P0133
- Powertrain
- Generic (SAE)
- Emissions
Diagnose soon
No safety risk and the engine drives largely normally. The practical cost is sloppier fuel control, worse economy, higher tailpipe emissions, and an inspection failure. It sits at moderate rather than low because the catalyst monitor may not run, and because degraded fuel control puts extra thermal load on the catalytic converter over time.
- Safe to drive
- Fine to drive. Expect worse fuel economy and an emissions failure until the sensor response is corrected.
O2 Sensor Slow Response (Bank 1, Sensor 1)
The front oxygen sensor on bank 1 still works, but it reacts too slowly. The computer timed how quickly it swings between rich and lean and the sensor is lagging behind. Sensors usually get lazy before they fail outright.
The module measures the sensor's response time — either the number of rich-to-lean and lean-to-rich crossings over an interval in steady closed loop, or the time taken to transition through the switch window after a commanded mixture change — and compares it against a calibrated maximum. Response degrades from ordinary age and thermal cycling, and much faster from contamination: silicone from non-sensor-safe sealant, phosphorus and zinc from burned engine oil, coolant from an internal leak, lead, or high-sulphur fuel. A heater that has weakened without failing outright also slows the sensor by leaving the element below its optimal temperature, and it will do so without ever setting a heater circuit code.
How the car detected itWith the engine at operating temperature and in steady-state closed loop, the module either counts switch transitions per unit time or forces a controlled lean and rich excursion and times the sensor's reaction. If average response time exceeds the threshold, the code sets, usually requiring the failure to repeat on a second trip before the light comes on. The monitor needs stable driving conditions to run, which is why this code can take several days of normal driving to reappear after clearing.
A trouble code records what a control module measured. It does not identify which part failed. Test before replacing anything.
- Safe to drive
- Fine to drive. Expect worse fuel economy and an emissions failure until the sensor response is corrected.
- Urgency
- Diagnose soon
- Will it pass emissions
- No
- If ignored
- Fuel control gets progressively less precise, so the mixture spends more time away from stoichiometric and the catalytic converter has more work to do cleaning up the excursions. Expect measurably worse fuel economy and higher emissions. The vehicle will fail an emissions inspection. Over a long enough period the extra thermal load contributes to catalyst degradation, and the sensor itself will usually progress to an outright failure code. None of that is urgent, but it steadily costs money.
- Check engine light
- Solid
- Clearing the light
- A completed drive cycle is normally required before the monitor re-runs and the light can clear.
- 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.
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 | Oxygen sensor aged | Very common | — |
| 2 | Sensor partially poisoned (silicone, lead, sulphur) | Common | — |
Ordering reflects how often each cause is responsible in general, not a probability for your vehicle. Confirm by testing.
In order. Each of these is cheaper or faster than what follows it, and each one can make the rest unnecessary.
The sensor waveform once the engine is fully warm
A cold sensor is legitimately slow, so judging it before full warm-up gives a false result. Once fully warm, in closed loop, and held at about 2000 to 2500 rpm, a healthy narrow-band upstream sensor crosses the 0.45 volt midpoint continuously — the usual benchmark is at least eight crossings in ten seconds — and reaches both below roughly 0.2 and above roughly 0.8 volts. Switching is legitimately slower at idle, so judge it at a steady elevated rpm. Narrow, sluggish swings confirm the code.
Heater circuit health
A weakened heater slows the sensor without ever setting a heater code. Checking heater current draw and how quickly the engine enters closed loop from cold is a cheap test that can explain P0133 entirely.
An exhaust leak upstream of the sensor
A leak ahead of the sensor lets ambient air in on exhaust pulses, which flattens the waveform and stretches transitions. It mimics a lazy sensor precisely and costs nothing but a careful listen at a cold start.
Oil or coolant consumption and any recent sealant use
These are what kill sensors early. If the engine burns oil, leaks coolant internally, or was recently assembled with ordinary RTV silicone, a replacement sensor will degrade the same way in months unless the source is fixed.
Bring the engine fully to operating temperature and into closed loop
Tools Scan tool live data
Confirm from live data that fuel system status shows closed loop and coolant temperature is at normal operating range before evaluating anything. This is the most commonly skipped precondition and it invalidates every measurement that follows.
ExpectedClosed loop status confirmed, coolant at normal operating temperature
Confirm whether this is a narrow-band or wideband sensor
Tools Service information, VIN
Check service information for the sensor type on this engine before judging any waveform. A wideband air-fuel-ratio sensor does not switch between 0.1 and 0.9 volts — it operates on current and reports lambda or equivalence ratio near 1.0, and its response is judged by how quickly lambda tracks a commanded mixture change. Narrow-band amplitude and crossing-rate rules do not apply to it.
