What bevel should I have when carving hardwood

Go short on hardwood, because a long bevel leaves the steel thin exactly where oak will break it. Then round the heel away so the edge still reaches the wood without lifting your elbow. That one change fixed a problem I carried for years, and it costs nothing but five minutes on a wheel.

Does your shoulder ache after a full day at the bench? Have you ever had a corner of an edge simply fold over in white oak? Somebody asked me this on a livestream and it deserves a proper answer, not a number.

Three gouges sit in front of me as I write. Same sweep, same size, all of them a number 3, and all three ground differently on purpose. One short. One long. One that is barely a bevel at all.

What the three grinds actually do

• Six millimetres of bevel, a quarter inch, is what I run on hard timber most days.
• Twenty millimetres, a little over three quarters of an inch, is the long grind, and it is fragile at the tip.
• The third tool has no heel left on it. I rounded it away, and that is the one I reach for.
• Short bevel plus a rounded heel gives strong steel and a low hand at the same time.
• Tools you strike keep their corner exactly as it came. Those need to sit high anyway.

What exactly is the heel of a gouge?

The heel is the back corner of the bevel, where the ground face meets the outside of the blade. Sight along any gouge and you will find it: a definite edge running across the metal, a step you can feel with a fingernail.

That corner decides how the tool sits in a cut. The edge does the carving, but the heel is what the tool rides on once the blade is buried, and riding is the word. A proud heel holds the blade up off the surface, and everything your arm does afterwards is a response to that.

Most people never look at it. Every argument about sharpening a carving tool is really an argument about the heel, whether the person talking knows it or not.

Why does a long bevel fail in hardwood?

Thin steel is the reason, and you can see it by eye if you hold the tool up to a window. Grind a bevel long and the steel tapers away to almost nothing. Beautiful in basswood. Genuinely weak in something harder.

Put that fragile tip into oak and the wood wins. Not always dramatically. Often it just folds a corner over, or takes a small bite out of the edge, and you spend the next twenty minutes at the wheel wondering what you did wrong. You did nothing wrong. That grind was simply wrong for the timber in front of it.

The grind	Length on a number three gouge	Good for	Where it lets you down
Short bevel	About a quarter inch, 6 mm or so	Hard timber	Forces the handle high, which reaches your shoulder by evening
Long bevel	Roughly 20 mm	Slices beautifully in soft timber while the hand stays low	Thin at the tip, and oak finds it
Short, heel rounded away	Still 6 mm	Hardwood by hand, hour after hour, with no raised elbow anywhere in it	Wrong under a mallet
The heel exactly as it left the factory	Whatever it arrived with	Struck work	Rides high the moment you try carving by hand with it

Which hardwoods punish a long bevel?

Oak and cherry are where I notice it first, though oak is not one wood and that matters. Brown oak runs a shade softer than the white American kind. Red oak punishes an edge as well. Lump them together and you will be surprised by the day one of them behaves differently from the last.

Cherry has its own habit. It carves sweetly and then, in one patch of wild grain, asks far more of the edge than the rest of the board ever did. That patch is where a long bevel gives up.

How hard is hard? Laboratories answer that with the Janka test, and species figures are published by the Forest Products Laboratory. Numbers are useful for buying timber. They will not tell you how a board behaves under a gouge, which changes inside a single plank as the grain wanders.

White American oak is the one I would warn a beginner about first. Red oak bites back differently, with those open pores that tear the moment your edge goes even slightly dull. The brown variety is the gentle relative of that family, and still nobody would call it soft.

How do you know your bevel is wrong?

Your body reports it before your eye does, so listen to the complaints in this order. First the shoulder, late in a long day. Then the elbow, which you will notice sitting higher than the elbow of the hand that is only holding the work.

After that, look at the surface you are leaving. A tool riding on its heel does not slice cleanly. It skids, then bites, then skids again, and the cut shows that stutter as a series of small flats where a continuous sweep should be.

The third symptom is the one people misread most often. You sharpen, you carve for a few minutes, the edge feels gone again, and you conclude the steel is poor. Sometimes the steel is poor. More often the grind is so long that the tip was never strong enough for the timber, and no amount of honing fixes a geometry problem.

None of this makes hardwood the enemy. Furniture wants hardwood panels, an acanthus leaf in oak holds a crispness basswood never will, and that crispness is worth the trouble. You simply grind for the material in front of you instead of grinding once and hoping.

What goes wrong with a short bevel?

Your arm goes wrong, slowly, over hours. Watch what a short bevel forces you to do: the heel stands proud, the edge will not reach the surface, so up comes the handle. I can raise a tool a long way and still not be into the material, and I mean a long way.

