P0667: TCM substrate temperature must agree with two other measurements

P0667 evaluates the credibility of the TCM substrate-temperature signal by comparing it with both fluid temperature and the controller's powerup-temperature sensor. It is a correlation fault: a value can remain within an electrical operating range yet disagree with the other measurements.

Two comparisons, temperature-dependent limits

The table uses different allowed differences for the fluid comparison and the powerup-sensor comparison. Those allowances also vary with the temperature region, so a single universal difference threshold would misrepresent the monitor. The failing relationship must persist for more than 300 seconds.

Evaluation is inhibited when vehicle speed is below 8km/h and accelerator position exceeds 20% for over seven seconds. It resumes only after that combination has been absent for more than 20 seconds. Transmission range and several speed or temperature-signal faults are also part of the enable logic.

Compare a synchronized record

A useful capture contains all three temperatures, range state, vehicle speed and accelerator position. Readings taken at unrelated times cannot show whether the required differences existed together. A short workshop observation may also be too brief to establish the documented persistence.

The pass criterion continues onto the next physical page and requires the differences to remain below their limits for over 70 seconds. Do not confuse a momentary agreement with that completed recovery. The document supplies neither component access instructions nor proof of which sensor is wrong; those decisions require the matching controller and transmission service information.

Source and applicability

GM 2010 10OBDG10 Hybrid Diagnostics, only the cited controller and monitor. These are historical calibration examples; individual vehicle, engine and hardware applicability is not established. GM PDF, page 204. GM PDF, page 205.