed to have a good ground. I will do more thorough continuity and resistance checks later today after I get off campus. Thanks Again!!!

 

What do you mean by 'all the monitors set' in terms of knock sensors?

 

A voltage of 57mV is not really worth anything as a reading, it's just background noise. It's the kind of voltage you would see if you put the two probes againsts your finger.

The signal the ECU expects from the Knock Sensor is a frequency rather than a voltage, so you would need a scope to read it (although I think some multimeters can read frequency?)

 

I don't disagree. Remember what I was measuring there was a noise on the cable coming from the ECU while being disconnected from the knock sensor. I was looking for a 5V ref. The noise just meant to me I could see a connection to the ECU.

But, I did just...after thoughtfully and carefully DISCONNECTING my battery ... Measure the resistance from pins 9 to 21 and 32 (ie through knock sensor 1 and 2 circuits). Got approximately 1 Mohm each. Same as I had for the knock sensors at the terminals. This indicates to me, the circuit isn't shorted. Maybe it changes when running, but no completely cut wire at least. So with the engine running I guess I need a scope or someone w one to look at the signal from the knock sensors. I guess I could also try to look at the Vac. Anyone know approximately the voltage the piezos kick out?

Power steering fluid was full and the alternator seemed tight. It is dark now so I didn't get to look at everything carefully. I didn't bring a torx set with me and it looks like the plug/coilpack cover is a t4 so I guess that is tomorrow.

 

Okay, so I thought of something. The original knock sensor code P0332 had freeze frame data that said the code pulled at 2079 RPMs. Maybe there is a resonance there actually causing a knock. The knock sensor harnesses seem okay with a continuous circuit through the sensors. I have tested the impedance on four different sensors and all of the read about 1 MOhm. What can cause a knock sensor code?

Fired the engine up and brought the engine up toward 2079 RPM. I am getting noise between 1800 and 2200 RPM. This sounds like it is coming from the timing chain again. Can the timing chain noise actually cause the knock codes I am getting of Circuit 1 Range and Performance, Circuit 2 Low Input Voltage? Or could it be that the engine is misfiring and causing knock codes. I sure seems that the engine performs well, but I can't seem to find anything wrong with the knock sensors and the knock sensor circuits. And I still don't seem to understand the pending B30F4 and P13B0 codes. Is this an ECU issue/Code Reader issue? These codes never seem to actually cause a check engine light on their own. But, they do pop up quickly and consistently within 20 minutes of driving after a reset.

How long does a typical jaguar xj6 take to get the evap and cat monitors set after a hard reset? And, shouldn't these set after 100 miles of driving? Maybe the B30F4 and the P13B0 codes are related to the or Evap Cat monitors.

 

I think your readings at the ECU plug do confirm that the circuits are good for the sensors. And as you say having tested 4 different sensors it's extremely unlikely that all are bad. So I'm also thinking there is an external factor involved here.

The 326 code in the Jaguar DTC guide has a slightly different definition as follows
"Knock sensing circuit 1 (cylinders 1, 2, 3) at maximum correction"

Possible Causes
Low coolant level
Poor quality fuel
Knock sensor harness wiring shield condition
(RFI interference)
Combustion chamber deposits (pre ignition)
Mechanical or background noise
ECM failure

So yes some external vibration could cause this code.


I had some time today so I did some testing on my own car on the knock sensors as it's an area I not too familiar with. Turns out my multimeter can read Frequency (and your photos suggest yours can too). So you could repeat the tests I did and see if you get similar value. Here's what I got (I have no codes and car runs fine)

All Tests are with Knock Sensor in place on engine block and connected to ECU. Readings taken by back-probing the connector (Bank 1 Sensor)

Blue is Ground at all times (even with ignition off)

Blue to Red (Ign Off)
V - 0.1mV
Ohms - 0.989 MOhm
Frequency - 0

Blue to Red (Ign On, Engine Off)
V - 0.4mV
Ohms - 0.989 MOhm
Frequency - 274.5 Hz

Blue to Red (Ign On, Engine On - Idle)
Frequency - 450-600 Hz (with occasional jumps higher)

Frequency increases as RPM increases. I took these Approx readings as the Frequency jumps around a lot.
740 RPM - 600 Hz
1150 RPM - 900 Hz
1500 RPM - 2.1 KHz
1900 RPM - 6.8 KHz
2100 RPM - 8.5 KHz


So it would be worth doing the above test and see if you get a dramatically different figure around 2100 RPM. If you do, and if you could repeat that with one of your replacement sensors, then I think that would pretty much confirm that the Knock Sensors are working OK and there is an external vibration cause or a failing ECU.

