LO2/RO2 (Left O2 Sensor/Right O2 Sensor):

In the single-cat 93 to 95's, there is an oxygen sensor located at the base of the driver's side exhaust manifold, and another just below the passenger side exhaust manifold, in the y-pipe branch. These sensors generate a small voltage, proportional to the difference between the amount of oxygen in the exhaust gasses, and the amount of oxygen normally occurring in the atmospheric air. If the combustion of the air/fuel mixture is chemically "perfect", it is called "stoichiometric". It has a composition of 14.7 pounds of air per pound of fuel, or a 14.7:1 air/fuel ratio, and the O2 sensor will generate a signal level of 0.450 volts (or 450 mV = millivolts).

When you first hook up your ScanMaster, and start your car, you will be convinced either the ScanMaster is broken, or your car is "possessed". Don't worry…… the unit is SUPPOSED to be flashing constantly changing O2 sensor values. The sensor is extremely sensitive at an A/F ratio of 14.7:1. Go very slightly richer than that (14.6:1) and the voltage can soar to 700mV. Go very slightly leaner than 14.7:1 (say 14.8:1) and the sensor output voltage plummets to 200mV. Since the PCM is constantly juggling the fuel up and down to keep it right at 14.7:1, the voltage is going to jump up and down too. So, you will see that, since the PCM updates it's calculations nearly 10 times per second, the O2 sensor voltages are going to be cycling rapidly back and forth between 100 and 800mV, and you might even see values as low as 050 mV, and as high as 950 mV. THIS IS NORMAL.

As described above, when you go WOT, the O2 sensors are no longer part of the control "loop", but they still report the A/F ratio by way of the voltage they generate. Note however, that this is NOT the primary function of the sensors, and they were never intended to be used this way. Because of this, they are NOT necessarily accurate at the kind of A/F ratios you see, or want at WOT. But, they can tell you ROUGHLY whether you are running rich (12:5 or richer) or lean (13.5:1 or leaner). And, by watching the voltages before and after a modification, you can measure whether the mod made the car run richer or leaner.

The output voltage of the O2 sensors is very "flat" under rich conditions… a very large change in A/F ratio only results in a very small change in mV's. And, under these rich operating conditions, the sensors become extremely sensitive to operating temperature. A change in operating temperature can alter the mV's as much as an actual change in A/F ratio - so be careful with the O2 sensor readings. They are only a "guide" at best.

A caution - if you have "reset" the computer by pulling the PCM-BAT fuse, or a cable off the battery, you should only look at the O2 sensor readings after you have driven for a few days, in order for the long term correction values to stabilize.

My ROUGH guidelines for evaluating the WOT O2 readings, as recorded on the ScanMaster:

Be careful. I have seen sudden transitions screw up the recorded values, and give you the max possible value of 999mV. This will usually be accompanied by a corresponding MPH value of "00".
The readings should be no more than 30mV apart. It is normal for the right and left volts to differ, but you can't see this on the ScanMaster default display anyway, since it RECORDS the average value of left and right, and MONITORS only the left sensor for continuous display. But if you use a Diacom or other recording software, the best readings will differ left to right by 5 or 10mV, and 20 to 30mV is not unusual. Anything over that is suspect. Either one sensor (or even both) is screwed up, or the engine has a problem that only shows up on one side of the motor.
My car ran best with Os readings in the high 800's… say 870-890mV. Keith at ws6.com seems to prefer readings as low as 820mV. Others like low 900's. All of this is for a normally aspirated (no supercharger or nitrous) motor. When you get into a major power adder, you want a richer mixture, and readings in the range of 925-950mV may be desirable for these motors.
If you ever run your car on a chassis dyno, and they hook up a wide-band O2 sensor for tuning, try to find out how your OEM sensor mV's compare to the wide-band A/F ratio. This will give you a baseline to tell you how "accurate" your OEM sensors are, and help you interpret future changes.
Remember, the O2 sensor can only measure OXYGEN. It can't measure fuel, or other components of the exhaust, so if you have a cylinder that misfires, there will be all the oxygen that was in that cylinder, and all the fuel that was in that cylinder, in the exhaust. The sensor will see that oxygen, and the voltage will drop, indicating "lean", but it is not really lean.

Same with an exhaust leak. If you have an exhaust leak before the sensor, the exhaust gasses can actually "suck" ("educt" is the technical term) air in through the leak, and the sensor will report the mixture as too lean, the PCM will add more fuel, and most of the fuel will be wasted…. Maybe just a little burned in the hot exhaust gas, reducing the oxygen. An exhaust gas leak can really mess up your computer's calculations, and cause the gas mileage to plummet.

At cold start, the O2 sensors are cold, and will not work. They have to heat up to 600oF before they work accurately. When you start your car, the ScanMaster will show about 450-500mV. As the car heats up, over a 2 or 3 minute period, the readings will gradually drop, going below 100mV. Then they will suddenly start to fluctuate, in the 100 to 800mV range, indicating the car is operating in closed loop.

A word of caution here - I hear a lot of people asking if they can just measure their O2 sensor voltages with a voltmeter. Everything I have read says the answer to this is probably "NO". First, the O2 sensor generates a small current. Many VOM's also apply a tiny current to the circuit. This current can damage the O2 sensors. The caution in the Howell Engine Development Manual states:

"If measuring sensor voltage with a voltmeter, make sure the impedance of the meter is at least 50megohms. And, the resulting voltage measurements can still be off by 200 - 300 mV's. "

I have never tried to use a voltmeter on an O2 sensor, so I don’t know how well it would work, but based on the above, it sounds risky.