Why Is My Epoxy Still Soft, Gummy or Tacky?
Uncured epoxy is a chemistry problem, and the chemistry is decided before the material leaves the bucket. The pattern of the softness tells you which mistake it was: a whole floor that never hardened is a ratio error, patches in an otherwise sound floor are a mixing error, and a floor that is merely slow is usually just cold.
Epoxy that never hardens is almost always a ratio or a mixing mistake, and the pattern of the softness tells you which. Uniformly soft across the whole pour is the ratio, because every part of that batch has the same wrong chemistry. Soft in streaks and patches with hard material between them is the mixing, because unmixed resin went down alongside properly mixed material. Gummy or rubbery rather than simply soft points instead at batch size, because a big pail overheats itself and the pot life collapses. A floor that is merely slow, rather than soft, is usually just cold, and that one fixes itself once the slab warms up.
Most likely causes
Off ratio
- The most common cause, and the only one that cannot be waited out
- Uniformly soft. A correctly mixed batch beside it leaves a sharp boundary
- Measuring by weight against a ratio stated by volume is a frequent route in, because Part A and Part B rarely share a density
- Fix: the material comes off. No amount of time creates the missing hardener
Not mixed thoroughly
- Soft streaks, swirls or patches in a floor that is otherwise hard
- Unmixed resin clinging to the pail wall, or sitting in the base radius where the paddle never reached
- Often visible as swirl marks before it is felt as softness
- Fix: scrape the sides and base, then transfer to a second clean pail and mix again. Repairs may be patchable if the surrounding material is genuinely sound
Too cold
- The whole floor is slow rather than soft, and it improves day on day
- Cure roughly halves in rate for every 10C or 18F lost, so a cold shop multiplies the stated time
- It is the slab, not the air. Concrete holds temperature long after a space has been heated
- Fix: warm the space and the slab, and give it time. This one does recover
Batch too big, or laid past pot life
- The one that shows up as gummy or rubbery rather than simply soft
- Epoxy makes its own heat as it cures, and a deep pail holds that heat in, so a big batch runs away with itself and the pot life collapses
- Material spread after it has begun to gel is being laid as a part reacted mass, so it neither wets the slab nor levels
- Fix: mix only what you can lay in the time you have, pour it out into a ribbon rather than leaving it standing in the bucket, and use a wide shallow vessel over a deep one
Moisture on the slab
- A cold slab under humid air condenses moisture the coat then has to cure through
- Often shows as a patchy, blotchy surface rather than uniform softness
- Fix: check the slab against the dew point before opening a pail, not after
Soft, gummy, or tacky? The words describe different faults
These get used interchangeably on site, but they point in different directions, so it is worth being precise about which one you have.
- Soft means the material dents and stays dented all the way through. That is a cure failure, and the distribution of it names the cause: everywhere is the ratio, in patches is the mixing.
- Gummy or rubbery means it has body but stays pliable, more like a cured elastomer than a floor. That is usually heat: a batch that overheated in the pail and was laid after it started to gel.
- Tacky means sticky to the touch. If it is sticky and soft underneath, it is the same cure failure. If it is sticky over material that is genuinely hard, the floor cured and something is sitting on top of it, which is a different and far cheaper problem.
The check that tells them apart
Press a thumbnail into an out of the way area and read what happens. Material that dents and stays dented has not cured, and the distribution of that softness is what names the cause. Material that gives slightly and springs back is cured but still gaining hardness, which is the cold case and needs nothing but warmth and patience.
If the surface is sticky or greasy but the material under it is genuinely hard, this is not a cure failure at all. That is a deposit sitting on top of a sound floor, it washes off, and the page you want is blushing. Getting this one wrong is expensive in the wrong direction: it is easy to strip a perfectly good floor because the top of it felt tacky.
What can be saved
Timing decides most of it. Material identified while it is still liquid can often be squeegeed and scraped off, which is a different order of work from what is needed once it goes leathery and then stays that way. If you catch an off ratio batch inside the pot life window, stopping immediately is worth a great deal.
What never works is coating over it. A fresh coat does not fix the layer underneath, it seals it in, and a hard skin over soft material has no bond strength. It prints under anything heavy and eventually lets go as a sheet, taking the good coat with it. Uncured material comes off first, every time.
Get the ratio right before the next batch
Enter the ratio printed on the label and the volume you need, and it splits that into Part A and Part B. Measure into graduated containers against those numbers rather than judging the pour by eye, and check whether the sheet states the ratio by volume or by weight before you start.
Splits into Part A 2 gal and Part B 1 gal
Open the live calculator →Preventing it on the next floor
- Read the mix ratio off the data sheet and check which basis it is stated on, volume or weight, rather than assuming volume
- Measure into graduated containers rather than judging by eye, and do not scale a ratio by memory
- Mix in a clean pail, scrape the walls and the base radius, then transfer to a second pail and mix again
- Fold rather than beat, and keep the paddle submerged, which also keeps air out of the film
- Take the slab temperature with an infrared thermometer, not the room temperature, and check it against both the product minimum and the dew point
- Mix only what you can lay inside the pot life, and remember pot life shortens as temperature rises
Frequently Asked Questions
Will tacky epoxy eventually harden if I just wait?
