GelvinGuitarsArchive

Free guitar building course · Part 2

Choosing wood,
and choosing
glue.

What genuinely matters in timber selection, what is folklore, and how to buy stock that is dry, stable and will not twist itself into firewood a year after you finish.

16 min read·Video 1:08:21

Watch on YouTube·Subscribe to the channel

Advertisement

The short version

Almost everything written about guitar wood is about species. Almost everything that actually matters is about properties. Is the board dry, is it stable, is it the weight you want, and is it strong enough to hold hardware under string tension. That is the whole game.

Get those right and a huge range of timbers will build you a good guitar. Get them wrong and the most famous, most expensive species in the world will still twist itself into firewood on you. So do not get hung up on the name on the label.

The tonewood argument

This is the most argued about subject in guitar building, so let me just tell you where I land and why, and you can do what you want with it.

On an acoustic, the wood species genuinely matters. The body is the instrument. It moves air, and what it is made of changes what you hear. No argument there.

On a solid body electric, it does not. And this is not a matter of taste, it is mechanical. Here is the actual chain:

  • The string vibrates and makes a standing wave along its length.
  • The pickup senses the change in its magnetic field as that steel string moves through it, and turns that into an AC signal.
  • That signal runs to your amp.
  • Nothing in that chain is acoustic. The pickup does not hear your body wood. It reacts to a moving piece of steel, full stop.

So on a solidbody, "this wood is warm, that wood is bright" is describing a mechanism that is not in the signal path anywhere.

What the species does change, reliably: how much the finished guitar weighs, how it balances on a strap, how it behaves under your tools, how it takes a finish, how stable it stays over the years, and how it looks. Those are all real and worth choosing for. That is a completely different claim from tone.

Plenty of people will fight me on this, and that is fine. Do not take my word for it, do not take anybody's. Build one and listen for yourself. It is your guitar.

The numbers, in one place

ItemFigureNote
Moisture content, goodUnder 10%Ready to use
Moisture content, wait12% or aboveNeeds longer to dry
Moisture meter costCheapThe best value tool in this whole list
Neck stock4/4 (1″)Usually plenty
Body stock7/4 or 8/4 (1 3/4″ or 2″)Bodies finish around 1 3/4″
Board foot1″ × 12″ × 12″Formula: (T × W × L) ÷ 144, all in inches
Cap or veneer thicknessMax ~3/8″1/2″ to 5/8″ if you are carving a top
Clamps for a body glue upThree minimumBig ones

Reading a board

Look at every face, including both ends. This is the single most useful habit you can build when buying wood. A board can look perfect from one side and be junk from the other, and the end grain tells you things neither face will, like where the core sits and how the board is going to move on you.

The core

A board cut through the pith, the dead center of the tree, will crack. I do not care how good the face looks. I had a redwood board that showed this perfectly: nice straight looking grain on top, splitting apart at the end, and the end grain showed the core running right down the middle of it.

Here is the kicker. That board is technically quartersawn. Quartersawn with the core still in it is still a bad board. Either side of the core would have been just fine. The core is the problem, not the cut.

Knots

A knot is where a branch was growing, and it is a different density than the wood around it. That causes you three separate problems:

  • It expands and contracts at a different rate than the rest of the board, so even after you sand it dead flush it comes back as the seasons change, sometimes pushing up like a little volcano.
  • It is much harder on your tools.
  • It does not cut clean. It chips and cracks under a router.

Twig sized knots are fine. Anything bigger, walk away, unless you specifically want that look, and if you do, make absolutely sure the board is properly seasoned first.

The worst neck I ever saw. A Fender neck came in and the owner figured it just needed a setup. It had a knot dead in the center of the neck, and that knot had put a permanent back bow into it that no amount of truss rod would pull out. The only real fix was pulling the fretboard and putting a two way rod in to fight it forever. He just replaced the neck. One knot, in the wrong spot, wrote off the whole thing.

