Randy Wazisname <randy@taylor.wyvern.com> writes:
> $150? Geeez, make your own. Parts needed:
>
> 1 anti-oil-fouler spark plug space for an old Ford (18mm plug). About $3
> 1 Universal single wire O2 sensor. About $30
> 1 analog or bargraph volt meter with a 2 volt (or so) scale. Digital
> will work, but hard to read, and sluggish. price varies
>
> [instructions deleted...]
>
> This style O2 sensor, as used in the K&N kit and early Bosch injection
> systems, puts out right at 1 volt when way rich, and 0 volts when really
> lean. Things are pretty much linear. Calibrate your volt meter accordingly.
> These sensors have to be over 600 degrees F to function properly. Below
> about 400 degrees, they don't function at all. This is why you want it as
> close to the engine as possible, and why all testing should be done dynamicly.
> There just isn't enough exhaust flow at idle to keep it warm.
>
> Okay, that's the basics. Would one of the E.E's on the list care to cook up
> a 1 volt bargraph display using $5 of parts from your local Rat Shack? :>
We did this once in the past, I think it was James TenCate who originally
brought up the subject, and here's what I wrote, briefly describing a simple
"bar graph" indicator:
----- Begin Included Message -----
On Apr 19, 9:23am, James A. TenCate wrote:
> Answer (from Car Craft magazine Tech Talk):
> "The O2 sensor acts more like a two-frequency switch than a proportional
> sensor. In other words, think of the sensor as being like a 'go/no-go'
> gauge, rather than showing the actual measurement...
A zirconium oxide oxygen sensor *is* a proportionally responding device, with
an output quite close to 500 mV at the stoichiometric air/fuel ratio (14.7:1).
But the output of the sensor goes from a minimum of around 200 mV or so (lean)
to a maximum of 800 mV (rich) over a very narrow air/fuel ratio range. So an
oxygen sensor will normally only tell you which side of stoichiometry your
car's engine is operating on *at that moment*. Even on a perfectly tuned
engine the output of the oxygen sensor will oscillate between min and max just
due to the transient nature of driving: acceleration, coasting, hills, etc.
For this reason, the ECMs on electronic fuel-injection systems are not set up
to try to maintain stoichiometry as measured at the oxygen sensor. They work
by adjusting the amount of fuel so that the amount of time that the sensor
indicates lean running is the same as the amount of time that the sensor
indicates rich, averaged over a long time interval (minutes).
What the magazine article says is essentially right, but you *can* use a
simple meter to check for proper running condition. You just have to monitor
the voltage over an extended period of time and visually interpret the
readings to get an indication which side of correct the engine is running at.
On Jan 25, 12:51pm, James A. TenCate wrote:
> Subject: Air/fuel mix meter update 1
>
> Just got through checking out the Cyberdyne air/fuel mix meter. It
> works just fine. The bar graph has 10 lights. 100 mV lights the
> first light, 200 mV lights the second light, etc. I put 300 mV
> on the meter and got three lights. The circuit drew less than
> 0.1 microamps so it probably won't load down the oxygen sensor.
The National LM3914 is a "bar graph display driver," essentially 10
comparators that allow an input voltage level to light up a bar graph. If
done right, this little thingy could work quite well. It has a relatively
high input impedance (~100 Megohms) and can be made to respond nicely to a 0
to 1V input signal.
----- End Included Message -----
Pat Vilbrandt Fluke Corporation Everett, Washington USA
pwv@tc.fluke.COM or: { uunet, uw-beaver, sun, microsoft }!fluke!pwv
|