Is frost on a dehumidifier normal?

Occasional frost is described in published maker guidance as common under certain conditions and not necessarily a sign that there is an issue. The useful question is not whether it is normal but where it sits against the published temperatures. The industry test standard keeps units above 62 degrees before measurement specifically to guarantee there is no frost on the coils, and in support of that rulemaking 13 units were tested with none entering defrost at any point. Frost in service is therefore a reading of the room.

What temperature does a dehumidifier freeze up at?

No published document we found gives a single frost temperature, and we are not going to invent one. What is published is the gap. The capacity label is measured at 65 degrees Fahrenheit; maker guidance gives the working floor as above 41 degrees; and the test standard keeps units above 62 degrees before measurement to ensure no frost on the coils. So the band where the coil can frost is the stretch of the advertised window that sits below the rating condition, and how far into it any particular machine frosts is not published anywhere we could find.

How do I stop my dehumidifier from freezing up?

Four published steps, in the maker's own order. Clean the filter first, because a loaded filter cuts the air moving across the coil and a coil with less air over it runs colder. Then unplug the unit and let the frozen moisture melt. Then check the room against the published window of above 41 degrees and below 96 degrees, because a space below the rating condition will do this again. If the space is genuinely cold, the published guidance points to a unit specified for low temperatures rather than a larger one.

Should I chip the ice off my dehumidifier?

No. The published remedy is to unplug the unit and allow the frozen moisture to melt, and the same guidance warns against using anything mechanical to speed that up. The coil is thin tubing, and the cost of a mistake there is a repair rather than a nuisance.

Why does my dehumidifier freeze up and stop collecting water?

Because frost costs you twice. Published discussion of the test procedure describes frost build-up on the evaporator coils as reducing performance directly, and as triggering periods of defrost during which a test unit may shut the compressor off. So the coil first stops passing air, and then the defrost cycle stops the machine making water at all. A frosting unit is not merely underperforming; part of its running time goes on recovering rather than dehumidifying.

My dehumidifier freezes up but the room is warm. What then?

Then airflow is the variable rather than the room temperature, and the filter is the first thing to check. The coil's surface temperature falls as the air moving across it falls, because less moving air means less heat carried back into the coil, so a blocked filter or a dust-packed coil face can put the coil below freezing in a room that feels warm. That reading of the mechanism is ours; the ordering that puts the filter first is the maker's.

Does a freezing dehumidifier mean it is broken?

Not on its own. Frost is evidence that the coil is cold and working, which is the opposite of a failed compressor. The observation that separates the two is on the coil behind the filter: frost means the cold is being produced and the room is the variable, while a coil that never feels cold at all points to the refrigerant or compressor path, and to service rather than troubleshooting.

Why is my dehumidifier freezing up?

Frost on the evaporator coil is not a fault. It is the machine reporting the temperature of the room, and the laboratory rulebook that produces the capacity label is built specifically to make sure frost never happens at the condition that number describes. Here is what the frost means, what it costs you, and what actually clears it.

Research review Goods Review Lab Last updated 17 October 2026 Frost, defrost and the pre-test conditions from the US DOE dehumidifier test procedure; the remedy from published maker guidance

In short

The short answer: A coil that frosts is not a broken machine. It is a machine running below the temperature its capacity was measured at, so moisture freezes onto the evaporator instead of draining off it. The unit is doing the same thing it always does; the room changed what happens at the end of the process.
The two temperatures: The capacity figure is measured at 65 °F and 60% relative humidity. Published maker guidance gives the window the unit is sold to run in as above 41 °F and below 96 °F. Everything between 41 and 65 is a band the unit is advertised to run in and the label does not describe.
The laboratory's own rule: The industry test standard requires that a unit is not exposed to below 62 °F during its run-in and pre-stabilisation period. The stated purpose is to ensure there is no frost on the coils before measurement. The rule exists to protect a number, not to protect your machine.
What the test showed: In support of that rulemaking the Department of Energy tested 13 units, and none entered defrost operation at any point during the test. Frost in service is therefore a statement about the room, not about the appliance.
What it costs: Twice. Frost blocks the air you are trying to dry from reaching the cold surface, so output falls. Then the defrost cycle that clears it switches the compressor off — so the unit spends part of its run doing neither of the two jobs you bought it for.
What to do about it: Unplug it and let the frozen moisture melt; the published guidance warns against using anything mechanical to hurry that along. Then check the filter, because the maker's own troubleshooting order puts the filter ahead of the frost.

