Why Is My Clear Coat Cloudy or Milky?
Three unrelated problems produce the same milky finish, and they need three different repairs. Blush lies on top of the film and washes off with water. Micro foam is millions of tiny bubbles trapped inside it. Moisture comes up through the slab, and it will do it again. Work out which one you have before you touch the floor.
A milky clear coat is three different problems wearing the same face, and the one thing they have in common is the thing that separates them: where the cloudiness sits. Amine blush lies on top of the film and washes off with warm water. Micro foam is millions of tiny bubbles cured inside the film, and no amount of washing touches it. Slab moisture works from underneath, and it is the only one of the three that gets worse. Start with the wash test, because it is free and it clears the most common cause off the list in a minute.
The three way test
Work through these in order. Each step only matters if the one before it came back negative, and the order is deliberate: cheapest and most likely first, most expensive to be wrong about last.
1. Wash a patch
- Warm water, a mild detergent and a white scrub pad, over an area about the size of your hand
- Dry it properly and look again in raking light, because a wet floor looks clear whatever is wrong with it
- Haze lifts and the coating under it is clear: it was amine blush, and blush is water soluble by definition
- Haze stays: the cloudiness is not on the surface. Go to step 2
2. Look into the film
- Hold a bright light low and rake it across the floor, and use a magnifier if you have one
- You are looking for structure rather than colour: discrete round voids suspended in the film
- Tiny bubbles you can resolve individually: micro foam, whipped in at the bucket and never released
- A soft haze with nothing to focus on: not foam. Go to step 3
3. Read the map
- Blush and foam are spread by how the floor was worked. Moisture is spread by the slab
- Haze that maps to the damp end of a building, a cold external wall, a crack, a joint or a slab edge is pointing at what is under it
- Confirm with a number, not a look: a relative humidity probe to ASTM F2170, or a calcium chloride test to ASTM F1869
- This is the branch worth being certain about, because it is the only one that keeps going
Branch one: amine blush
The curing agent in an epoxy is an amine, and if it reaches the surface while the air is damp it reacts with water and carbon dioxide instead of with the resin. What that reaction leaves behind is a salt, which is why the fix is water rather than solvent, and why a floor can be perfectly sound underneath something that looks ruined. Conditions during cure drive it: cold slabs, damp air, and a poorly ventilated space where moisture has nowhere to go.
The controlling number is not the humidity on its own, it is the gap between the slab and the dew point of the air. Cross that gap and moisture condenses onto the surface the coating is trying to cure on.
Was the slab far enough above dew point?
Blush is a conditions failure before it is anything else, and conditions are the one piece of evidence that disappears within a day. Enter the air temperature, the relative humidity and the slab temperature to get the dew point and the spread you actually had.
The industry minimum spread is 5F, and this site flags anything under 9F as marginal
Open the live calculator →The published minimum is 5F of spread between the slab and the dew point, which is the figure SSPC-PA 3 and ASTM D3276 rest on and the one most data sheets repeat. The 9Fthreshold above it is our own caution margin rather than an industry rule, and it exists because a slab that is 5F clear at nine in the morning is often not clear at four in the afternoon. If you were working inside that margin, blush is a reasonable explanation.
Humidity has a separate say. The cure time calculator raises a blushing flag once the air is above 85% relative humidity, because several polyaspartics start hazing there regardless of what the slab temperature is doing. Treat that as a second independent check rather than as a substitute for the spread.
Two different humidities, and mixing them up sends you to the wrong branch. The figure above is air humidity, measured with a hygrometer held in the room, and it belongs to the blush branch. The slab readings further down are in situ relative humidity, measured with a probe sunk into a hole drilled in the concrete, and they belong to the moisture branch. They are different instruments measuring different things, and a floor can be fine on one and failing on the other.
The repair is the cheapest on this page. Wash it off with warm water and a mild detergent, rinse, and let it dry fully. Do not reach for solvent, which spreads a water soluble film rather than lifting it, and do not sand before washing, which grinds a contaminant down into the profile you have just opened. The full method, including what to do when the recoat window has already closed, is on the blushing page.
Branch two: micro foam
Micro foam is air that was beaten into the mix and never got out again. It is the same defect as bubbling, only at a scale where you read it as a haze rather than as domes, and it shades into pinholes when the bubbles do reach the surface and the film is too stiff to close behind them. All three are the same story told at different sizes.
