Moisture content and equilibrium moisture content: two different numbers
Two numbers decide whether a table top stays flat or a stair tread opens a gap. The first is the moisture content (MC) of the wood itself. The second is the equilibrium moisture content (EMC): the moisture content the wood will settle at in a given climate, the point at which it neither gains nor loses water.1 Problems start when the two are far apart, because the wood will close that gap whatever anyone wants.
How moisture content is calculated
Moisture content is the mass of water in a piece of wood divided by the mass of the same piece when completely dry, the "oven-dry" mass, times 100.1 Because the reference is the dry wood and not the wet piece, the figure can exceed 100%. Green timber often does: the FPL lists average green moisture contents for American species of 55% in the heartwood and 72% in the sapwood of American beech, 64% and 78% for white oak, and 46% and 44% for white ash, while very wet species pass 200%.1 The formula matters when you compare figures: firewood is often quoted on a wet basis, as a share of the original mass, which gives a lower number for the same log.2
Free water, bound water and the fibre saturation point
Water sits in wood in two places. Free water fills the cell cavities; bound water is held inside the cell walls. The moisture content at which the walls are saturated but the cavities are empty is the fibre saturation point. It averages about 30%, although it varies by several percentage points between species, between pieces and with the measuring method.1 Above that point, losing water changes almost nothing except weight. Below it, wood shrinks as it dries and swells as it takes up moisture, and its strength properties change. Everything in this article happens below the fibre saturation point; how timber gets there is covered in our article on kiln drying versus air drying.
Equilibrium moisture content: what the air does to wood
Wood that is kept out of rain and away from heat sources takes its moisture content from the relative humidity (RH) and temperature of the surrounding air. Daily swings mostly affect the surface; seasonal swings reach the whole cross-section. Coatings such as varnish or paint slow the exchange, but they do not stop it.1
The EMC table: temperature × relative humidity
The US Forest Products Laboratory (FPL) has published the relationship between EMC, humidity and temperature since before 1920. The current version is Table 4-2 of the Wood Handbook, calculated from an equation that fits the historical data.1 The rows below cover the temperatures at which wood is normally stored, machined and used.
| Air temperature | 20% | 30% | 40% | 50% | 60% | 65% | 70% | 80% | 90% |
|---|---|---|---|---|---|---|---|---|---|
| 4.4 °C (40 °F) | 4.6 | 6.3 | 7.9 | 9.5 | 11.3 | 12.3 | 13.5 | 16.5 | 21.0 |
| 10.0 °C (50 °F) | 4.6 | 6.3 | 7.9 | 9.5 | 11.2 | 12.3 | 13.4 | 16.4 | 20.9 |
| 15.6 °C (60 °F) | 4.6 | 6.2 | 7.8 | 9.4 | 11.1 | 12.1 | 13.3 | 16.2 | 20.7 |
| 21.1 °C (70 °F) | 4.5 | 6.2 | 7.7 | 9.2 | 11.0 | 12.0 | 13.1 | 16.0 | 20.5 |
| 26.7 °C (80 °F) | 4.4 | 6.1 | 7.6 | 9.1 | 10.8 | 11.7 | 12.9 | 15.7 | 20.2 |
| 32.2 °C (90 °F) | 4.3 | 5.9 | 7.4 | 8.9 | 10.5 | 11.5 | 12.6 | 15.4 | 19.8 |
The 0.1% precision only shows trends: real values vary between species and with sorption history, by a point or two. Source: Glass & Zelinka (2021), Wood Handbook FPL-GTR-282, chapter 4, Table 4-2. Values derived mainly for Sitka spruce; °F rows as published.
Reading the table: humidity matters, temperature barely does
Two patterns stand out. First, relative humidity drives almost everything. At about 21 °C, wood sits near 4.5% MC at 20% RH, 9.2% at 50% RH and 20.5% at 90% RH. Second, temperature hardly matters in the indoor range: at 50% RH the EMC is 9.5% at 4 °C and 9.1% at 27 °C.1 That is why a hygrometer is the most useful instrument in a wood store, and why a heated room is dry for wood: warming the air lowers its relative humidity.
Show the data
| Relative humidity | Equilibrium moisture content |
|---|---|
| 20% | 4.5% |
| 25% | 5.4% |
| 30% | 6.2% |
| 35% | 6.9% |
| 40% | 7.7% |
| 45% | 8.5% |
| 50% | 9.2% |
| 55% | 10.1% |
| 60% | 11.0% |
| 65% | 12.0% |
| 70% | 13.1% |
| 75% | 14.4% |
| 80% | 16.0% |
| 85% | 17.9% |
| 90% | 20.5% |
Furniture targets of about 8% correspond to roughly 40–45% RH. Source: Glass & Zelinka (2021), Wood Handbook FPL-GTR-282, chapter 4, Table 4-2.
