How to Tell if Concrete Is Sealed
The water drop test answers one question well and misses one treatment entirely. A slab can drink water and still be treated, and it can bead water with nothing on it at all.
Drop water on it in several places. Bare concrete darkens and drinks it within seconds; if it beads, something is repelling it. Then keep going, because that test answers one question only. Follow it with a blade and a scratch. A blade separates a film sitting on the slab from a treatment soaked into it, which decides whether removal is even a coherent idea. A scratch catches the case the water test cannot see at all: a silicate densifier is not a repellent, so a densified slab drinks water and is still a much harder floor to grind than the one you priced.
The water test, and exactly where it stops
The water drop test earns its place. It costs seconds, it needs nothing but a bottle, and it is the check this site recommends everywhere prep is discussed, from the four prep gates to why floors peel. Water should darken the concrete and soak in within seconds. If it beads, sits proud, or pulls back into droplets, a coating poured onto that spot will behave much the way the water just did.
What it is measuring, though, is narrower than what people ask of it. It measures whether the surface will accept a liquid. It does not say what is stopping the liquid, it does not say how thick that something is, and it does not say whether the something can be ground off in one pass or three. Those are separate questions with separate answers, and the answers change the quote.
It also has two ways of being wrong, in opposite directions. It can clear a slab that has been treated, and it can condemn a slab that has not been. Both are covered below, and the first one is the expensive one.
Four things get called sealed, and they are not the same problem
“The slab is sealed” is used for four quite different treatments. They were applied for different reasons, they sit in different places relative to the surface, and they defeat different tests. Sorting out which one you have is most of the diagnosis, because it decides the method, the depth and the time.
- Topical sealers are a film on the concrete: acrylic, urethane or epoxy, applied to keep stains and water out. They are the mental image most people have of a sealed floor, and they are the easiest of the four to identify
- Cure and seals are a film too, but applied to fresh concrete for a completely different purpose: holding mix water in long enough for the slab to cure properly. Nearly every new slab has one
- Penetrating sealers are usually silanes or siloxanes. They go into the pore and make it water repellent from the inside, so there is no film to find, nothing to peel, and nothing a stripper can usefully attack
- Silicate densifiers are not sealers at all in the sense the others are. They react with free lime in the concrete and fill the pore with hardened material to reduce dusting and raise wear resistance. They are not water repellent
That last distinction is the one the industry itself is clearest about and site visitors are least likely to expect: a densifier is why polished concrete still needs a separate guard or sealer afterwards. If the densifier repelled anything, the guard would be redundant. It is the same reasoning the grit sequence guide works through from the polishing side, where a densifier is a step you want rather than a discovery you dread.
What each test can and cannot see
Run the water drop first because it is free and it narrows things fast. Then pick the follow up from what it told you, rather than running all of them on every floor.
| Treatment | What it is | Water drop says | What confirms it |
|---|---|---|---|
| Topical sealer | An acrylic, urethane or epoxy film sitting on top of the concrete. | Beads. The film is doing the job it was bought to do. | A blade lifts a curl and leaves a clean edge where the film stops. |
| Cure and seal | A film sprayed onto new concrete to hold water in while it cures. | Beads wherever it survived, and often nowhere else. | The same blade test, run at the walls and under anything that stayed covered. |
| Penetrating sealer | A silane or siloxane lining the pore. Nothing sits on the surface. | Beads, sometimes strongly, with no film there at all. | A blade gives you nothing, yet water still beads. That contradiction is the answer. |
| Silicate densifier | Reacts with free lime and fills the pore with hardened material. | Soaks in. The test clears a slab that has been treated. | A screw skates instead of throwing powder, and a trial patch cuts slowly. |
Three of the four bead water and one of them does not, which is why the water drop cannot be the whole diagnosis. Nothing in this table is a measurement or a product specification. It is a sorting order for a floor you are standing on, and a trial patch settles anything it leaves open.
The blade test is the one that does the most work for the least effort. Press a sharp scraper or a knife into the surface at a shallow angle and try to lift something. A film gives you a curl or a flake and leaves a visible clean edge where it ends. A penetrating treatment gives you nothing at all, because there is nothing there to lift. That sounds like a null result and it is not: a slab that beads water and yields nothing to a blade has told you the treatment is inside the concrete, which is a different job with a different depth.
A solvent spot narrows down which film you have when that matters. Put a little acetone or xylene on a rag, leave it on a small area of the film for a minute, and look. Many acrylics and acrylic based cure and seals soften or go tacky; a cured epoxy or urethane will largely shrug it off. Do this in a spot nobody will look at, with ventilation, and away from ignition sources. Treat the result as a hint rather than an identification, because product chemistry varies and a floor may carry more than one layer.