Raise the handle and the elbow follows it up. After the elbow comes the shoulder, every time. Do that for eight hours a day, every day, and you will feel it. You do get used to it, in the way people get used to any small daily damage, but tired is tired and your body keeps the score.

Which is exactly the trap. Hardwood needs short steel for strength, and short steel makes you carve with your shoulder in the air. For a long time those two facts sat in front of me like a choice.

How do you grind the heel off a gouge?

Grind that corner away. A smooth curve goes where it used to be. That is the whole trick, and it is the single most useful thing on this page.

Look at my third tool and no heel survives at all. There is no step and no corner left, only a soft rounding running back into the body of the blade. The bevel is still short, so there is still real metal behind that tip. But nothing stands proud any more, and the blade settles into the cut while my hand stays low.

The elbow stays down, the shoulder stays down, and nothing aches at seven in the evening. I get the strength of a short grind and the reach of a long one out of the same piece of steel, and the only thing I gave up was a corner that was never doing me any good.

It is a wheel job, and a light one. Most of the shaping work is described properly in the sharpening course, because doing this badly is as easy as doing it well, and the second attempt costs you more steel than the first.

When should you keep the heel?

Keep it on every tool you strike, and here is the geometry nobody explains. Striking a gouge means hitting the very top of the handle. On mine that target is about three inches, three and an eighth really, somewhere near 80 mm of exposed handle.

You cannot slide a tool along the surface of your work and swing at the same time without the head catching your work. Up comes the handle regardless. A softened corner buys you nothing there, because the mallet has already forced the tool into the position this whole trick exists to avoid.

Hardwood work does not call for a light mallet either. Mine runs two pounds, a full 32 ounces, near enough 900 grams, and that mass is doing the work so my arm does not have to.

Does this apply to chisels too?

Every edge that enters wood at a low angle has the same argument going on behind it, so a flat chisel answers to it exactly as a gouge does. Length of grind against strength of tip, and height of heel against the height of your own hand.

What changes is how visible the fault is. Across a wide flat edge a proud heel lifts the whole blade evenly, and you feel it as the tool refusing to take a fine shaving. On a curved sweep only part of the heel touches, so the tool rocks, and that rocking is easy to blame on your hand instead of on the grind.

The same reasoning runs under the grain question as well. Working uphill into the fibres asks more of the edge than working with them, and a long fragile grind fails on the uphill stroke first. Fix the geometry and a surprising number of grain problems quietly stop being problems.

Where did this question come from?

A student asked it on a livestream, after watching me demonstrate the long bevel I use on most of my tools, and it is a better question than it looks. Do I shorten the bevel for harder wood? Yes. That single word would have been a poor answer on its own, though, because the interesting part is what you do afterwards.

Questions like that are why the live sessions are worth sitting through. Somebody watches carefully enough to notice a contradiction between what I said last week and what my hands are doing this week, and then everybody learns something. The same thing happens in the livestream archive and in the techniques lessons, where most of the material began as somebody's awkward question.

For hardwood by hand I keep two gouges within reach at all times. One carries the rounded heel. The other still has its factory corner, left just as it came. Working between them tells me more about a board in five minutes than any hardness chart will, and if the ordinary one starts skidding I know the timber has hardened up and the rest of the session belongs to the rounded tool.

How many differently ground tools do you need?

Two, for hardwood, and I keep exactly that on the bench. One gouge with the heel rounded away for hand carving, one untouched for the mallet. Same sweep, same size, ground for two different jobs.

People assume a working carver grinds every tool the same way. Walk past my bench and you would find the opposite: sweeps ground for the timber and the method, not for a rule out of a book. The number on the handle tells you the shape of the edge. It tells you nothing about the bevel behind it, and the bevel is where the day is won or lost.

Start with one tool. Pick one gouge you use constantly and take its corner off, carve a hardwood panel with it, and pay attention to where your elbow sits. Your arm will tell you inside an hour whether I am right.

That kind of small correction is most of what separates a comfortable carver from a sore one. It is also the sort of thing nobody writes down, which is why I keep a whole course about tools and why which brand you buy matters far less than what you do to the steel afterwards. A cheap gouge with the right grind beats an expensive one carrying the factory corner, which is roughly what I found testing the BeaverCraft set as well.

So take a gouge you trust, look at the heel, and decide whether it is earning its place. If it is not, take it off. And if you want somebody to walk you through the grinding rather than guessing at it, pick a course and begin.

Shall we carve?