Grain and where it belongs

Grain slightly off straight is fine. Dead straight is ideal. What you are really after is consistency, because big changes in the grain pattern across a board are exactly where cracking starts.

Tighter, straighter grain gives you a stronger, more stable neck. A spot where the grain opens up and gets coarse is perfectly good for a body, but keep it out of a neck.

Fretboard timber

Ebony is my preference. Very dense, very tight, tough, with natural oils in it. Holds frets like nothing else. It comes coated in wax to keep moisture out and shows up nearly seasoned. Those white streaks you see are normal and harmless, and you can stain them out if they bug you. It is hard on tools, but worth it.

Rosewood is dense and tight too, but nowhere near ebony, and the quality is all over the map.

Species does not equal species. Two boards of the same wood, cut from the same tree, can act completely differently. I have a rosewood board that weighs about half what a comparable piece of ebony does and gives a little when you squeeze it. That board is going to dent under hard hammer ons and wear badly. That is one board's problem, not rosewood's problem. Judge the board in front of you, not the name.

For a fretboard you want density. If it feels spongy in your hand, it is going to dent.

Most timbers will work as long as they are strong enough to take the string tension. The exception is anything with coarse open grain. Oak has been used and it works, but the grain feels rough under your hand unless you fill it completely, so why bother.

The weight rule, and its one exception

  • Bodies and necks: the board should feel light for its size. Light for its size means dry, and it makes a lighter guitar, which is what players want. Heavy for its size means there is still water in it.
  • Fretboards: exact opposite. Here you want it to feel heavy for its size. That is density, and density is what holds your frets and resists denting.

Buying lumber

Quarters

Hardwood thickness gets sold in quarters of an inch:

CalledThicknessUsed for
4/41″Necks, usually enough
6/41 1/2″
7/41 3/4″Bodies
8/42″Bodies, with material to spare

Those are nominal numbers, not real ones. A board sold as 4/4 might actually measure 7/8″ or even 3/4″. Measure it before you buy, every single time. Assuming 4/4 means a true inch is exactly how people end up with a neck blank that is too thin to use.

Board feet

A board foot is a piece 1″ thick by 12″ by 12″. The formula is:

board feet = (thickness × width × length) ÷ 144, everything in inches.

Run it on a board that is 9 ft long, 10″ wide and 2″ thick:

  • 9 ft × 12 = 108″ long
  • 108 × 10 = 1,080 square inches
  • 1,080 ÷ 144 = 7.5 board feet if it were 1″ thick
  • × 2 for the real thickness = 15 board feet

Multiply that by whatever your supplier charges per board foot and you have got your price, and on a good hardwood that number gets big in a hurry. Suppliers measure nominally in their own favor too, so an 11 1/2″ board gets charged as 12″. Lean toward buying the bigger piece, since you are paying for it either way.

What to check before you hand over money

  • A slight bow along the length is normal and fine, as long as it is not severe.
  • Cupping across the width, walk away.
  • Expect to mill big boards yourself. Resaw one face straight on a bandsaw or table saw, then work off that reference.

When you find the right board, buy it. It is rare to find a piece that is exactly what you want. Usually one end is gorgeous and the other tapers, cracks or twists. When a genuinely good one turns up, do not gamble on finding its twin next month, because you probably will not.

Buying big and milling your own is where you actually save money. A pre made fretboard blank is not much on its own, but a single block of ebony gives you three of them for roughly the price of one blank, so that one is close to a wash. Rosewood and maple you come out well ahead, since the boards are cheap and you get several fretboards out of each.

Moisture

Buy a moisture meter. They are cheap, you push two probes into the board, and it takes all the guesswork out of the single most important property of the wood. Easiest money you will spend.

  • Under 10%, good to use.
  • 12% or above, leave it longer.

Here is what moisture actually does to you down the line:

  1. Extra weight in the finished guitar.
  2. Warping, twisting and general instability.
  3. Loss of sustain, since the water damps the vibration.
  4. Shrinkage after you machine it. You route everything to fit perfectly, the wood dries, and now nothing fits. That one hurts.

