OBD-II trouble code
P2139: Throttle/Pedal Position Sensor 'D'/'F' Voltage Correlation
Which hardware this code compares is not fixed: the letter-to-device mapping is the manufacturer's choice, so D and F may be two tracks inside one pedal or two channels in separate assemblies. That one lookup decides the diagnosis, because the cheap shared-connector explanation applies in the first case and is unavailable in the second.
Quick facts
- System
- Powertrain
- Category
- Throttle / Idle
- Severity
- High severity
- Drivable
- Usually safe to drive short-term
- Repair cost range
- $0 – $750
- DIY difficulty
- Intermediate DIY
Browse every code in P2100–P219A, or start from the full code library.
What does P2139 mean?
P2139 sets when the module compares the 'D' and 'F' position channels and finds that the relationship between them has moved outside tolerance for long enough to be more than noise. Both signals are inside their own electrical limits. Neither would set a low, high or circuit code on its own. What has failed is the agreement between them.
The important structural fact about this particular pairing is one that does not apply to the other correlation codes in the family, and it should be the first thing anybody working the code understands. A/B compares two channels inside one throttle body. D/E compares two tracks inside one accelerator pedal. In both of those cases the two channels being cross-checked are physically co-located: they sit in the same housing, they are fed from the same connector, and on many designs they share a ground path and sometimes a reference. That shared plumbing is why a single connector fault, one corroded ground or one damaged pigtail can throw a correlation code without either sensing element being defective at all. It is the most common cause of those codes and it is the first thing a good technician checks.
D and F are the pairing where co-location cannot be assumed either way, and the honest answer is that the letter-to-device mapping is set by the manufacturer rather than by the standard. On some applications D and F are two tracks within the same accelerator pedal assembly, sharing a connector exactly as D and E do. On others F is a separate device — a throttle-body channel, an auxiliary or hand throttle, a secondary control on a chassis-cab or upfitted vehicle — with its own connector, its own harness run and its own reference and ground. The wiring diagram for the vehicle settles which, and nothing else does.
That one lookup splits the diagnosis in two, and it is nearly free. If the diagram shows D and F sharing an assembly and a connector, treat the code exactly as you would a D/E: one corroded terminal, one bad ground or one damaged pigtail can throw it with both sensing elements perfectly healthy, and that shared circuit is the first and cheapest thing to check. If the diagram shows two independently supplied devices, the shared-circuit explanation is genuinely unavailable, and the disagreement has to come from one of three places: an offset developing inside one of the two devices, a resistance fault on one device's own circuits skewing its reading, or a calibration mismatch between two devices that were never taught to agree in the first place.
That third possibility deserves its own emphasis because it produces a P2139 on hardware that is entirely healthy. Two separate devices have to be brought into agreement by a learned relationship stored in the module. Replace either one without performing the relearn the platform calls for, fit a non-equivalent part whose output curve differs from the original, or restore a module to defaults without re-teaching it, and the two channels will track perfectly well while disagreeing by a constant offset — which is exactly what a correlation code is looking for. A P2139 that appears within days of pedal, throttle body, auxiliary control or module work is a calibration problem until proven otherwise, and no amount of testing the individual channels will find it.
One asymmetry is worth carrying into the repair. D is a pedal track and, on most implementations, part of the pair the module uses to form the driver's actual torque request. F is a third channel whose job is largely to confirm. So when the two disagree and the module has to decide what to trust, it is usually the auxiliary channel that gets discounted, which is why a P2139 often arrives with a car that still drives — and why a P2139 arriving with a reduced power event points towards the pedal side rather than the third channel.
Common causes
- Shared connector, ground or reference fault, where this platform puts D and F in one assembly
- Relearn or calibration procedure not performed after a pedal, throttle body or auxiliary control was replaced
- Non-equivalent replacement part whose output curve differs from the original
- Offset developing inside one device's sensing element as it wears
- High resistance on one device's own ground or reference circuit skewing its reading
- Module restored to defaults or reprogrammed without re-teaching the channel relationship
- Mechanical wear or play in the control carrying one of the two channels
- Obstruction limiting travel on one control so it cannot reach expected values
- Corroded terminal on one of the connectors, shared or independent
- Harness damage on one device's circuits only
- Failed sensing element in either the pedal or the device carrying F
Symptoms
- Check engine light with a stored correlation code between two channels
- Reduced engine power warning on applications where the disagreement reaches the commanding pair
- Vehicle that drives normally because the auxiliary channel was the one discounted
- Code appearing within days of pedal, throttle body, control or module replacement
- Two position traces that track together but sit a constant distance apart on a graph
- Throttle response that feels slightly out of step with pedal effort
- Auxiliary or hand throttle control that no longer produces the expected response
- Cruise control unavailable while the fault is stored
- Code that returns after clearing without any change in how the vehicle behaves
- Normal idle and normal starting, since neither depends on the comparison
Diagnostic steps
- 1.Confirm from the wiring diagram whether D and F are in the same assembly or two different ones. The mapping is manufacturer-specific and both arrangements exist, and that single fact decides the whole job: co-located means the shared connector, ground and reference get checked first; separate means that explanation is unavailable and calibration moves to the front.
- 2.Ask what work has been done recently before testing anything. Pedal, throttle body, auxiliary control or module replacement inside the last few weeks makes calibration the leading theory regardless of what the channels measure.
- 3.Graph both channels together through several full, slow sweeps of the control. A constant gap between two otherwise parallel traces is a calibration or part-equivalence problem; a gap that opens at one point in travel is wear.
- 4.Verify each device's own reference and ground separately, since they are independently supplied. A voltage drop on one device's return shifts only that channel and reads as disagreement.
- 5.Check part numbers against the original specification if either device has been replaced. An output curve that differs slightly produces a permanent offset on hardware that tests perfectly.
- 6.Perform the platform's relearn or idle relearn procedure where one has not been done, and re-evaluate before condemning anything.
- 7.Inspect both connectors independently. They are different connectors in different environments, so finding one clean says nothing about the other.
- 8.Check the mechanical side of whichever control is exposed. Restricted travel, a trapped mat, a bent linkage or a worn stop keeps one channel from reaching its expected values.
- 9.Note whether the vehicle has lost power. A derate alongside this code suggests the disagreement is affecting the commanding pedal side rather than the confirming channel.
- 10.Recheck across a full sweep after any repair and confirm the two channels now hold their learned relationship end to end, not just at rest and at full travel.
Repair cost
$0 – $750
This is one of the few codes in the family with a genuine zero-parts outcome: where the cause is a missed relearn after other work, the fix is a procedure, and a shop that caused it will usually perform it at no charge. Diagnosis otherwise runs $120 to $250, and includes confirming from the wiring diagram whether the two channels are independently supplied. A circuit or terminal repair on one device is $90 to $280. An accelerator pedal assembly is $180 to $450 fitted. A throttle body is $250 to $600. An auxiliary or hand throttle control with its harness is $250 to $750. Where a non-equivalent part is the cause, the cost is fitting the correct one plus the relearn.
Estimate your repair
Run the numbers for your vehicle
Open the Repair Cost Estimator with accelerator pedal position sensor replacement 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.