Likely, you are familiar with at least the basic concept of an oxygen sensor. An oxygen sensor — or O₂ sensor as it’s sometimes called — detects the amount of oxygen in your exhaust, post-combustion, in order to determine your air-fuel ratio. Oxygen sensors come in two variations: narrowband, which are on many OEM EFI systems, and utilize a 0- to 1-volt signal, and wideband, which utilizes a 0- to 5-volt signal, making them a far more precise option.

In most modern EFI systems, the ECU will take the readings from the oxygen sensor and use that to adjust the actual fueling of the engine. This is called “closed-loop” operation, and creates a loop where the ECU takes a reading from the oxygen sensor, determines that the engine is rich or lean, and then commands either more or less fuel based on the readings it receives. This happens many times per second to maintain the targeted air fuel ratio. The precision of the adjustments depends on how robust the system is, and whether wideband or narrowband sensors are used within the system.

An oxygen sensor electrically measures the amount of oxygen present in your exhaust, in order to generate an air-fuel ratio for the ECU. They come in both narrowband (0-1V) and wideband (0-5V) variants, with the latter offering much more precision and resolution.

In order to get the readings the ECU needs, the oxygen sensor can be located in a number of areas. Typically you’ll see an oxygen sensor after the merge of an exhaust manifold to read an entire bank of cylinders at once. On a vehicle equipped with a catalytic converter, you’ll usually find another sensor downstream of the converter. Since a catalytic converter’s whole job is to convert exhaust gasses and remediate unburnt hydrocarbons, that will provide a less-than-accurate reading for tuning the air-fuel mixture. The post-cat sensors exist solely for emissions and monitoring the efficiency of the catalytic converters.

As such a crucial sensor in the mix, keeping it happy and healthy is important. An oxygen sensor can fail for a number of reasons, the most common of which is fouling. Leaded gasoline can foul oxygen sensors in a hurry, as can just a sustained overly rich condition. In that same vein an improperly located sensor (usually below the horizontal axis of the exhaust tube it’s mounted in) can fall victim to condensation and unburnt fuel. There are other sensor-killing substances in use in your car, such as ethylene glycol in your coolant (if you blow a head gasket) or uncured or non-sensor-safe silicone from say cheap RTV gasket maker.

The factory narrowband oxygen sensor readings are on the second line from the top, and are left and right sensors in an LS1. The graph is zoomed in to represent about 1.1 seconds of actual time. The stoichiometric point is 450mV.

We recently ran into a situation with our LSX Magazine Project Honey Badger, where after sitting for a number of years, and running incredibly rich before it was parked, we struggled to get the cammed LS1 under the hood to run right. After datalogging the situation, we realized that we either had dirty injectors or bad oxygen sensors. As a Hail Mary before throwing parts at it, we added some of AMSOIL’s Upper Cylinder Lubricant, and gave her the old Italian tune-up, making a few hard pulls to the rev-limiter.

To our surprise, not only did the problems go away, the readings from out oxygen sensors became much more consistent, bank-to-bank. While that could be an indication of an unstuck injector or an oxygen sensor being cleaned out, the fact is the system runs as it should once again. The oxygen sensors properly read combustion gasses and relay that to the ECU, so it can command, the proper fuel from the injectors. If there is a hiccup in any one part of the loop, your engine becomes very unhappy, very fast.