Moisture Content in Timber Floors

When to sand and when to wait. Moisture content is the single most important number on a timber floor job, and getting it wrong is the most expensive mistake in the trade.

What moisture content actually means

Moisture content (MC) is the weight of water in the timber expressed as a percentage of the weight of the oven-dry wood. A board at 12% MC contains water equal to 12% of its dry mass. Timber is hygroscopic -- it constantly absorbs and releases moisture to match the surrounding air. In a stable indoor environment, it reaches a point called equilibrium moisture content (EMC) where it stops gaining or losing moisture. That is the state the floor needs to be in before anyone puts a sander on it.

Target MC ranges for Australian conditions

Australia covers a wide range of climates, and the target EMC shifts accordingly:

• Coastal NSW, QLD, VIC (temperate to subtropical): 10-14% MC is typical for hardwoods in a lived-in home with reasonable ventilation.
• Inland and arid regions: 8-12% MC. Lower ambient humidity means lower equilibrium.
• Air-conditioned environments year-round: 8-10% MC. Constant air conditioning strips humidity from the air, and the timber adjusts downward.
• Engineered timber: Follows the same principles but generally starts drier from the factory. Check the manufacturer specification -- most ship at 8-10% MC.

The exact number matters less than consistency. If every board on the floor reads within 2% of every other board, the floor is stable. If there is a 5% spread between boards in different areas, something is wrong -- either the boards were not acclimatised, or there is a moisture source underneath.

How to read a moisture meter

Two types of meter are common in the trade:

• Pin meters (resistance type). Two pins driven into the timber measure electrical resistance between them -- wetter timber conducts more. These give a reading at a specific depth, which is useful. Drive the pins at a slight angle to get a reading from the middle third of the board, not just the surface. Surface readings alone are unreliable because the top dries faster than the core.
• Pinless meters (capacitance type, like a Tramex). Press the pad flat on the timber and it reads a volume of wood beneath it. Fast and non-destructive, good for scanning a whole floor quickly. Less accurate on dense Australian hardwoods -- calibrate against a pin meter reading to be safe.

Whichever meter is in use, take readings in at least 8-10 locations across the floor, including near external walls, near bathrooms, in the centre of the room, and near any suspect areas (discoloured boards, boards near plumbing). Record the readings. If there is a dispute later about whether the floor was ready to sand, those numbers are the evidence.

Why sanding too early causes cupping

A newly installed floor over a concrete slab that is still curing will absorb moisture from below. The underside of each board is wetter than the top, so the edges curl upward -- that is cupping. If the floor is sanded flat while it is cupped, the high edges are removed. When the slab eventually dries and the moisture equalises, the boards try to flatten out. But the timber that was sanded off the edges is gone. The result is crowning -- the centre of each board now sits higher than the edges, and the floor looks wavy under any kind of side light. The only fix is to sand again, and by now there may not be enough timber left for another full sand.

This exact sequence -- sand too early, cup, crown, re-sand -- is the most common and most expensive callback in the floor sanding trade. A moisture meter and patience prevent it entirely.

Concrete slab moisture testing

The slab underneath a timber floor is just as important as the timber itself. A slab that is still releasing moisture will push that moisture into the boards from below indefinitely.

• In-slab relative humidity probes (ASTM F2170): The gold standard. Drill holes to 40% of the slab depth, insert probes, wait 72 hours for equilibration, then read. Target: below 75% RH for most timber installations. Some adhesive and coating manufacturers specify 70%.
• Calcium chloride test (ASTM F1869): A surface test. Measures moisture vapour emission rate (MVER) from the top of the slab. Target: below 3 lbs per 1,000 sq ft per 24 hours. Less reliable than in-slab probes because it only measures the surface, and a dry surface can hide a wet core.
• Plastic sheet test: Tape a 1 m square of polyethylene to the slab and leave it for 24 hours. If moisture condenses underneath, the slab is too wet. This is a rough screening test, not a measurement.

If the slab fails, a moisture barrier is required before timber goes down. Epoxy-based barriers (applied directly to the slab) or polyethylene sheet barriers are common. The barrier type depends on the installation method -- glue-down timber over epoxy, floating or nailed timber over sheet.

When to walk away

Some jobs are not ready. If the moisture readings on the timber are above the target range and trending upward, or if the slab is reading above 75% RH and the builder says "it'll be fine, just sand it," that is the moment to stop. Document the readings, explain the risk in writing, and do not sand. The cost of a delayed start is nothing compared to the cost of sanding a floor twice because it cupped.

If the client or builder insists, get a signed acknowledgement that the moisture readings were above specification and that the decision to proceed was theirs. Protect the business.

After sanding: seal the system

Once the floor is sanded at the correct MC, prime and topcoat promptly. An unsealed freshly sanded floor is wide open to absorbing moisture from the air or from a spill. Apply Bona Prime Intense on high-tannin species or Classic UX on standard species, then two coats of topcoat (Traffic HD, Traffic GO, Mega, or Wave 2K). A fully sealed floor is far less reactive to short-term humidity changes than a bare one.

Related: Process FAQs →