Wood Movement Allowance Calculator
Every guide tells you to slot the screw holes and open them up a bit with a rasp. None of them says how much. This works out the actual length, from your panel, your species, your climate — and where each fastener sits, which is the part that changes the answer most.
Movement is anchored, and adds up outward
A panel does not swell as a block. It expands about whatever point holds it fast, and a point at distance d from that anchor travels d multiplied by the species coefficient and the moisture change. That single line is the whole tool, and it carries a consequence worth sitting with: a top fixed at its centre moves only half its total change at each edge, while the same top fixed along one edge moves the entire change at the other. Two builds with identical wood and identical weather need slots of quite different lengths, purely because of where they were pinned.
It also means the slots in one piece are not all the same. A fastener near the anchor barely needs relief; one at the outer edge needs the lot. Cutting them all to the outermost length is harmless but fussy; cutting them all to the innermost is how a top splits in its first winter. Work out the travel at each position and cut accordingly.
Start from the raw dimensional change with the Wood Movement Calculator, and size the stock to buy with the Lumber Calculator.
Frequently Asked Questions
How long should a slotted screw hole be for wood movement?
Long enough for the travel at that particular screw, which is not the same for every screw in the piece. The slot needs the shank diameter plus the distance the wood will shrink plus the distance it will swell, measured from the position it occupies on the day you build. A 36-inch walnut top in a heated house changes width by about three tenths of an inch across the year, so a screw at the outer edge of a centre-anchored top needs roughly an eighth of an inch of travel either side of the shank — and a screw three inches from the centre needs barely a sixteenth of that.
Why does it matter where the screw is?
Because a panel does not move as a block — it expands and contracts about whatever point is holding it fast, and the travel at any point is proportional to its distance from that anchor. Fix a top at its centre and each edge moves half the total change; fix the same top along one edge and the far edge moves the whole of it. This is the part almost every guide leaves out when it says to slot the holes, and it is why cutting every slot the same length is either wasted work in the middle or a split top at the edge.
Can I just glue a tabletop to the apron?
Not across the grain. Glue does not stretch, and a cross-grain glue line holds the panel at a width it will not stay at, so the panel splits rather than the joint letting go. This is the single most common way a first solid-wood table fails, and it usually shows up in the first winter when the heating comes on. Fix the top at one point and let everything else slide: buttons in a groove, figure-eight fasteners, or screws through slots cut to the lengths worked out here.
Does it matter what time of year I build?
It changes which way the wood will move, not how much slot you need. Build in a dry spell and the piece is near the bottom of its range, so almost all the travel is swelling; build in a damp one and it is nearly all shrinkage. Either way the slot has to cover the whole range, so its total length is the same — what changes is where the screw sits inside the slot on the day you drive it. Building at the middle of the range is the tidiest option, since the screw starts centred and has room both ways.
How accurate are these figures?
They are planning estimates, and the uncertainty is in the wood rather than the arithmetic. The coefficients are published shrinkage figures and the seasonal ranges are conventional values for each setting, but your actual equilibrium moisture content depends on your house, your finish and how long the stock acclimatised in the room it will live in. Wood brought straight from a cold store moves more than any table predicts. When the answer is marginal, cut the slot longer — nothing is lost by a slot that never reaches its end.
A planning estimate from published shrinkage coefficients and conventional seasonal moisture ranges. Your own conditions vary; when the answer is marginal, cut the slot longer.