Ice maker faults: module or water supply, and how to tell

What the ice in the bin is telling you

An automatic ice maker runs a fixed sequence, and every complaint you can have about one belongs to a single step in it. The inlet valve, usually in the machine compartment behind the grille or at the back of the cabinet, opens for a set number of seconds and the mold fills. The water freezes. A thermostat on the mold senses that the ice is cold enough to release, the mold heater warms the metal for a moment so the cubes let go, and a motor turns the ejector blades around to sweep them into the bin. A shutoff arm, or a bin sensor on the later generations, stops production when the bin is full. Knowing that sequence turns a vague complaint into a location, and the shape of what lands in the bin is usually enough to place it without picking up a tool.

• Small, thin or hollow cubes, correctly shaped but short: the mold received less water than the fill was timed for, which is a supply fault every time
• One fused sheet or a slab across the whole mold: too much water arrived, usually an inlet valve passing slowly when it should be shut
• A mound of ice climbing up from the back of the mold: water freezing in the fill tube before it ever reaches the mold
• Full, properly formed cubes sitting in the mold and never leaving it: the fill is fine and the harvest has stopped, which is the module
• A dry mold in a freezer that is otherwise hard: either no water is arriving, or the valve is never being asked to open
• Correct cubes arriving too slowly: neither fault, and usually a freezer temperature or airflow question instead

The test that settles it, and the one thing not to do

Pour cold water into the mold by hand, slowly, stopping below the top of the dividers, then shut the door and leave it alone for a couple of hours. If a full set of properly formed cubes turns up in the bin, the module has just proved every part of itself: the thermostat sensed, the heater released, the motor and gear train turned, the ejector swept. Whatever is wrong sits upstream of the mold, and nothing inside the head needs buying. If the water freezes into a solid block and stays there, or nothing moves at all, the fault is in the head and the water side is a red herring. Two cautions go with that test. Do not overfill, because a mold flooded past the dividers gives you a fused sheet and a false answer. And never turn the ejector blades by hand: they are driven through a plastic cam and gear train, and forcing them, particularly backward, is how a module that needed a thermostat becomes a module that needs replacing.

• Cold water, poured slowly, stopping just below the top of the dividers
• Freezer door shut, and two hours of patience before you conclude anything
• Never rotate the ejector blades by hand, and never pry ice off the mold with anything metal
• If you open the freezer and find the mold already full of water, you have your answer without running the test at all
• None of this applies to an undercounter clear-ice machine, which has no mold and no ejector: those build ice by flooding a plate from a reservoir and fail in an entirely different way

If it is the water side

The most useful thing to understand about a fill fault is that the fill is timed, not measured. The valve opens for a fixed number of seconds and whatever arrives in that window becomes the cube, so a supply that is weak rather than dead produces a perfectly regular cube that is simply too small. The appliance is behaving exactly as designed, with less water than the design assumed, which is why nothing about it looks broken from the front. It is also why the sentence people offer on the phone, that the water is on and the kitchen faucet runs fine, proves nothing. The faucet is a wide-open line with nothing in its way. The ice maker is fed through a small-bore tube, a screen, usually a filter, and a valve seat, and any one of those can halve the delivered volume without the sink ever noticing. There is a version of this common wherever the ice maker arrived after the house did. A retrofitted feed is often taken off a saddle valve, a clamp that pierces the pipe and seals against a washer, and the hole it makes is narrow when new. Corrosion closes it further, a little at a time, with nothing to mark the change. The cubes shrink across a couple of years, slowly enough that nobody connects the ice to a fitting behind the cabinet.

• Pressure read at the inlet valve itself, static and again while the valve is open, rather than judged at a faucet
• The screen in the valve inlet, where scale and grit collect first and which is cheap to clear
• Filter age, including an inline filter under the sink that the house has forgotten about
• The valve solenoid coil: an open coil means the valve never opens and the mold stays dry
• The line back to the wall: kinks behind the cabinet, a partly closed stop, a corroded saddle valve tap on an older install
• Whether the feed comes off a reverse osmosis unit, which delivers water at far lower pressure than a straight cold line
• The fill tube, where a partial ice plug produces a short fill and a mound of ice underneath it at the same time

If it is the module, and where the money goes

The head above the mold holds a small motor and a plastic gear train, a cam that drives both the ejector and the shutoff, a thermostat that reads mold temperature to start harvest, a mold heater that releases the ice, and a switch that calls for the next fill. They fail in recognizable ways. An open mold heater leaves perfect cubes welded in place. A stalled motor leaves them formed and unmoved. A thermostat that never closes means the cycle simply never begins, so the mold sits full and frozen with nothing else happening around it. Confirming which one is meter work at the module connector, and that is genuinely a technician's job rather than false modesty about it: the head carries line voltage inside a freezer compartment, and its test points are not somewhere to experiment. What matters to an owner is the order of spending. The inlet valve is among the cheapest parts on the appliance and the supply line is plumbing. The module, or on much of the newer equipment the complete ice maker assembly, is a mid-priced part and is increasingly sold as one piece rather than as serviceable internals. Fitting a new assembly to a line that was never delivering enough water gets you exactly the ice the old one made, and that is the most common way this single fault gets paid for twice. If the test points at the head rather than the supply, that is ordinary work: ice maker repair in Rancho Bernardo, San Diego covers both halves of it, on the legacy mechanical makers in the older streets and on the clear-ice machines in the newer kitchens across this corner of San Diego County.

• Perfect cubes stuck fast in the mold: suspect the mold heater
• Cubes formed and never swept out: suspect the motor or the gear train
• Mold full, frozen, and nothing ever starting: suspect the thermostat
• Water in the bin, or one fused slab: suspect the inlet valve passing when it should be shut, not the module
• Prove the water side before buying the head, because on newer units you are buying a whole assembly on the strength of that conclusion