[Diy_efi] Digifant - pinning down a detail

[Mike]

At 02:56 PM 10/1/2003 +0800, you wrote:
>I gather it's because of the drastic rate of retard that occurs when
>knock is detected. Something like 3 degrees immediately, whereas the
>advance (back to "imminent" knock) is done in smaller increments
>of 0.5 degrees over a longer time. Maximum retard due to knock is
>supposedly 15 degrees.

Ok then, you are saying you 'feel' this 3degrees and thats the
problem in lower power output or does the ECU 'keep going' even
after this first 3 degrees, presumably because it still gets a
knock signal and again until its done the full 15 degree retard
and its *this* level of retard you feel by virture of the
consequential lower power output when this process has completed ?

>A small perturbation results in at least 3 degrees of retard on one
>cylinder. Because the timing on all cylinders is not allowed to
>differ by more than 6 degrees (IIRC), the event may retard other
>cylinders by a smaller amount. That's why I think the torque drops
>so sharply.

Its really hard for me with my setup to notice with any sort
of reliabilty a 5 degree retard, I am speculating you are
feeling the full 15 degree retard. Because if it really is only
the 3 degrees you mention and afterall you've got a modded
cam - I'd leave it at that and wouldnt bother going to the
trouble to design another ECU for 3 deg...

If its the full 15 degrees in fairly quick succession then
I would think the cause is a fault in the sensor, grounding,
wiring etc etc Instead of it doing it from new, hence try to
clean up the knock signal a bit first by something not
intrusive such as a good suppression cap - to try that first,
if that doesnt make any change then I'd suggest something
else - on a different thread to avoid confusion ;)

>Historically speaking, as the car has exhibited similar behaviour
>since Feb 1992 (when it was "new"), I don't see degradation as being
>the issue. If there is excessive ignition noise causing the problem,
>then it existed when the car left the factory.

Could have be a manufacturing SPC issue from day one, extreme
std dev from their SPC normal curve, tolerable but not optimum.

I guess you've had the same 'bosch amp' and same knock sensor
in there from day one ?

rgds

mike


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi