P0139

Powertrain
Generic (SAE)
Emissions
Safety

Severity

Diagnose soon

Downstream sensors do not directly control fueling on most vehicles, so driveability is usually unaffected and there is no safety risk. A lazy sensor mainly costs you catalyst monitoring. But if the underlying cause is a genuinely rich mixture, raw fuel is entering a hot catalyst and washing cylinder walls, and that damages expensive components over time.

Safe to drive: Usually safe short term. Get it looked at now if it smells strongly of fuel, runs rough, or the oil level is rising.

Standard definition

O2 Sensor Circuit High Voltage (Bank 1 Sensor 2)

In plain English

The oxygen sensor behind bank 1's catalytic converter is responding too slowly — when the mixture changes, it takes longer to follow than the computer allows. That is the standard definition of this code. A few manufacturers instead use this number for a signal stuck high, so confirm which one your scan tool and the manufacturer's service information give before testing — the two versions need completely different tests. Either way, the code is a detection, not proof the sensor has failed.

The generic SAE definition of P0139 is slow response from bank 1's post-catalyst oxygen sensor; high voltage on that same sensor is P0138. Several manufacturers nonetheless use P0139 for a persistently high signal, so confirm the definition your service information gives for this vehicle before diagnosing. Either way the subject is the post-catalyst sensor on bank 1. Slow response means the sensor eventually reaches the correct value but lags beyond the allowed transition time, which is typically an aged or contaminated element. Where the manufacturer-specific high-voltage version applies, note that a zirconia sensor cannot chemically generate much more than about one volt on its own, so a reported value meaningfully above that is electrical in origin, not a mixture problem.

How the car detected it

The ECM monitors the post-catalyst sensor with the engine fully warm and in closed loop. Behind a working catalyst that sensor normally sits fairly steady in the upper part of its range with only lazy movement. The generic slow-response version of the code measures how long the sensor takes to follow a mixture change the ECM commanded or observed, and fails it when the transition time exceeds a calibrated threshold. The manufacturer-specific high-voltage version instead fails the sensor when its value exceeds a calibrated ceiling for a set period — the threshold is calibration data and varies by manufacturer, commonly somewhere between about 1.0 and 1.5 volts.

A trouble code records what a control module measured. It does not identify which part failed. Test before replacing anything.

Can I keep driving?

Safe to drive: Usually safe short term. Get it looked at now if it smells strongly of fuel, runs rough, or the oil level is rising.
Urgency: Diagnose soon
Will it pass emissions: No
If ignored: If the sensor or its wiring is at fault, you lose catalyst monitoring and will fail emissions testing. If the mixture really is rich, expect fuel dilution of the engine oil, fouled spark plugs, shortened catalyst life, and eventually a catalyst efficiency code such as P0420.
Check engine light: Solid
Clearing the light: A completed drive cycle is normally required before the monitor re-runs and the light can clear.

Symptoms you may notice

Increased fuel consumption
Possible rough idle or hesitation

Many vehicles show no symptom at all beyond the warning light. The absence of a symptom does not mean the fault is not real.

Likely causes

These are the causes this code can have. For the order to work through them in, see what to check first.

#	Cause	How often	Component
1	Engine running rich (leaking injector, high fuel pressure)	Very common	—
2	Sensor signal wire shorted to voltage	Common	—
3	Downstream O2 sensor failed	Common	—

Ordering reflects how often each cause is responsible in general, not a probability for your vehicle. Confirm by testing.

What to check first

In order. Each of these is cheaper or faster than what follows it, and each one can make the rest unnecessary.

