OBD-II trouble code
P2223: NOx Sensor Heater Sense Circuit Low (Bank 2)
The heater feedback voltage at the second nitrogen oxide sensor is below the acceptable floor. The cause worth ruling out first is not a short but a ground offset, because the reference these sensors share is only wrong while current is flowing — which means the fault is invisible to every measurement taken with the heater off.
Quick facts
- System
- Powertrain
- Category
- Auxiliary Emissions Controls
- Severity
- Low severity
- Drivable
- Usually safe to drive short-term
- Repair cost range
- $100 – $950
- DIY difficulty
- Intermediate DIY
Browse every code in P2200–P229F, or start from the full code library.
What does P2223 mean?
A voltage is never an absolute thing. It is always a difference between two points, and a controller reading a feedback line measures it against whatever it is using as zero. If that zero moves, the reading moves with it, and the controller has no way of telling the difference between a signal that fell and a reference that rose. On an aftertreatment harness where both nitrogen oxide sensors commonly share a ground path back to the chassis, that distinction is the most useful thing to know about this code.
Here is the mechanism. A corroded or loose shared ground has resistance. Resistance does nothing measurable until current passes through it, and then it develops a voltage across itself, which lifts the local ground at the sensor above true chassis ground. The controller, comparing the feedback line against its own reference, sees the gap between them shrink and reports the line as low. Nothing is shorted, nothing is broken, and the entire fault exists only during the moments the heater is drawing current.
That is why the most common way this code is misdiagnosed is by measuring correctly at the wrong time. Continuity on the ground reads fine with the ignition off. Resistance to chassis reads fine. The connector looks clean. Everything passes because nothing is loaded, and the fault was never present in any of those tests. The measurement that finds it is a voltage drop taken across the ground path while the heater is actually on: put the meter between the sensor's ground terminal and a known good chassis point, command the heater, and watch the reading. A ground that should read close to zero and instead climbs into the hundreds of millivolts under load has just explained the code.
The pattern in the stored codes reinforces this, and it is worth checking before opening anything. Because the ground is shared, an offset caused by that shared path tends to affect both sensors, so seeing the same kind of low feedback fault at both addresses points firmly at the common connection rather than at either sensor. A fault at this address alone moves the odds back towards this sensor's own connector and its own branch of the harness.
When the fault really is local, there is a second distinction worth drawing, and it takes one step. Watch the live feedback value while the heater is commanded on and off. A value pinned at zero and completely unresponsive to the command is a hard connection to ground — something is holding the line down and it does not care what the controller does. A value that is low but not zero, and that changes when the command changes, is a resistive fault: the line is still connected to what it should be connected to, through more resistance than belongs there. Those two findings send you to different places. The first means looking for where the conductor is touching metal or where a connector is bridged; the second means looking for a corroded terminal or a poor splice in the line itself.
The things that produce either are unglamorous. Terminals at the tail of an aftertreatment harness live in salt, water and grit, and the failure is usually a green terminal or a joint that has quietly gone high-resistance rather than anything dramatic. A conductor that has worn through its insulation against a shield or a chassis rail produces the hard version. Both are inexpensive to fix once found, which is the encouraging part of this code.
As for consequences, there are almost none in the short term. This circuit reports on the heater and drives nothing, so the element keeps working and the vehicle keeps running normally while the controller complains that it cannot verify what it commanded. The real costs are an emissions monitor that will not complete and, on a diesel, the aftertreatment inducement sequence that eventually begins on distance rather than on anything the driver notices.
Common causes
- Corroded or loose shared ground connection, which only develops a voltage offset while the heater is drawing current
- High-resistance terminal or splice in the feedback conductor, leaving the line connected but through more resistance than it should have
- Feedback conductor worn through its insulation and touching a heat shield, bracket or chassis rail
- Salt, water and grit corrosion at a connector mounted low in the road spray at the tail of the aftertreatment harness
- Bridged terminals inside a flooded connector, connecting the feedback pin to a ground pin alongside it
- Chassis ground stud that has loosened or corroded beneath its eyelet, common on vehicles that have had exhaust work
- Internal fault in the sensor's own electronics reporting a low heater status
- Damaged repair splice from an earlier harness repair that was crimped rather than soldered and sealed
Symptoms
- Check engine light with no change in how the vehicle drives
- The same kind of low feedback fault stored at both nitrogen oxide sensor addresses, which indicates the shared ground rather than either sensor
- Emissions readiness monitors that will not complete
- Heater status reported as low or zero on a scan tool while the sensor's nitrogen oxide data looks entirely normal
- The fault appearing or worsening in wet weather and easing when the vehicle has been dry for several days
- Occasional unrelated codes from other devices that share the same ground point
- On diesels, the eventual staged inducement warnings if the fault is left unresolved
Diagnostic steps
- 1.Check whether the same fault is stored at the other nitrogen oxide sensor address. Both together point at the shared ground; one alone points at this sensor's own branch.
- 2.Measure voltage drop across the ground path with the heater actually drawing current — between the sensor ground terminal and a known good chassis point. This is the test that finds an offset, and no unloaded measurement will.
- 3.Inspect the chassis ground stud and eyelet for the aftertreatment harness, cleaning it to bare metal and retorquing it if there is any corrosion under the terminal.
- 4.Watch the live feedback value while commanding the heater on and off. Pinned at zero and unresponsive means a hard connection to ground; low but moving means a resistive fault.
- 5.For the unresponsive case, trace the feedback conductor for chafe points against shields, brackets and the chassis, and check the connector for bridged terminals.
- 6.For the resistive case, measure voltage drop along the feedback conductor itself and check every terminal and splice in the run.
- 7.Unlatch the connector and examine it under good light for salt deposits, moisture and green corrosion, testing terminal retention rather than judging by appearance.
- 8.Inspect any previous harness repairs in the run. A crimp that was never sealed is a standard source of a slowly rising resistance.
- 9.Clear the code and drive a full cold-start cycle so the heater is commanded before judging the repair.
Repair cost
$100 – $950
This code is usually cheap to resolve once the right measurement is taken. Diagnostic time is $100 to $250, and the loaded voltage-drop test is where that money earns its keep. Cleaning and retorquing a corroded chassis ground is often under $100. Repairing a terminal, a splice or a chafed section of the feedback conductor runs $80 to $400, and most of these end there. Replacing the sensor assembly is $250 to $900 fitted and belongs at the end of the list, because the element is frequently heating perfectly while this fault is stored.
Estimate your repair
Run the numbers for your vehicle
Open the Repair Cost Estimator with wiring harness / circuit repair preselected. Adjust labor rate and vehicle category to fit your situation.
DIY vs shop
This is an intermediate DIY job. It usually involves diagnostic steps, specialty parts, and some careful work in tight spaces. If you have the tools and a service manual or trustworthy video for your specific vehicle, it is achievable in a weekend. Otherwise, a competent independent shop will be faster.