Sensor sensibility

Jan. 1, 2020
Most automotive repair professionals have a working knowledge of the operation of the basic oxygen sensor. However, newer wideband designs are more sophisticated and often misunderstood by even the most experienced technicians.

The world owes a debt of gratitude to the oxygen sensor. The air quality in our major centers has improved over the last several decades, despite significant population growth and millions more cars travelling on our nation’s road system.

Like this article? Sign up for our enews blasts here.

Certainly, a host of technologies have helped to achieve the near-zero emissions of the modern automobile. However, many of these technologies would not have had the same impact (or might not have worked at all) without the feedback function provided by the oxygen sensor.

The oxygen sensor has experienced a metamorphosis of its own. From the earliest one-wire design in the mid-1970s to the latest wideband versions, oxygen sensors are more capable than ever of providing critical information to the vehicle powertrain management system. This increased capability has also benefited the diesel engine, as it has enabled the use of advanced diesel emission control technologies.

Most automotive repair professionals have a working knowledge of the operation of the basic oxygen sensor. However, newer wideband designs are more sophisticated and often misunderstood by even the most experienced technicians. Since these sensors are becoming more common as time goes by, both gasoline and diesel engine techs need to get educated on their inner workings.

The Nernst Cell
Most oxygen sensors (except for the relatively obscure titania sensor) operate on principles discovered in the late 1800s by Professor Walter Nernst. The basic oxygen sensor is a galvanic cell (better known as a Nernst Cell) made up of a gas-tight ceramic electrolyte sandwiched between two platinum electrodes. The ceramic electrolyte is composed of zirconia and yttrium, and is exposed to exhaust gas on one side and reference (atmospheric) air on the other. In the case of thimble sensor configurations, exhaust gases flow across the outer surface of the thimble and atmospheric air is on the inside.

When the sensor is cold, the ceramic electrolyte acts as an electrical insulator. However, when its temperature rises to 650oF (350oC), the electrolyte becomes a conductor. At this point, oxygen molecules at the sensor’s inner surface are catalyzed by the platinum electrode and pick up free electrons to become ions. The oxygen ions travel through the hot electrolyte to the electrode at the outer (exhaust) surface, where they give up their excess electrons to the other platinum electrode and are released as oxygen molecules. The exchange of electrons with the oxygen ions is the electrical potential (voltage) that is used by the vehicle ECM to determine oxygen concentration in the exhaust gases.

So why do the oxygen ions flow from the inner surface to the outer surface and not the other way? The answer has to do with a difference in reaction rates between the two electrodes. The inner electrode sees atmospheric air with a relatively high oxygen concentration, so its reaction rate is always high. On the exhaust side, however, the reaction rate varies depending on the oxygen content in the exhaust gases. With a lean exhaust, oxygen is relatively plentiful and thus the reaction rate at the outer electrode comes close to that of the inner electrode. Under these conditions, the electrical difference (voltage) across the two electrodes is low (about 0.10 volt).

In contrast, a rich air-fuel mixture has very little leftover oxygen in the exhaust gases. This makes for a much slower reaction at the outer electrode, which causes oxygen ions to transfer from the inner electrode at a faster rate and thus develop a higher electrical difference between the electrodes (about 0.90 volt). Again, the difference in the reaction rates at the electrodes is what is responsible for the voltage developed by the oxygen sensor.

The classic Nernst cell is also known as a switching sensor. This is because it generates a low voltage in the presence of a lean mixture, a high voltage in a rich mixture and switches rapidly between the two as the mixture transitions through stoichiometry. A major limiting factor in Nernst cell operation is that it cannot tell how rich or how lean the air-fuel mixture is, which creates challenges when the engine is run at a mixture other than stoichiometric.

Another important point is that this same Nernst cell can also be used as an oxygen pump by sending an electric current through it. Oxygen ions can be made to flow through the electrolyte in either direction, depending on the polarity of the current. These effects will be discussed in more detail later in this article.