Which definition the manufacturer's service information gives for P0139 on this vehicle
Slow response and stuck-high are different faults with opposite test paths, and the generic definition is slow response. Establishing this first stops you running the wrong half of the diagnosis.
Long term fuel trim on bank 1
This splits the diagnosis in one glance. Strongly negative trims mean the engine really is running rich. Normal trims alongside a high or lazy downstream reading point at the sensor or its wiring instead.
Whether the reported voltage is physically possible
A zirconia cell cannot generate more than about one volt on its own, so a value meaningfully above that is electrical in origin — either the signal wire picking up voltage from a powered circuit, or high resistance or an open in the sensor's signal return, which raises the voltage the ECM sees. Test both before buying a sensor.
The downstream sensor connector for water intrusion and corrosion
It usually sits low under the vehicle. Moisture bridging the heater feed into the signal terminal is a classic cause of an impossibly high reading.
The upstream sensor data and every companion code
If the upstream sensor also reports rich and trims are negative, the fault is in fueling, not in the downstream circuit.
Fuel pressure and injector behaviour, but only if trims are negative
A leaking injector or a regulator holding pressure high adds fuel the ECM never commanded, and it is a common cause of a rich exhaust worth ruling out early.

How a shop diagnoses this

Confirm what this code means on this specific vehicle
Tools Manufacturer service information, scan tool with OEM definitions
Look up the manufacturer's definition rather than a generic list. The generic SAE meaning is slow response, but some manufacturers use this number for high voltage, and the two require different tests. Guessing wastes the rest of the diagnosis.
Expected
A definitive statement of whether the monitor failed on response time or on voltage level.
Record freeze frame and full live data
Tools Scan tool
Capture the upstream and downstream signals together with short and long term fuel trims, coolant temperature, load and fuel pressure if the ECM reports it.
Expected
Data that either supports a real rich mixture — negative long term trim, warm engine, fuel-rich upstream signal — or contradicts it.
Measure the sensor's transition time
Tools Scan tool with graphing, propane enrichment tool or vacuum source
With the engine fully warm and in closed loop, force a rich-to-lean and lean-to-rich transition — propane at the intake, then an induced vacuum leak, or a commanded fuel-trim change where the scan tool supports it — while graphing the downstream sensor. Compare the transition time against the manufacturer's threshold. Also check the sensor tip on removal for silicone, coolant or oil contamination, which is what usually makes a downstream sensor lazy.
Expected
The sensor follows the commanded change within the manufacturer's stated response time. A sensor that reaches the correct value but lags well beyond spec, with supply, ground, signal and heater circuits all good, is the slow-response failure.
Unplug the downstream sensor with the engine running
Tools Scan tool
This is the decisive sensor-versus-circuit test for the high-voltage version. With the sensor disconnected the ECM should report only its own bias voltage.
Expected
A steady reading near 0.45 volts. If it still reads high with the sensor unplugged, the sensor is innocent and the fault lies in the harness or the ECM.
Test the signal circuit for a short to voltage
Tools Digital multimeter, wiring diagram
With the sensor and the ECM both disconnected, check the signal wire for voltage and for continuity to the heater feed or any other powered circuit in the same loom. Inspect where the harness passes near the exhaust or a suspension component.
Expected
No voltage on the signal wire and no continuity to any powered circuit.
Voltage-drop the signal return circuit
Tools Digital multimeter, back-probe leads
High resistance or an open in the sensor's ground path raises the apparent signal voltage seen at the ECM, producing a falsely high reading with a perfectly good sensor. This is the other electrical cause of an above-one-volt value and is missed far more often than a short to voltage.
Expected
Under about 0.1 volt of drop across the signal return circuit.
Verify the coolant temperature sensor reading
Tools Scan tool, infrared thermometer
A coolant sensor reporting a colder engine than reality keeps the ECM in warm-up enrichment indefinitely, which produces a genuinely rich exhaust with no other obvious fault. This is a scan-tool-only check and takes minutes.
Expected
Scan tool coolant temperature matching actual engine temperature within a few degrees once fully warm.
Check the evaporative purge system
Tools Bidirectional scan tool, vacuum gauge
Command the purge valve closed and check for vacuum at the purge line. Consider whether the canister has been saturated with liquid fuel by repeated over-filling of the tank. Like the coolant check, this costs nothing and opens nothing up.
Expected
No purge flow while the valve is commanded closed.
Chase the rich condition if the trims say it is real
Tools Fuel pressure gauge, spark plug socket, fire extinguisher
Only after the two checks above. Relieve fuel system pressure before connecting a gauge, work away from ignition sources and hot exhaust components, and have a fire extinguisher to hand — fuel rails hold pressure after shutdown and can spray. Measure running fuel pressure against specification and check whether it holds after shutdown. Pull spark plugs and look for a uniformly sooty pattern. A rapid pressure bleed-down points at a leaking injector or a failing check valve.
Expected
Fuel pressure within the manufacturer's specification and holding for several minutes after shutdown.
Repair what testing indicted, then prove it
Tools Scan tool
Clear the code and complete a full drive cycle. The catalyst and oxygen sensor monitors must run before you can call the repair verified.
Expected
Monitors reporting complete with no pending or confirmed codes.

