Sika Floor Adhesive Mistakes: 14 Failures, $43K, and What I Do Now

After 14 documented floor failures in eight years—roughly $43,000 in wasted material, labor, and tear-out—I've reached an uncomfortable conclusion: the chemistry was rarely the problem. Our Sika adhesive and coating jobs failed when we ignored the substrate, chose the wrong product, skipped the open-time window, or took mixing shortcuts. The product doesn't fail first. The process around it does.

I'm not sharing this from a desk. I've handled installation and coating orders for a flooring contractor since 2017, and I keep our team's mistake log: fourteen significant mistakes, about $43,000 in budget that we had to eat. The most expensive one was a 3,100-square-foot wood floor we installed and later pulled up because nobody tested the slab moisture first.

If you only take one idea from this post, take this: read the current datasheet, test the actual substrate, and respect the working time.

Sika floor adhesive: the 'one product' trap

When someone types sika floor adhesive into a search bar, they usually want one answer. The real answer is a catalogue. There are adhesives engineered for wood, for rubber, for vinyl, for tile, and for insulation boards. If your crew grabs whichever pail is closest, you've already introduced the first variable of a failure.

In 2020, I let a crew use leftover SikaBond MS wood-floor adhesive for a rubber gym floor. The logic was that a flexible adhesive would handle the movement. Five weeks later, the rubber lifted at the seams, and the peeled sheets showed almost no adhesion to the rubber backing. The wrong adhesive formed a film that simply didn't couple with that rubber compound. Replacing it cost about $3,800. A look at the adhesive selector chart—which lists compatibility with specific floor coverings—would have stopped the job before it started.

Flexible doesn't mean waterproof

The most expensive entry in my failure log happened in early 2022. The label on our pails said SikaBond MS 14kg flexible wood flooring adhesive. That product is excellent for engineered wood and parquet in dry interior conditions, and it's one of the few adhesives I trust for long glue lines. But it is not a moisture barrier.

The general contractor told us the concrete slab was a month old and 'dry enough.' Our moisture meter sat in the van. We glued 3,100 square feet of white oak directly to that slab. Eight months later, the floor was cupped, several planks had popped loose, and the owner wanted it gone. The repair cost us $9,700.

A flexible wood-floor adhesive absorbs small movements in the wood caused by humidity changes. It cannot stop vapor pressure from a wet slab. If there is any doubt about the substrate, test it and write the number in the job file. 'Flexible' describes the joint, not the water resistance of an untested substrate.

Will acetone take off paint? The question that cost us a floor

We get a version of this question almost every month: will acetone take off paint? The short answer is sometimes, but not the way people hope. Acetone will soften or dissolve many uncured paints, some lacquers, and a few solvent-based alkyds. It will not strip a fully cured two-part epoxy or polyurethane floor coating in any practical sense, and it is a poor way to prepare concrete for a new coating.

In 2021, one of our crews tried to remove an old oil stain from a concrete slab with acetone before applying a new epoxy coating. A rag made the stain look lighter, so they went ahead. Three weeks later, the coating had fish-eyes and pinholes in exactly that area. The acetone didn't remove the oil; it spread the contamination deeper into the pores and left a thin film that ruined adhesion.

Use acetone for cleaning uncured resin and tools, not for substrate prep. If you need to remove an old coating or contaminant, grind it, blast it, or use a stripping product specified for that finish. Then do a small adhesion test before you coat the whole floor.

Epoxy resin color techniques: consistency beats fancy swirls

Most articles about epoxy resin color techniques focus on metallic swirls and wave effects. Those are fun, but the harder skill is making two batches of the same color look identical after cure.

We tinted a custom light-gray epoxy for a commercial floor in two mixes. The second batch looked slightly 'richer' in the bucket, so the installers added a little more colorant. After cure, the difference was obvious: a visible band across the middle of the room. The customer didn't need a color meter to notice it, and the re-coat came out of our pocket.

What we do now is mechanical: dispense colorant by weight, disperse it into the resin thoroughly before adding hardener, mix at a controlled speed for the same time on every batch, and pre-blend the entire calculated amount of material before starting the pour. If a repair patch is needed later, match it like a print job: a color difference of Delta E below 2 is rarely visible, 2 to 4 is noticeable to a trained eye, and anything above 4 will be seen by almost anyone.

Metallic and mottled finishes have their own rules. The fastest way to ruin one is to apply the second bucket after the first has started to set, creating a hard edge. When you're working with reactive resins, plan the pour in one continuous operation or accept that you will see the joint.

PPG powder coating adalah? And why powder is not a floor coating

A customer in Indonesia recently asked about recoating steel railings and mezzanine structure. Our local contact forwarded the question: 'PPG powder coating adalah?' The phrase means roughly 'what is PPG powder coating?' It's a fair question.

PPG is a large coatings manufacturer, and powder coating is a dry finishing process, not a liquid paint you brush or roll. Plastic powder is sprayed onto an earthed metal part, then melted and cured in an oven at around 200°C. That is why powder coating is hard, even, and popular for metal components: it cures at temperatures that no on-site concrete floor or wood substrate can tolerate.

Powder coating and liquid floor coatings are different manufacturing processes, not competing versions of the same paint. On a concrete floor, use a liquid-applied system from the Sikafloor range—epoxy, polyurethane, or polyaspartic—that cures chemically at room temperature. If a specification mixes the two, stop and clarify before you order materials.

What I stopped doing after $43,000 in rework

The failure log changed our process more than any training video did. Today, every project, even a small repair, gets a short pre-work checklist:

• specific Sika floor adhesive and datasheet match the floor covering and substrate
• moisture reading or adhesion test recorded in the job file
• total batch quantity calculated, colorant weighed, and mixing tools ready
• open time and minimum working time written on the board in the room

That list lives in an app now. Over the past 18 months, it caught 47 potential issues before they became callbacks. Our rework rate dropped from about 7% to under 2%. Some installers say checklists slow them down. So does tearing out 3,000 square feet.

Where my advice stops being useful

This is not official Sika installation guidance. I'm an installer, not a Sika employee, and the company hasn't paid for this article. Product names and formulations change, and labels can differ by country. What was true of our pails in early 2025 may not be true for the product sold in your region or under an updated datasheet.

I also don't have a complete explanation for why one moisture-related failure appears in two weeks and another takes eight months. My best guess is that the rate of vapor emission, wood species, board width, and room temperature all influence it. If you have a better theory, I'd genuinely like to hear it.

The $43,000 number is real, and the checklist is free. Start your own mistake log before the next project, not after it.