How do you know when screed is dry enough for flooring?

A screed becomes walkable, then takes light traffic, then takes following trades — and none of those milestones say anything about how much water is still inside it. A floor can carry a scaffold tower and still be far too wet to bond vinyl to. The moisture that matters for a floor covering is the moisture deeper in the section, which migrates to the surface after the floor is sealed by an impermeable finish.

That is why sealing a finish onto an untested floor is such an expensive mistake: the covering traps the remaining moisture, which then attacks the adhesive bond, lifts the finish or causes discolouration weeks after everyone has left site. Programme estimates — the kind set out in screed drying times — tell you when to plan the test, not whether to skip it.

It does not mean the humidity of the room. A subfloor hygrometer test works by sealing a small volume of air against the screed surface and isolating it from the ambient air, then leaving it until that trapped air reaches equilibrium with the moisture in the screed. The reading is the relative humidity of that confined volume — in effect, the humidity the screed itself produces.

Pointing an ambient hygrometer at the room, or waving a meter over the surface, measures something entirely different. That distinction is the single most common source of false confidence on site.

Current F. Ball guidance notes that BS 8203, BS 5325 and BS 8201 recommend hygrometer testing, and states that where a series of tests does not exceed 75% RH the base can be considered dry enough for most floor coverings — with 65% RH given for solid timber and the wood supplier consulted. BS 8203:2017, the current code of practice for the installation of resilient floor coverings, is a floor-covering standard: it sits alongside, rather than replaces, what the screed manufacturer says about its own material.

For traditional sand and cement screed and cementitious flowing screeds, the sealed hygrometer test against the flooring criterion is the usual route to a decision, with the ≤75% RH benchmark the most frequently encountered figure for resilient coverings.

Manufacturers publish their own targets. Current Cemfloor guidance for moisture-sensitive flooring gives a final moisture target of 2.5% CM or 75% RH, and current Cemfloor technical guidance says electronic encounter meters may be used as a guide but confirmation should be by the specified hygrometer or carbide method before floor coverings are applied. Those are Cemfloor's figures for a cement-based screed — they are not a general cementitious rule, and the percentage associated with anhydrite systems does not transfer to them.

There is no reliable universal conversion between %RH and %MC for cementitious screeds. Test by the method the criterion is written in. Where drying speed is part of the programme, see fast-drying screed.

Calcium sulphate screeds behave differently and are tested differently. Current Gyvlon post-installation guidance states that electronic resistance meters reading from the surface downwards vary in accuracy and should be used only as a rough wet-or-dry guide; that RH testers are a British Standard test for most screeds, but that BS 8204-7 notes they are not accurate for calcium sulphate screeds at high moisture content above about 85% RH; and that for a definite value the BS 8204-7 recommendations should be followed — the CM carbide method to BS EN 13892-10, or oven drying at 40 °C — to establish percentage moisture content.

Gyvlon also publishes an approximate conversion for its own material: roughly 75% RH corresponding to about 0.5% moisture content, 80% RH to about 0.7%, 85% RH to about 1.0% and 90% RH to about 1.5%. Those correlations are approximate and product-specific — a guide to interpretation, not a substitute for the required test.

On criteria, published Gyvlon finishing guidance has given typical maximum CM values of around 0.5% for impermeable floor coverings and around 1.0% for moisture-permeable coverings. But the product matters: Gyvlon's specialist E2C products can allow bonded finishes at materially higher residual moisture under their own system rules. So the first question is never “what is the anhydrite limit?” — it is “which product is this, and what does its system require?” See liquid screed for how the families differ.

A controlled commissioning or heat-up cycle can assist drying, and on many projects it is part of the plan. Two conditions apply. First, it can only be done where the screed manufacturer's guidance permits it, and only after the initial curing period the screed requires has been respected — heating a young screed can damage it permanently. Second, the heating regime itself should follow the screed and heating system requirements rather than an improvised schedule.

Final moisture testing should then be carried out under the conditions and by the method the flooring and screed systems require. We do not publish a fixed “heating off for X days, then test” rule, because that period is set by the systems involved, not by a general convention. And a heated floor that feels warm and dry underfoot is still telling you nothing measurable. See screed over underfloor heating.

