When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Fighting a knock sensor code for 2 years — and still can’t beat it. P0333, S-Type 3.0
Hi!
I have a persistent issue with my car, and I’ve been struggling with it for over two years. I've done a lot of work trying to bring the car to perfect condition — many things have been fixed or replaced — but this one issue just won’t go away.
The problem:
I’m getting error code P0333, which refers to the knock sensor 2 circuit high input.
When this happens, the car goes into “restricted performance” mode and limits the RPMs to 3000.
When the error occurs
The error always shows up at cold start — if the car has been sitting for several hours. The error goes away after some time, and the duration depends on outside temperature:
In cold winter, it may take 30+ minutes;
In hot weather, it usually disappears within 10 minutes.
If I’m driving on the highway for more than ~2 hours, the error may come back, and it goes away again if I turn the engine off for some time.
The error sometimes shows up during heavy rain or after driving through water.
Occasionally, it can appear randomly, but that’s rare.
Additional notes:
The error can be reset via ELM or Jaguar IDS software, but it comes back after few seconds.
The check engine light doesn’t always illuminate when the error appears.
If I reset the code with the engine off, it disappears — but comes back within seconds after starting the engine.
The code does not appear with just ignition on — only after actual engine start.
What has been done:
Replaced the knock sensor with multiple original and compatible sensors (all tested — signal is correct);
Wiring from the knock sensor to the ECU was checked — everything is fine;
Fuel injectors cleaned;
Engine flushed with oil treatments multiple times (to remove deposits);
Brand new OEM alternator installed;
Engine mounts replaced with new OEM parts;
Tried running 95, 98, and 100 octane fuel — made no difference;
Engine ECU was re-flashed with the latest official firmware — no change;
Spark plugs replaced with correct new ones.
The engine itself runs smooth and clean. There are no signs of knocking or any other error codes. Electrical systems are working fine. Occasionally, in hot weather, I get a temporary ABS/DSC fault, but it disappears by itself and timing seems unrelated to the knock sensor error
Car: Jaguar S-Type 2003 Engine: 3.0 V6 (AJ30)
I'm extremely grateful for any help, ideas, or advice you may have.
This is literally the last remaining issue with the car — everything else is in great shape.
I’ve already done so much trying to fix it that I’m out of ideas, and I’d really appreciate any suggestions.
Did you physically check the wiring end to end out of the loom. The reason i ask because i had the exact same problem on a chevy truck. I did a resistance check and it was fine. My last resort was to isolate the wire out of the loom and found it broken. It would only act up when vibrations jiggled it. Other than that i would suggest a scan tool that can read the live knock sensor data and with a passenger monitoring it drive around until you get the failure and see if the data changed also. Just to see if there really is an issue or maybe a false positive. Seeing you have ids software maybe check flight recorder or datalogger or even use it to monitor the sensor while driving
Last edited by scottjh9; Jun 19, 2025 at 03:52 PM.
The only thing else i can think of is carbon buildup that is too much for additives to remove. A scope down the sparkplug hole may shed some light
Thanks for the suggestion! I'll give it a try to be sure
However - how does it explains why the error occurs if I drive through a reasonably deep puddle?
Did you physically check the wiring end to end out of the loom. The reason i ask because i had the exact same problem on a chevy truck. I did a resistance check and it was fine. My last resort was to isolate the wire out of the loom and found it broken. It would only act up when vibrations jiggled it. Other than that i would suggest a scan tool that can read the live knock sensor data and with a passenger monitoring it drive around until you get the failure and see if the data changed also. Just to see if there really is an issue or maybe a false positive. Seeing you have ids software maybe check flight recorder or datalogger or even use it to monitor the sensor while driving
Is the sensor head secured by a single bolt like this example?:
Image shamelessly stolen from RockAuto
If so, my first thought is the bolt could be bottoming out in a blind hole. The bolt would feel tight but is not properly clamping the sensor to the engine. The bolt could be too long, you’ve got debris in the bottom of the blind hole, or the replacement sensor is too thin.
The OBD code list has this as the first possibility:
If that’s not it, how about this oddball theory? Could the other knock sensor be confusing the poor computer? No idea how knock is monitored, but maybe the computer compares both sensors. If it sees a big difference, it flags what is deemed the bad sensor. But if the other sensor is sending an inaccurate signal, the poor computer may misinterpret the wrong one as being out of tolerance. If this doesn’t sound too far-fetched, it may be worth gambling on replacing the “good” sensor.
The reason i ask coils is that there are many threads on here that the end problem was bad or aftermarket coils that were a last resort and fixed the problem that on the outside was not even coil related. I for one am one of those. I chased a tranny issue for 3 years and new oem coils fixed it
Which knock sensor has been replaced? Between the wiring diagrams and service manual, I see an apparent disagreement with the labeling. The wiring diagram shows sensor #2 is located on the right side of the engine block (viewed facing forward), above the starter. However, in the service manual, if you search for fault code P0333, it labels the corresponding sensor as LH (not #1 nor #2) and says it is located on top of the engine block, under the big intake plenum.
I'm not sure which location is correct. Perhaps you've been messing with the wrong sensor due to bad information?
Wiring diagrams and OBD codes near the top of this page, and the service manual near the bottom:
That would be my suggestion too.
Are you sure you changed the correct knock sensor?
This brings up a point I post often usually about O2 sensors?
If your going in and replacing something that comes in pairs consider replacing both at the same time.
As the knock sensors are on both sides I think it's false economy to do all that mechanic work and then only replace the one sensor.
Think about if for a minute?
Both sensors have seen exactly the same use and history and one has now failed.
When will the other one fail?
.
.
.
Yeah. We gave him a lot to check out. I think he will find the issue and let us know. I am not familiar with the v6. I know on the 4.2 the sensors are in the valley in line with the crank
Thanks a lot for all the suggestions and information — I honestly didn’t expect this much attention to the post, and I really appreciate it. Most of the hands-on work like removing the intake manifold and checking the electrical system is being done by my auto electrician — I’m not quite handy enough to tackle that myself.
The reason I’ve been quiet these past few days is that my next visit to the shop got delayed, so I haven’t been able to follow up on some of the points you all mentioned. Plus, it was the weekend, and I figured the poor guy who’s been working on my car for weeks wouldn’t be too thrilled if I called him on a Sunday with more knock sensor questions
Sorry for the silence — I’m sure I’ll get more info this week and will update you all. Just didn’t want to flood the thread with a bunch of "Thanks, I’ll check that" messages that might clutter things up for others reading this in the future.
That would be my suggestion too.
Are you sure you changed the correct knock sensor?
This brings up a point I post often usually about O2 sensors?
If your going in and replacing something that comes in pairs consider replacing both at the same time.
As the knock sensors are on both sides I think it's false economy to do all that mechanic work and then only replace the one sensor.
Think about if for a minute?
Both sensors have seen exactly the same use and history and one has now failed.
When will the other one fail?
.
.
.
Thank you!
What I can say for sure is that we’ve definitely replaced both sensors and checked the wiring thoroughly. The top sensor (AJ85676) has actually been replaced three times already, each time with the correct torque applied — but it hasn’t made any difference.
The only thing i can think of about the puddle is maybe the hole that creates it and vibration
That could explain things, but I’ve had to drive a couple of times on a flooded — but flat — road (not close enough to cause a hydrolock), and unfortunately, the error still showed up