"Season it for two years" is folklore. Seasoning just means drying. You do not need a rule about how much time has passed when you can measure the actual number. When the meter reads low, the wood is ready, whether that took two years or six months.

One thing to watch: after you cut a board, test the fresh end grain right in the center. The inside of a board stays wet a lot longer than the faces do, because the faces have been sitting out in the air.

Judging a board: some real calls

Let me run through some actual boards. The reasoning matters more than the verdicts, so pay attention to the why:

  • Walnut, no. Looked fine from one face, had problem spots on the other. I only caught it because I checked both sides.
  • Cracking maple, no. Once a crack shows up, the grain around it is already weak, and machining it just opens it further.
  • Ebony with a small crack and a knot forming, no. You could use it in a pinch, but it is a bad call. It is going to curve as it dries.
  • Heavily figured bird's eye maple, works, but not for a customer. Every one of those "eyes" is a twig that started and stopped growing, so the board is full of tiny knots. Worth knowing the standard you hold yourself to, and that it can sit higher than just "will it work."
  • Lightly figured bird's eye maple, yes. Cut a year ago and still dead flat. Stability beats figure.
  • Canary wood with a flaw, yes. Because I can shift where the fretboard sits on the board so the flaw lands right at the centerline where it gets cut away. Look at what you can get out of a board, not just at the board.

Safety: sensitizers and toxins

Research any wood before you buy it. Some species are sensitizers. Repeated exposure, dust on your skin, in your eyes, in your lungs, can make you permanently allergic to them. Some woods carry real toxins. And sensitization tends to spread: get reactive to one species and you can find yourself reacting to a bunch of others too.

And this is not just the exotic stuff. Mahogany, maple and oak are all worth protecting yourself from. Gloves, respirator, eye protection, dust extraction. Every time.

It hits whoever plays the guitar too. Toxic oils in a fretboard can leave a player breaking out in blisters months down the road, and it almost always gets blamed on a nickel allergy when it was the wood the whole time.

People come to a custom build wanting purple heart, leopard wood, snake wood, all of it. Look every one of them up first, both for your own shop and for whoever is going to be holding that guitar for hours at a stretch.

Oily woods and glue

A high natural oil content is great for the player, since that fretboard will go a long time before it ever needs oiling. It is a headache at glue up, though.

Gluing an oily board is like trying to glue a stick of butter to a piece of wood. The glue just will not grab.

The fix: scrub the glue face with acetone and a rag until no more oil comes off on the rag, then glue it immediately. Wait too long and the oil migrates back out of the rest of the board into the face you just cleaned, and you are right back where you started.

Cocobolo and several of the rosewoods are the usual suspects. As a rule, the more exotic the wood, the more likely you are dealing with this.

Figure, spalt and burl

Spalting is fungus. Those black lines everybody finds so beautiful are a mold that has worked its way through the wood, and it weakens the cell structure everywhere it has been. You can build with it, but treat it as the compromised material it is. And wear your protection cutting and sanding it, because you are machining mold.

Burl is closer to a tumour, an abnormal growth, usually down near the base of the tree.

The general principle: anything that departs from the normal grain is usually weaker right there. Knots, spalt, burl, heavy figure, the visual interest and the structural weakness come from the same cause. A lot of what we call a beautiful piece of wood is really a tree that was sick.

That does not mean avoid it. It means use it where strength is not its job, as a cap or a veneer over a solid substrate underneath. Keep caps to about 3/8″, or 1/2″ to 5/8″ if you are carving a top.

Every grain has a weakest line

A flamed maple veneer tears easily along the flame and resists tearing across it. It fails along its weakest path like a zipper. All grain does this, flame just makes it easy to see.

In practice, this is why flamed, quilted and bird's eye maple chip and tear so easily. Do not run them through a planer, ever. Use a drum sander, or work them by hand.