Four things put the air there, and all four happen at the bucket rather than on the floor:
- Mixing too fast. A drill run flat out with a whisk or a paddle lifted near the surface folds air in continuously. Slow, and keep the paddle buried
- Skipping induction. Many products want the mixed material to stand for a stated period before it goes down, and one of the things that happens in that time is entrained air rising out
- Pouring deeper than the rating. A thicker pour gives a bubble further to travel and gels sooner in the middle, so what is in there stays in there
- Cold material. Cold resin is more viscous, and a more viscous mix holds onto air. This is why the same technique that works in July foams in November
The repair is the one people least want to hear. The bubbles are cured into the film, so nothing lifts them and nothing hides them. A clear recoat leaves them visible through the new layer, which is why recoating without diagnosing first wastes a second set of material. The floor is abraded back far enough to remove the foamed layer and rebuilt, or it is accepted as it is. Getting the mixing right is much cheaper than either, so worth being precise about on the next batch.
Branch three: moisture from the slab
Concrete holds and passes water vapour, and a coating is a barrier laid across the top of that. Where the slab is pushing more vapour than the coating can tolerate, the moisture arrives at the underside of the film and clouds it, usually blotchily and usually in the places you would predict: the damp end, the cold wall, the crack, the joint, the edge.
This branch behaves differently from the other two, and that difference is the diagnosis. Blush is over once cure finishes. Foam is fixed in place the moment the film gels. Moisture keeps going, because the slab keeps doing what it was doing. Haze that returns after a wash, spreads over weeks, or turns into blisters is this branch and not the others.
Test the slab, then judge the reading
A cloudy floor over a wet slab is an early symptom, and blistering is what follows. Enter a relative humidity or calcium chloride result and the tool judges it against the bands that decide whether a slab is fit to coat, needs a barrier, or needs the source found first.
Under 75% is clear for most epoxy systems, and everything above it needs a decision
Open the live calculator →A reading under 75% in situ relative humidity is clear for most epoxy systems, and if your slab reads there then moisture is very unlikely to be what clouded the floor. Look again at the first two branches. Above it the answer stops being a single number and starts being a conversation with the product data sheet, and high enough readings call for a barrier or for finding the source before anything else is coated. The bands and both test methods are set out on the moisture vapor barrier guide.
The repair depends on the reading rather than on the haze. Recoating is the wrong move here and worth naming plainly: another layer is another barrier, so the vapour pressure that caused the cloudiness builds against a thicker film and comes back as blistering or as loss of adhesion. A floor over a slab that tests high comes off, the slab gets a barrier sized to the reading, and the system goes back down on top of that.
If it is none of the three
Two other things get called cloudy and are not on this page. A film that is hazy and also soft, tacky or printing under a boot is a cure problem rather than a clarity problem, and belongs on epoxy not curing. A floor that has gone amber or yellow rather than milky changed colour after it cured, which is a different mechanism with a different cause, and belongs on yellowing.
If none of that fits what is in front of you, start from the floor rather than from a guess and work back: identify your coating defect is organised by what you can see from standing height.
Preventing all three on the next floor
- Test the slab before you quote. A number from a probe or a calcium chloride kit is the only thing that turns the moisture branch from an argument into a fact, and it costs almost nothing compared with removing a floor
- Take the slab temperature, not the room temperature. Concrete holds cold long after the air in a space has been warmed, and the slab is the surface the dew point rule is about
- Log the conditions as you coat. Air temperature, slab temperature and relative humidity, written down. All three vanish within a day, and without them a blush diagnosis is guesswork
- Mix slowly and let it stand. Paddle below the surface, the induction time the data sheet asks for, and the rated thickness rather than a deeper pour
- Ventilate while it cures. Moving air carries away the moisture that feeds blush, and a sealed room during a damp spell is the condition that produces it
Frequently Asked Questions
What is the fastest way to tell the three apart?
Ask where the cloudiness sits, because each cause puts it in a different place, and one test answers it.
Wash a patch about the size of your hand with warm water and a white scrub pad, then dry it and look again. If the haze lifts and the coating under it is clear and sound, it was amine blush lying on the surface, and blush is water soluble by definition.