Why the coil freezes at all

The evaporator coil is meant to run cold. That is the whole mechanism: air is pulled across a surface colder than the air, the water in it condenses on that surface, and the water then drips into the tank or down the drain. A coil that is not cold removes nothing.

So the machine is deliberately built to sit below the air temperature. In a warm room that still leaves the coil above freezing and the water drains away. In a cold room the same coil, working the same way, drops below freezing — and then the water does not drip off it, it sticks to it. Frost is not a second failure mode bolted on to the design. It is the ordinary mechanism of condensation arriving at the wrong temperature.

There is a second lever on the same coil, and it is why airflow matters. The coil's surface temperature falls as the air moving across it falls, because less moving air means less heat transferred back into the coil. A loaded filter is the usual way that happens in a machine that used to work. That reading is ours; the ordering it explains is the maker's.

Four published temperatures, one scale

Same machine, same document set. The shaded stretch is the part of the advertised operating window that sits below the condition the capacity label was measured at — which is the stretch where the coil can frost.

Where each published figure sits, on a 0–100 °F axis

Working floor — maker guidance 41 °F

No-frost floor — the laboratory's own rule 62 °F

Label measured here — and best performance begins 65 °F

Working ceiling — maker guidance 96 °F

050100 °F

The four temperatures are the published ones. Drawing them on a single 0–100 °F axis so they can be read against each other is our arrangement, and it is the point of the figure. A basement that sits at 50 °F is comfortably inside the window the unit is sold for, and twelve degrees below the point the laboratory will not let its own test units go below. Nobody advertises that gap, because the two figures come from different documents written for different readers.

What the test procedure does to keep frost out of the number

This is the part that turns a common household nuisance into something you can read. The people who write the test that produces the pint figure know about frost, and their published discussion says so directly. They are not describing your basement; they are describing a measurement, and frost is treated as contamination of it.

Frost as the laboratory sees it

The rule they wrote: The current industry test standard requires that a unit is not exposed to temperatures below 62 °F during the run-in and pre-stabilisation period — the stage before measurement begins.
Why, in the department's own words: The provision is intended to ensure there is no frost build-up on the evaporator coils prior to testing, which could directly reduce performance during the test or result in periods of defrost during which a test unit may shut off the compressor, resulting in a further reduction in measured efficiency and performance.
So frost is a two-part loss, on the record: That sentence is the whole page in one line. Frost is not one problem. It reduces performance by sitting on the coil, and then the defrost cycle that clears it takes the compressor offline — a second loss stacked on the first.
What they observed: In support of the rulemaking, 13 units were tested and none of them entered defrost operation at any point during the test. On that evidence the department concluded that defrost operation is uncharacteristic of dehumidifier operation under its test procedure.
And then they declined to adopt the rule: Because frost does not develop in their test, the additional test time and requirements would not change the performance the procedure measures, so the provision was judged unduly burdensome. Read that the right way round: the rule exists to keep frost away from a number, and the reason it was dropped is that frost never comes near the number anyway.

Put the two halves together and the conclusion is not really about frost at all. The label on the box describes a frost-free machine in a frost-free condition. The unit you own was sold into a window that reaches down to 41 °F. If your coil is frosting, you have found the seam between those two facts, and it runs the length of the published material without ever being drawn.

The cost of frost is paid twice

The first payment is the coil. Frost is an insulating layer between the air and the cold surface the air is supposed to meet. As it builds, less of your room's air actually reaches the metal, so the machine keeps running, the fan keeps turning, and the amount of water coming off falls away. Nothing about that is visible from the front of the unit.

The second payment is the compressor. The published description has the defrost cycle shutting the compressor off. While that cycle runs, the unit is removing no water at all — it is spending your electricity getting itself back to the state it should have been in.

