I can't stress enough there is nothing you're going to learn from "testing" an oxygen sensor. "Air / fuel ratio" sensor is the same thing. The only people who test O2 sensors are the engineers who design the systems and want to see what kind of results are being reported so they can program the computer's software to look for the normal responses. We don't have access to their hundreds of thousands of dollars worth of equipment, and we wouldn't know how to interpret the results.
A properly-working front oxygen sensor will switch between "rich" and "lean" about two times per second once the sensing element reaches 600 degrees F. A digital voltmeter takes a reading, analyzes it, then displays it while it takes the next reading. It might measure one pulse at 0.85 volts, for example, then miss the next 0.27 volt, 0.74, 0.32, and 0.68 volt pulses, then take its next reading at 0.29 volts. Digital meters do not average their readings to smooth them out. You would see about every fifth or every eighth reading, and they would be bouncing around wildly too fast to see or to mean anything.
Some people mistakenly think they can use the " 20 AC Volts" scale, but most meters don't have an AC Volts scale that low. Also, AC voltmeters are only accurate when reading a clean sine wave at 60 hertz, (house current). The O2 sensor's wave forms are not even close to a sine wave, and they are much slower than 60 hertz. AC voltage rises smoothly up to its maximum voltage, then it goes to a maximum negative voltage. The meter flips over the negative part of the wave form so it is actually measuring only positive pulses. Oxygen sensor wave forms never go negative. It is actually pulsing between a lower positive voltage and a higher positive voltage, with 0.50 volts being the approximate cutoff point between "rich" and "lean". In the "AC Volts" range, the meter would see only the positive part of what it expects to be a true sine wave. The circuitry is designed to block DC voltage, so even though the voltage is changing rapidly, being only positive pulses, it looks like DC voltage, and it gets blocked. Your meter very likely would read 0 volts AC. What does that tell you when the sensor is actually working just fine?
There is one place where it can be of value to see what the oxygen sensor is seeing, but it requires an oscilloscope, similar to what I used for tv repair. Once you are taught what to look for in the wave form, you can determine if a misfire is being caused by a fuel problem, a spark problem, or a compression problem. This was taught by a fellow who owns a company that specializes in diagnosing the one in a hundred cars that no one seems to be able to figure out. Most of his customers are other repair shops. For the other 99 cars, there are other tests that usually let us figure out the same thing.
The goal of the Engine Computer is to get a perfect fuel / air mixture. For the purposes of this sad story, that would be 0.50 volts from the O2 sensor. The problem is oxygen sensors measure just that, ... Oxygen in the exhaust gas. They do not measure unburned fuel. If the exhaust goes a little lean, the sensor's readings would show that, and the computer would respond by adding a little fuel. If the mixture went a little rich, the sensor can't see the extra fuel, so it wouldn't be known or adjusted for. Instead, the computer purposely drives the mixture too lean, then too rich, about two times per second. When it goes too lean, the unburned oxygen is stored in the catalyst in the catalytic converter. A fraction of a second later when the mixture is too rich, the unburned fuel mixes with that stored oxygen, and is burned to clean it up. If the system were able to hold a steady 0.50 volts, there wouldn't be that stored oxygen in the catalyst when it was needed to address a slightly rich condition. That's why the mixture has to be constantly bouncing between "rich" and "lean". If the computer is satisfied with what it sees from the O2 sensor, the sensor is working. The best you can hope to do is read the voltages on a scanner. Over time the computer will average the mixture out so it is a perfect mixture.
The downstream, or rear oxygen sensor after the catalytic converter is even less exciting. If the converter is doing its job, the exhaust gas coming out will be slightly lean for a long time, then eventually will switch to a little rich. The switching rate could approach once every two or three minutes. The computer is looking at the two cycles per second from the front sensor compared to the real slow switching rate for the rear sensor. The only way that can happen is if the converter is working. When the converter stops doing its job, no change takes place in the composition of the exhaust gas as it passes through it. That means the rear sensor would report the exact same thing as the front sensor, and the switching rates would be identical. That would be for a totally failed, (or removed) catalytic converter. It doesn't even have to get that bad to set a diagnostic fault code. When the converter begins to have reduced effectiveness, the switching rate of the rear sensor will increase, to perhaps once a minute. That's not enough yet to do anything, but the computer is watching that. As the converter get even less effective, say from lead contamination, or pellets breaking loose and being expelled, the switching rate for the rear O2 sensor increases still more. There is a magic point where the computer decides too little change is taking place in the converter, and it knows that by the increased switching rate of that rear sensor. The point of value is the fault code for "catalytic converter efficiency" is set thanks to the greatly increased activity of the rear oxygen sensor.
There is always a long list of conditions that must be met for a fault code to set, and one of those conditions is certain other codes can't already be set. In this case, the computer is commanding a constantly changing rich and lean fuel condition by varying the length of time it pulses the injectors open, then it expects to see signal voltages from the front oxygen sensor confirming those mixtures. If the front sensor doesn't respond as expected, the computer sets a fault code for that sensor related to the defect it detected. With a known defect for the front sensor, the computer has nothing it can trust to compare the rear sensor's readings to, so it will not set a fault code for the rear sensor, even if it isn't working. This is one of those frustrating cases where you finally bring the car in and have the front sensor problem repaired, then the self-test for the rear sensor resumes, a problem is finally detected for it, the new fault code sets, and the Check Engine light turns right back on again. You incorrectly think the mechanic didn't diagnose the problem correctly or he didn't repair it properly, but he had no second fault code to know there was a second problem. This is frustrating for car owners and for mechanics, but it can be avoided by having the first problem repaired right away before the second problem has a chance to develop.
The bottom line is to have a fault code for "catalytic converter efficiency", you have to have two properly-working oxygen sensors. There's over a dozen fault codes related to failed oxygen sensors, and they mean very different things, or types of failures. The Engine Computer is real good at finding problems with the O2 sensors, but you don't have any fault codes indicating that. Your plan of attack is similar to an engine severely overheating and spewing coolant out every gasket, and you want to solve the problem by testing the coolant temperature sensor that is telling you the engine is too hot. You need to address the cause of the overheating. Your computer has already told you the catalytic converter isn't doing its thing. That's what you need to address, not the sensors that are reporting the condition.
Saturday, October 14th, 2017 AT 10:45 PM
