Cross Grain

A cutting list says the top is 300 mm wide. It cannot tell you it will be 305 mm in August, that the customer's heated house will pull it to 294 mm in February, or that the cleat you drew across the back will split it either way. Cross Grain works those three things out, and then tells you what to do about them. COUNTS FROM THE ROOM, NOT FROM A RULE OF THUMB • Wood ends up at whatever moisture content the air around it holds. The app calculates that with the Hailwood–Horrobin sorption model published in the USDA Wood Handbook — it reproduces the Handbook's own table to within a twentieth of a percentage point • A heated house in winter is far drier than most people expect. That is when panels split, not in the damp half of the year • Set the shop you build in and the room the piece is going to. They are different problems and the app keeps them apart THE MOVEMENT NOBODY SEES COMING • The seasonal swing is the one everyone knows about, and it repeats every year • The one-way settling is the one that ruins pieces: build a 900 mm oak top in a damp shop at 14% and deliver it into a house holding 7%, and it arrives 900 mm wide and stays at about 890 mm. It never comes back • Both are reported, separately, with the direction named — swelling and shrinking fail differently HOW IT WAS CUT IS WORTH ABOUT A FACTOR OF TWO • Flat-sawn boards move roughly twice what quarter-sawn ones do across the width. Same tree, same width, and the only difference is the angle the growth rings cross the face • Quarter-sawn stock is stable in width and moves through the THICKNESS instead — which is what catches a tenon that has to stay in its mortise • The species table can be ranked for the cut you are actually buying, and by which timbers reward quarter-sawing at all. Padauk barely cares; red oak cares enormously THE VERDICT IS ABOUT THE JOINT Six millimetres of movement is nothing in a groove and a split panel in a glued frame. So the app asks how each board is held, and answers accordingly: • Floating in a groove — the groove depth that keeps it engaged at its narrowest • Breadboard end or slotted screw — the slot length the fastener has to travel • Buttons and z-clips — fine, and what would spoil it • Screwed through round holes, glued into a frame, or a cleat glued across the grain — the three that fail, why, and what to do instead. One tap swaps the joint DRAWN, NOT DESCRIBED The home screen is the board itself, drawn to scale and redrawn as you drag a slider from the driest point of the year to the dampest, with the end-grain section showing how it was cut. The movement is exaggerated so it is visible on a phone — the factor is printed beside the drawing and can be set back to true scale, because a drawing that quietly lied about its scale would be worse than no drawing. EVERYTHING ELSE • The sorption curve for any temperature and humidity, and the inverse: what a room would have to hold to keep wood at a target • How long a board of a given thickness takes to acclimatise — it goes with the SQUARE of the thickness, so doubling the thickness quadruples the wait • The width to mill today so it finishes on size once it settles • A cutting note for the wall above the bench, CSV of every board with the derived columns, and a full JSON backup 100% OFFLINE. NO ACCOUNT. NO TRACKING. Everything stays on your device. There is no sign-up, no server and no analytics. The app declares exactly one permission — INTERNET — used solely to open our privacy policy page from Settings. Our Play Data safety declaration is "no data collected, no data shared", and it is accurate. The shrinkage figures are published species averages, not measurements of the board in your hand. The app says so wherever it shows one: treat every number as the size of the problem rather than a promise. Free. No ads. No in-app purchases.