# The Material

Hewn Shop Notes · https://hewn.build/shop-notes/the-material

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 Material." Hewn Shop Notes, hewn.build/shop-notes/the-material. Drafted 2026-04-24; AI-drafted, not yet checked.

---

Every plan was written by someone who assumed they could get flat, straight lumber at the specified thickness, in reasonable widths, without garbage in the working area. That assumption is correct often enough that people stop questioning it. Then they stand at the bench with a board already committed and realize the shelf is going to come out 5/8" thick instead of 3/4".

That is not a woodworking problem. That is an assessment problem. And it is solvable in fifteen minutes, before the saw turns on.

## The Yield Question

Your plan calls for a 3/4" shelf, 9" wide, 36" long. You have a 4/4 board. Nominal one inch, which should finish somewhere between 7/8" and 15/16" after S4S. Plenty of room.

Now look at the board. It has 3/16" of cup across the face. That's a gentle bow, nothing dramatic. But the face jointer run takes the high edges down to the low center. That's 3/16" gone. You're at 11/16" rough, which planes to maybe 5/8" finished. That board will not make the shelf.

The math was available before you touched the jointer: rough thickness minus defect depth minus planer allowance. If the result is less than the plan's finished dimension plus a hair of tolerance, the board won't yield. Pick a different board or buy thicker stock.

Twist works the same way. A board that winds 3/16" across its length loses that from the high corners in face jointing. For long parts, the twist depth is material cost and the yield arithmetic has to account for it. For short parts, you can crosscut the board into manageable sections and joint those individually. That's a legitimate move. Not every bowed board is waste. But it's a decision you make before the first cut, not after.

## The Width Problem

A lot of plans were written when wide single boards were common and cheap. They're less common now. A plan calling for a 12" door panel or an 11" apron is often assuming a supply your local yard doesn't carry at a reasonable price.

The answer is usually a glued panel. This is not a compromise. A well-made edge glue-up is as strong as the wood around it, and under most finishes the joint disappears.

But it's a decision, not a default. For painted work, the glue line genuinely doesn't matter. For clear-finished case sides and shelves where grain continuity isn't the point, a tight glue-up in a species without prominent figure is invisible. For show surfaces, a table top, a set of matched drawer fronts, a book-matched panel, the glue line changes what the piece looks like. That's a design question, not a technical one. The plan doesn't answer it. You do.

One thing the plan never mentions: a glued panel needs time. Twenty-four hours minimum before joinery. If that panel is a case side with dados for shelves, the case side doesn't exist as a workable part until the glue cures. That sequence step belongs in your build plan before you start.

## The Grade Map

A check that runs 3" into the end of a 48" board is nothing if the part doesn't start at the very end. A check that runs 6" into a board where your longest part is 6" and begins at the end is a sourcing problem. The question is always where the defect sits relative to the parts you need.

Lay the cut list against the actual board. Mark defects. Map parts. Where defect and part overlap, you either adjust the layout or replace the board for that part. Five minutes at the lumber stack prevents the discovery that costs half a day after crosscutting.

Knots follow the same logic. A sound, tight knot on a painted shelf underside is nothing. A knot where you're cutting a mortise is structural. That position is not suitable for that part. The knot didn't become a problem at the saw. It was always a problem. The assessment finds it first.

Sapwood is a visibility question. In walnut and cherry, the color difference between sapwood and heartwood is dramatic. If sapwood on a clear-finished show surface isn't intentional, it needs to be cut around. Account for the sapwood zone before you allocate stock.

## The Supply Gap

This is different from species substitution. Post 109 covered swapping one species for another. The supply gap is when you need white oak and the yard has white oak, but only in 4/4, and the leg blanks call for 8/4.

For thickness, laminating two layers of 4/4 works for many parts. It does not work for show surfaces where a centered glue line will be visible under clear finish. For those, source thicker stock from a different supplier or build in the lead time.

For species that appear intermittently, buy ahead. Hard maple in 8/4 for leg stock is not something most BC interior yards carry consistently. If you have a project that needs it, buy it when it shows up, not when you need it.

## Before the First Cut

All of this happens before the first cut. That is the point. The yield calculation, the width decision, the grade map, the supply response. These are the steps between the cut list and the saw. They are not overhead. They are the work.

The plan is an argument. The material is what you have to make the argument with. Read the material before you commit to the cut. That's how you find out whether you can honor the plan's terms, or whether you need to adapt them before the saw turns on.