WPSLPGExhaust Gas Oxygen (EGO) sensor Information

[Steve Schwartz]

Well hello!

I am going to jump in with my first post to this list. No Fear!

This oscillation, duty-cycle, frequency stuff is IRRELEVANT.  This
discussion lacks a clear question!  What are we trying to accomplish?  I saw
mention of WOT performance in open-loop mode. Is this what we want the EGO
sensor to tell us?  If so, indeed, we need only 2 LEDs - RICH and LEAN.
Note that this is in OPEN-LOOP, i.e. no feedback path to drive the mixture
toward stoichiometric.

If, as in the example at http://www.cnnw.net/~fourty/uses.html, we are in
CLOSED-LOOP, i.e. an operational EFI system, the oscillation is caused by
the control algorithm varying the mixture to stay as close as possible to
stoich.  The injector pulsewidth is constantly being adjusted to make the O2
sensor cross the stoich point: leaner if too rich, and richer if too lean.
Every time the mixture crosses the stoich point, the EGO will peg at max. or
min.  Of course when in closed loop, and you look at the signal with your
handheld DMM, you will get some intermediate number, based on the integral
of the signal, approx. the duty cycle of the signal shown in the above
webpage example.  This is the low-pass filter mentioned below.  This number
means NOTHING!  Nor does the frequency domain info that could be calculated
from a FFT of the signal. IRRELEVANT!

THERE IS NOTHING INTRINSIC TO THE EGO TO MAKE IT OSCILLATE, HAVE A CERTAIN
FREQUENCY OF OSCILLATION, OR HAVE A SPECIFIC DUTY CYCLE.  It just reflects
the control algorithm of the EFI's ECM or PCM, which, BTW, is NOT
'bang-bang'.  Rather, it is strictly proportional control, with two-state
sensing...

As has been pointed out, the EGO's response NEAR STOICH varies with
temperature, pressure, and construction, and NO TWO WILL BE IDENTICAL, let
alone any two installations: placement (hence, temp.) or backpressure.

In fact, it is possible to have a second EGO in addition to the one in the
closed loop system, which reads permanently RICH or LEAN, depending on where
it is installed!

A lot more useful info could be gained from the injector pulse-width:  at
least the ECM knows if it's being successful with control.  A maxed-out duty
cycle (assuming no system malfunction) might suggest it's time to move to
larger injectors, an auxiliary injector (ex.,turbo system), etc..  This
certainly has limits, but contains SOME useful info.  For true AFR, ain't no
free lunch...

So, let me ask the following:

What are we trying to do?

Best,
Steve, the FNG.





> -----Original Message-----
> From: owner-diy_efi at diy-efi.org [mailto:owner-diy_efi at diy-efi.org]On
> Behalf Of Tom Meagher
> Sent: Wednesday, August 02, 2000 10:00 PM
> To: DIY_EFI at lists.diy-efi.org
> Subject: Re: WPSLPGExhaust Gas Oxygen (EGO) sensor Information
>
>
> I found the following site helpful in visualizing the behavior of the EGO
> sensor.
>
> http://www.cnnw.net/~fourty/uses.html
>
> Oscillation duty cycle, averaged with a low pass filter to produce a DC
> voltage is evidently the operative parameter.  Perhaps you could
> sample the
> signal with the PIC and do FFT's on it to produce additional
> information to
> better characterize AFR.
>
> Does anyone know the physics behind the oscillating output?  Is the
> frequency related to engine speed, etc?  Instantaneous pulses of gases
> traveling through the exhaust pipe?  Something more subtle I suspect...
>
> Thanks,
> Tom Meagher
>
>
>
>
> ----- Original Message -----
> From: <ext-peter.betts at nokia.com>
> To: <DIY_EFI at lists.diy-efi.org>
> Sent: Wednesday, August 02, 2000 2:58 AM
> Subject: RE: WPSLPGExhaust Gas Oxygen (EGO) sensor Information
>
>
> > LONG QUESTION, but please help :-)
> >
> > >From reading the information sent by Darren (not included here) I am
> rather
> > dissapointed regarding the AFR meters.
> >
> > The opinion seems to be that the AFR meters with a 10 LED bar graph
> > representing a voltage range between 0 -> 1 V are a waste of time. One
> only
> > needs a 3LED system showing RICH, STOCH and LEAN.
> >
>
>
> ------------------------------------------------------------------
> ----------
> To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without
> the quotes)
> in the body of a message (not the subject) to majordomo at lists.diy-efi.org
>

----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org