o2 meter

[Orin Eman]

> I just built the 10 LED O2 meter today (
> http://www.students.tut.fi/~eppu/dev/EGO-bar.html ), and it worked great
> on the testbench. There was a note in the instructions that I should
> ground it only at the o2 sensor. What is the best way to ground it
> there? Do I attempt to solder a lead right onto the sensor? Probably not
> since the solder would just melt... Should I try to get one of those
> rings that the o2 sensor would screw through and crimp the ground? My
> headers are Jethot coated also, so I don't know how good of a ground it
> would be. Thanks in advance for any help :)

You want to get as close as is reasonable.  The idea is to have a
low resistance path between the circuit and the O2 sensor.  You can
find a voltage drop of 0.2 volts or so between other grounding points
and the exhaust manifold.

My car's ECU grounds to the intake manifold and I seem to remember
there was a .18V difference between there and the ECU itself.

So, you can either find somewhere that doesn't have any voltage
drop or add an op-amp circuit to compensate.

I grounded my meter at the ECU which means the scale is displaced
somewhat.  Unfortunately, if the true O2 voltage is 0V, the drop
in the wiring causes it to read -0.2V relative to the ECU ground.
It doesn't actually matter too much since the change in the O2
voltage means a lot more than what it actually is... ie
shifting back and forth between rich and lean is good and means
the ECU is operating closed loop.  If all my LEDs go out,
it means fuel is cut off (deceleration).  If one of the 'richer'
LEDs is on, it usually means that the ECU is running open loop
due to WOT or boost conditions.  If it wasn't showing rich under
WOT or boost (actually > about 1.2 bar absolute) I would worry.

Orin.