Why Industrial Floor Coatings Fail: A Diagnostic Guide

The two mechanisms that dominate floor coating failure

A coating is a thin film asked to stay attached to a thick, porous, moving mineral slab for years. Almost every failure is a failure of that relationship rather than a failure of the resin chemistry itself. Two mechanisms dominate: moisture within or beneath the concrete, and preparation of the concrete surface before the coating went down. Published sources disagree about which is the more frequent culprit, and on real projects the two are usually tangled together — a poorly prepared surface tolerates far less moisture than a well-prepared one. Treating either as "the" single cause leads to the wrong repair.

Moisture-driven failure is a pressure problem. Water vapour rising from the ground or from concrete that has not finished drying reaches the underside of an impermeable film and cannot leave. Where the film is weakest it lifts. In Singapore and the wider region this is not an edge case: slabs sit on warm, wet ground year-round, ambient humidity is high enough that slabs dry slowly, and older slab-on-grade construction frequently has no functioning damp-proof membrane at all. A slab that has been in service for twenty years and "has never had a problem" has simply never had an impermeable film on it before.

Preparation failure is a mechanical problem. Concrete arrives with laitance — a weak, fine, cement-rich skin left by finishing — and often with a curing compound, sealer, oil or rubber deposit on top of that. A coating bonded to laitance is bonded to something that is itself barely attached. Mechanical preparation removes that layer and opens the pore structure so the primer can key into sound material. Sparco's TDS for Epoxy Bonding Primer #100 states the substrate must be dry, sound, clean and free from oil, grease, loose material and other bond-inhibiting materials, prepared mechanically by ball blasting, milling or diamond grinding, with weak concrete removed. Those are not stylistic preferences; they are the conditions under which the stated performance was measured.

The five families of failure — and the symptoms that reveal them

It helps to sort failures into families before diagnosing a specific floor, because the families call for genuinely different investigations. Bond failure is separation, and it is either adhesive (the coating leaves an interface cleanly — the substrate, or a previous coat) or cohesive (the material itself tears, either the coating splitting internally or concrete coming away on the back of the lifted film). Moisture-driven failure shows as blistering, osmotic bubbles, or wholesale debonding under a floor that otherwise looks intact. Mechanical and substrate failure comes from the slab: dusting or low-strength concrete, live cracks propagating through the film, or movement at a joint the coating was carried across.

Application failure covers everything that happened on the day. Contamination reintroduced between preparation and priming. Film thickness applied outside the range the product was formulated for. A recoat window missed, so the previous coat cured too hard for the next one to key into. Application when the substrate was within a few degrees of dew point, so condensation formed on the surface, or amine blush formed on a curing epoxy and was overcoated rather than removed. Each of these produces a floor that looks correct on handover and separates months later.

Service-condition mismatch is the quietest family. Nothing was done wrong; the wrong system was specified. A general-purpose epoxy under hot washdown, a coating in a bund exposed to a solvent it was never tested against, or a thin roller coat under steel-wheeled traffic will all degrade on schedule. Softening, discolouration, chemical attack and rapid wear-through belong here, and no amount of surface preparation would have prevented them. Selection depends on traffic, chemical exposure, moisture and downtime, and should be confirmed through technical review before specification.

How to diagnose a failed floor in the right order

The single most common diagnostic error is jumping to a cause before establishing where the separation actually occurred. A floor that has debonded from the concrete, a floor whose two coats have separated from each other, and a floor whose concrete has torn away beneath a perfectly bonded film are three different problems that look identical from standing height. Finding the failure plane is what turns a guess into a diagnosis, and it decides everything downstream — whether the slab is the patient or the coating is.

Work from the visible to the invisible. Photograph and map the affected areas first, because the pattern carries information: failure across the whole floor points at the slab or the system; failure confined to a bay points at a day's work; failure along a wash-down channel points at exposure. Then take physical samples. Cut a square through the coating with a diamond blade, lever it up, and turn the piece over. Only after you know the plane does moisture testing become meaningful, because a moisture result on a floor that failed from contamination will simply confirm that concrete contains water.

