You are not logged in! This article is in Preview mode!
Click
Secure Login to login and view this article in its entirety.
If you are not a member please feel free to preview this article.
However please note that much of the content will be unavailable including some images;
How to test an Oxygen Sensor
There are several types of oxygen sensors, but for the
purposes of this article I will be discussing the standard heated oxygen sensor
available on most vehicles today. This article will not include any information
on the newer wide band or A/F Ratio sensor. Most technicians already
understand the purpose of todays oxygen sensors, but may not necessarily know
how to properly test them for performance. Proper oxygen sensor diagnosis is
particularly important when dealing with certain trouble codes such as P0171
System Lean or P0420 Catalyst Efficiency. It is important to understand when
the oxygen sensor is reporting a lean or rich condition properly, and when it
is malfunctioning and giving false data. A P0171 could be and often is caused
by a faulty O2 sensor, but is probably more often caused by a true lean
condition. The same holds true for P0172 System Rich. Perhaps one of the most
misdiagnosed codes is P0420. While there are many things that can cause this
code besides a faulty catalytic converter, the upstream oxygen sensor is
perhaps the most common. Yes, I said upstream, not downstream. I will explain
more on this later, but first lets go over some oxygen sensor basics (without
getting too technical).
An oxygen sensor should be more appropriately named an "oxygen demand" sensor. The thimble of the standard zirconia sensor is like a miniature catalytic converter.
When the sensor is up to full operating temperature of about 300 C, and the system is under a rich condition, the negative ions of oxygen on the atmospheric side of the thimble want to travel to the exhaust side to complete the reaction.
This movement in oxygen atoms creates a voltage up to about 1 volt. When the exhaust gases are lean, there is little to no need for any of the oxygen ions on the atmospheric side of the thimble, therefore the sensor produces little to no voltage.
Atmospheric Ports and Skewed Sensors
This brings us to an important part of the oxygen sensor
that many technicians do not know exists. The atmospheric port(s) is a small
hole or holes that allow atmospheric air into the oxygen sensors main chamber.
These holes are very small and often cannot be seen. Many oxygen sensors(almost all newer models) use
the electrical wiring entry points or the wiring itself as atmospheric ports. The fresh atmospheric
air sits on one side of the main chamber with the exhaust gas on the other
side. When the oxygen sensor reaches operating temperature of around 300
degrees Celsius, it will begin to produce a voltage based on the difference in
oxygen content on each side of this chamber.
If the atmospheric ports become dirty and clogged with
mud, oil or other road grime, the oxygen sensor will begin malfunctioning. As
less and less oxygen can enter the atmosphere side of the chamber, the oxygen
content difference between the two chambers will be lower causing a lower
voltage output. This is generally referred to as a lean biased sensor. With a
lean biased sensor, the vehicles computer will begin adding fuel (increase fuel
trims) in an attempt to maintain an average of 450 millivolts. Since the
atmospheric ports are partially blocked and less atmospheric oxygen can reach
the external part of the chamber, the oxygen content in the exhaust stream must
also be lower than normal to provide enough of a difference to produce a
voltage. This will result in vehicle that is running rich and has high fuel
trims since the computer thinks the vehicle is actually running lean.
Take a look at the scangraph of this 1990 Chrysler
Lebaron. The vehicle was failing the state emissions test with very high HC and
CO readings. With the black exhaust and fouled spark plugs it was very easy to
determine that it was running extremely rich. When the vehicle was first
cranked and the fuel control was in open loop, it actually ran well with little
or no excess smoke from the exhaust. Even when the vehicle first achieved
closed loop everything was fine, but as the engine compartment began to heat up
and oil from a leaky valve cover began to drip on the oxygen sensor, the
atmospheric ports quickly became clogged and the fuel trims began a steady
climb.
Even though the fuel trims are rising and the amount of
oxygen content in the exhaust stream is very low, the oxygen sensor is still
reading as if there is a lean condition because very little oxygen can reach
the atmospheric side of the sensors main chamber due to the blocked ports. Remember, the sensor creates a voltage based
on the difference in oxygen content between the two chambers. If
the external chamber has low oxygen content, and the internal chamber has low
oxygen content, very little voltage will be produced.
If the ports become completely blocked, it is possible to
have more oxygen in the exhaust stream than on the atmospheric side of the
sensor. This in turn causes the sensor to work in reverse and begin to produce
a negative voltage. Take a look at the scope capture below of a dodge pick-up
with a system lean code stored. As with the Lebaron above, this truck was
actually running very rich, not lean.
With the engine RPM raised to 2500 and propane added, the
maximum voltage achieved was 600 millivolts. When the propane enrichment was
dropped, the voltage then dropped well into negative values at around 400mv.
Another quick test to check for a skewed oxygen sensor is
to watch the sensor voltage and fuel trims while testing with an exhaust gas
analyzer. If the oxygen sensor is switching or is pegged at a low voltage and
the fuel trims are positive yet the exhaust gas shows the vehicle is running rich,
suspect a lean biased sensor. Be sure to verify there are no exhaust leaks that
can affect the sensor readings and skew the exhaust gas readings. Using a gas
analyzer is a good way to verify an oxygen sensor is the cause of a fuel trim
problem and is not just reacting to a problem.
A good oxygen
sensor should always be capable of producing at least 850 millivolts and should
never drop below 0. Below is another lean biased oxygen sensor from a Honda
Accord.
Once again the maximum voltage reading is too low with
artificial propane enrichment at around 700mv. When the propane is removed the
voltage drops below 0 to around 200mv as seen below.
Using the Downstream Sensor for Reference
The downstream oxygen sensor can be an effective tool to
diagnose a skewed upstream sensor. Since it is unlikely that the both upstream
and downstream sensors will fail at the same time, check the downstream sensor
to see if the upstream is lying to the PCM. Even though the downstream is
usually more of a flat voltage than a waveform, it does still react to the
overall fuel mixture and generally is a better indication of true Lambda than
the front sensor is. Take a look at the example below from a 1996 Dodge
Caravan.

The upstream sensor is staying below 200mV even though the
short term and long term fuel trims are at +25% each. This could be due to a
very lean condition, or it could just be a bad oxygen sensor. Notice that the
downstream sensor is pegged at 940 mV indicating a very rich condition, which
is exactly what would be expected with such high fuel trims if there were no
true lean condition present.
In the example above the upstream sensor did finally wake
up but was still skewed to the lean side as seen in this next capture. The
upstream still spends most of the time below the 450mV mark, but will spike
higher occasionally. The downstream sensor is still pegged at full rich
indicating that the upstream is lying to the PCM about the air/fuel mixture.

You are not logged in! This article is in Preview mode!
Click the "Secure login" link in the upper right to login and view this article in its entirety.