BTW, in your first photo Mileage on dash shows as 111111. Is that just coincidence or is it stuck like that?

 

If you hold a long screw driver from where the upper tensioner bolts to head, to your ear, Do you get the same noise but louder then at either sensor? Does your upper tensioner have a arrow that points up in the cap? What happens if you strap and insulate your spare sensor to someplace else say the intake manifold? But not physically touching the engine.

Let us know.

Thank you.
Larry Louton

 

Thank you sincerely for doing this. Here are my numbers to the right of yours. As you can see they are significantly different. But... I don't know what if means to have the dvm measure frequency. Is it monitoring the V vs t profile and fitting. Anyway, it will let the numbers speak for themselves.

Blue to Red (Ign Off)
V - 0.1mV. 0.0V
Ohms - 0.989 MOhm. 1.010 MOhm
Frequency - 0. 0.0Hz

Blue to Red (Ign On, Engine Off)
V - 0.4mV. 0.0V
Ohms - 0.989 MOhm. 1.011 MOhm
Frequency - 274.5 Hz. 900 Hz

Blue to Red (Ign On, Engine On - Idle)
Frequency - 450-600 Hz (with occasional jumps higher). 3.3 kHz

Frequency increases as RPM increases. I took these Approx readings as the Frequency jumps around a lot.
740 RPM - 600 Hz. 3.3 kHz
1150 RPM - 900 Hz. 12.5 kHz
1500 RPM - 2.1 KHz 17.7 kHz
1900 RPM - 6.8 KHz. 25.5 kHz
2100 RPM - 8.5 KHz. 26.5 kHz

 

Wow. That's very different. It's possible that this sensor is bad (but I think unlikely). If you can, swap the sensor for one of your spares and repeat the test. If you get the same result then it's nailed on that the sensors are good (and I think the ECU is good) and that you actually have a significant problem with a noise/vibration from somewhere in the engine. Check out Larry's good suggestions as a possible next step.

I am slightly puzzled by your reading with engine off which I would expect to be the same as mine if it was actually an engine noise/vibration issue. Did you take this reading after the engine settled? I found that things like the fuel pump running, IACV setting or whatever is going on on first key turn caused higher readings, then it settled to a static 274.5 after a few seconds.


Your numbers make sense in terms of the 326 code. According to the Electrical Guide, allowable values for Frequency are between 0Hz and 20 KHz (the guide actually says 20Hz max but that's clearly a mistake). So you're maxing out around 2000 RPM as per the Freeze Frame data.

 

Thanks for the good suggestions Larry. Yes, the upper tensioner has an arrow that points up. I am almost sure the noise at 2000 rpm is chain slap and yes, it is noisier at the upper chain tensioner. Do you know how I know if the noise is coming from the upper chain tensioner or the lower chain? I am tempted to buy the newer style tensioner, but they are $125, and my purchase of oxygen sensors and knock sensors hasn't solved anything. The noise is restricted to occur between about 1800 rpm and 2200 rpm. Here is a video of the noise that I made yesterday night.


I haven't tried to strap the spare sensor to the intake yet. I am curious to do something like that.

The frequencies I had on my dvm did go up with the fuel pump, but the 3000hz was after calming down at idle. The engine runs smoothly and nicely other than the 2000hz noise. Also, the frequencies of 25khz are really high. Humans can only hear up to 20 and even 10khz is a really high pitch. But again, I don't know exactly what the DVM is measuring and so I don't put too much stock in these numbers yet. I could borrow an oscilloscope from the computer lab, but they run on AC and I don't have a place to work with AC. I do have someone with a snapon scanner who said he would take a look this weekend.

