# The Dovetail Angle Is a Look, Not a Strength

Hewn Shop Notes · https://hewn.build/shop-notes/the-dovetail-angle-is-a-look-not-a-strength

AI-drafted, not yet checked: Drafted by Chip, the writing system Hewn runs on Claude, from Hewn's craft corpus. Jonathan Salomons has not read this note yet, and no outside reader has checked it. Read it as a shop note: check a number before you cut to it.

Cite as: "The Dovetail Angle Is a Look, Not a Strength." Hewn Shop Notes, hewn.build/shop-notes/the-dovetail-angle-is-a-look-not-a-strength. Drafted 2026-06-03; AI-drafted, not yet checked.

---

Every dovetail tutorial hands you the same rule. 1:6 for softwood, 1:8 for hardwood. It gets stated like a load rating, with the quiet threat that getting it wrong weakens the joint. Cut too steep and the short grain at the tip snaps. Cut too shallow and it slips. So you set your bevel gauge to a number a stranger told you was structural, and you lay out your tails to it like it matters.

For a glued joint, it doesn't. And there's a clean old test that settles it.

In 1958 the Forest Products Research Laboratory ran the experiment. Walker cut glued dovetails across a wide range of slopes, from quite shallow to quite steep, and pulled every one of them to failure. They all let go at the same load. They all let go the same way, shearing along the short grain right at the corner of the socket. The angle made no measurable difference to any of it. That's the headline, and it deserves to be said as flatly as the rule it kills: in a properly glued dovetail, slope ratio does not change the strength.

The reason is mechanism. What holds a glued dovetail isn't the geometry of the wedge. Glue locks the joint in every direction, so the interlock stops doing the structural work. What's left to fail is the wood itself, across the grain, at the re-entrant corner of the socket where the fibers are weakest and the stress piles up. That limit is set by the tensile strength of the pin board across the grain. It's the same whatever angle you cut. Change the slope and you've changed how it looks. You haven't touched what breaks it.

So why do the limits exist at all? Because there are two, and they're real. Go much below about 7° and you lose the mechanical lock, the bit of wedging that holds the joint together before the glue grabs. Go much above about 15° and the tail tip turns to fragile short grain that crumbles when you knock the joint home or when it takes a knock in service. Both limits are about assembly and abuse. Neither is about service load.

And here's where the old hardwood-softwood split finally earns a basis, just not the one it claimed. Hardwood is several times tougher across the grain than softwood. That's why it survives a steeper tail without the corner blowing out during assembly. The tradition encoded a real difference. It just filed it under strength when it belonged under toughness, and under service when it belonged under the knock-together.

So cut the angle you like. Lay the tails out with a sliding bevel set by eye if that's the look you want. Stay roughly between 7° and 15° so the joint locks and the corners survive the mallet, and inside that range the joint does not care about the rest. The angle is yours to choose. Choose it for how it reads across a drawer front, not for a strength it was never giving you.