Why Do Epoxy Floors Peel in Bloomington? Four Local Failure Modes
September 9, 2026
A peeling floor in Bloomington almost always traces to moisture, prep, salt or temperature, and each one has a test or a rule that would have caught it before the first coat.
Failure one: water coming up through the slab
Concrete on damp ground passes water vapor upward, and a bonded film lives with the pressure that vapor builds. When the pressure wins, the coating blisters or lets go in sheets with the underside white or wet. Bloomington's slabs sit on three grounds, a sand plain in the northeast, low hills across the west and the Minnesota River valley along the south, so two garages a mile apart can test differently.
Two ASTM tests settle the question before primer goes down. ASTM F1869 seals a calcium chloride dish under a dome for 60 to 72 hours and reports moisture vapor emission in pounds per 1,000 square feet per 24 hours; the common limit is 3 pounds. ASTM F2170 drills to 40 percent of slab depth, seats an RH probe, waits 24 hours and reads the humidity inside the concrete; typical limits are 75 to 80 percent, with the room at service conditions for 48 hours first and three probes per first 1,000 square feet plus one per additional 1,000.
The dish reads only the top 1/2 to 3/4 inch of the slab; the probe reads the middle, where most of the moisture lives. Either result describes that spot on that day: a slab that passes in February, with 0.87 inches of normal precipitation, can fail in June, the wettest month at 4.58 inches, after a 51.2 inch snow season has melted into the ground. White, crusty efflorescence on a basement slab is the same signal in mineral form, and there a coating is the wrong answer until the water is managed; a coating is not waterproofing for liquid water rising from below.
Failure two: a surface the coating never bonded to
The second failure peels differently: the film comes up clean and glossy on the back with a skin of gray dust or old sealer attached. That is a bond failure, and ICRI Technical Guideline 310.2R-2013 exists to prevent it. It defines concrete surface profiles CSP 1 through 10, from nearly flat to peaks deeper than a quarter inch, and matches each to a method: grinding at CSP 1, acid etching at 2, light shot blasting at 3, medium shot blasting at 5, heavy abrasive blasting at 6.
Thin-film coatings want CSP 1 to 3, high-build systems CSP 3 to 5, and both data sheets referenced here call for CSP 2 or better by diamond grinding. Acid etching reaches a number on the chart but does not remove curing compounds, old sealers or the weak laitance on a hard-troweled 1960s slab, which is why manufacturers list etching, sanding and pressure washing as unacceptable. Concrete grinding and surface prep is where a Bloomington job spends its hours; on a 1950s rambler garage with three layers of old paint it can outlast the coating itself.
Failure three: five months of brine
Bloomington sits inside one of the country's most heavily salted road networks: MnDOT spread 190,758 tons of salt and 16,119,540 gallons of brine statewide in 2025-26, and the Minnesota Pollution Control Agency estimates about 365,000 tons a year across the metro. A share arrives in wheel wells every evening from November to April.
On bare concrete the winter carry-in guide describes the result: chloride keeps the surface saturated, water expands about 9 percent when it freezes in the pores, and the paste scales, pits and dusts. A coating keeps brine on top, where a squeegee removes it; the failure comes when a coating goes over a slab already loaded with chloride and weakened paste, and the concrete keeps flaking under the film.
Failure four: a cold slab, or a hot tire
Coatings cure by chemistry, and the chemistry slows as concrete cools. The MSP normals put November's high at 41.9 degrees F and its low at 27.7, with 20.9 freezing nights; a standard epoxy wants the slab at 50 degrees F or warmer, and its 8 to 12 hour recoat window assumes it. Miss the window on a cold slab and the second coat bonds only where the first was scuffed, so sheets lift between layers rather than at the concrete.
Summer has the opposite failure. July's normal high is 83.4 degrees F, and a tire that has run 494 at highway speed comes home hotter still. A thin, soft film, especially a one-part garage paint, softens under it and comes up in a tread pattern when the car pulls out. Full cure on the data sheets runs 7 days, and parking early is the self-inflicted version.
Reading the failure before fixing it
The back of a peeled flake tells most of the story.
- Blisters, wet undersides, white powder: moisture. Test with F2170 before any recoat and expect a mitigating primer if the slab fails; published systems are rated to 20 pounds and 99 percent RH.
- Clean, glossy peel with dust on the back: prep. Grind to CSP 2 or better and rebuild from bare concrete.
- Pitting that continues under the film: salt-damaged concrete. Grind past the weak paste, patch, then coat.
- Lifting between coats, soft spots, tire prints: temperature or cure. Strip the coat and reschedule for a warm slab.
What a Bloomington scope should say
The fix starts on paper, for a tuck-under in Oxboro or a basement beneath a Penn-American split. A scope for garage floor coating or a below-grade floor should name the moisture test and its pass mark, the target CSP and the method that reaches it, the slab start temperature and the return-to-service times. A contractor offering epoxy flooring across Bloomington and the surrounding suburbs should have those numbers ready; a price without them is a guess.
More from the blog
Quick Answers
Why is my epoxy garage floor peeling after one winter in Bloomington?
Most first-winter failures are moisture or prep. Chloride carried in from November to April keeps the slab surface saturated, and if the coating went down over unground concrete, old sealer or a slab that never passed an ASTM F2170 or F1869 test, the film had no bond to hold. Check the back of a peeled flake: dust or sealer means prep, wet or white means moisture.
What moisture level is too high for epoxy?
The common published limits are 3 pounds per 1,000 square feet per 24 hours on the ASTM F1869 calcium chloride test and 75 to 80 percent relative humidity on the ASTM F2170 in-slab probe; one polyaspartic data sheet prints 3 pounds and 80 percent. Above those numbers a standard primer is likely to blister, and the job needs a moisture-mitigating primer rated for the reading or a different floor altogether.
Can a failed epoxy floor be recoated instead of stripped?
Only when the existing coating is bonded and the failure is cosmetic or limited to the topcoat. If a scraper lifts sheets, if the underside is wet or dusty, or if pitting continues in the concrete beneath, the floor must be ground to sound concrete and rebuilt. A recoat over a failing layer inherits every one of its problems and adds one more layer to remove later.