Technological Evolution
As with any enduring technology, oxygen sensors have undergone an evolution in both their capability and sophistication. Bosch developed the original one-wire sensors that depended solely on exhaust gas heat to bring them up to working temperature. These sensors were slow to go into closed loop and often went cold and returned to open loop during idle conditions. Bosch addressed this problem by developing the heated oxygen sensor in 1982, incorporating an electric heating element, which extended directly into the thimble itself.

With an external source of heat, the oxygen sensor achieved light off faster and also could be designed for minimal direct exposure to exhaust gases and potential contaminants. When they were first introduced, three and four-wire sensors were built with electric heaters and thimble-type elements.

Another major milestone was achieved with the introduction of the planar oxygen sensor in the 1998 Volkswagen Beetle. Instead of a thimble-
type construction, the sensor element was built on a rectangular profile with the various elements of the sensor laminated on one another. This design had numerous advantages, including faster response time, lower power consumption and ability to reach operating temperature faster. Planar sensors also were simpler to manufacture, which created opportunities for more complex oxygen sensors to be designed. While thimble-type oxygen sensors are still common, planar construction has become the dominant technology.

As time went on, OEMs developed more complex fuel control strategies that required accurate measurement of air-fuel ratios over a broad spectrum. The basic Nernst cell was not able to meet these requirements. While it could tell whether the mixture was lean or rich, it could not determine how lean or rich. Despite the challenges, the Nernst cell wasn’t obsolete yet.

Latest and Greatest
Enter the wideband oxygen sensor. These are known by many names, including universal exhaust gas oxygen sensors, linear AF sensors, wideband lambda sensors, wide range AF sensors, etc. What, if any, is the difference between them? According to Dave Ehle and Paul Schriro of Delphi Product and Service Solutions:

“A similar analogy would be asking the question if black cherry or chocolate are both ice cream. Yes, they are, but they are different flavors of ice cream. The same principle applies to linear AF sensors, wideband oxygen sensors, wideband lambda sensors and wide range AF sensors. They are the same technologies with different implementations.

“The difference is the developers of each sensor, OE tier one suppliers, have implemented the same technology slightly differently; each employs a slightly different scale of air/fuel (A/F) ratio to sensor output (scale from the stoichiometry point or 14.7:1 A/F ratio or Lambda = 1.”

A wideband oxygen sensor essentially is a combination of two Nernst cells, sandwiched together in a planar element. One of the Nernst cells is used to measure oxygen content in the exhaust gases, serving the same purpose as a conventional oxygen sensor. However, it is important to note that this cell is not directly exposed to the exhaust gases. Instead, the gases must make their way through a diffusion passage before reaching the sensor’s detection cavity.

The second Nernst cell (known as a pumping cell) is a secondary path for oxygen to move between the exhaust gases and the detection cavity. Directly controlled by the PCM, the pumping cell can move oxygen ions in either direction, depending on the polarity of the current sent to it. The PCM also controls the rate that oxygen is transferred by the amount of current it sends to the pumping cell (typically less than 3 mA).

The key to understanding how a wideband oxygen sensor works is to remember that the PCM is always trying to keep the sensor’s Nernst cell at stoichiometry (~450 mV). For example, if there is a lean exhaust (too much oxygen) in the detection cavity, the pumping cell will transfer oxygen out of the cavity to maintain a stoichiometric reading at the Nernst cell. The exact opposite happens with a rich exhaust, as the pumping cell will move oxygen from the exhaust gases into the detection cavity. The PCM determines the air-fuel ratio based on the polarity and amount of current it sends to the pumping cell to maintain a stoichiometric reading at the Nernst cell.

Oxygen Sensor Failures

While the original oxygen sensors had a life expectancy of 20,000 miles, the newest designs can easily last 100,000 miles or more. However, they are not bulletproof and will suffer premature failures if the exhaust environment becomes hostile. According to Bosch, the most common failure is thermal shock caused by water droplets contacting the heated ceramic element during cold starts. This can crack the electrolyte and render the sensor non-functional.