There is a published consequence of that second payment which is worth more than it looks. A comment filed on the rulemaking argued that a unit which enters a defrost cycle needs a four- or six-hour test rather than a two-hour one to be judged fairly, because a defrosting unit's output over a short window is not representative of it. The same logic applies to your own judgement: if your unit is frosting, watching it for twenty minutes tells you nothing at all about how much water it can produce.

What you are seeing, what is actually happening, and the first thing to check
What you are seeing	What is actually happening	First check
Ice or frost visible on the coils behind the filter	Moisture is freezing onto the evaporator instead of draining off it, so the coil is below freezing	The temperature of the room, then the filter
Runs, goes quiet for a stretch, runs again, little or no water	Defrost. The compressor is off for the length of the cycle, so no water is being made during it	Whether the coil was frosted before the pause
Frost appears after months of use, not from new	The coil is running colder than it used to, which usually means less air is moving across it	The filter, then the fan and the face of the coil
Frost in a room that feels warm	Airflow, not room temperature, is the variable — a blocked filter or a dust-packed coil can drop coil temperature anywhere	The filter first, before you assume the room

Frost, or a dead compressor?

From a metre away these two look identical: the fan is turning, the machine is humming, and the tank is dry. They are separated by one observation, and it is on the coil rather than on the panel.

Look behind the filter. If there is frost or ice on the coil, the machine is cold and working; the room is the variable, and the rest of this page applies. If there is no frost and the coil never feels cold at all, the cold is not being produced, and that is the refrigerant or compressor path — a service question rather than a troubleshooting one. Our own pages stop there, and so should a troubleshooting list.

One more thing mimics a failed compressor and is worth knowing before you conclude anything. Maker manuals state that the compressor circuit has an automatic three-minute delayed start if the unit is switched off and straight back on, to protect it. Switch it off, switch it on, and you will hear nothing for three minutes. That is the machine protecting itself, not the machine failing.

What actually clears it

Unplug it and let the frozen moisture melt. That is the published remedy, and the same guidance warns against using anything mechanical to speed it up — the coil is thin tubing and a screwdriver is how people turn a nuisance into a repair.
Clean the filter while you are in there. The maker's own troubleshooting order puts the filter ahead of the frost, and the reading above is why that ordering makes sense rather than being arbitrary.
Then look at the temperature of the room, because that is the actual variable. Published maker guidance is explicit that below the ideal figure the unit still runs and simply pulls less moisture out of the air, and that the appliance should not be run at all outside its stated range.
If the space is genuinely cold, the answer is a different kind of machine rather than a bigger one. The published guidance for a cold space points to a unit specified for low temperatures. From the equipment side, a restoration supplier's sizing tool switches its recommendation from a refrigerant unit to a desiccant unit below 60 °F — and labels that figure a planning number for drying jobs rather than a product rating. We are quoting it as exactly that, because it is the only published temperature boundary we found that is attached to a machine-type decision at all.
Raise the room temperature if you can. It is the cheapest item on this list and the only one that addresses the cause instead of the symptom. Frost is a temperature report; the only thing that turns the report off is the temperature.

What this page does not cover

A compressor that has failed or lost its refrigerant. Same signature — fan turning, no water, no frost — and the answer is service rather than anything on this page.
A unit that never powers on at all. That is a power or switch fault, and no amount of coil inspection will find it. The four states this complaint can be, and which one you are in, are sorted on dehumidifier not working.
Sizing the unit in the first place. That is a different question with its own published table, covered on the calculator and on the pints page.
Water arriving through a wall or a floor. No setting, no size and no machine type wins that argument. A unit that runs continuously and never catches up is describing a drainage problem rather than a capacity one.

