[Diy_efi] Bernd: UEGO sensor question

[bcroe at juno.com]

On Sun, 14 Jul 2002 09:41:45 +0800 (WST) Bernd Felsche
<bernie at innovative.iinet.net.au> writes:
> bcroe at juno.com tapped away at the keyboard with:
> 
> > The Vs output is like a battery, a very low current one.  If 
> > it were a variable resistance, the pull up or pull down 
> > resistor value would be critical (which it is not) like the 
> > pullup on a water temp thermistor.  
> 
> That makes sense... I've been trying to model it as a voltage source
> at the core and it doesn't hang together in terms of response under
> various scenarios.
> 
> If the cell is modelled as a small current source (variable, as it
> depends on the ratio of partial pressures of oxygen) and a series
> resistor (temperature dependent) then that explains the voltage
> behaviour under "load".

I haven't specifically measured this.  But a resistor in series with 
a current source makes no sense (a current source is infinite 
impedance already).  Putting the resistor in parallel is OK, but 
that is also exactly equivalent to a voltage source with the same 
resistor in series.

> Add a small capacitor in parallel with the current source and one
> can approximate the behaviour when the cell is "exercised" by
> external, possibly-alternating voltages.
> 
> A physical model, to perhaps use as a "simulator", may be more
> difficult; the amount of current generated is miniscule; in the
> range of low micro- if not nano-Amps.

That current needs to be in the 10s of microamps for the circuits 
we have been using.  A sensor model is difficult in that it varies 
over the operating range.  I have settled for finding a stable PID 
loop at the more difficult points, then backing off speed 20% for 
sensor variations.  I have used a 2 HZ square wave to shift Ip 
for tuning the loop.  This is not as ideal as a "square wave" gas 
content, but far easier to do and apparently adequate.

Bruce Roe

> /"\ Bernd Felsche

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