Moisture testing should follow the recognised methods rather than a hand-held meter reading alone. ASTM F2170, Standard Test Method for Determining Relative Humidity in Concrete Floor Slabs Using in situ Probes, measures internal slab RH; ASTM F1869, Standard Test Method for Measuring Moisture Vapor Emission Rate of Concrete Subfloor Using Anhydrous Calcium Chloride, measures surface emission. Both are test methods only — 75% internal RH and 3 lb/1,000 ft²/24 h are commonly specified as maxima, but the acceptance limit belongs to the coating manufacturer or the project specification, not to the standard. Sparco's TDS for Epoxy Bonding Primer #100 caps maximum permissible concrete substrate moisture content at 5%, and requires a minimum pull-off strength of the prepared substrate of 1.5 N/mm².

Common mistakes that turn one failure into two

Recoating over a floor that has already debonded is the most expensive of these. A new coat applied over an existing film that is no longer attached inherits the old bond line, and the failure returns with a thicker, harder layer on top of it. The new material is not the problem; it is sitting on a plane that was already broken. If a sample lifts with a clean underside, the existing system is not a substrate, whatever it looks like from above.

Blaming the coating when the slab is the problem is the second. When the underside of a peeled piece carries concrete, the bond beat the concrete — the adhesion was stronger than the material it was bonded to. Changing the coating product will change nothing. The response is substrate reinstatement: remove the weak material, rebuild with a repair mortar such as Sparco Epoxy Mortar, a three-pack non-shrink, steel-trowel-applied system used for concrete floor repair, patching and reinstatement, and then re-prime onto sound concrete.

The third and fourth are procurement mistakes rather than technical ones. Skipping moisture testing because the slab "looks dry" ignores that a slab can look and feel dry at the surface while carrying high internal RH a few millimetres down; ACI 302.2R-06, Guide for Concrete Slabs that Receive Moisture-Sensitive Flooring Materials, exists precisely because surface appearance is not evidence. And specifying by price per square metre without a separate line item for mechanical preparation invites the tenderer to economise on the one operation the coating depends on. A prep line item that survives value engineering is worth more than a premium topcoat.

A checklist to walk the floor with

Before commissioning any investigation, a facility manager can gather most of the information a specialist will ask for in a single walk. Take a torch, a chalk marker, a phone camera and a coin or key for scratch-testing. Mark and photograph every distinct defect type separately rather than photographing the floor as a whole, and note what happens above each area — a leaking line, a wash bay, a forklift turning point — because exposure explains pattern.

The aim is not to reach a verdict on the walk. It is to establish whether the failure is localised or systemic, whether moisture is plausibly involved, and whether the slab beneath is sound. Those three answers determine whether you need a moisture survey, an adhesion test programme, or simply a competent repair to a damaged bay.

• Is the failure across the whole floor, confined to one pour or bay, or following an exposure zone such as a wash channel or loading bay?
• Tap the coating with a coin or hammer handle — does it sound hollow beyond the visibly failed area? Mark the hollow boundary; it is usually larger than the visible one.
• Lift a loose piece and turn it over. Is the underside bare and glossy, powdery, or carrying concrete?
• Scratch the exposed slab. Does it dust, or does the coin skid across sound material?
• Are blisters present, and do any of them contain liquid when pierced?
• Do cracks in the coating line up with slab joints, saw cuts, or column lines?
• Is there standing water, a leaking service, ground-water ingress at a loading bay, or a monsoon-exposed edge nearby?
• Can anyone produce the original specification, the TDS, the preparation method used, and the applied coverage per coat?

From diagnosis to repair strategy

Once the failure plane is known, the repair strategy follows from it rather than from the size of the failed area. If the coating debonded cleanly from concrete, the coating must come off wherever the same preparation was used — which is usually the whole floor — and the substrate must be mechanically prepared to expose sound concrete. If concrete came away with the coating, the slab is being reinstated before anything else happens. If two coats separated from each other, the failure is confined to the intercoat plane, and the lower coat may still be a serviceable substrate once abraded.

Moisture changes the sequence. A slab with high internal RH and no damp-proof membrane will re-fail any impermeable film, however well prepared, until the vapour is either allowed out or held back. That means either a system that tolerates the moisture level, a moisture-mitigation layer, or accepting the risk in writing. This is a specification decision, not a site decision, and it should be confirmed through technical review with the moisture test results in hand.

Whatever the route, the acceptance criterion is measurable rather than visual. Sparco's TDS sets a minimum pull-off strength of the prepared substrate of 1.5 N/mm² and a maximum permissible substrate moisture content of 5% before priming with Epoxy Bonding Primer #100. Testing to those figures before the first coat is applied costs a fraction of what it costs to discover, eighteen months later, that they were never met.

Frequently asked questions