Re: Mixture guage

[Simon Favre <favres@domain.elided>]

James Seabolt wrote:

>I just installed one of those K&N fuel/air mixture guages and I think I
>either have it hooked up wrong or something strange is going on.  I
>hooked it into the factory oxygen sensor on my 1980 Fiat Spider.

The aftermarket O2 meters are supposed to be connected to an independant
sensor, I believe.  Read on.

>When the engine is cold, the guage reads 14.7 and stabalizes.  I would
>think that at first the guage would act erratic because the sensor is
>not hot and it should later stabalize after it heats up but it acts just
>the opposite.

The sensor can heat up a lot faster than the temp gauge in the car.
Does your car have a 3-wire sensor?  These are self-heating.  They get
hot almost as soon as you turn the car on.  The EFI system in your car
is probably doing something different to the mixture depending on engine
temperature.

>The guage has about 7 leds which ranges from something like 18.0 to
>12.0.  During a lean condition none of the lights are suppose to
>function.  During a rich condition (12.0) all of the lights come on.

This is a function of the design of the sensor box.  I would hazard a
guess that the K&N unit is based on the same LM3914N chip that most of
the others are.  It is possible to make the chip operate as a bar or dot
display with or without blanking.  Wes Ingram also sells one.  I prefer
to have either extreme leave the last LED in the chain turned on.

>After the sensor is hot, the guage "bounces" back and forth like the
>lights on an equalizer.  It's almost a perfect rhythm.  It's as if the
>sensor is detecting an extreme lean condition then an extream rich
>condition, then lean, then rich and so on.

You have to realize that a normal EFI system uses the O2 sensor like a
switch.  The output of the sensor is extremely non-linear.  Over its
normal range, it is linear enough, but the slope is pretty steep around
the ideal mixture.  What the EFI system does is let the mixture wander
back and forth across that boundary looking for the transitions at the
sensor.  By keeping the mixture excursions balanced on either side of
the transition point of the sensor, the system keeps the AVERAGE mixture
quite correct.  The behavior you are seeing is probably normal for your
system.

>Under acceleration, the guage starts to stabalize for a brief second.
>It doesn't stay on long enough for me to tell exactly but I think it's
>around 13.5 .
>
>Under deceleration, the lights go completely out for about a second or
>two.

Normally, acceleration (WOT) should result in a slightly rich mixture.
Taking your foot off the gas pedal results in a lean condition.  It
sounds like you are confirming this, but the display is tough to
interpret in a closed loop system.  It may not be possible to have the
EFI system run the car normally and still get meaningful info out of the
K&N O2 meter, because of the switching action described earlier.

Isn't this sort of meter intended to be installed on a carbureted car
for tuning purposes?