memcals

[Robert W. Hughes]

> I have a box full of knock sensors.  Would anyone care to dream up a
> simple method of telling the tuned frequency and sensitivity of each
> sensor?  A swinging weight and one of those handheld wondermeters?  I
> could probably borrow one long enough to do some testing.

You might try connecting sensors to the sound card in your computer and
recording a gentle tap at 44kc. Then run this record through one of the
FFT programs. If the sensor is tuned you will see a prominent peak but
if it is wideband you will just get a noisy flat line. The only one I
have tested was for an 86 Buick GN and it was flat, wideband. When
installed on a 2.8L V6 it showed a small signal down around 50-100 hz at
idle which was the actual engine running and a prominent about 1kc wide
hump centered at about 5600hz which was the combustion chamber signal.
Presumably this part of the signal would be much larger for a detonation
event. Incidentally, when doing the tap test, gentle is important. The
sensor can generate a surprisingly high voltage if you really hammer it.
-- 
Robert W. Hughes (Bob)
BackYard Engineering
29:40.237N, 95:28.726W or perhaps 30:55.265N, 95:20.590W
Houston, Texas "The city with too much Oxygen"
rwhughe at ev1.net
----------------------------------------------------------------------------
To unsubscribe from gmecm, send "unsubscribe gmecm" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org