Runs In Diagnostic Mode But Not Normal Cycle

Why this matters

An appliance that runs every component flawlessly in service mode but stalls or refuses to run a normal cycle is one of the most informative fault patterns in the trade, and one of the most misread. Techs see the motor, valve, pump, and heater all fire in diagnostic mode and conclude the hardware is good and the board is bad. Sometimes. But far more often, the difference between the two modes is exactly the diagnosis: service mode bypasses the sensors and interlocks that a normal cycle depends on. The component that fails to run normally is being held off by a sensor input, a safety, or a missing condition that the diagnostic mode ignores. This tree turns that split into a fast diagnosis.

Symptom presentation

In the manufacturer's service or test mode, every output cycles on command: the motor spins, the valve fills, the pump drains, the heater energizes, the fans run. But under a normal user-selected cycle the machine does nothing, stalls partway, or skips a step. A washer fills in test mode but will not start a wash. A dishwasher runs the pump in diagnostics but stalls at fill. A dryer heats on test but blows cold in a normal cycle. The hardware is clearly capable; something in the normal control logic is blocking it.

Quick checks

Understand what service mode bypasses. Diagnostic modes typically drive outputs directly, ignoring most sensor feedback, level switches, thermistors, lid/door interlocks, and cycle-progression logic. A normal cycle requires all of those inputs to be valid before it advances. So the question becomes: which input does the normal cycle need that diagnostic mode skips? Confirm the door or lid switch is making (a normal cycle will not run with an open interlock, but test mode often will). Confirm water level, temperature, and pressure sensors are reading plausibly. Pull any stored fault codes from the normal-cycle attempt; the normal cycle will log the sensor or interlock it stalled on, while test mode will not.

Isolation tree

The fault lives in the inputs that normal mode checks and diagnostic mode ignores.

1. Does the normal cycle start at all, or stall at a specific step?

   • Will not start. Suspect a gating interlock or a global enable: door/lid switch, child-lock, control-lock, a "remote start" or delay setting, or a primary input the cycle validates at entry.
   • Starts then stalls at one step. The input required to advance past that step is failing. Map the stall point to its sensor.

2. Map the stall point to the input it waits on.

   • Stalls at fill. The cycle waits for a level or pressure signal it never receives. Test mode opened the valve regardless; normal mode needs the pressure switch or flow meter to confirm level. Suspect the pressure switch, air dome/tube, or flow meter.
   • Stalls at heat. The cycle waits for a temperature rise the thermistor never reports. Test mode energized the heater open-loop; normal mode needs the thermistor feedback. Suspect the thermistor or its wiring.
   • Stalls at drain or spin. The cycle waits for an empty signal or a balanced/lid-locked condition. Suspect the pressure switch reading "empty," the lid lock, or the balance sensor.
   • Stalls at motion. The cycle waits for a tach or speed feedback it never gets. Suspect the motor tach/Hall sensor (the motor spins in test but the cycle cannot confirm speed).

3. Confirm the suspected input. Read the sensor the stalled step depends on and compare to a plausible value. A pressure switch stuck "full" blocks fill-complete; a thermistor reading open blocks heat-confirm; a tach with no signal blocks motion-confirm.

The pattern is consistent: the output works because test mode commands it directly; the cycle fails because the feedback that gates the next step is missing.

Confirming diagnosis

Confirm by reading the gating input live during a normal-cycle attempt. Start a normal cycle and watch the specific sensor the stall maps to: the pressure switch contacts as water rises, the thermistor resistance as the heater runs, the door/lid interlock continuity, the motor tach output as it spins. The input that does not change, or reads implausibly, while its output is working is the fault.

Cross-check against the stored fault code from the normal-cycle attempt, which names the timed-out condition (long fill, no heat rise, no speed feedback, door fault). When the named condition matches the sensor that fails to respond, the diagnosis is confirmed: a sensor or interlock fault, not a dead output and usually not the board. Only when every gating input reads correct during a normal cycle and the cycle still will not advance do you suspect the control logic itself.

A practical guard against a wrong board call: before condemning the control, wiggle-test and ohm the harness between the suspect sensor and the board with the connectors seated as the cycle runs. An intermittent connector or a chafed lead reproduces a sensor-feedback fault that ohms fine at rest and looks like a board failure. The board reads what the harness delivers; a broken feedback path makes a good board behave as if the input is dead. Verify the wiring continuity end to end before the board is on the parts list.

Remediation

• Interlock/door/lid fault: Replace or adjust the door/lid switch or lock assembly; clear control-lock or child-lock settings.
• Pressure/level fault: Clear the air dome and pressure tube of clogs, repair leaks in the tube, or replace the pressure switch or flow meter.
• Thermistor fault: Replace the thermistor and repair its harness; the heater is fine, the feedback was not.
• Tach/speed fault: Replace the motor tach/Hall sensor or repair the harness so the cycle can confirm motion.
• Confirmed control logic: Only after inputs verify good, replace the control board.

References

• UL 749, Standard for Household Dishwashers
• UL 2157, Standard for Electric Clothes Washing Machines and Extractors
• UL 2158, Standard for Electric Clothes Dryers
• DOE, 10 CFR Part 430 Subpart B, energy conservation test procedures for major appliances