Last week I ran across a write-up on how to check the O2 sensor. Unfortunately, I can't tell you exactly where I found it, but I was searching all the "tech info"s on TTORA, IH8MUD, YOTATECH, 4x4, etc. and there it was.
And here it is:
Oxygen Sensor Test
From Faq
Testing O2 Sensors
Dick Foster/NorCal TTORA
Here is how to test that O2 Sensor before you go and plunk your hard earned money down on the parts counter for a new one. For this test you will need a standard run of the mill propane torch like those sold at hardware stores and home centers. Use propane and not MAPP gas. The MAPP gas would probably be OK but I have not used it and everything I have read says to use propane. You will also not a digital multimeter or DMM. Use a digital one and not an old analog style meter because analog meters load the output of the sensor too much to get a good reading. The input resistance of the meter (impedance) has to be high. All that really means is that means the meter does not take a lot of current from the circuit under test. You'll also need two small insulated alligator clips to connect the meter leads to the pins in the sensor plug without shorting the pins to one another.
First remove the O2 sensor that you want to check. Let some PB Blaster soak on the nuts and studs for a while so you don't twist the studs off trying to get the nuts off. Unplug and remove the sensor.
The first thing to do is to examine the sensor. The tip should be clean and gray. If it's very black and sooty, you've been running rich and that may be the reason your CE light is on. This can become a vicious cycle with the ECU, engine and O2 sensor. The O2 sensor gets a little sooty from running rich and causes the output of the sensor to drop so the ECU makes the engine even richer causing more soot to build up on the sensor. If you have a lot of soot on the sensor, just brush it off of the outside of the sensor tip and proceed the test will most likely clean the rest off. If the sensor tip is white looking you have probably used silicone somewhere that is not sensor safe and the sensor has been poisoned. You can test it to be sure but silicone will kill an O2 sensor pretty quick. I think it's the acidic acid that does them in but I'm not sure on that. If the sensor tip looks red, maybe someone has been using leaded race gas and has probably killed the O2 sensor by doing so. If the sensor has whitish looking scale deposits all over it, look for a water leak such as a bad head gasket or something of that sort. The sensor is probably toast in that case from contamination from the antifreeze in the coolant.
Now look at the outside part of the sensor housing, you should see four small vent holes up toward the top of the sensor near where the wires come out. These are the reference ports and they must be clean. You should see white ceramic just inside the holes in the tube. If you have all of some of the ports caked over with mud, grease or something else this could be your problem with the sensor. Clean the sensor housing and make sure the openings are clear and you see white ceramic.
Check the wires for breaks and missing insulation. The sensor is not going to work if any of the wires are broken or the leads have been shorted to something.
Once the sensor passes the visual it's time to check it electrically. Clamp the sensor by one of the ears in a vise or some vise grips so that you can heat the sensor with the torch and not burn yourself or something else. Now look closely at the wires on the plug end, you should see two black wires. The black wires are for the heating element. Set the meter set to measure Ohms, use the low ohms scale if you meter has to be set manually, when you measure the two black wires, you should read something like 12 ohms of resistance. DO NOT repeat DO NOT allow the meter leads touch the other two pins in the connector when the meter is set to measure ohms. If you do you could ruin the sensor so be careful. Some say this is not true but some do, it's probably cheaper to be careful. If it does not pass the ohm meter test i.e. open circuit, the sensor is bad so you need a new one. Don't bother with the rest of the test unless you're curious. The O2 sensor element has to be hot to operate; the heater is there to get it hot as the engine is warming up and for prolonged periods of idle. If you look at the wires again you should see two other colored wires one white and one blue or maybe they are some other color. Set the meter to measure DC Volts, a low scale like 2V if your meter does not have auto-scaling and has a dial for that function. Now hook the meter leads to the white and blue wires. Never mind the polarity for now we are just going to measure how much voltage the sensor puts out and in this case the DMM doesn't care. Use the small insulated alligator clips to connect the meter leads to the connector pins. You are going to need to have your hands free to operate the torch. DO NOT short the pins to anything, or again you may ruin the sensor. Once you have the meter hooked up securely so the connection is good and the pins are not shorted it's time to test. Light the torch, now heat the perforated sensor end, you can even get it a little red, remember it has to be hot to operate. With the sensor tip engulfed in the torch flame the output should go to 1V or very close to 1V. As soon as you move the flame away from the sensor the reading should drop to 0V or very close to 0V. Do this a couple of times not letting the sensor cool too much in the process and verify that the change in voltage rapidly follows adding and removing the flame. What's happening is that the torch flame is consuming all the oxygen and you are exercising the sensor just as it's designed to operate. Now heat the sensor for a while, maybe 30 seconds to a minute don't get it too bright red hot or anything but you do want to see that the reading stabilizes at about a volt when the sensor is in the flame and that it does not become erratic or drop out. I've had one go intermittent on me when it was really hot with sustained engine loading like pulling a hill. It turns out that is a common failure mode for O2 sensors. Usually they just sort of wear out and the output voltage is too low and it's sluggish to react to the change is oxygen level. I think anything over 0.8V is supposed to be OK but I would probably change any sensor that output less than 0.85V or so. In fact I'd start shopping at 0.9V. Sometimes they can get sluggish or have low output from carbon deposits that you will probably burn off with this test. If it checks out OK, put it back in your truck and look for another problem. Some common problems to look for are exhaust and vacuum leaks. Air entering the exhaust system will make the sensor output as if the mixture was running rich (excess oxygen) and cause the ECU to run the engine too lean. Remember that soot problem from running rich I talked about earlier too? It's a real good idea to put some anti-seize on those studs when you put the sensor back in. It will make removing it a lot easier the next time around.