I think I am going to change the plugs etc tomorrow and pull out the top timing tensioner again. I have to do some more reading.

 

I wanted to come back to this. The did have a low overflow coolant level. I did too it up yesterday and I did notice the old owner had three small water bottles in the trunk. Possibly coolant leak. But why would low coolant levels cause the predetonation timing to be at Max correction. Noise from water pump or hot temps causing carbon buildup in cylinders to preignite fuel before spark? Anyway, it is worth mentioning that I added water ( maybe 1.5L) to top up the coolant reservoir.

I think fuel quality should be pretty good by now. On my third tank of premium.

Knock sensors have original shielding on harnesses so I doubt much interference, but I guess I could wrap in Al foil.

Combustion chamber deposits, I just don't know. I don't have a garage to work in so I want to avoid pulling the head.

Mechanical or background noise. Matbe the timing chain thing could really cause vibrations picked up by the knock sensors and interfering with what is "expected." There was however no real difference that showed up w the DVM on frequency measurements. I think I need to get a scope.

Also, the knock code isn't currently showing and a real hinderance to tests is it takes about 80 miles of driving to have it show.

The evap and cat monitors never seem to set. How long should they take if all was perfect with the car? Current status is the pending P13B0 and B30F4. No knock codes currently.

 

I guess we might all have been guilty of overlooking the obvious, ie that the knock sensors were actually doing their job correctly, and this does strike me as a fertile ground for investigation.
Chain slap at the Rev range you mention is not uncommon, even with new chains and tensioners, but it could be a question of degree. Difficult to tell from your video. Have you performed an oil change? The reason I ask is that the viscosity of the oil affects the effectiveness of the tensioners. There would be no harm and not too much expense in changing the oil to ensure it is fresh and the correct grade. Thereafter, it might be worth removing the top tensioner to check that it doesn’t have any blockage in the oilways and generally looks ok. The lower tensioner requires removal of the timing cover and is much more involved, so I would leave that for now, and if you are going to look at it, it would then be worth replacing chains and tensioners etc while you have it off.
As you have spare knock sensors, I quite like the idea of trying to run the engine with them connected to the loom but free from the engine block.
All fluids should be at their correct levels, and I wouldn’t ignore the transmission, and if you haven’t already, I would freshen the plugs.

 

There is another document that would be useful for some further reading. It's an "Electrical Diagnostic" manual for the Engine Management system.

There is some more info on the Knock sensor codes, but also look at the "Drive Cycle" section which describes what should be done in order for all checks to be made and codes set if required. It might save you some time and fuel.

It's available in the very hard to use stickies. Let me guide you. Here's a copy of the link
https://www.mediafire.com/?pp7qlydtsp6sx

After extracting the RAR file, look in the "Content" folder. There is a file called "pewyekd8pdf.pdf". That's the one you need.

 

Is the knock sensor harness threaded in and out of the inlet manifold and over long?
I know noting, but remember XJRengineer (worked on design of AJ16) talking about the knock sensor harness length being very important (for correct function) but (if you have had to unthread them form the inlet manifold to remove it) the temptation to shorten it, to make iinlet manifold removal easier the next time is strong. This might explain the difference frequency reading when the engine is off?

 

The one thing you would not want to do to make inlet manifold removal easier is to shorten any of the wires which are in the loom which passed through between its branches. It is a total PITA to reconnect them in the supercharged cars, and being shorter would make it worse. Thread drift, but worth clearing up!

 

Hi Everyone,
Sorry for the slow response this weekend. I worked a lot and didn't have much time to spend on the car. The car has a new filter and the recommended 5W50 oil in the engine. I haven't reset the computer since pulling the harness to check the frequency and the resistance of the knock sensor circuits. I have put over 100 miles on it since then and the CAT and EVAP monitors are still not set. Interestingly the knock sensor codes haven't come back yet. Eventually (20 miles after reset) or so the pending P13B0 and B30F4 codes came up on my innova 3160 scanner. Unfortunately, these are pending codes and so they disappear as soon as I turn the car off. I had originally thought that Autozone read the same four codes, but now I realize that Autozone's scanner only pulled the knock sensor codes for me. That makes me think the weird format of these codes are a problem with my Innova scanner (I have 2 different ones and both read the same P13B0 and B30F4). I also think that these codes are probably the thing keeping the CAT and EVAP monitors from setting. So what have I done?