ExpectedConfirmed sensor type and the correct data PID to watch
Graph the sensor signal at a steady 2000 rpm
Tools Lab scope preferred, or scan tool with fast graphing
Hold a steady elevated idle and watch amplitude and transition speed, not just the average value. Look for whether the trace reaches both extremes and how long it dwells at the midpoint. A lab scope shows this far better than a scan tool's sample rate.
ExpectedFull-amplitude swings with sharp transitions through the switch window
Force rich and lean excursions and time the response
Tools Propane enrichment tool, lab scope
Add propane at the intake to drive rich, then use a snap throttle and decel fuel cut to drive lean. Time the sensor in each direction. Slow to rich, slow to lean, or slow both ways narrows the cause — asymmetric response often points at contamination on one part of the element.
ExpectedPrompt movement to both extremes on a healthy sensor
Test heater operation and closed-loop entry time
Tools Multimeter or current clamp, scan tool
Measure heater current or resistance, and time how long the engine takes to enter closed loop from a genuine cold start. A slow entry with no heater code stored still indicates a weak heater dragging down sensor response.
ExpectedHeater within spec and rapid closed-loop entry from cold
Inspect for exhaust leaks ahead of the sensor
Tools Smoke machine or stethoscope, light, mirror
Check the manifold, flange gaskets, and any welds upstream of the sensor, ideally cold when cracks are widest. Smoke into the exhaust makes small leaks visible.
ExpectedNo leak path between the head and the sensor
Review fuel trims before blaming the sensor
Tools Scan tool live data
Large trims in either direction mean an underlying mixture problem is dragging the sensor toward one extreme and limiting its ability to swing. Trims beyond about plus or minus 10 percent are worth investigating and should be resolved before judging sensor response.
ExpectedTrims within a modest range around zero
Remove and inspect the sensor tip for contamination
Tools Oxygen sensor socket, light, penetrating oil
The deposit tells a story. A white or grey chalky coating suggests silicone or coolant. A black sooty coating suggests a persistently rich mixture. A shiny brown or metallic glaze suggests fuel additives or lead. Each finding sends you somewhere different before you fit a new sensor.
ExpectedA clean grey-tan element, or a deposit that identifies the contaminating source
Replace and confirm over a full monitor run
Tools Scan tool with readiness display
Only once wiring, heater, exhaust integrity and mixture are all sound should the sensor be replaced. Afterwards, clear codes and drive under steady conditions for several trips so the monitor can actually run before declaring the fix good.
ExpectedFast switching restored, oxygen sensor monitor complete, no code return
| Repair | Only after you confirm | Typical cost | Who |
|---|---|---|---|
| Replace the upstream oxygen sensor | Measured slow response to forced rich and lean excursions, with heater in spec, no exhaust leak upstream, and normal fuel trims | $170–$520 | DIY |
| Repair an exhaust leak upstream of the sensor | Smoke or audible leak located between the cylinder head and the sensor | $180–$950 | Hard |
| Repair the sensor heater circuit, fuse, or relay | Heater current or resistance out of spec, or a slow closed-loop entry time from a cold start | $90–$420 | DIY |
| Correct the underlying mixture fault driving the sensor to one extreme | Fuel trims outside plus or minus 10 percent, with a located vacuum leak, fuel delivery shortfall, or MAF error | $150–$750 | DIY |
| Address the contamination source — oil consumption or an internal coolant leak | Documented oil or coolant consumption together with a matching deposit on the sensor tip | $400–$3,000 | Shop |
On a straightforward job this is a sensor and half an hour of labour. The range widens sharply when the sensor is seized in an old exhaust bung, when an upstream leak turns into manifold work with broken studs, or when the real cause is an engine consuming oil or coolant. Spend on a quality sensor here specifically — this code is about response speed, which is where budget sensors fall short first.
- Judging the sensor before the engine is fully warm and in closed loop. A cold sensor is genuinely slow and will look failed when it is fine.
- Judging switching speed at idle. Switching is legitimately slower there — evaluate at a steady 2000 to 2500 rpm against the eight-crossings-in-ten-seconds benchmark.
- Replacing the sensor without asking why it aged early. Oil consumption, an internal coolant leak, or the wrong silicone will kill the replacement the same way.
- Reading P0133 as a catalytic converter code. This is the sensor ahead of the converter, and this code says nothing about converter efficiency.
- Using ordinary RTV silicone on the intake, valve cover, or exhaust during a related repair. Only sensor-safe sealant belongs anywhere near an engine with oxygen sensors.
- Clearing the code and rechecking immediately. This monitor needs specific steady-state conditions and can take several days of normal driving to run again.
- Buying the cheapest available sensor. Response speed is exactly the characteristic that separates a quality sensor from a poor one, and this code is about response speed.
- Oxygen Sensor Upstream
- 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.