In Front of the Flywheel - October/November - 2018

Oxygen Sensors, Part I

Continuing with our recent theme of inputs, this month we’ll cover oxygen sensors. There’s a lot of information you need to know to understand oxygen sensor operation and testing. As a result, we’ll cover basic sensor testing this month and some more advanced theory in the next issue.

The Oxygen Sensor’s Job

An oxygen sensor’s main purpose is to provide the PCM with a feedback signal indicating how well it’s fueling the engine. The PCM wants to maintain a 14.7:1 air-to-fuel ratio (for gasoline) and it’s constantly making corrections to the amount of fuel it delivers to stay near this point.

To accomplish this, the oxygen sensor sends a voltage signal that varies with air/fuel ratio in the exhaust. A high air/fuel ratio indicates that the engine is too rich and the PCM needs to back off fuel delivery.

Conversely, a low air/fuel ratio indicates that the engine is running lean and the PCM will need to add more fuel. As long as the engine isn’t misfiring, that description is adequate for now. We’ll address a more detailed description of oxygen sensor operation in part II. For now, let’s move forward with that description.

In the case of a zirconia sensor — the most common type of sensor you’re likely to encounter — a rich condition generates a high voltage. High is a relative term, but in the zirconia sensor’s case that would be about 1 volt.

As the engine runs leaner, the sensor’s voltage decreases to near 0 volts. A typical operating range for a functioning sensor and an operational fuel delivery system would range from about 100 to 900 millivolts, or 0.1 to 0.9 volts.

The PCM uses this voltage signal to determine how well the engine was fueled. The PCM is constantly making corrections to fuel delivery and the oxygen sensor’s signal will vary in this range.

Oxygen Sensor Testing with a Scan Tool

Because scan tool communication has become faster over the years, and scan tool graphing has also improved, observing graphed data PIDs is often good enough to make a pass/fail judgment on an oxygen sensor. The engine runs at about 2000 RPM to get the engine, and more importantly the oxygen sensor, up to operating temperature.

Ideally, the PCM should be operating in closed loop and trying to control fuel delivery. Once the oxygen sensor signal is switching back and forth, snap the throttle a few times and you can observe the results on your scan tool.

When you snap the throttle, the PCM should add fuel and cause the oxygen sensor signal to go high. When you release the throttle, the PCM cuts back on fuel delivery and the oxygen sensor signal should go low.

Take a look at the results as I snap the throttle on a 2008 Ford Mustang with many oxygen sensor codes (figure 1). Basically, we’re forcing the engine rich and lean and observing the feedback from the sensor.

What we’re shooting for is in excess of 800 millivolts on the rich side and less than 175 millivolts on the lean side, without ever dropping below zero volts. If the sensor meets these criteria, we know the oxygen sensor is capable of generating the appropriate voltage for the given condition.

In the case of the Mustang, the oxygen sensor codes weren’t due to oxygen sensor failure. The root cause was a lean condition that affected the oxygen sensor signals.

Another way to accomplish this is to use a propane enrichment tool to force the engine rich. You add propane to the intake, forcing the engine to run rich as you observe the sensor signal. Then you remove the propane, forcing the engine lean, while you observe the signal again. You should expect the same results when performing the test this way. This technique is also a great way to observe oxygen sensor operation using an oscilloscope.

On some vehicles, Mode $06 is an additional way to take a look at the results of the PCM’s oxygen sensor tests. Some manufacturers display this data pretty well while others don’t. When using Mode $06, I tend to look at the data and, if the results aren’t acceptable, I continue with an additional test: either scan tool or scope testing.

In this case (figure 2), bank 1 sensor 1 set a code, and we confirmed it was an issue. The PCM’s mode $06 data suggests that bank 2 sensor 1 isn’t far behind. In this case, it’d be reasonable to sell both upstream oxygen sensors to resolve the issue and avoid future trouble.

Oxygen Sensor Testing with an Oscilloscope

Oxygen sensor testing with an oscilloscope, or DSO (Digital Storage Oscilloscope), is the best method to test an oxygen sensor. But, due to the advances in OBD systems, this style of testing is required less frequently. Back in the OBD-I days, this was a test I performed at least once a day. Late-model OBD-II vehicles require this test much less often, but it’s still a valid test you can use to confirm oxygen sensor operation when their performance is questionable.

  • Connect your DSO positive lead to the oxygen sensor signal wire.
  • Connect the negative lead to a sensor ground; not the chassis or battery ground.
  • Set the voltage scale to show 0 to 1 volt on the display.
  • Start with a time base of 500 milliseconds per division, or 5 seconds on the screen.
  • Start the engine and run it around 2000 RPM to get the oxygen sensor warmed up and the PCM into closed loop.

At this point, you’re ready to begin your testing.

To confirm whether the oxygen sensor can report a rich condition, artificially richen the engine with propane. You’d expect a functioning oxygen sensor to generate in excess of 800 millivolts (figure 3).

Once you force the engine rich and confirm the oxygen sensor’s voltage (figure 4), force the engine lean. By removing the propane, or creating a vacuum leak if necessary, you can observe the lean voltage generated by the oxygen sensor. Our lean goal voltage is less than 175 millivolts, without dropping below zero volts. If it drops negative, consider the sensor to be faulty and need replacement.

To be honest, an oxygen sensor will almost never fail this test. A completely dead oxygen sensor generates no voltage and a failing one will not be far behind.

The last part of the test is response time. You can accomplish the previous two steps, rich and lean, without a scope, but not response time. A voltmeter or scan tool just aren’t fast enough to observe what you need to know. This last portion of the test requires a quick “blip” of propane into the intake manifold and you’ll measure the response time of the oxygen sensor in milliseconds.

Here’s how it works: When the oxygen sensor is generating a low voltage, quickly add propane. What you’re looking for is the rate at which the oxygen sensor responds. That is, how much time does it take the sensor signal to move from 300 millivolts to 600 millivolts? A passing sensor will accomplish this task in less than 100 milliseconds. As the sensor ages, the time to switch will increase. Good luck testing that with a voltmeter or scan tool!

You can perform oxygen sensor testing either way you choose, but oscilloscope testing is obviously the better option, due to the amount of detail it provides. Once the sensors have tested good, you can trust the fuel trim numbers generated by the PCM for a wide range of failure diagnostics.

In the next issue of GEARS, we’ll explore how the oxygen sensors really work and how they affect fuel trim numbers.

Engine or electrical diagnostic issues you’d like to see addressed? Let Scott know. Send him an email to and you just may have your question covered in a future issue of GEARS Magazine.