Tech To Tech
Getting To The Bottom Of Low O2 Sensor Readings
The March Tech to Tech, "Learn to begin at the end," covered an initial diagnostic procedure using the comparison of oxygen (O2) sensor, injector pulse and tailpipe readings. In that article I promised to elaborate on the possibilities you might experience using this technique. Let’s begin with some of the most common causes of low O2 sensor readings.
First, keep in mind that O2 sensors must be tested with the engine at 2500 RPM, the engine warmed up and the O2 sensor up to operating temperature. Now let’s go over some low O2 sensor readings, their related injector and tailpipe readings and the underlying causes (in descending order from the most common to the least common culprits).
Condition One
O2 sensor voltage low (lean), injector pulse wide (rich) and tail-pipe rich. Causes:
- Dead O2 sensor. This is a sensor that has failed for any number of reasons. It produces
no voltage, remembering low voltage = lean signal (L=L). The computer responded with
a rich command. Solution? Replace the O2 sensor and look for the reason for the failure.
- False air at the O2 sensor. This air can come from a misfiring cylinder, giving the O2
sensor a false indication. False O2 can also come from a malfunctioning AIR injection
system dumping air onto the sensor. The false O2 can also come from exhaust leaks.
Leaking exhaust manifold gaskets, cracked exhaust manifolds, and even an improperly
tightened O2 sensor can cause this problem. Remember, the exhaust manifold is not
always under pressure, the pulsing nature of the exhaust causes negative pressure waves
to draw air into the manifold. Sometimes, this problem is so severe that you may think
the O2 sensor is dead.
- Short to ground in the O2 circuit. This usually occurs from mechanical damage to the
wire. The most common problem we have seen is the O2 wire crimped under a valve
cover when the gasket was replaced. Some O2 sensor wires are shielded with a braided
shield coaxially over the signal wire. You can tell by the wire being considerably thicker
and less flexible. We have found cases where the signal wire was shorted to the shield,
usually from mechanical damage. Never use a piercing wire tap to test a signal through a
shielded wire! Misrouted and burned O2 sensor wires can also be a problem.
- An open in the O2 circuit. This one can be tricky! Suppose that the O2 signal wire
connection at the computer is open. Testing anywhere along the O2 wire is going to show
a proper signal (rich). However, the computer is seeing zero volts (lean). If the computer
gives scan data, then this diagnosis can be easier, but many vehicles don’t supply this
data as of yet. OBD II will help here.
This issue of poor connections at the computer is more common than you might think and makes for a difficult diagnosis. Keep in mind that the short problem above affects the whole circuit, but an open may affect only part of a circuit when testing the signal. This also brings up another critical point I try to cover in all my classes. Test circuits as close to the end as possible. This means that on inputs, test as close to the computer as possible! On outputs, test as close to the output actuator device as possible. A good output from the computer does no good if there is an open in the wire between it and the device it’s supposed to operate — and vice versa for an input!
- The O2 sensor is too cool to respond. This should not occur if proper warm-up procedures are used. This becomes a problem with heated O2 sensors that are way downstream that may not depend on exhaust heat. OBD II requires O2 monitoring after the catalytic converter. These O2 sensors are dependent on the heater to keep them at operating temperature. Heater circuits are high current circuits and may or may not be controlled by the computer. A failure in the heater circuit is not uncommon and will cause the O2 sensor to not respond.
Condition Two
O2 reads low (lean), pulse width wide (rich) and the tail-pipe lean. Causes:
- Low fuel pressure. Fuel pump failures are a common occurrence. Fuel pumps work all
the time under adverse conditions: constantly working against a plugged fuel filter, low
fuel causing overheating, and contaminates attacking the fuel pump windings and pump
assembly. Also, the pump is a motor and motors don’t last forever. To properly diagnosis
a fuel pump, you must test pressure while the problem is occurring! Many fuel pumps
heat up and the bearings start to bind after 20 minutes to two hours of driving. When
disassembled, the brush holders will be burned up from too high a current draw from not
operating at proper RPM. The fuel pump is certainly the most common low pressure
failure, but not the only one. It’s also important to check fuel filters, fuel pressure
regulators and the entire pump-side delivery system (pinched lines, etc.).