The curve also explains why furniture targets cluster around 8%. That is the EMC of air at roughly 40–45% RH, the range of a heated home in winter. A board at 12% MC is in balance with about 65% RH, which is a humid summer, an unheated building or a covered yard.
Why the table is a guide, not a law
The FPL is explicit that the 0.1% precision of its table only illustrates trends. The values were derived mainly for Sitka spruce, a softwood, and real EMC varies between species and between pieces of the same species. Species differences are small at low and moderate humidity and become considerable at high humidity, where wood rich in extractives tends to have a lower EMC.1
History matters too. Wood that is drying down to a given humidity settles at a higher moisture content than wood that is taking up moisture to the same humidity. This effect is called sorption hysteresis. The table values sit midway between the two, which makes them a practical compromise when you do not know which way the wood is moving.1 In practice, read the table as "plus or minus a point or two".
European sources agree closely. The German adhesive industry association (IVK) publishes an EMC table for 20 °C that gives 7.5% at 40% RH, 9.1% at 50% and 12.0% at 65%, within a few tenths of the FPL values.5
Indoor climate through the year
Indoor humidity is not fixed. In Central and Eastern Europe it falls in winter, when cold outdoor air is brought in and heated, and rises in summer. A German timber expert, writing for the Deutsche Handwerks Zeitung, describes 55–60% RH as normal indoors in summer, while winter airing already pulls rooms down to 40–45% RH, and lower with more airing; he warns that humidifying well above 50% in winter risks mould.6
Read against the EMC table, that cycle moves furniture wood between roughly 8% and 11% moisture content every year. A dry, well-heated winter at 30% RH takes it down to about 6%; a humid summer at 65% RH brings it up to about 12%.1,6
Wood cannot be dried "beyond" this cycle. The best a manufacturer can do is deliver it at the middle of the range its customers' rooms will impose, and design joints and panels that tolerate the swing on either side.
How much wood moves when its moisture changes
Below the fibre saturation point, wood shrinks and swells differently in each direction. Tangential movement, along the growth rings, is about twice the radial movement, across the rings. Longitudinal movement, along the grain, is small: 0.1–0.2% from green to oven-dry for most species, although juvenile wood and reaction wood can reach 2%.1
Movement per percentage point: European values
For European species, the Bavarian State Institute of Forestry (LWF) publishes differential shrinkage, the change in dimension per one percentage point of moisture content between 5% and 20% MC. For beech (Fagus sylvatica) it is 0.41% tangential and 0.20% radial; for oak (Quercus robur) 0.36% and 0.16%; for lime 0.24–0.32% and 0.15–0.23%.3 For ash (Fraxinus excelsior), the same institute gives 0.38% and 0.21%.7 Other compilations quote slightly different figures, so treat the second decimal with caution.5
Worked example: a 500 mm panel from 6% to 12%
Width change is width × coefficient × change in moisture content. A 500 mm wide flat-sawn beech panel that goes from 6% to 12% MC therefore swells by 500 × 0.0041 × 6 ≈ 12.3 mm. The same panel in quarter-sawn beech swells by about 6 mm. A more typical seasonal swing of three points, from 8% to 11%, still moves the flat-sawn beech panel by about 6 mm.3
Show the data
| Flat-sawn (tangential) | Quarter-sawn (radial) | |
|---|---|---|
| Beech | 12.3 mm | 6.0 mm |
| Oak | 10.8 mm | 4.8 mm |
| Ash | 11.4 mm | 6.3 mm |
Real boards are rarely perfectly flat- or quarter-sawn; an edge-glued panel with mixed ring orientation falls between the two values. Source: Coefficients from Grosser & Ehmcke (2016), LWF Wissen 78, Table 3 (beech 0.41/0.20, oak 0.36/0.16 % per % MC) and Grosser (2001), LWF Wissen 34, Table 3 (ash 0.38/0.21). Calculation by SARAL.
The FPL uses the same method with its own dimensional change coefficients for American species. Its tangential coefficient for American beech, 0.00431 per point of MC, is almost identical to the European value. The FPL adds that real boards are rarely perfectly flat-sawn or quarter-sawn, so a calculation with the tangential coefficient will probably overestimate the width change.2 An edge-glued panel with mixed ring orientation sits between the two bars of the chart.
Why the cut matters
Beech moves a lot by European standards, which is why beech panels and tops need either a stable indoor climate or construction that lets them move. Oak moves less, and quarter-sawn stock of either species roughly halves width movement. How the cut is chosen at the saw, and what it costs in yield, is the subject of our article on plain, quarter and rift sawn timber.
