[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]

Re: O2 sensor readout (Air/Fuel ratio indicator)



>   > I have done some testing with a portable 4 - gas (O2, HC, CO, CO2) 
>   ...
>   On many cars, Audis among them, full throttle is a special sensed condition
>   that disables the closed loop and provides enriched mixture for maximum power.
>   On my 4000SQ, this is a microswitch on top of the throttle body.

> I've often wondered about that -- the "sensor" (aka microswitch) feeds into
> the engine computer. Does the computer truly go "open loop", or does it in-
> stead switch its internal "target" closed-loop point? I.e., in true "open
> loop" the computer would nominally switch the Freq Valve "off" (or "on", I
> never can remember how many nested pressure actions and counter-actions are
> going on out there), although it might instead simply switch to a pseudo-
> "open loop" mode running the Freq Valve at 25% (or 75%, or ???). On the
> other hand, if it stays "closed loop", then it would simply switch the "tar-
> get" mixture setting, using the Freq Valve to maintain the desired WOT mix-
> ture, say 13:1, rather than so-called-Ideal 14.7:1.

> Anyone know fer sur?

According to "Bosche Fuel Injection and Engine Mangement" by Probst:

"when the engine is cold, the thermoswitch is closed.  The control unit 
sends s fixed, slightly-rich duty-cycle signal of 60% to the lambda control
valve. ..."

So for K-jet with lambda, in warm-up, its what you're calling "pseudo open
loop".  This on page 23 in chapter 5.

While it doesn't call out the method for enrichment on K-jet lambda systems,
it does say that for KE-jet the pressue actuator current is increased by
a fixed amount, and that open loop implies ignoring the lambda sensor input.
That's roughly the equivalent of sending a fixed duty cycle to the frequency
valve (lambda control valve).

Since lambda sensors behave as they do, you're suggestion  of holding in
a closed loop around a different control point wouldn't seem to be possible;
by the time the mixture reaches 14.5:1 the sensor is basically pegged near its
1-volt "too rich" output.  These things are really tightly centered around
stoich (14.7).

There's a zillion incarnations of the three basic version of K-jet, but 
hopefully this made some sense...

>					-RDH

Walter


--
Walter Meares		Intermetrics, Inc.	walter@inmet.inmet.com
Information Systems	733 Concord Ave		Cambridge MA 02138
(617) 661-1840