Repair Control Board vs Sensor vs Wiring Intermittent Decision Matrix

Why this matters

Intermittent appliance faults that clear on reset and return mid-cycle are the calls most likely to end in a wrong, expensive part swap, because the failure is not present when the tech has the panel open. The reflex is to condemn the control board, which is usually the costliest and least often the actual culprit. The real distribution is heavily weighted toward connections and sensors: a chafed harness, a back-out connector, a corroded terminal, or a sensor drifting near its threshold causes the large majority of intermittents, with the board a distant third. This matrix gives a disciplined order of suspicion and the evidence that promotes or demotes each candidate, so the decision rests on measurement and failure-pattern logic rather than on which part is easiest to order. Used correctly it converts a guess into a defensible call you can stand behind on a flat-rate callback.

The three candidates and their signatures

Wiring/connection faults are the most common intermittent cause. Signature: the fault correlates with vibration, temperature, or mechanical movement; wiggling a harness reproduces or clears it; an affected connector shows discoloration, corrosion, or a backed-out pin; the same symptom appears across unrelated functions that share a harness or ground.

Sensor faults are the second most common. Signature: a single function misbehaves in a way consistent with a wrong reading (a thermistor reading falsely cold suppresses cooling; a level sensor satisfying early underfills); the sensor measures out of range only at certain temperatures or after warm-up; the fault tracks a physical condition (cold, hot, wet) rather than random timing.

Control-board faults are the least common but real. Signature: the function fails despite a confirmed-good sensor signal arriving at the board connector and a confirmed-good harness and ground; a board component shows visible damage (a swollen capacitor, a burnt relay, a cracked solder joint reflowing with heat); the fault follows the board when swapped to a known-good unit.

Decision matrix

Use this order and these gates.

First, reproduce or characterize the fault. If you cannot make it appear on site, monitor with a data logger or leave the unit in a service-test loop. Never condemn a board on a fault you have not observed or measured.

Second, suspect wiring/connection. Inspect every connector in the affected circuit for seating, corrosion, and back-out. Flex the harness while the function runs. Check grounds and bonding. If wiggling reproduces the fault, or a connector is visibly degraded, the fault is here. Repair the connection and re-test before going further.

Third, suspect the sensor. Measure the sensor against its temperature-resistance table at multiple temperatures, including the condition where the fault appears (cold-soak the unit, run it hot, wet the area). A sensor in range at room temperature can drift out of range only at its operating extreme. If the sensor reads wrong at the fault condition, replace it.

Fourth, suspect the board only after the first three are cleared. Confirm the correct input arrives at the board connector and the board still fails to act correctly, or the board shows physical damage, or the fault follows the board on a known-good swap. Only then replace the board.

Evidence that promotes the board

Promote the board to prime suspect when: the sensor signal is confirmed good at the board's own connector and the board still misreads or fails to drive the output; a board component is visibly damaged or a solder joint reflows under a heat-gun test and temporarily fixes the symptom; or the identical board in a known-good unit cures the fault on swap. Absent at least one of these, the board stays demoted behind wiring and sensors regardless of how convenient it would be to blame it.

Evidence that demotes the board

Demote the board (do not replace it) when: the fault correlates with vibration or harness movement; a connector is corroded or backed out; a ground is high-resistance; the symptom appears across multiple functions sharing a harness or ground; or a sensor measures out of range at the fault condition. Any one of these explains the intermittent without the board and should be corrected and re-tested first.

On callbacks and documentation

Because intermittents resist single-visit confirmation, document the measurements that drove the call: connector condition, harness-flex result, sensor readings at the fault condition, and the board's input-versus-output behavior. If the fault cannot be reproduced, set expectations with the customer that the visit established what is not failing and may require a return when the fault reappears with the unit running normally. A photographed corroded connector or an out-of-range sensor reading is a defensible repair; a board swapped on a hunch that returns is a callback you absorb.

References

• ANSI / UL 60730 (Automatic Electrical Controls) for control reliability references.
• NEC Article 422 (Appliances) for appliance circuit, grounding, and connection requirements.
• ANSI / AHAM appliance performance standards (HRF-1, HLW-1, DW-1) for the function-specific behavior used to characterize sensor faults.
• IPC-A-610 (Acceptability of Electronic Assemblies) for solder-joint and assembly defect references used in board inspection.
• OEM service manuals for the specific appliance: sensor resistance tables, harness diagrams, and board input/output test points on the model tag.