Power cuts, surges and what they do to a built-in

What actually happens while the power is off

Nothing electrical, which is the first thing worth understanding. In a clean outage the damage is thermal and hydraulic. The cabinet warms at a rate set almost entirely by how full it is, because the frozen mass inside a freezer is the thermal store doing the work once the compressor stops. How long the food inside stays safe is a published figure worth looking up separately; what matters mechanically is that a packed freezer coasts about twice as long as a half-empty one. Meanwhile the frost that was on the evaporator melts, and that water has to travel down the drain trough and out to the pan. A drain already carrying a partial restriction fills at exactly that moment, and the returning cold sets it solid. The evaporator then keeps a little more ice each cycle than the defrost can clear, and the fault that gets reported in November began in a September outage nobody connected it to.

• A full freezer holds roughly twice as long as a half-empty one, because the food is the store
• Evaporator frost melts and drains, and a partly restricted drain refreezes solid on the way back
• Ice in the bin fuses into a slab, which jams an ejector arm when the machine restarts
• Every hour the doors stay shut buys more than any other decision available to you

The restart is where the stress is

A compressor is at its most vulnerable in the second the power returns. On a long outage the refrigerant pressures in the system have equalized and the motor starts against almost nothing. On a short one, the flicker of a line fault clearing and a recloser slamming back on, they have not, and the compressor is asked to start against a system still holding pressure. It stalls, drawing locked-rotor current until the thermal overload opens and takes it off line. That is the protection working, and it is why the control on most modern refrigeration will not let the compressor try again straight away. If your unit sits there doing nothing for a few minutes after an outage, that pause is a feature. Three flickers in a minute are harder on a compressor than one clean outage lasting a day, and a sustained brownout is harder than either, because a motor fed low voltage draws more current for the same work and may not come up to speed at all.

• One clean outage of hours is gentler than three flickers in a minute
• Restarting against unequalized pressure stalls the compressor until the overload opens
• A lit display over a silent machine room, for the first few minutes, is the restart delay and not a fault
• Low voltage is harder on a motor than no voltage at all
• Portable generators sag when a well pump or an air conditioner starts on the same supply

The order to check things in when the power comes back

Work from the supply inward and do not judge the appliance for several hours. Start at the breaker panel, because a built-in is normally on a dedicated circuit and its receptacle sits behind or above cabinetry where you will never reach it. Then look at the control: is it lit, and is it showing an alarm. High temperature alarms latch, meaning they are a record of what happened while the unit was dark rather than a live report of what is happening now. Then wait. A cabinet full of warm food takes hours to pull back down, and a compressor that is on its restart delay is not a compressor that has failed. When you do judge it, judge it with a probe on a shelf rather than off the display. Then look for ice spread across the floor of the freezer, which is the drain signature, and empty the ice bin of whatever fused together while it was warm.

• Breaker first: the receptacle behind a built-in is not reachable, so the panel is the switch
• Read the panel for a latched alarm, which records the outage rather than diagnosing the appliance
• Give it several hours before deciding it is not cooling, then measure at the shelf with a probe
• Ice spread across the floor of the freezer, which is the signature of a drain that set solid on the way back down
• Empty the ice bin of the fused slab before the machine tries to harvest into it
• Listen for the condenser fan at the grille, which tells you the machine compartment woke up

Sags and surges do different damage than an outage

A transient is a different event from a loss of supply. Lightning, a recloser reconnecting a faulted line, or a large motor dropping off nearby puts a voltage spike on the same conductors your electronics are sitting on. What is exposed is not the compressor, which is a robust induction motor with its own overload, but the low voltage side: the control board, the user interface board behind the display, and on newer refrigeration the module that drives a variable-speed compressor instead of switching it directly. Those are the failures that look dramatic from the kitchen. A dark panel. A unit that does nothing at all. A display showing something that makes no sense. A fan running continuously with no cooling behind it. They are also, genuinely, uncommon compared with the number of times they get blamed.

• Transients arrive at the moment of restoration as often as at the moment of failure
• The exposed parts are the low voltage ones: control board, interface board, compressor drive module
• A panel-mounted surge protective device is the only practical protection for an appliance whose plug you cannot reach
• Plug-in strips protect what you can plug into them, which does not include a fitted built-in
• Non-inverter generators deliver a waveform that electronics tolerate rather than enjoy

When it really is the board

The board gets condemned because it is the one part nobody can assess by looking at it. The honest sequence runs the other way. Confirm the supply voltage is what it should be, confirm the unit is actually receiving it, and then test what the board commands before you touch the board. If the control is calling for the condenser fan and the fan is not turning, the measurement is taken at the fan, and a board delivering the correct voltage into a dead motor is not the fault, it is the evidence. When a board has genuinely failed there is usually something to find: a scorched trace, a swollen capacitor, a relay welded closed, a burn mark visible once it is out of the housing. A board that tests clean and looks clean, on a unit with an unexplained symptom, is a diagnosis nobody has finished yet. Boards also vary by generation on this equipment, so the model and serial string is worth more to the person taking the call than any description of the symptom.

The faults that arrive a week later

This is the part that costs money, because the connection is easy to miss. The drain iced during the restart, so a slow overflow starts inside the cabinet and the first anyone sees of it is water at the kick plate ten days on, by which time it is in the cabinet base rather than on top of the floor. The evaporator ices a little further each cycle, so the compartment drifts warm over two weeks while the panel goes on showing the figure it was set to. The ice maker never resumed and nobody noticed until the bin ran out. A wine cabinet came back with one zone running and the other not. All of that is ordinary post-outage work and all of it is cheaper caught early, which is why the outage itself is worth mentioning at the start rather than only the symptom in front of you.