Wideband O2 control

Bernd Felsche bernie at perth.dialix.com.au
Sat Apr 1 01:04:39 GMT 2000


Garfield Willis writes:

>(2) Q: what kind of interesting controller phenom do you get into if
>you're actually building your own controller from scratch, have access
>to the code, and want to equip it to run closed-loop WOT via wideband O2
>sensing.

Seeing that the O2 is customarily used as an "error" device in the
feedback loop, one could use the error output to adapt fuelling to
changes in the engine itself. e.g. You don't have to manually
re-program the operating maps for increments (one hopes) in airflow
and fuelling. An extrapolation of that is that the ECU will be able
to tell you if there was an improvement - it's an adjunct to a
road-dynamometer function.

With closed-loop control, we can also ensure that we're getting
maximum power when required. At present, the fuelling assumes a
certain engine condition and most likely, the actuall fuelling is
too lean or too rich for maximum power.  We can use lean-burn in a
controlled manner (e.g. during cruise) to reduce fuel consumption.

On the emissions side; knowing the amount of air pumped, we can
determine the amount of "penalty" accumulated in an emissions cycle.
So the engineer can not only control the amount of emissions outside
of the cat. conversion window, but also take subsequent measures to
ensure that the engine operates within the permissible limits for
the entire cycle... it may be considered "cheating" by some, but
then so is choosing appropriate gear ratios!

It's not like we're deliberately recognizing the emission cycle and
operating in a compliance mode.

"Ideal" AFRs are well known; for performance, emission (usually in
conjunction with cat. convertors, EGR, secondary air pump, etc) and
fuel consumption. A completely-different sort of ECU, using for
example neural networks (NN) would make it easier to adapt to any
engine given just a tiny subset of operating parameters. The ECU's
NN then alters the controllables to achieve the required goal(s).
I understand that research is already under way with that approach.

-- 
Real Name: Bernd Felsche
    Email: nospam.bernie at perth.DIALix.com.au
     http://www.perth.dialix.com.au/~bernie - Private HP
----------------------------------------------------------------------------
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




More information about the Diy_efi mailing list