How Much Will That Board Move? And What Is a Board Foot?

Two numbers decide whether a piece of furniture survives its first winter: how much the wood will move, and whether the joinery lets it. The first is calculable to a useful precision from published data. The second is a design decision that the first one should drive.

Why wood moves at all

Wood is hygroscopic: it exchanges moisture with the air until it reaches equilibrium. That equilibrium point is the equilibrium moisture content, and it is set by the relative humidity and the temperature of the room, not by how long ago the board was dried. A kiln-dried board brought into a damp workshop will take moisture back on.

Below about 30% moisture content — the fibre saturation point — water is bound inside the cell walls, and adding or removing it changes the cell dimensions. Above that point the extra water sits in the cell cavities and the wood does not change size. This is why all movement calculations stop at 30%.

These come from the Hailwood-Horrobin equation as published by the USDA Forest Products Laboratory, which is a sorption model rather than a lookup table, so it handles any temperature and humidity you put into it. The 65% row landing on exactly 12.0% is not a coincidence of rounding — that pairing is the reference condition a lot of published wood data is quoted at.

Grain direction changes the answer by a factor of two

Wood does not shrink equally in all directions. Tangential shrinkage — around the growth rings, which is the width direction of a flatsawn board — is typically about twice the radial shrinkage across the rings. Longitudinal shrinkage along the length is so small it is usually ignored.

That two-to-one ratio is also the reason flatsawn boards cup. The face nearer the bark shrinks more than the face nearer the heart, so the board curls away from the heart. Quartersawn stock, where the rings run through the thickness, moves less and stays flatter — which is why it costs more.

Species matters as much as sawing. Eastern white pine flatsawn moves less than black cherry flatsawn, and quartersawn white pine moves a third of what quartersawn white oak does. Black walnut is the interesting row: it has an unusually low tangential-to-radial ratio, about 1.4 rather than 2, so quartersawing it buys you less than quartersawing oak.

The arithmetic

Buying the wood: what a board foot is

Hardwood is sold by the board foot, which is a volume, not an area or a length. One board foot is 144 cubic inches: a piece one inch thick, twelve inches wide and twelve inches long, or any other combination that multiplies to the same volume.

That third column is the National Hardwood Lumber Association standard for lumber surfaced on two sides, so 13/16 inch really is what surfaced 4/4 is supposed to measure rather than a yard shortchanging you. Thicker stock than 8/4 follows the same idea but check with your supplier rather than assuming. The important column for your bill is the middle one: you pay on the nominal thickness even though you take home the third.

Buying rough and surfacing it yourself changes this. Rough kiln-dried 4/4 is allowed to measure as little as 15/16 inch in the grading area under the NHLA kiln-dried rule, and once you have flattened a board with any cup in it you can easily finish under 3/4 inch. Plan your thicknesses from what you can actually get out of the board, not from the nominal figure.

Putting the two together

• Buy rough stock oversized. You will lose thickness to flattening and width to straightening an edge, and the board foot price is charged on what you carried out of the yard.
• Let the wood acclimatise in the room it will live in before you cut joinery — days for thin stock, longer for thick. Sticker it so air reaches all faces.
• Measure your shop and your home humidity across a year if you can. Designing for a 65% to 25% swing when your house actually runs 45% to 35% means over-engineering everything.
• Choose quartersawn for wide panels and tabletops where movement matters, flatsawn where it does not and the figure is what you want.
• Orient boards in a glue-up with the growth rings alternating if you want a flatter panel, or all the same way if you would rather the whole panel cup gently than ripple.

Humidity drives movement; temperature barely does

The equilibrium moisture content table earlier in this guide was for 70°F. It is natural to wonder whether a cold garage or a hot summer changes the picture. Running the same sorption equation at three temperatures shows that it does, but only slightly.

A fifty-degree swing in temperature moves the equilibrium point by about one percentage point at most. A thirty-point swing in humidity moves it by six or seven. So when you plan for wood movement, the number to find out is the range of relative humidity the piece will live through. A cheap hygrometer left in the room for a year tells you more than any published average.

How movement scales with width

Movement is proportional to width, so the figure for a 12-inch board is easy to scale. Here is flatsawn white oak, the species that moves most among the common hardwoods in this guide, across the full summer-to-winter swing used above: 65% down to 25% relative humidity at 70°F.

A 36-inch dining table top in flatsawn white oak can change width by more than three quarters of an inch over that swing. That is the figure a woodworker has to design around. It also explains why narrow parts, such as a 3-inch rail or leg, can be joined rigidly without trouble: a sixteenth of an inch spread across a joint is absorbed without damage, while three quarters of an inch is not.

When you build decides which way the wood will move

Movement is not symmetrical around the day you build. It is measured from the moisture content the wood has when you cut the joinery, so the season you build in decides whether the piece will mostly expand or mostly shrink afterwards.

Built at a middling humidity, this panel has nearly twice as much room to expand as to shrink. A frame-and-panel door built snug in that shop would be forced apart in August. The rule that follows is simple and often ignored: build in dry conditions and leave extra room for expansion; build in humid conditions and expect the piece to shrink, which opens gaps rather than cracking frames.

