🚀 go-pugleaf

In article <#+8tqjm@rpi.edu> richtt@aix.rpi.edu (Timothy G. Richter) writes:
>I need to determine the output of an engine knock sensor.  As far as I
>know, the knock sensor is a piezo-electric "microphone" which "listens"
>for knock.  Knock occurs at 6.5kHz, and the knock sensor is (I'm guessing)
>a narrow-band pass filter at 6.5khz.  Using this assumption, I'd expect
>to get a 6.5kHz sine wave at the output when the sensor is exposed to knock.
>I can't get engine under load to create knock, so I'm going to try using
>a speaker.  Does anyone know how these sensors operate and what the output
>is?  This is going to be connected to a microcontroller (68HC16) to help
>with fuel injection and spark timing.
>


   I'm no expert on engine knock, so wait for the experts.  In any case,
my guess is that engine knock produces a very complicated waveform, with
many harmonics.  After all, if it were just a 6.5Khz sine wave, it 
probably wouldn't cause damage to the engine!  6.5Khz might be optimized
for some characteristic of the block.
   I would recomend getting your engine to knock a bit.  Use real cheap
gas (Sunoco 86 octane), or advance the timing a few degrees.  That's a 
simple reversible matter and it won't cause engine damage unless you 
drive that way continuously.  If you can't adjust anything, try flooring
the pedal on a cold morning while going up hill, shortly after starting
the car.  The choke should be set so you'll get a rich mixture and it
might knock.  Try taping the sound.
   Let me know about the results you get.  A properly working engine knock
sensor might have saved my last Shitty Chevy.  On the other hand, it was
a Celebrity with a Fiero 4 cylinder engine so nothing could save it...
Good luck.
							Danube