cheap EGA

[Brad Sheridan]

>Ed, this topic comes up periodically.  I routinely
>slam those little devices because they're notoriously
>inaccurate.  But, OTOH, they are quick and do yield
>some data -- albeit it questionable.  They should
>be accurate at stoich -- they absolutely are NOT off
>of it -- but you can get a seat-of-the-pants feel
>for what's going on, even if the data is inaccurate
>(i.e. it may not be 16:1, but it IS lean!!).  So,
>my amended opinion -- it's vaporware that can give
>useful information if you know that that's what it
>is -- i.e. know its limitations.

  I've felt they are of very limited use as well, but someone pointed out
something to me recently that sounds reasonable. Although 02 sensor output
voltage doesn't correlate with a specific AFR, the output voltage is
something that can be made repeatable, ie. if you get best power with .8V
output at WOT one day, trying to tune the car close to that value the next
time will give give good power, assuming you haven't changed any hardware
on the car. Sounds plausible, I've seen other people mention doing
something like this.

>
>You can make your own too.  RatShack used to have
>LED bargraphs, and Mouser, Digi-Key, et al do have
>them, that take zero to one volt or so as an input.
>All you'd have to do would be to use one of them
>would be an op-amp buffer (to achieve the needed
>high impedance required by the EGO) and a voltage
>divider or gain/attenuator circuit to drive the
>LED bar-graph.  Total cost including EGO should
>be less than $40.

Use an LM3914. Its has input buffering and is perfect for this app. Most of
the schematics I've seen on the web use this. Between this the 10 LED bar
graph, and a few resistors, $10 is about what it should cost. I've built
one, even made a circuit board for it. Its a very simple circuit, but of
very limited use. I've heard that many of the $100+ O2 sensor monitors use
the LM3914.

Brad