Turning the numbers into joinery allowances

• **Tabletop fasteners and buttons.** The slot in the apron or the button's tongue has to let the top move by the full calculated amount across its width, and the movement is split between both edges if the top is fixed at the centre. For the 36-inch white oak top, fixing at the centre means each edge moves by about half of 0.83 inches in the extreme case.
• **Breadboard ends.** Pin the tenon tightly only in the middle. Elongate the holes towards each edge in proportion to the distance from the centre, so the outer pins can slide by most of the half-width movement.
• **Frame-and-panel.** Size the panel so its edges do not bottom out in the grooves at the humid extreme, and so the groove still covers them at the dry extreme. For the red oak panel above, that means leaving roughly 0.15 inch of total expansion room across the width, split between the two grooves, if you built it at 40% RH.
• **Cross-grain glue joints.** Avoid gluing a wide board across the grain of another. If a design needs it, keep the cross-grain glue line short, or glue only a small section at the centre.

Choosing a species with movement in mind

When a piece has a wide solid panel that cannot easily be allowed to move, species choice becomes a design tool. On the same 36-inch top and the same 65% to 25% swing, flatsawn black cherry moves about 0.56 inches against white oak's 0.83. That is a third less movement from a change of species alone, before you consider quartersawing.

The shrinkage coefficients in the table earlier in this guide are the whole comparison. A lower tangential figure means less movement in flatsawn boards; a lower radial figure means less in quartersawn ones; and a ratio of tangential to radial close to 1 means less cupping. Eastern white pine is the stable end of the list, white oak and sugar maple the mobile end.

Checking the wood before you cut

All of this assumes the wood has reached equilibrium with the room it will live in before you cut the joinery. If it has not, it will keep moving after assembly, regardless of how carefully the joints were designed. A pin or pinless moisture meter tells you where a board is; the equilibrium table tells you where it is heading.

• Measure the moisture content of several boards on arrival and again after a week or two stickered in the room. When the readings stop changing, the wood has settled.
• Compare the settled reading with the equilibrium moisture content for the humidity in the room where the finished piece will live. If they are far apart, wait, or move the wood to somewhere closer to that condition.
• Check readings in the middle of a board as well as the ends. End grain exchanges moisture faster and gives misleading readings on a freshly cut board.
• Moisture meters are calibrated for particular species and temperatures, and most have a correction chart. Use it; a reading that is a couple of points off is the same size as the movement you are trying to plan for.

From cutting list to board feet: a worked order

Here is a cutting list for a small hardwood table turned into a lumber order. The dollar figure at the end is illustrative; use your yard's price per board foot.

The waste allowance is the line people leave out and then regret. Rough boards have checks at the ends, knots, sapwood and defects that have to be cut around, and flattening a cupped board costs thickness. How much to allow depends on the grade and on how selective you are about grain; many woodworkers use a figure somewhere between 20% and 40%. At an illustrative $10 per board foot, the difference between buying 56 and 70 board feet is $140, which is a lot cheaper than a second trip to the yard for one more board in a matching colour.

Softwood is different: nominal sizes

Dimensional softwood lumber from a home centre is not sold the way hardwood is. It is named by nominal size, and the actual dimensions are smaller: a dry, surfaced 2×4 measures 1½ by 3½ inches. It is usually sold by the piece or by the linear foot rather than by the board foot.

If you ever need a board foot figure for softwood, check which convention the supplier is using. An 8-foot 2×4 is 5.33 board feet by its nominal size but only 3.5 board feet of actual wood. Hardwood yards price on the rough nominal thickness, as described above; a softwood price quoted per board foot may be using either, and the difference is more than a third.

A 36-inch tabletop in six species, both ways

Putting species and sawing together gives the comparison that actually decides a design. Here is a 36-inch-wide solid top in each of the six species used in this guide, flatsawn and quartersawn, across the wide 65% to 25% swing, plus the flatsawn figure for a house that stays between 45% and 35%.

Three things stand out. Quartersawing roughly halves movement for most species, and cuts it by two thirds for white pine. Black walnut is the exception: quartersawing saves only 29%, because its tangential and radial shrinkage are unusually close, so paying the premium for quartersawn walnut buys less stability than it does in oak. And the climate column shrinks every figure to a fraction of an inch. In a well-controlled house, even the most mobile top on the list moves less than a quarter of an inch, which is well within what ordinary tabletop fasteners handle.

What a moisture content percentage actually means

Wood moisture content is expressed as a percentage of the wood's oven-dry weight, not of its current weight. That is why it can exceed 100% in very wet green wood, and why the arithmetic is slightly different from most percentages you meet.

Oven-drying a sample is the reference method that moisture meters are calibrated against. It is destructive and slow, so woodworkers rely on meters day to day, but it is worth understanding because it explains why meter readings are quoted the way they are, and why a board at 8% and one at 12% differ by four percentage points of oven-dry weight, not by four percent of what the board weighs now.

Sources

• USDA Forest Service, Forest Products Laboratory, Wood Handbook: Wood as an Engineering Material (General Technical Report FPL-GTR-282) — the species shrinkage percentages, the fibre saturation point and the Hailwood-Horrobin sorption coefficients: https://research.fs.usda.gov/treesearch/62200
• Species shrinkage figures in the table above are green-to-ovendry values from that handbook. Trade names covering several botanical species (mahogany, ipe) were checked against the specific species the handbook lists rather than the trade name.
• National Hardwood Lumber Association grading rules, for the quarter thickness system, the standard surfaced (S2S) thicknesses and the kiln-dried rough thickness allowance: https://www.nhla.com/

Where to go next

If the piece involves resin as well as timber, epoxy resin mixing and curing covers why a 1:1 ratio by volume is never 1:1 by weight and why pour depth decides whether it cures clear. Wood moving inside a cured resin pour is exactly the case where the movement figures above stop being academic.

Common questions