Embarrassingly, I have focused so much on the knock codes that I haven't yet even changed the plugs. Yesterday I pulled the coil cover and when I pulled the coils off the plugs found plenty of oil. Plug #1 was entirely covered by oil so that I couldn't even see the top of the plug. Plug #2 and six also had some oil in the plug wells. I cleaned the oil out as best I could and pulled all the plugs and installed new ones. Drove another 40 miles last night. P13B0 and B30F4 came back up after 20 miles or so. I just ordered a valve cover gasket and spark plug seals from ebay. (https://www.ebay.com/itm/Jaguar-95-9...AAAOSw0QFXDQeE).

I think that I should also probably order a timing chain tensioner. Now that I know the noise is there I can easily here it when I am driving and accelerating at around 2000 RPMs. Is there anyway to know whether it is the upper tensioner which is easy to replace or the lower tensioner which requires a lot of work to do?

My thoughts:
Regarding the knock sensors I am not sure what is setting them off. I am not an expert, but I don't think that extended lengths on the wiring harness should matter too much. If they were passing much current this could affect resistance and possibly impedance in the frequency spectrum, but the resistances of my knock sensor circuits are the same at the ECU as they are at the knock sensor terminals. I guess extra length could cause more noise if shielding isn't good. According to B1mcp's post above the P0326 code said that the engine was at maximum correction. I think this means that the engine was retarding the spark to the ECU limit. Maybe carbon buildup in the cylinders is causing a glow and preignition? Is the oil in the spark wells affecting the spark? The P0332 code I don't have details for. But since the knock sensor codes hasn't come back (yet) I plan to focus on the pending codes until it does.

Another thing I noticed is that coil #1 is a new an aftermarket part (it doesn't match 2 through 6). The car runs really smoothly though. I could try to pull coils one by one while the engine is running and make sure that each causes the engine to change RPMs. This is how I have checked coils in the past. Should I do this? Would it cause harm? Could I damage the coils?

It seems like what is really necessary is to get a different code reader to look at these pending codes before I tear into a fairly big job like changing both tensioners and timing chain belts, though I have seen that this can be done without pulling the head.

Thanks again for all of your comments and help. I will go through them and try to download the suggested materials. Thank you. Thank you. Thank you.
Nathan

 

What ever you do soak up the oil first before removing spark plug. It's called hydro lock. Did you try unplugging your spark knock sensors? Installing your spare sensors to the harness and securing to intake manifold in such a way so as not to touch intake or block. if it can't hear the noise then it would not be a fault of the knock sensors. The chain tensioner is closest to the front sensor which is the one causing the predominant DTC flag I presume?

Thank you.

Larry Louton

 

Just to be clear, a long cable for knock sensors is GOOD. Only bad if it’s been shortened in the past!

 

I did soak up the oil before removing the plug. Fluids aren't very compressible so yes, I was concerned about having it leak into the cylinder. I imagine this (hydro lock-though I haven't heard the term) could cause damage to the engine. I haven't yet had the chance to remove the knock sensor and attach them (in isolation) elsewhere. I can try soon. The trouble with testing different fixes with the knock sensor is that it takes a lot of driving (100+) miles to get the codes to come back. The car now has about 600 more miles on it since my first post and it isn't legal for me to be driving it. As soon as the knock codes come back (P0326 and P0332) I can do a hard reset and try the sensors in isolation. With no vibration signals though, wouldn't the whole purpose of the knock sensors (used to time the ignition advance) be gone?

 

I did have the top tensioner out before and my removing it and cleaning it seemed to make the noise chain slap noise go away. I think I will do this again tonight now that I have changed the oil and I am sure that I have 5W50 in the engine. I still have the "old style" tensioner in the top I think.