The 1975 introduction of catalytic converters made this a historic automotive year. Ushered-in was an all-new automotive ideology. Back in 1975, catalytic converters were routinely misunderstood, misdiagnosed, and over-replaced, but they survived, emissions laws tightened, and cars became remarkably cleaner.
Today, the distinction of being routinely misunderstood, misdiagnosed, and over-replaced has been passed on to the lowly oxygen sensor.
Absolute fuel control is mandatory for peak mileage, better performance and lower emissions. Engineers working toward better fuel control knew that measuring engine exhaust might tell how much fuel should have gone into the engine. The question: How could what just happened be used to control what will happen?
This sounds a lot like modern fuel injection! For years, fuel injection systems had been available from Europe and Asia, but they were single-purpose designs for performance, not emissions or economy.
U.S. manufacturers took a completely different approach toward designing better fuel management systems and slipped headfirst into automotive purgatory with horribly problematic feedback carburetors and jury-rigged emissions systems. Collectively, they acted as if oxygen sensors didn’t exist.
Meanwhile import manufacturers designed, built, sold, and perfected oxygen sensor-controlled electronic fuel injection and gobbled up market share. Domestic cars suffered. Fuel economy was nonexistent, and performance was an unspoken, dirty word! But ultimately, EPA rules forced everyone to adopt fuel injection and oxygen sensors.
Oxygen sensors are simultaneously both sophisticated and simple. Scientists knew certain crystals could produce a small voltage when exposed to oxygen. Not exactly useful information until an engineer theorized that mounting one of these crystals in a car’s exhaust might provide an electrical measurement of exhaust oxygen content.
This is still mostly useless information unless you understand that exhaust oxygen is a direct indicator of how much fuel entered the engine. Or, a higher percentage of fuel entering the engine means less oxygen in the exhaust leaving the engine. Conversely, a lower percentage of fuel means more exhaust oxygen.
An oxygen sensor is really a battery that produces voltage when there’s more oxygen on one side of its crystal than the other side. That crystal is mounted in a housing with one side exposed to exhaust gas and the other side exposed to the atmosphere. This allows an oxygen sensor to convert fuel mixture readings into electrical signals.
A rich mixture means less oxygen at the exhaust side of the crystal and a lot on the atmospheric side causing the sensor to produce a voltage of about 900 millivolts. This voltage is directed to the car’s on-board computer, which commands the injectors to deliver less fuel leading to significantly more oxygen in the exhaust. Now exhaust oxygen and atmospheric oxygen are nearly identical and the voltage drops from 900 millivolts to 100 millivolts.
When the computer sees the lower voltage, it commands a rich mixture. Again there’s less oxygen in the exhaust, and the voltage swings back to 900 millivolts. This back and forth switching from rich to lean is continuous and takes place in fractions of a second.
Voltage that oscillates between two extremes results in an average halfway between the two. Because this oscillating voltage controls fuel delivery, mixture is also an average and produces a ratio of 14.7 parts air to one part fuel, the ideal for catalytic converter efficiency.
Being so simple, why are oxygen sensors misdiagnosed 90% of the time? Oxygen sensor codes, that’s why. Codes do not literally tell what’s wrong with a vehicle. Codes only tell what is being affected by what’s wrong and what test to perform to find the real problem.
Think about this. A severely dirty air filter restricts how much air can enter an engine. Less air with the same fuel causes a rich mixture. The computer tries to compensate by cutting back fuel to match the restricted airflow but often can’t. Vehicle computers have an absolute limit to how far they can adjust fuel mixture.
Also, a computer can’t control parts not under its control, and air filters are not computer controlled. The dirty filter causes a continuously rich mixture making the oxygen sensor stall. It’s unable to switch because there is always too little oxygen in the exhaust. Result? High sensor voltage, no switching, an oxygen sensor code, yet the sensor is fine.
Naturally, a bad oxygen sensor can set an oxygen sensor code, but so can other sensors: A clogged PCV system, leaky or dirty injectors, leaking fuel pressure regulator, high fuel pressure, bad vacuum hoses, intake manifold leak, etc. Anything that causes a continuously lean or rich fuel mixture can set an oxygen sensor code.
Oxygen sensor codes do not automatically indicate a failed sensor! An oxygen sensor code means perform the test that matches the code to determine if a sensor has failed or something else is affecting the sensor. Proper procedures offer dividends through a reduction in downtime, fewer repeat check engine lights, fewer repeat repairs per vehicle, and less money wasted by replacing good sensors.
Pat Goss is the resident master technician for Motorweek TV (PBS), a columnist for the Washington Post, and the host of radio and cable shows discussing vehicle maintenance. He also was a consultant to the Prince George’s County, MD Police Department on fleet maintenance, and he is the president of Goss’ Garage in Seabrook, MD. He can be reached at firstname.lastname@example.org.