It depends entirely on which of the two situations you are in, and they need opposite responses.
If the material is cold, yes. Cure is a chemical reaction and reactions slow down as temperature drops, roughly halving in rate for every 10 degrees Celsius or 18 degrees Fahrenheit lost. A floor at 10C that a data sheet specified at 20C may take several times the stated time, but it will get there, and warming the space is the whole fix.
If the ratio was wrong, no, and waiting only costs you time. Off ratio epoxy is not slow cured epoxy: there is unreacted resin or unreacted hardener with nothing left to react with, and no amount of time creates the missing molecules. A floor that is uniformly soft after twice the stated cure time at the right temperature is not going to recover.
How do I tell a ratio error from a mixing error?
Look at how the softness is distributed, because the two mistakes leave different patterns.
A ratio error affects the entire batch equally. Every part of what came out of that bucket has the same wrong chemistry, so the floor is uniformly soft, and if you poured several correctly mixed batches alongside it you will see a sharp boundary at the edge of the bad one.
A mixing error leaves streaks, swirls and patches. Unmixed resin clinging to the side of the pail or sitting in the corner where the paddle never reached goes down as part of the coat but never had its share of hardener, so you get soft spots and often visible swirl marks in an otherwise hard floor.
That distinction matters because the repairs differ. A ratio error means the whole area comes off. A mixing error may be repairable in patches, if the sound material around them is genuinely sound.
Can I coat over epoxy that has not cured?
No, and it is the most expensive mistake available at this point. A new coat does not fix the layer underneath, it traps it. You end up with a hard skin over soft material, which has no bond strength, prints under anything heavy, and eventually fails as a sheet, taking the good coat with it.
Uncured material has to come off. How much work that is depends on how soft it is: material that is still genuinely liquid can often be squeegeed and scraped off while it is wet, which is far easier than what is required once it goes leathery. If you have identified the problem within the pot life window, acting immediately is worth a great deal.
Does measuring by weight instead of volume cause this?
It does when the two get mixed up, and it is a common route to an off ratio floor. Most epoxy mix ratios are stated by volume, and Part A and Part B usually have different densities, so a ratio that is correct by volume is not the same numbers by weight.
The practical rule is to use whichever basis the data sheet states, and to check which one it is rather than assume. If the sheet says 2:1 by volume, measure volume. If it gives a weight ratio, weigh it. A 2:1 volume ratio measured out as 2:1 by weight is off by however much the two densities differ, which is easily enough to leave a floor soft.
Can mixing too big a batch make the floor soft or gummy?
Yes, and it is the cause people reach for last because it feels like the opposite of a problem. Epoxy cures by an exothermic reaction: it generates its own heat, and that heat speeds the reaction, which generates more heat. In a pail the mass holds that heat in, so a large batch runs away with itself.
Two things follow. The pot life you were counting on collapses, sometimes to a fraction of the figure on the sheet, and the material can begin to gel while it is still in the bucket. Anything spread after that point is being laid as a part reacted mass rather than a liquid, and it neither wets the substrate properly nor levels out. It commonly cures gummy, rubbery or unevenly hard.
The fixes are all about getting the heat out. Mix only what you can lay comfortably within the pot life, pour the mixed material out into a ribbon on the floor rather than leaving it standing in the pail, and use a wider shallower vessel rather than a deep one. Remember that pot life shortens as the ambient temperature rises, so a summer batch size is not a winter batch size.
How cold is too cold to lay epoxy?
Read the number off the data sheet rather than working from a general figure, because products vary widely and some are formulated specifically for low temperature work. What is consistent is which temperature matters.
It is the slab, not the air. A concrete floor holds temperature far longer than the air above it, so a space heated to a comfortable working temperature that morning can still have a slab several degrees colder, and it is the slab the coating is in contact with. Measure it with an infrared thermometer rather than reading the room thermostat.
Also watch the direction of travel. A slab that is still cooling pulls the coating temperature down with it, and a cold slab is what drives condensation onto the surface, which is a separate problem that also stops a coat curing properly.
The surface is sticky but the floor underneath is hard. Is that the same problem?
No, and it is worth separating because the fix is completely different and much easier. If the material has genuine hardness under a sticky or greasy film, the epoxy has cured. What you have is something sitting on top of it.
That is usually amine blush, a reaction between the curing agent and moisture or carbon dioxide in the air, and it is a surface deposit rather than a cure failure. It is water soluble, so it washes off, and the floor underneath is sound. It becomes a real problem only if you coat over it, because nothing bonds to it.
A true cure failure dents under a thumbnail and stays dented. If it does not, read the blushing page instead of this one.