The scratch test is borrowed from bond vs concrete hardness and used here for a different purpose. Drag a masonry screw across the surface. Soft, untreated concrete marks easily and throws powder. A hardened surface lets the screw skate and leaves a faint line. On that page it is a tooling question. Here it is the test that catches the densifier the water drop cleared.
A trial patch outranks all of them, for the same reason it does on the removal page. Grind a small area and watch what happens: what comes off, how fast, what the swarf looks like, and whether the machine is cutting or polishing. Every test above is a proxy for that, and the patch takes minutes.
The false clear: a slab that drinks water and still costs you
This is the case worth carrying away from the page, because it does not produce a coating failure. It produces a job that runs over.
A silicate densifier leaves the surface absorbent. Water still darkens the slab and still soaks in, often just more slowly than on untreated concrete, so the test everybody trusts returns a clean pass. What the densifier changed was hardness, and hardness is what decides how fast your tooling opens the floor. The slab you quoted at a normal rate is now a slab that resists, and the bond you chose for it may be the wrong one, which turns a slow pass into a glazed set and a pass that is not cutting at all.
Nothing on the pour record warns you. Compressive strength is not surface hardness, and a densifier is applied long after the concrete is specified. The tells are the floor’s history and the floor itself: an unusually smooth, slightly glossy sheen on an old industrial slab, a warehouse that was polished or dust proofed years ago, a screw that skates. Any of those, and the honest move is a trial patch before the rate goes on a quote.
The false alarm: a bead that is not a sealer
The opposite mistake is cheaper but not free, because it sells a customer a removal they do not need and puts hours in a quote that will not be worked.
- The slab is cold. Water sits on a cold surface far longer than on a warm one, and a slab in an unheated building in winter is a slow absorber that looks like a sealed one
- The slab is already wet. Concrete that is damp, or that has not finished drying, has nowhere to put your drop. Testing right after a rinse or a wash guarantees a confusing answer
- The surface is burnished. A slab troweled hard and late has a dense, closed skin with very little open pore at the top. It can hold a bead for a while with nothing applied to it, and it is a real prep problem, just not a chemical one
- Something is on it that is not a sealer. Wax, floor polish, oil or a silicone residue all repel water, and they are removed and diagnosed differently. Oil in particular has to come out before grinding rather than after, or the machine drives it deeper
Test again on a dry area at working temperature, then reach for the blade. If nothing lifts, nothing is sitting on the slab, and the question becomes whether the surface is treated inside the pore or simply closed up. Both come back to a grinding plan rather than a stripping plan, which is a much easier conversation to have at quote time than at week two.
One floor is not one surface
Wherever the diagnosis lands, run it in more than one place, because these treatments are almost never uniform. Sealers get applied to the areas somebody cared about. Cure and seals are designed to break down under UV and traffic, so a bay that has been driven and lit for two years can read perfectly clear down the middle and still be sealed along the walls, in the corners, and under anything that has been sitting there since the slab was poured.
That asymmetry is worth stating plainly on a quote. A floor that tests clear in the open and sealed at the perimeter is not a clear floor with a small exception; it is a floor whose plan has to cover the worst area, because the coating will fail at the perimeter and take the reputation of the whole job with it. The prep pillar makes the same point about contamination generally: one good spot in the middle of the room does not clear the room.
Reprice the grind once you know what is on the slab
A positive result does not add a step so much as change the first pass, which now has removal work ahead of it. Put your own measured rate in rather than the seeded one, because a sealed or densified floor is exactly the case where the assumed rate is what costs you.
The seeded rate assumes bare, untreated concrete. A trial patch is the only thing that gives you the rate for this floor.
Open the live calculator →What changes when the answer is yes
Finding a treatment rarely changes the method, since the answer is almost always mechanical either way. It changes the size of the job, and four things move.
- The first pass is doing two jobs. It has to take the treatment off before it can start opening the slab, so it runs slower than a first pass on bare concrete and it wears tooling faster
- Acid etching leaves the option list. Etching works by dissolving cement paste, and it cannot reach paste that a film is holding it away from. On a sealed slab it is not a weaker choice, it is not a choice
- The depth is set by where the treatment is. A film comes off the top. A penetrating sealer means grinding until you are past the concrete it lives in, and that is a deeper cut than a profile pass
- The moisture question reopens. A film has been holding water inside that slab for as long as it has been there. Take it off and the slab is free to give the moisture to your coating instead, so the moisture gate has to be tested on the prepared surface rather than assumed from the old one
That last one is the one that turns a prep problem into a stop. Gate four is the only one of the four that more grinding cannot fix, and an old sealer is a common reason a slab has never had to show what it is passing upward.