Questions this page answers

Why is my dehumidifier freezing up?: Because the evaporator coil is running below freezing. The coil is meant to sit colder than the air so moisture condenses on it, and in a cold room that same coil drops below freezing and the moisture freezes onto it instead of draining off. The capacity label is measured at 65 degrees Fahrenheit and 60 percent relative humidity, while published maker guidance gives the operating window as above 41 degrees and below 96 degrees, so a room between those two figures is inside the window the unit is sold for and below the condition its number describes.
Is frost on a dehumidifier normal?: Occasional frost is described in published maker guidance as common under certain conditions and not necessarily a sign that there is an issue. The useful question is not whether it is normal but where it sits against the published temperatures. The industry test standard keeps units above 62 degrees before measurement specifically to guarantee there is no frost on the coils, and in support of that rulemaking 13 units were tested with none entering defrost at any point. Frost in service is therefore a reading of the room.
What temperature does a dehumidifier freeze up at?: No published document we found gives a single frost temperature, and we are not going to invent one. What is published is the gap. The capacity label is measured at 65 degrees Fahrenheit; maker guidance gives the working floor as above 41 degrees; and the test standard keeps units above 62 degrees before measurement to ensure no frost on the coils. So the band where the coil can frost is the stretch of the advertised window that sits below the rating condition, and how far into it any particular machine frosts is not published anywhere we could find.
How do I stop my dehumidifier from freezing up?: Four published steps, in the maker's own order. Clean the filter first, because a loaded filter cuts the air moving across the coil and a coil with less air over it runs colder. Then unplug the unit and let the frozen moisture melt. Then check the room against the published window of above 41 degrees and below 96 degrees, because a space below the rating condition will do this again. If the space is genuinely cold, the published guidance points to a unit specified for low temperatures rather than a larger one.
Should I chip the ice off my dehumidifier?: No. The published remedy is to unplug the unit and allow the frozen moisture to melt, and the same guidance warns against using anything mechanical to speed that up. The coil is thin tubing, and the cost of a mistake there is a repair rather than a nuisance.
Why does my dehumidifier freeze up and stop collecting water?: Because frost costs you twice. Published discussion of the test procedure describes frost build-up on the evaporator coils as reducing performance directly, and as triggering periods of defrost during which a test unit may shut the compressor off. So the coil first stops passing air, and then the defrost cycle stops the machine making water at all. A frosting unit is not merely underperforming; part of its running time goes on recovering rather than dehumidifying.
My dehumidifier freezes up but the room is warm. What then?: Then airflow is the variable rather than the room temperature, and the filter is the first thing to check. The coil's surface temperature falls as the air moving across it falls, because less moving air means less heat carried back into the coil, so a blocked filter or a dust-packed coil face can put the coil below freezing in a room that feels warm. That reading of the mechanism is ours; the ordering that puts the filter first is the maker's.
Does a freezing dehumidifier mean it is broken?: Not on its own. Frost is evidence that the coil is cold and working, which is the opposite of a failed compressor. The observation that separates the two is on the coil behind the filter: frost means the cold is being produced and the room is the variable, while a coil that never feels cold at all points to the refrigerant or compressor path, and to service rather than troubleshooting.

Where these numbers come from

The fixed test ambient of 65 °F dry-bulb and 60% relative humidity for a portable unit, the 62 °F floor a unit may not be exposed to before measurement, the description of frost build-up on the evaporator coils as something that can directly reduce performance and trigger periods of defrost during which a unit may shut off the compressor, the conclusion that defrost operation is uncharacteristic of dehumidifier operation at the 65 °F test condition, and the figure of 13 units tested with none entering defrost are all from the US Department of Energy's dehumidifier test procedure rulemaking and the discussion published with it; the procedure itself is 10 CFR 430 Appendix X1, and the industry standard the discussion refers to is AHAM DH-1-2022. The operating window of above 41 °F and below 96 °F, the note that performance is reduced below 65 °F, the statement that occasional frost build-up is common under certain conditions and not necessarily a sign that there is an issue, the remedy of unplugging the unit to let the frozen moisture melt, and the warning against using anything mechanical to speed that up are from published maker support and manual documentation, read from two separate manufacturers whose figures agree. The automatic three-minute compressor delayed start is from the same maker manual documentation. The 60 °F boundary between refrigerant and desiccant equipment is from a restoration equipment supplier's sizing tool, which labels it a planning figure rather than a product rating. The single-axis arrangement of the four temperatures, the ordering of the checks, the two-part framing of what frost costs, and the reading of why a loaded filter matters are our own working and are labelled as such. Nothing on this page is a test result, and no number here was estimated.