Konck sensor signal monitoring in a different way

Hi Guys,

I have a friend with a Subaru EJ257 engine which was recently rebuilt, and unfortunately, the mechanic and the tuner are different, so there are some fingerpointing at every step of the story.

The car has a Pectel ECU and the tuner recently did a quick checkup and says there is a problem with cyl#3 and the mechanic needs to fix it. He doesn't say what the problem is, but he bases his statement on the knock monitoring of the ECU. More precisely the frequency of the knock sensor signal at idle for all the cylinders. You can see a screenshot of the realtime monitor from the ECU and as you can see cyl#3 has a little higher frequency signal then the other three cylinders.

Now, me personally never saw a knock sensor signal used like that, but the mechanic, who is also a friend of mine is in trouble and has not much clue what to look for. He did a leakdown test of the engine to start with, and that was obviously perfect as the car just completed its first 500ish miles to break in.

Do you have any idea/suggestion on what the suspected problem can be or what to check? Or how the knock sensor signal frequency is impacted by any potential problem?

Thanks,

Béla

Signals of the sensor will differ on each cylinder depending on its mounting position, the closer to a cylinder the higher the amplitude.

Personally, I would not use the idle noise level, as described in your scenario, to diagnose a mechanical problem. IF number 3 cylinder in fact did have a piston rocking / bearing problem, it would be severe and one would hear it clearly by listening by ear, it would mean that there is severest damage.

If the motor is mechanically sound (as it seems to be in your case), proceed with harmonizing all signal levels by adjusting the gain values for all cylinders during a ramp run with ignition timing set to a value known to not cause detonation. This way the knock control system will show the normal mechanical noise and the knock warning threshold will be set slightly above that.

If a ramp run cannot be done at this point in time, I suggest to record the base noise level by free revving the motor to get some idea of the normal noise pattern until you get the opportunity to do a ramp run on the dyno or whilst driving on the road.