- Plugged injectors. Many techs think the O2 sensor has unlimited authority to increase or decrease fuel. Nothing could be further from the truth. The O2 sensor is given different authority programmed into the computer based on block learn cells. These cells use air- flow sensor/manifold absolute pressure (MAP) sensor and RPM parameters.
If the O2 sensor had complete authority to add fuel, consider what would happen if it died and falsely told the computer the engine was lean all the time. The injector would be commanded to keep adding fuel until the engine stalled or the engine hydrolocked. In actuality, the computer can make fuel adjustments based on an O2 reading within specified windows of authority. When this authority is exceeded, the O2 sensor can no longer compensate.
Most injection systems will have a pulse width at idle of two to three milliseconds (ms). As injectors become plugged, the O2 commands more on-time to compensate. Most systems I have seen hit the authority wall at 5.5 ms. At this point, the computer can no longer compensate and the tailpipe goes lean.
Condition Three
The O2 sensor low (lean), injector pulse narrow (lean), tailpipe lean. Causes:
- A bad computer input. This is the one that seems to throw many techs. The key is that the O2 sensor command and the injector command are not opposite. However, if the O2 sensor is in command, then the injector pulse should do the opposite. In what cases might the O2 sensor not be in command? The coolant temperature sensor (CTS), the air charge temperature sensor, the throttle position sensor (TPS), and the MAP sensor or air flow sensor all have command authority over the O2 sensor. For instance, if the TPS tells the computer that the throttle is closed when in fact it is well open, then not enough fuel will be delivered, causing a lean condition. Remember, if the O2 sensor and injector are not opposite, then the O2 is not in command.
Condition Four
The O2 sensor reads a negative voltage (yikes!), the injector is commanded either rich or
lean and the tailpipe is reading either rich or lean! Cause:
- Bad O2 sensor. A cracked O2 sensor can give a negative voltage reading.
Other Weird Problems
- Bad grounds. A possible bad ground reference won’t change the true O2 reading, but
may cause the computer to misinterpret the data.
- Alternator diode problems. This can cause strange or nonexistent computer O2
readings. Remember, the O2 sensor may supply a good signal, the computer just fails to
interpret it correctly. This possibility should have been eliminated in pretests. The AC
voltage tested at the battery should never be over 250mV on a computer-controlled
vehicle.
- One more tough problem I have seen throw more than one shop for a loop is bad coolant! When coolant gets old it becomes corrosive — either acidic or basic. Depending on water and engine composition, the engine can become a battery with the same or opposite polarity of the car’s battery. Usually, I have seen the engine produce negative voltages as high as 1/2 volt negative. This wreaks havoc with all systems, causing irrational computer responses. If the coolant looks bad, replace it! Don’t waste valuable diagnostic time on basic maintenance problems that should already have been addressed! I strongly recommend using distilled or deionized water in the cooling system. Most tap water does not meet manufacturers recommendations for purity to be used in a cooling system.
Next month, we’ll look at causes of high O2 sensor readings and their related conditions.
About the author: Mark Warren is the owner of Mark’s Auto, an Automotive Service Association (ASA) member shop in Tucson, Arizona. Warren discusses underhood diagnostics and other service considerations monthly in Tech to Tech. If you have a comment for Mark Warren, you may write him c/o AUTOINC., P.O. Box 929, Bedford, TX, 76095-0929, or fax to (817) 685-0225. Mark’s email address is 76727.3631@compuserve.com (no period at end).
TechTips || Stat Corner || News Briefs
News Briefs Extra || Taking The Hill || Directions
A Team Approach || EPA Proposes || Tech To Tech
Testing Shortcuts || Paint Preparation || Convention Puts Members
Around ASA || More Info About AutoInc. || Chairman’s Message
AutoInc. Magazine ®, Vol. XLIV No. 6, June 1996