Target moisture content for furniture, flooring and joinery
The FPL's rule is simple: install wood at the moisture content it will have in service. That keeps the seasonal change, and therefore the movement after installation, as small as possible.2 The targets below are the published reference values.
| Use | Average | Individual pieces | Source |
|---|---|---|---|
| Interior woodwork, flooring, furniture, trim (most US areas) | 8% | 6–10% | FPL, Table 13-2 |
| Same uses, dry heated regions (US south-west) | 6% | 4–9% | FPL, Table 13-2 |
| Same uses, damp warm coastal regions | 11% | 8–13% | FPL, Table 13-2 |
| Exterior siding, trim, sheathing, laminated timber | 12% | 9–14% | FPL, Table 13-2 |
| Hardwood for furniture, flooring, cabinets, millwork (kiln target) | 6–8% | – | FPL, ch. 13 |
| Most hardwoods, drying rule of thumb | 6.5–7.0% | small spread | FPL-GTR-118 |
| Wood in living spaces (German adhesive industry) | ≈8% | – | IVK TKH-3 |
| Trim and flooring in buildings heated only occasionally | 12–18% | – | FPL-GTR-118 |
| US hardwood lumber exported to Europe (typical, as shipped) | 10–13% | – | FPL-GTR-118 |
EN 942 (joinery) gives its own guide values in an informative annex; they are not reproduced here because the standard is paywalled. Source: Bergman (2021), Wood Handbook FPL-GTR-282, chapter 13, Table 13-2 and text; Denig, Wengert & Simpson (2000), FPL-GTR-118, chapters 4 and 9 and Table 1.4; IVK (2024), TKH-Merkblatt 3.
Furniture and interior woodwork
For interior woodwork, flooring, furniture and trim in most of the United States, the FPL recommends an average of 8% at installation, with individual pieces between 6% and 10%. In very dry heated regions the average drops to 6%, and in damp coastal regions it rises to 11%.2 Hardwood for furniture, cabinets and millwork is normally kiln dried to 6–8%, slightly below the service level, because wood picks up a little moisture during processing, transport and installation.2 The FPL's hardwood drying manual goes further: as a rule of thumb, most hardwoods should be dried to an average of 6.5–7.0% with only a small spread.8 The German adhesive industry works on about 8% for wood used in living spaces.5
Flooring and stairs
Flooring and stair parts fall into the same interior category. What goes wrong with them is usually not the delivered moisture content but what happens on site. The FPL's checklist for finished flooring: measure random pieces on arrival with a meter, deliver only after masonry and plaster have dried, install the heating first, open the bundles inside the building, and keep the house about 8 °C warmer than outside for about three days before laying.2 Our article on solid oak stair treads covers the product side.
Joinery, windows and exterior work
Exterior siding, trim and laminated timbers have a recommended average of 12% (9–14% for individual pieces) in most of the United States.2 Trim and flooring for buildings that are heated only occasionally should be dried to 12–18%.8 In Europe, EN 942 on timber in joinery contains an informative annex, "Guide to moisture content in joinery", that sets values by service condition, from external joinery to rooms with continuous heating.9 The standard is paywalled, so specify its class in the contract and ask your supplier which value they work to.
Gluing: the 2-point rule
Pieces glued together must be at almost the same moisture content. The IVK's guide for dispersion (PVAc) adhesives allows a difference of no more than 2 percentage points between parts, and the FPL applies the same 2-point limit to laminations in a glued member.2,5 Otherwise each lamella moves differently after pressing, which stresses the glue line and distorts the part. This matters for every edge-glued panel.
Why "kiln dried" is not a moisture specification
Kiln drying sets a starting point, not a permanent state. Wood that leaves a kiln at 8% starts moving towards the EMC of wherever it is stored. The FPL measured how fast: solid-piled hardwood kiln dried to 8% or less, kept under a good pile roof in a yard in warm, humid weather, gained about 2 percentage points per month during the first 45 days and about 1 point per month after that. In open sheds the rate was about 1 point per month, and in closed sheds about 0.3 points. Dry material in a ship's hold typically picks up about 1.5 points during a normal voyage.2
Imported timber is not always dried to furniture levels either. The FPL's drying manual lists a typical final moisture content of 10–13% for American hardwood lumber exported to Europe.8 That is fine for further processing, as long as the buyer knows it and re-dries or conditions the wood.
