We have pulled up more failed epoxy in San Francisco than in any other city we work. Almost none of it failed because the resin was bad. It failed because nobody found out what the concrete was doing before they poured over it. That is the whole story of coating an old slab in this town, and it starts with a test, not a sales pitch.

Our full local coverage lives on the San Francisco service area page, and this same protocol runs on every garage floor and commercial project we take here.

Why Old San Francisco Concrete Is Its Own Animal

A slab poured in the 1920s or 1930s was poured to different expectations. Plastic sheeting under concrete did not become standard practice until well after most of the Sunset, the Richmond, Bernal Heights, and the older Mission blocks were already built. That means the concrete sits directly on soil, and soil holds water. Whatever is in the ground has an open road into the bottom of your floor.

Then add the water table. Big stretches of this city sit on sand and old fill, close enough to the ocean and the bay that groundwater never gets far away. Add ninety-some years of settling, which opens hairline routes through a slab that was monolithic when it cured. A floor in that condition can look bone dry, sweep clean, and still be moving a steady amount of vapor upward every single day.

What Moisture Actually Does to Epoxy

Epoxy is a sealed membrane. That is the point of it and also the trap. When you bond a sealed film to concrete that is still passing vapor, the vapor arrives at the underside of your finish and cannot leave. Pressure builds in that thin layer between resin and slab. It has to go somewhere, so it goes at the bond.

You see it first as small blisters, usually in the middle of a bay or along a wall where the slab is coldest. Then the blisters spread and connect. Then a whole sheet loses adhesion, and the floor peels up in pieces that look almost intact on the back side, which is the giveaway. A coating that peels off clean was never really stuck. It was sitting on top of a moisture problem waiting for its turn.

Alkalinity rides along with that moisture and attacks the bond chemically at the same time. So a floor that failed this way did not wear out. It was undone from below while the top still looked new.

The Tests, Explained Plainly

There are three ways we get an answer, and they are not equal.

Calcium chloride test. A small dish of dry salt gets sealed under a plastic dome on bare, prepped concrete and left alone for a stretch of days. The salt pulls in whatever the slab gives off. We weigh it before and after, and the weight gain tells us how much vapor is coming through a given area over a given time. It is a good read of what is leaving the surface right now.

Relative humidity probes. We drill down into the slab and set a sensor at depth, then let it come to equilibrium. This one tells us what the concrete is like on the inside rather than just what is escaping the top today. On an old San Francisco slab we prefer it, because surface conditions on a foggy week and a dry week can read very differently while the interior stays honest.

The plastic sheet check. Tape a square of clear plastic down tight, leave it a day or two, come back and look for condensation or a darkened patch of concrete. It is free and anyone can do it. It also only ever tells you yes. A clean sheet does not mean a dry slab. It means the sheet did not catch it. We use it as an early flag, never as clearance to coat.

What We Do When the Number Comes Back High

A failing reading is not the end of the project. Most of the time it means a moisture mitigation primer goes down first. That is a dedicated barrier coat built to sit between wet concrete and your finish, absorbing the vapor drive so the decorative layer above it never feels it. It costs more and it adds a day. It is also the difference between a floor that lasts and a floor we come back to.

Sometimes the slab is worse than a primer can carry. Standing water in the crawl, a broken exterior drain, downspouts dumping against the foundation, a failed membrane on a below-grade wall. In those cases we say so and we walk, or we tell you what has to get fixed first. Coating over an active water source is taking your money for something we know will fail. We would rather lose the job than own that floor.

The Fog Belt and the Sunny Side Are Different Underground

Out toward the ocean, in the Sunset and the Richmond, the marine layer keeps ambient humidity high for long stretches and the ground never really dries out between wet seasons. Slabs there tend to read wetter and stay wetter, and they are also more likely to be the older uninsulated pours.

The eastern and southern neighborhoods that get more sun, places like Bernal, Potrero, and the Mission, dry differently at the surface, which is exactly where people get fooled. A slab in Potrero can feel warm and dry underfoot in September and still carry real moisture at depth. Same city, ten minutes apart, two different answers. That is why we test the actual address instead of applying a rule of thumb by neighborhood.

A Quote With No Moisture Conversation Is a Red Flag

If someone walks your San Francisco floor, quotes a price, and never mentions testing, they are pricing a coating and hoping. Ask any contractor what test they run, when they run it, and what number would change their plan. A real answer sounds specific. A vague one tells you they intend to grind and go.

We give a firm price after we know what the slab is doing, not before. On our end that means a site visit, prep, the test, and then the plan. It is slower at the front and it is the reason our floors stay down.

A 1930s Sunset Slab Versus a 2005 SoMa Loft

On a 1930s Sunset garage we assume no vapor barrier until proven otherwise. We plan for an RH probe rather than trusting a surface read, we budget the schedule for mitigation primer, we open up and repair the cracks that have formed over ninety years of settling, and we grind to a deeper profile because that old concrete is often softer and dustier at the top than it looks.

On a 2005 SoMa loft the pour almost certainly has a barrier under it and a known mix design. We still test, because we have found leaking building envelopes and plumbing over newer slabs plenty of times, but the expected answer is clean and the prep leans toward profiling a hard, dense surface for grip. Two very different jobs, one non negotiable first step.

Frequently Asked Questions

Do you moisture test every San Francisco slab?

Yes. Every slab, every project, residential or commercial. On older San Francisco concrete the odds of an unbarriered pour are high enough that skipping the test is not something we are willing to do.

How long does moisture testing add to the schedule?

The test itself needs the slab prepped and then left undisturbed for a period of days so the reading reflects real conditions. We build that into the timeline up front rather than treating it as a delay.

My slab looks dry. Do I still need it tested?

A dry looking surface tells you very little. Vapor moves through concrete without leaving anything visible on top, and San Francisco slabs read differently depending on the week and the neighborhood. Looks are not a measurement.

What is a moisture mitigation primer?

A dedicated barrier coat applied to the bare slab before the decorative system. It takes the vapor drive from below so the finish above it never feels the pressure that causes blistering and peeling.

Would you ever turn down a San Francisco floor?

Yes. If the slab has an active water source, standing water in a crawl space, a broken drain, or a failed below grade membrane, we tell you what has to be repaired first rather than coating over it and watching it come apart.