Where to go next
Once you know which of the four you have, removing sealers and curing compounds turns that into a plan, and the useful part is how little of it is a removal job: a thin film comes off with the profiling pass, a penetrating sealer has to be cut past, and a densifier cannot be removed at all. For the whole sequence this diagnosis sits inside, how to prep concrete for epoxy covers the four gates in order. If the film is thick enough to be its own operation, how to remove old epoxy and paint covers why removal and profiling are two lines on a quote rather than one.
If you were considering etching, grinding vs acid etching explains why a sealed slab removes it from the table rather than merely disfavoring it. For the hardness problem a densifier leaves behind, bond vs concrete hardness covers the tooling side. And for what happens when a treatment is missed entirely, why epoxy floors peel is the failure this whole page exists to avoid.
Frequently Asked Questions
What is the fastest way to tell if concrete is sealed?
Drop water on it in several places and watch. On bare, absorbent concrete it darkens the slab and soaks in within seconds. If it beads, sits proud or pulls back into droplets, something is repelling it, and a coating will do much the same thing the water just did.
What the test does not do is tell you what that something is, and it does not catch everything. It answers one question: whether the surface will accept a liquid. A silicate densifier does not repel water at all, so a densified slab can pass this test cleanly and still be a very different floor to grind than the one you priced. Treat the water drop as the first question rather than the whole diagnosis.
Can a slab pass the water test and still be treated?
Yes, and it is the case worth knowing about, because it costs you money rather than causing a coating failure. A silicate densifier is not a film and it is not a water repellent. It reacts with free lime in the concrete and fills the pore structure with hardened material, so the surface stays absorbent and water on it will still darken the slab, often just more slowly than on an untreated floor.
What it does change is hardness. A densified slab resists tooling in a way nothing on the pour record will warn you about, so the profile you planned arrives later and at a higher tooling cost than the rate you quoted. The tell is not the water test. It is a scratch test and a trial patch.
My slab beads water. Is there definitely a sealer on it?
Not definitely, and the cheap explanations are worth ruling out before you price a removal. A slab that is cold, already damp, or simply has not finished drying will take water slowly or not at all. A hard troweled, burnished surface can be dense enough to hold a bead for a while with nothing whatsoever applied to it. Both look exactly like a sealer for the thirty seconds you spend watching a drop.
Test again on a dry area at room temperature, and back the water test with a second one. Try to lift something off with a blade. A film gives you a curl, a flake, and a visible clean edge where it stops. If nothing comes off and water still beads, you are looking at a penetrating treatment or a very dense surface rather than a film, and those are different jobs with different answers.
Does it matter whether the sealer is topical or penetrating?
It matters because removal is only a coherent idea for one of them. A topical sealer, whether acrylic, urethane or epoxy, sits on the concrete, and taking it off is a question of getting under it. A penetrating sealer, usually a silane or siloxane, is in the pore rather than on the surface, so there is nothing to peel and nothing to strip. You reach it by removing the concrete it occupies, which means grinding deep enough to be past it.
The blade test separates the two in about a minute, and the answer changes the tooling, the depth and the time. It also changes what a chemical stripper can do for you, which on a penetrating treatment is very little.
The slab is new. Should I assume there is a curing compound on it?
Assume yes unless somebody can show you otherwise. Curing compounds go onto most new slabs as a matter of course, they are designed to form a film and stay put while the concrete cures, and the crew that placed the floor rarely records which product was used.
The part that catches people out is that many of them are built to break down under UV and traffic over a number of weeks. So a slab can read perfectly clear in the middle of a bay where the light and the forklifts have been, and still be sealed along the walls, under racking, and anywhere that stayed covered. Test the open floor only and you will find the compound with your coating instead of with your water bottle.
If the slab is sealed, does that change anything beyond grinding harder?
It changes three things on the quote. The profiling pass now has removal work ahead of it, so the first pass does more than open the surface and takes longer than the rate you would use on bare concrete. Acid etching stops being worth considering at all, because acid works on cement paste and cannot reach paste it is held away from. And a treatment applied unevenly means the floor is not one surface, so the plan has to cover the worst area rather than the spot you happened to test.
The moisture question also reopens. A film that has been holding water inside the slab has been doing that job for as long as it has been there, and once it comes off the slab is free to give that moisture to your coating instead.