If it does not lift, the cloudiness is inside or underneath the film, and the next look is with a bright light held low and raked across the floor, ideally with a magnifier. Discrete tiny bubbles suspended in the film mean micro foam. A soft blotchy haze with no visible bubbles, which maps to damp areas, cold spots, cracks or slab edges, means moisture.
Do the wash test first every time. It is free, it takes a minute, and it removes the most common cause from the list before you spend anything.
Can I just recoat over it?
Not until you know which one it is, because recoating fixes one of the three and makes the other two worse.
Over blush, a recoat bonds to a contamination layer rather than to the coating, and that is one of the most common reasons a topcoat peels months later. Wash the blush off first and the recoat is fine.
Over micro foam, a recoat hides nothing. The bubbles are in the cured film underneath and they stay visible through a clear topcoat, so you have paid twice for the same defect.
Over slab moisture, a recoat is actively harmful. Another layer of coating is another vapour barrier, so the pressure that caused the haze builds against a thicker film and comes back as blistering or delamination rather than as a cosmetic problem.
Is a cloudy clear coat always the installer’s fault?
Two of the three usually are, and the third often is not.
Micro foam is a technique problem almost every time. Mixing too fast or with the wrong paddle whips air in, skipping the induction period means it never releases, and pouring the mix thicker than the product is rated for traps what is left. All three are choices made on the day.
Blush is a conditions problem, and conditions are the installer’s responsibility to check. The information needed to avoid it is on the data sheet and on a hygrometer, so coating a slab that was close to dew point is a decision rather than bad luck.
Slab moisture is the one that is often nobody’s fault on site, provided the slab was tested. A slab that reads dry and later pushes moisture usually has a building problem behind it: a failed vapour retarder, a leak, or drainage running the wrong way. A slab that was never tested is a different conversation.
Why does solvent not clear a hazy floor?
Because none of the three causes is an oily residue, which is the only thing wiping with solvent actually removes.
Blush is a salt formed when the amine curing agent reaches the surface and reacts with water and carbon dioxide in the air instead of with the resin. Salts dissolve in water, not in solvent. Solvent softens the blush and spreads it, so you often finish with a thinner haze over a wider area plus solvent residue to deal with.
Micro foam and slab moisture are not on the surface at all, so there is nothing there for a solvent to reach. Wiping a floor that has bubbles cured into it changes nothing except the gloss.
Warm water and a mild detergent is the correct first move, and it is also the cheapest one.
The haze came back after I washed it off. What does that mean?
It means you have found the useful piece of evidence, so do not treat it as a failed wash.
Blush that is removed does not return on a floor that has finished curing, because the reaction that produced it is over. Haze that comes back is either a floor still curing in damp conditions, in which case it stops once cure completes and the conditions improve, or it is not blush at all and you cleared a light surface film that was sitting over the real problem.
Moisture from the slab is the cause that reliably returns, because the driver never went away. If the haze reappears in the same places, and those places are the damp end of the floor, the cold wall or the line of a crack, test the slab before doing anything else.
Does a cloudy clear coat weaken the floor or is it only cosmetic?
It depends entirely on which of the three it is, which is the whole reason this page exists.
Blush on its own is cosmetic. It becomes structural the moment you coat over it, because the next coat bonds to the blush rather than to the epoxy.
Micro foam is mostly cosmetic but not entirely. A film full of tiny voids has less material in it than the specification assumes, and the voids concentrate stress, so a badly foamed topcoat wears through sooner than a sound one.
Moisture is the serious one. Haze from slab moisture is an early symptom rather than the problem, and what follows is usually blistering, then loss of adhesion over the affected area. Treating it as a cosmetic complaint is how a cloudy floor becomes a failed floor.
How do I stop it happening on the next floor?
Each cause has a different control, and none of them is expensive next to a recoat.
For blush, coat when the slab is comfortably above dew point rather than marginally above it, and watch the humidity as well as the temperature. A slab cooling towards the dew point overnight is the classic case.
For micro foam, mix slowly with a proper mixing paddle rather than a whisk, keep the paddle below the surface, let the mix stand for the induction time on the sheet so entrained air can release, and pour at the rated thickness rather than deeper.
For moisture, test the slab before you quote, not after you coat. A relative humidity probe or a calcium chloride test is the only thing that turns a guess into a number, and the number decides whether the job needs a barrier.