The consequences are expensive. The same manual estimates that probably 75% of manufacturing problems in furniture and cabinet plants are related to incorrect moisture content, and that in nearly all cases the wood is too wet.8 It also points out that "6–8%" is ambiguous: does every piece have to fall in that range, or only the average? Does it apply to the core or to the whole piece?8 A good order answers those questions.
Measuring moisture content: oven-dry method and meters
In Europe, the moisture content of a piece of sawn timber is measured under the three parts of EN 13183. Part 1 is the oven-dry method, the reference. Part 2 covers estimation with an electrical resistance meter, and Part 3 estimation with a capacitance meter.4 Parts 2 and 3 are "estimation" methods: they are calibrated against the oven-dry result and inherit its authority only when they are used correctly.
The oven-dry method (EN 13183-1)
The principle is the same everywhere. The FPL procedure takes full cross-sections 25 mm long, at least 500 mm from the end of the board because ends dry and wet faster. Each piece is weighed at once, dried at 101–105 °C until its mass no longer changes noticeably between weighings three hours apart, which takes 12–48 hours, and weighed again.2 It is slow and destructive, but it is the method every meter is checked against, and kiln operators use it on their kiln samples.
Resistance (pin) meters (EN 13183-2)
Pin meters measure the electrical conductance between two needles driven into the wood. Wetter wood conducts better. Some practical points from the FPL:2
- Range. Readings are reliable from about 6% to 30% MC. Above 30% they are qualitative only.
- Depth. Insulated pins read only at their tips. In a drying board, which is wetter inside, driving them to one-fifth to one-quarter of the thickness gives roughly the average moisture content of the piece. Uninsulated pins read the wettest point along their length, so a board that has been rained on looks wetter than it is.
- Temperature. Add about 0.5 percentage points for every 5.6 °C the wood is colder than the meter's calibration temperature, and subtract the same when it is warmer.
- Species. Apply the species correction supplied with the meter.
Capacitance (pinless) meters (EN 13183-3)
Pinless meters measure the dielectric properties of the wood through a pad pressed on the surface. They leave no holes and are fast, which makes them the usual tool for checking flooring and finished parts. Their field is strongest near the surface, so the reading is dominated by the outer layers and may miss a wet core. They need a density or species setting, and their reliable range is about 4% to 30%.2
How accurate is a meter, and how many pieces should you measure?
Used properly on a lot of timber that has been dried to a uniform moisture content below fibre saturation, the average of corrected meter readings should be within one percentage point of the true average.2 A single reading on a single board proves little. The FPL recommends testing at least 10% of the pieces of an item; if the average is within one point of the target and every piece is within the individual limits, the lot is probably satisfactory.2
How we manage moisture at SARAL
How to specify moisture content: a checklist for buyers
- State a target and a tolerance. For example, "8% average, all pieces within 6–10%", following the FPL interior values. Say whether the tolerance applies to the average or to every piece.
- Name the measurement method. Oven-dry samples (EN 13183-1), resistance meter (EN 13183-2) or capacitance meter (EN 13183-3), with the species and temperature corrections used.
- Ask about conditioning. If the timber will be resawn or glued, ask whether the kiln charge was equalised and conditioned. Both steps are explained in our kiln drying article.
- Check storage and transport. Covered, closed storage and wrapped packs keep the moisture content the kiln achieved. Open yards do not.
- Measure on arrival. Test at least 10% of the pieces before you machine, glue or install them, and keep the wood stickered in the room where it will be used if it needs to adjust.
- Design for the swing. Allow for a seasonal change of three points or more in heated interiors. Panels, tops and treads need fixings that let them move.
Frequently asked questions
What moisture content should wood be for furniture?
For furniture used in heated rooms, aim for an average of about 8%, with individual pieces between 6% and 10%; hardwood for furniture is normally kiln dried to 6–8%.2 In a very dry, continuously heated interior, the lower end of the range is safer.
What is the equilibrium moisture content of wood at 50% humidity?
At about 20–21 °C, wood in equilibrium with 50% relative humidity has a moisture content of about 9.2%. At 40% it is about 7.7%, and at 60% about 11%.1 Species and sorption history can shift these values by a point or two.
Is a pinless moisture meter accurate?
Yes, within its limits. Capacitance meters read reliably between about 4% and 30% MC once they are set for the species or density, but they mostly measure the outer layers, so a wet core can go unnoticed. For drying stock, an insulated pin meter driven to a quarter of the thickness, or an oven-dry sample, tells you more.2
Is 12% moisture content too high for indoor furniture?
For heated interiors, yes. Wood at 12% is in balance with about 65% relative humidity.1 In a room that runs at 40–45% in winter it will dry towards 8% and shrink as it does so, opening joints and gaps.1,6
