|
When I bought my current car (1986 Daihatsu Skywing, so no knock sensor or EMS) it suffered from occaisional run-on after switching off, and an intermitant sharp tap at idle which I tentatively diagnosed as pre-ignition, though I don't really know what that sounds like.
I rigged up an IV line into one of the many rubber hoses around the carb which was subject to inlet manifold vacuum, and had it sucking distilled water (actually aircon condensate, but hopefully close enough) into the inlet tract.
The line needed to be piched quite tightly to avoid drowning the engine at idle, but could be opened up at higher revs. Over a couple of days operation, with occaisional static sessions at higher revs, the run-on and tapping noise stopped, and havn't recurred.
I understand the water works by combining with the incandescent carbon to form "water gas" H20 + C > H2 + CO, which then burns.
If I had to do it again I'd leave the carb alone and introduce the water via the air cleaner.
If this IS pre-ignition you should deal with it asap, since its potentially more damaging than detonation.
With regard to your two theories, they both seem to be testable. I'd say, however, that the second one (knock sensor in the wrong hole) doesn't sound a very likely explanation, since I believe detonation causes the whole engine block to "ring".
This implies that detection wouldn't be very sensitive to precise sensor positioning, which wouldn't in any case generate a "false positive".
|