Handheld meters are genuinely useful, which is part of the problem — they produce a confident-looking number in seconds. What they actually read is an electrical property at or near the surface, and that property is affected by the material and its mix, the depth the instrument senses, surface condition and residual laitance, temperature, salts, reinforcement or pipework below, and the instrument's calibration for the material in front of it.

The surface of a screed also dries first. A floor can read dry at the surface while holding substantial moisture 20 mm down — which is precisely the moisture that will migrate upwards once an impermeable covering seals the floor.

Used properly, a meter is a survey tool: run it over the whole floor, find the wet zones, and put the definitive tests there. It becomes a definitive result only where the flooring or screed system explicitly accepts that method.

For the substrate side of tiling, see preparing screed before tiling. For resin systems, see resin flooring and resin screed.

First, do not treat the failed reading as an obstacle to be argued around. It is information, and it is cheaper than a failed floor.

The practical options are: improve the drying environment with ventilation and air movement, and dehumidification where the building is closed enough for it to work; use the underfloor heating system as a drying aid where the screed manufacturer's guidance permits it and initial curing requirements have been met; accept a programme delay and re-test; or, where the substrate genuinely cannot reach the criterion in time, consider a moisture-management or surface damp-proof system.

That last option is product- and system-specific. These systems exist for substrates above normal flooring thresholds, but they are approved for defined conditions and readings, and installing one is not permission to ignore a wet substrate. The flooring or resin manufacturer sets the allowable readings and the preparation required.

What not to do: run direct-fired heaters into a closed building, force-dry a screed that is still curing, or lay a finish on the basis of a date. Rapid uncontrolled drying causes its own defects.

• Spread readings across the floor rather than relying on one test to speak for the whole area.
• Include the thickest or deepest zones — depth drives drying, so those areas often lag behind.
• Include perimeters, doorways, corners and anywhere air movement has been poor.
• Include areas that look, feel or read wetter on a screening survey, and any area that was covered or protected during drying.
• Include cold areas, north-facing rooms and spaces that have been closed up.
• Avoid areas that are unrepresentative for the wrong reason — a recent spillage or a locally damaged patch — unless you are deliberately investigating them.
• Record where each test was taken, so results can be repeated and compared rather than argued about.

How many tests, and how they are distributed, should follow the method and acceptance arrangements the project and the relevant standards set out for the covering being installed. Agree that with the flooring contractor before testing starts rather than afterwards — the same principle that applies to tolerances in BS 8204 screed standards. Deeper zones deserve particular attention; see screed depth.

• Screed type and, where relevant, the specific product and manufacturer.
• Installation date, and the date the screed became accessible.
• Nominal and, where known, average depth, plus the build-up (bonded, unbonded, floating, over underfloor heating).
• Underfloor heating commissioning status and the heat-up regime used, if any.
• Ventilation, temperature and site conditions during the drying period, as far as they are known.
• Test method and device used, including model and calibration status where applicable.
• Date and time of each test, and the equilibrium period allowed for hygrometer readings.
• Test locations, described or marked on a plan.
• Results as measured, including any readings that failed.
• The criterion being tested against, and the finish, adhesive or system it comes from.
• Who carried out the testing, and for which company.

A record like that turns a moisture question into a documented decision. It also protects everyone involved when a floor covering fails months later and the conversation turns to what was known and when.

The criterion comes from the specification and the manufacturer of the finish, adhesive or system being installed. The decision to proceed sits with the party installing that finish — in practice the flooring or tiling contractor, who should confirm the substrate meets their product and system requirements before bonding anything moisture-sensitive to it.

Our part is to give you what we know and to be straight about it: the screed type and product, the installation date, the build-up and nominal depth, the drying guidance that applies to that system, and the conditions the floor has been left in. That supports a proper test — it is not a substitute for one, and we do not certify a floor as ready for every covering.

If a programme depends on a floor being coverable by a particular date, raise it before the screed is specified rather than afterwards. Request a commercial quotation and send the programme with the drawings.

• Screed drying times
• Preparing screed before tiling
• Screed depth
• BS 8204 screed standards
• Sand and cement screed
• Liquid screed
• Fast-drying screed
• Screed over underfloor heating
• Resin flooring and resin screed
• Request a commercial quotation