What the display cannot tell you about the cabinet

Setpoint, sensed air, and the gap between them

Depending on the generation, the number on the front is the setpoint, the sensor's current reading, or the two alternating. Even when it is live, it is the temperature of air at one point, and that point was chosen for controlling the machine rather than for describing your groceries. Air at the return is not air at the top shelf, and neither is the same as the temperature of a dense jar that has been sitting in the door since Tuesday. A gap between the two is normal and not a defect. It becomes a fault when it grows, and it grows for a small number of reasons: a thermistor whose resistance has drifted with age, ice forming over the sensor or the evaporator serving it, or a fan that has stopped circulating, leaving the sensor in a pocket of cold air while the rest of the compartment quietly is not.

• Setpoint and actual are not the same number, and some panels only show the first
• A sensor reads air at one location, chosen to control the machine
• Dense food lags air temperature by hours, which is why a single spot reading misleads
• A widening gap has three usual causes: a drifted sensor, ice over it, or air that has stopped moving

The faults that never change the number on the display

The expensive faults on this equipment are the quiet ones, and they are quiet precisely because the control is still winning. A condenser matted with dust holds temperature perfectly right up until the afternoon it cannot. A gasket leaking at one corner shows up as longer run time and frost on the evaporator long before it shows up as a warm compartment. A defrost drain that froze shut puts water into the cabinet base with no alarm anywhere in the system, because nothing in the machine is watching for it. An ice machine with a restricted coil halves its output rather than stopping, so nobody reports a breakdown. Early refrigerant loss looks like a unit that simply runs more than it used to. In every one of those cases the display reports success, and it is buying that success with run time nobody is measuring.

• A matted condenser holds temperature until the day it does not
• A gasket leaking at one corner shows as run time and frost, not as a warm reading
• A frozen defrost drain floods the cabinet base with no alarm at all
• A restricted ice machine halves output instead of failing
• Early refrigerant loss reads as a unit that just runs more

Run time is the number you want, and no panel gives it to you

What an experienced technician is really estimating on arrival is duty cycle: what fraction of the hour the compressor is running to hold the setpoint. Nothing on the front of the appliance reports it, but an owner can approximate it well enough to be useful. Sit in the kitchen for an evening and note when the unit is audible and when it is not. A built-in that never stops running in a comfortable room has something wrong with it even if every compartment reads perfectly, and a unit that used to cycle and now runs continuously has changed, which is itself the diagnosis. Long runs are normal after a big grocery load, after a door has been held open, during and after a defrost cycle, and on a hot day.

• Note when the compressor is audible over an evening - that is a usable duty cycle
• Continuous running in a comfortable room is a fault even when the panel reads correctly
• A change in run time is more informative than any single temperature
• Long runs after groceries, a held-open door, a defrost cycle or a hot day are normal

Measuring the cabinet properly, which takes a day and costs nothing

Put a probe or a simple dial thermometer in a glass of water on the middle shelf. Water has thermal mass, so it reports what food experiences rather than what the last door opening did, and that single choice is the difference between a reading worth acting on and a number that changes every time somebody wants a drink. Leave it 24 hours and read it in the morning before the door has been opened. In the freezer, tuck the probe between packages instead of hanging it in open air. In wine storage, take it at rack height in each zone, top and bottom. Then write the reading down beside the setpoint that produced it, because one without the other says nothing, and repeat the whole exercise a week later before concluding anything at all.

• Fresh food: a probe in a glass of water on the middle shelf, 24 hours, read before the first opening
• Freezer: between packages, not hanging in air
• Wine: at rack height, top and bottom of each zone
• Record the setpoint alongside the reading - either number alone is meaningless
• Repeat a week later; one reading is a snapshot, two are a pattern

What the display is genuinely good for

None of this makes the panel useless. It is the only part of the appliance that records events, and events are information a thermometer cannot give you. Door-ajar warnings, high-temperature indications and, on the generations that keep them, a sign that power was lost while nobody was home are all real evidence, particularly in a house that stands empty for stretches. A panel that has gone blank, become unresponsive, or is showing something it has never shown before is telling you about power and about the control itself rather than about temperature. The useful discipline is to write down exactly what it shows, whether it flashes or holds steady, whether it clears after power is cycled, and what the appliance was doing when it appeared. Indications differ between generations, so guessing at a meaning from a forum post is how people buy boards they did not need.

• Door and high-temperature warnings are events, and events are what a thermometer cannot record
• A power-loss indication matters most in a house that is not occupied continuously
• A blank or unresponsive panel is a power and control question, not a temperature one
• Write down exactly what is displayed, whether it flashes, and what the unit was doing

What to do when the two numbers disagree

A difference of two or three degrees, measured properly over 24 hours and confirmed a week later, is real and worth acting on. There are only two families of explanation. Either the sensor is telling the board something untrue, or air has stopped moving and both numbers are correct in different places. Check the cheap things first: the grille clear and the condenser fins visible to a light, the gasket tested around the full perimeter with a strip of paper, no ice over the evaporator cover, and nothing inside the compartment parked against a vent. One more piece of narrowing helps on Sub-Zero built-ins specifically. The fresh food and freezer compartments run as independent sealed systems with no air passing between them, so a freezer holding perfectly beside a fresh food side that is not has already excluded half the machine. And do not fix a disagreement by moving the setpoint: the control is obeying the same sensor either way, and all that changes is where the error lands.

• Confirm the difference twice before acting on it
• Grille, condenser, gasket, evaporator ice and blocked vents, in that order
• One compartment right and the other wrong narrows the fault by half on a dual-refrigeration built-in
• Never compensate with the setpoint - it hides the fault and can freeze produce