EGOR

[Robert Harris]

Lets say you are a bullet maker specializing in exotic loads.  You get paid a
visit by some Transylvanian Villagers in need of silver bullets to slay some
blood sucker and his evil assistants.  You zip out to your hidden silver mine
in the desert, scoop up the good stuff and take it back to the shop.  Mission
accomplished.

How does this relate to a wide range EGO?   When you went back to your mine,
you didn't care EXACTLY where it was, nor were you concerned about PRECIOSLY
where you dug up the silver.  The only thing you cared about was getting back
to the silver field.

The same thing applies to using a wide range EGO for engine control.

There are only three ratios of interest:

Lean Best Power - the mixture where the maximum power is obtained per unit
mass of fuel.

Rich Best Power - the mixture where the maximum power is obtained per unit
mass of air.

Stoic.  The point where all available oxygen is combined with all available
fuel at the time and point measured.

The first two are important for maximum power and economy - the last simply to
satisfy the catalytic converter and the rectally inserted cranium clean air
bureaucrats.  But, it does lay about halfway between the two best power
points.

Stoic is determined primarily by the fuel and little else.  The others vary
considerably from engine to engine, altitude, temperature etc.  Both are
determined empirically from actually tuning the engine and optimizing the
results.  In other words - they are not magic numbers.

If you record the ego output at these two points, you have a nice means of
getting back to these points.  The actual numbers are meaningless - as long as
they are repeatable and of usable scale.

Precision is a fruitless anal exercise in futility.  At either point you can
vary the mixture a few percent and the difference will be lost in statistical
noise.

There are a number of wide range EGO systems or projects in existence.  Most
calibrate the EGO to a known output curve.  Free air calibration, trim
resisters, or a special patch to the ECU seem to be in the lead.  All of them
assume an anal need to be exact.

But suppose for a moment, that you reverted to the good pre-computer
engineering practice of good enough is good enough - get on to something else
that matters?  It is reasonable to assume that the published output curves are
close to reality, and that if one were to peg some part of the curve to a
known standard, and correct the output accordingly, the rest of the curve
would be good enough for engine control and replacement of sensors with
similar sensors as they deteriorated.

And that can be accomplished easily with little defecation complication by
simply used a standard EGO as a reference.  The standard EGO is highly
accurate, repeatable and reliable at detecting Stoic crossing.  Simply by
recording the wide range EGO actual output at each crossing, and then applying
a correction to the output to bring it to the standard curve, you can
effectively continuously calibrate the wide range EGO to well within the good
enough limits.  And it should be tolerant of reasonable sensor aging and
replacement.  Not much is needed.  A second standard EGO sensing the same
exhaust stream as the wide range EGO, some minute computing power and a
handful of silicon should do it.

Good enough self calibration for engine control and getting back to the best
power points after they are determined by other much more accurate means.

Thimk about it.

----- End of forwarded message from Robert Harris -----
----------------------------------------------------------------------------
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