Possible repairs and what they cost

Repair	Only after you confirm	Typical cost	Who
Replace the bank 1 downstream oxygen sensor	The sensor's measured transition time exceeds the manufacturer's threshold, or the ECM reports its correct bias voltage with the sensor unplugged and the sensor still reads high — in either case with supply, ground, signal and heater circuits all tested good	$130–$380	DIY
Repair a shorted, chafed, or water-damaged harness section or connector	The ECM still reports a high signal with the sensor disconnected, and voltage or continuity to a powered circuit is measured on the signal wire	$100–$400	DIY
Repair the sensor signal return or ground circuit	Excessive voltage drop measured on the signal return, with the reported value normalizing once the circuit is repaired	$100–$350	DIY
Repair or replace a leaking fuel injector	Fuel rail pressure bleeds down abnormally after shutdown and a cylinder shows a wet, fuel-fouled plug, with strongly negative long term fuel trim	$250–$1,200	Hard
Replace the fuel pressure regulator or clear a restricted fuel return	Running fuel pressure measures above the manufacturer's specification	$150–$700	DIY
Replace an evaporative purge valve stuck open	Vacuum is present at the purge line while the ECM has the valve commanded closed	$120–$400	DIY
Replace the engine coolant temperature sensor	Reported coolant temperature disagrees with actual measured engine temperature, keeping the ECM in warm-up enrichment	$100–$300	DIY

The sensor is the cheap end and frequently the wrong answer. Confirming the definition and running the unplug test cost nothing and tell you whether to buy one at all. If negative fuel trims point at fueling, budget for diagnostic time before parts — chasing a rich condition by replacing oxygen sensors is a common way to spend several hundred dollars and still have the code.

Common mistakes with this code

Replacing the downstream sensor when the real fault was upstream fueling. The code comes straight back.
Assuming this code means high voltage. The generic definition is slow response, and running the short-to-voltage tests on a lazy-sensor fault finds nothing.
Treating an above-one-volt reading as proof of a short to voltage. A high-resistance or open signal return raises the voltage the ECM sees just as effectively.
Expecting the downstream sensor to affect driveability. On most vehicles it does not, so a rough idle alongside this code is either a separate problem or the same rich condition.
Drying out a water-filled connector and reconnecting it without sealing the entry point.
Overlooking a saturated evaporative canister after repeated overfilling of the fuel tank.
Replacing the catalytic converter. This code does not test catalyst efficiency.

Components involved

Oxygen Sensor B1s2
Oxygen Sensor Wiring
Catalytic Converter Bank1

Sources

Definition: Generic code set (SAE J2012 / ISO 15031-6) via the OBDexopen database, dedicated to the public domain under CC0-1.0.
Diagnostic guidance: Written and reviewed in-house. See our editorial policy.
Review status: Reviewed

Found something wrong? Tell us — we publish corrections.

Which definition the manufacturer's service information gives for P0139 on this vehicle

Long term fuel trim on bank 1

Whether the reported voltage is physically possible

The downstream sensor connector for water intrusion and corrosion

The upstream sensor data and every companion code

Fuel pressure and injector behaviour, but only if trims are negative

Confirm what this code means on this specific vehicle

Record freeze frame and full live data

Measure the sensor's transition time

Unplug the downstream sensor with the engine running

Test the signal circuit for a short to voltage

Voltage-drop the signal return circuit

Verify the coolant temperature sensor reading

Check the evaporative purge system

Chase the rich condition if the trims say it is real

Repair what testing indicted, then prove it