Contamination is also a major issue, with silica, glycol, carbon and oil residue being the main culprits. Obviously, an oxygen sensor will not survive if excessive coolant or motor oil makes its way into the exhaust system. The motor itself needs to be in good condition in order for an oxygen sensor to live out its useful life.

Old age is another possible reason for an oxygen sensor failure. As mentioned before, they don’t last forever, and it is possible that the root cause for a failure is the poor thing finally wearing out.

The final result is that the PCM is able to estimate air-fuel ratios from as rich as 10:1 up to very lean (air). This means that the engine can be kept in closed loop fuel control when it is run at a mixture other than stoichiometric, such as when cruising using a lean mixture or passing a vehicle with a WOT burst. Wideband oxygen sensor technology also enables diesel emission control technologies such as the NOx adsorber catalyst (NAC) found in 2007 and newer Dodge/Cummins pickup diesels.

Diagnosis First, how can you tell that an oxygen sensor is of the wideband persuasion and not a conventional Nernst cell? One way is to take a careful look at the sensor’s connection to the vehicle wiring harness. A wideband sensor will often have at least five wires running into its connector, which will likely have a trim resistor built into it. Also, you won’t find a wideband oxygen sensor instal
led after the catalytic converter, as that level of capability is not required to monitor catalyst operation.

There is no doubt that onboard diagnostics and scan tools are both getting better as time goes on. With that in mind, a capable scan tool will be your best friend when diagnosing wideband oxygen sensors. Looking at the Parameter Identifiers (PIDs) at Key On Engine Running (KOER), you will know for sure you aren’t dealing with an ordinary oxygen sensor because the units are often reported as lambda (stoichiometric = 1). You’ll also note that the “switching” you’re used to seeing doesn’t look the same with a wideband sensor. Try graphing the lambda PID alongside the pumping current PID. Note that with a functional sensor, these graphs follow each other very closely.

It should be noted that some technicians use inductive current probes to view a wideband sensor’s pumping current on an oscilloscope. Due to the low currents involved, ordinary probes won’t work; you’ll have to purchase a special microamp-capable unit from a reputable manufacturer. With this technology in hand, you can view how a wideband sensor responds when the technician drives the mixture lean or rich, and compare the pumping current to other waveforms.

If the vehicle has a Check Engine light illuminated, retrieve Diagnostic Trouble Codes (DTCs) and check service information for related Technical Service Bulletins (TSBs) and/or pattern failures. If there are no DTCs or pending codes, the Mode $06 stuff you learned in the past can really come in handy. Oxygen sensor test results are supposed to be reported in Mode $05, but it’s becoming more common for them to show up in Mode $06 instead. Hopefully your scan tool supports interpretation of the Mode $06 data, otherwise you will have to get out the manufacturer’s service information to figure out which Test IDs (TIDs) and Component IDs (CIDs) are related to the oxygen sensor. If the sensor is coming close to failing its tests, it might be time to replace it despite a lack of DTCs.

If contamination has been a contributing factor in an oxygen sensor failure, the source must be identified and the customer made aware before going further. You may wind up telling the customer that the engine needs new head gaskets, or even that the engine should be replaced. The customer still could nix the high dollar repair, but at least you aren’t misleading them that a new sensor is going to fix their problem.

We also should extend the benefit of the doubt. Before condemning an oxygen sensor as bad, we need to make absolutely sure that no leaks exist in the exhaust system upstream from the sensor. A cracked manifold or leaking gasket can alter an oxygen sensor’s operation and might even cause false catalyst efficiency DTCs. Your shop’s smoke tester can earn its oats here and will find other exhaust leaks at the same time.

Once the repair is complete, you also will want to investigate whether your scan tool supports functional testing of the oxygen sensor. If so, you may be able to use this to verify your repair without having to run the monitor.

With new technology comes the need for education. One inexpensive way of going about this is to spend some time looking at a healthy wideband oxygen sensor’s data with your scan tool. That way, you will have a much better idea of what a bad sensor looks like when it comes your way. These informal education sessions can pave the way to accurate repairs and loyal customers.

Have articles like this sent to you weekly by signing up for our enews blasts here.