1

My vehicle has a manual transmission, and occasionally I'll accidentally lug the engine briefly, when shifting into 2nd gear a little too soon after taking off. When I do this, I hear a sound (seemingly from the passenger side of the engine) that sounds a lot like marbles being shaken inside a metal can.

As soon as I hear this, I immediately take my foot off the gas and down-shift back to first. But, isn't this the sort of thing that the knock sensors are supposed to detect? If so, why doesn't my check-engine light come on? (I have a bluetooth OBDII scanner, and there are no codes.)

UPDATE: I forgot to mention, my knock sensors sit in the intake valley of the engine. I've tried fooling the engine into thinking there is knock by hitting the intake manifold with a wrench, but nothing happened (no CEL, no change in engine noise indicating change of timing.)

Engine is a Toyota 5VZ.

1

Knocking will not necessarily cause a CEL. The knock sensor is there to inform the ECU that the engine is knocking, and then the ECU will deal with the knocking. Knocking will on trip the CEL if the ECU can't deal with the knocking.

As for your fooling the ECU, the type of knocking you are trying to induce with a wrench will likely not cause a CEL. Knock sensors and the ECU can tell the difference between engine legit engine knocking, weird vibrations in the engine bay, road noise, loud thrash metal being played and your wrench. They don't just listen for knocking of any sorts, they generally listen for:

  • a specific type of knocking (it terms of sound frequency and direction)
  • a specific timing (on the compression stroke)
  • a specific rhythm (engines will knock once in a while, and that's ok, but if it's knocking on every single compression stroke, that's a problem the ECU needs to take care of, and if it can't take care of it, the CEL will come on.)

Given the engine RPMs (even at idle), it is almost impossible for you to cause knocking sounds with a wrench on a modern engine that will trick the ECU into thinking it's actually the engine that is knocking. The engine is doing 1000 rpms, your arm can maybe do 120 rpms.

  • After thinking about it more, I think I mis-remembered some of the instructions I read for testing the knock sensor. I think I should have measured the output of the sensor for AC voltage while tapping on the engine, rather than tapping while the engine is running (and knock sensors connected.) – Ryan V. Bissell May 30 '17 at 17:17
  • Also: I believe that tapping the engine block will yield a similar-enough noise to that of pre-detonation, because in both cases you are ringing the engine like a bell. – Ryan V. Bissell May 30 '17 at 17:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.