Glue

GlueVerdict
Titebond II or IIIWhat I use. Stronger and more water resistant than Original.
Titebond OriginalCheapest, works fine, plenty of builders use nothing else. Just less water resistant.
Hide glueGreat for acoustics, vintage work, and anything you might need to take apart later, since heat releases it. You do not need it on a solidbody.
Polyurethane, the foaming kindNo. Read the label. Some familiar brands sell both a wood glue and a polyurethane. Get the wood glue.
Epoxy, spray adhesiveNo.

Grab Titebond II or III and quit thinking about it. The few extra cents buys you water and heat resistance you may as well have.

How strong is a glue joint, really

I put this to the test myself, on a guitar I built years ago and never finished. Hit it hard right at the neck joint:

The glue did not fail. The wood broke. And here is the part I love: that joint turned out to be hide glue, which is weaker than the PVA glues above it, and it still outlasted the mahogany around it. I broke a second joint with a lot of force and it did the exact same thing.

So the answer to "is the glue strong enough" is that a proper glue joint is stronger than the wood on either side of it. The joint is not your weak point.

Which raises the obvious question about all those imported guitars that snap at the scarf joint. Two causes: the wrong glue, and over clamping. Squeeze so much glue out that the joint is starved and it fails, no matter what glue you used.

Gluing up a body

  1. Joint both mating faces flat. A jointer is the right tool. A table saw cut cleaned up with a sanding block also works, and that was my method for years with good results.
  2. Dry fit and check the seam. You should see one clean line and no gaps.
  3. Spread an even coat. Not too much, not too little. A cheap flux brush does it, a paint roller is faster if you are doing several bodies. Buy the brushes by the fifty and throw them away after.
  4. Slide the two halves against each other to transfer glue evenly onto both faces.
  5. Three big clamps minimum. Work from the center out.
  6. Watch for the halves shifting. A clamp that is not square will push them out of line, so fix the clamp instead of wrestling the wood by hand.
  7. Couple of hours to working strength.

You want some squeeze out, not all of it. A little glue escaping the seam tells you the pressure is right. Cranking it down harder pushes the glue out of the joint and starves it, which is the exact failure mode in those mass produced scarf joints. Too much clamping pressure is a real way to make a weak joint, so ease off.

Fewer glue lines is better where it is practical, a one piece body over a two piece, a one piece neck over a laminate. But do not kid yourself that glue seams change the tone of a solidbody, because they do not.

Reading grain to predict how a board will move

This is the part most builders only learn the hard way, after it has already gone wrong on them.

The light colored earlywood is the absorbent part. It acts like a sponge, taking on moisture and swelling. The dark latewood barely moves, and behaves more like a thin sheet of metal.

Now glue a sponge to a sheet of metal and get it wet. The sponge swells, the metal cannot, and the whole thing curls. That is exactly what a board does.

  • Flatsawn tends to cup across the face as the growth ring arcs try to flatten out.
  • Quartersawn mostly just expands and contracts in width instead, which is why it is more stable.

Why this shows up in your frets

You set your frets perfectly flush with the edges of a finished fretboard. Then the seasons change:

  • Humidity rises, the board swells, and now the fret ends are sitting inside the edge.
  • The air goes dry, the board shrinks, and the fret ends stand proud and catch the player's hand.

That is the sharp fret ends complaint you hear about, and it is almost always wood that was not properly seasoned, sometimes with a weak finish on top of it.

No finish is truly waterproof. A thick coat of nitro or urethane might get you to 98 or 99 per cent, but moisture still moves, just slower. You cannot seal your way out of wet wood, and do not try to fudge it in your dimensions to compensate either. The fix is upstream: use properly dried wood in the first place. That is the whole reason the factories kiln dry.

Watch the builds on YouTube

Full build videos, start to finish, including the parts that went wrong.

Build series · New video most months

Tools and pages that go with this

Fret calculatorString tension

Last reviewed: August 2026

Written by Will Gelvin

Former owner and luthier at Gelvin Guitars, builder of custom instruments, and inventor of the patented Variable Inductance Pickup. I built this archive to explain not only what works, but why it works. About my experience and approach.