Knife sharpening, from dull to working sharp

What "sharp" actually means

An edge is the line where two surfaces meet. Sharp means that line is narrow — a few microns across — and clean, with no folded metal hanging off it. Dull means the line has grown wide and rounded, either worn away or rolled over.

That definition tells you exactly what sharpening is: you are not adding anything. You are removing metal from both sides until the two bevels meet at a line again instead of at a landing strip. Stones, angles, and systems are all just methods for doing that repeatably.

It also explains the most common frustration on the subject. Someone spends twenty minutes on a fine stone and the knife is no sharper, because the bevels never met — they were smoothing the shoulders of an edge that still had a flat on top. Fine grit cannot fix that. Only coarser grit and more patience can.

Working sharp versus showing sharp

A blade that shaves arm hair and a blade that survives a day of work are not always the same blade, and this is where a lot of sharpening advice quietly misleads people.

A very fine, very acute edge cuts beautifully and gives up quickly on anything abrasive — cardboard, rope, dirty wood, hide with grit in it. A slightly coarser edge left at a medium grit, with microscopic teeth still in it, feels less impressive on paper and outlasts the refined one by a long way on exactly that material. Kitchen knives want the first. A work knife usually wants the second.

The angle conversation

Angle is the biggest decision in sharpening and the one most people never actually make — they just follow whatever the factory left behind.

Two numbers get used and they are not the same. Per-side angle is what you hold against the stone. Included angle is both sides added together. A knife "sharpened at 20 degrees" per side has a 40-degree included edge, and half the arguments on this subject are two people quoting different numbers at each other.

• 12 to 15 degrees per side — kitchen and fine slicing. Wonderful cutters, and they roll or chip the moment they meet bone, a staple, or a cutting board made of the wrong thing.
• 17 to 20 degrees per side — the general-purpose range, and roughly where most decent folders and fixed blades leave the factory.
• 20 to 25 degrees per side — hard use, chopping, and anything that gets treated worse than a knife should be. Less delicate in fine cuts, far more durable.
• Machetes, axes, and anything that meets wood at speed live above that range, usually with a convex edge rather than a flat bevel.

Two rules about angle that save hours

First: keeping a factory angle is dramatically less work than changing one. Reprofiling a blade to a new angle means removing metal along the whole bevel on coarse grit — a session, not a touch-up — and people who set out to "just try a lower angle" on a Tuesday evening usually regret it by ten o'clock.

Second: if an edge keeps rolling or chipping in normal use, the angle is too acute for what you are doing with it. Two or three more degrees per side is the fix, and it costs you far less cutting performance than you expect. The reverse is also true — if an edge feels like it is wedging rather than slicing and it never chips, you have room to go thinner.

The burr is the whole game

There is exactly one reliable signal that you have sharpened far enough on a side, and it is the burr: a thin lip of metal pushed over to the opposite face, formed at the moment the two bevels finally meet.

The process is the same on every system and every abrasive. Work one side until you can feel a burr along the entire edge, heel to tip. Feel for it by dragging a fingertip off the edge, away from it, never along it. Then switch sides and raise the burr the other way. Then remove it with lighter, alternating passes, and finish.

Nearly everything that goes wrong in sharpening is a burr problem. No burr means the bevels never met, and the edge is still flat on top no matter how long you have been at it. A burr everywhere except near the tip means the tip is under-sharpened, which is why the tip is the dull part of most otherwise sharp knives. And a burr that will not come off — the knife feels sharp, then dull, then sharp again within a minute — means it needs lighter passes or a strop rather than more grit.

Stones, plates, and what grit means

Grit numbers are not standardised across systems, so a 1000-grit water stone, a 1000-grit diamond plate, and a 1000-grit belt are three different surfaces. Treat grit as a rough ladder rather than a specification.

• Coarse, roughly 200 to 400 — reprofiling, repairing chips, and reviving something genuinely dull. Removes metal fast, which is as dangerous as it sounds on a blade you like.
• Medium, roughly 600 to 1000 — where most real sharpening happens, and a perfectly honest place for a working edge to stop.
• Fine, 2000 and up — refinement. A smoother, more slicing edge, at the cost of some of the bite that helps on rope and cardboard.
• Diamond plates cut every steel, stay flat, and need no soaking, which is why they have taken over for modern high-hardness steels. Water stones cut fast and feel wonderful, dish with use, and need flattening. Oil stones are slow, forgiving, and nearly indestructible. Ceramic rods are maintenance tools, not sharpening tools.
• The pull-through carbide "sharpeners" at hardware counters do not sharpen — they tear metal out of the edge at a fixed angle. They will make a dull knife cut again and take years off the blade doing it.

Freehand versus a guided system

Freehand means you hold the angle. It is fast once learned, works on any blade shape including curves and recurves, costs almost nothing, and takes genuine practice before the results stop varying.

A guided system clamps the blade and fixes the angle mechanically. It is far more repeatable, it is the honest recommendation for anyone who sharpens a few times a year, and it is slower to set up and more awkward on long or oddly shaped blades.

• Sharpen occasionally? A guided system will beat your freehand results, because the variable that ruins freehand is inconsistency and a clamp removes it.
• Sharpen often? Learn freehand anyway. The muscle memory transfers to a stone on a tailgate, which is where you will eventually need it.
• The trick that teaches freehand fastest is a marker. Color the bevel, take three passes, and look. Ink gone at the shoulder means your angle is too low. Ink gone only at the very apex means it is too high. Ink gone evenly means you have found it — now hold that.

Stropping, and what it actually does

A strop is leather or firm fabric, usually loaded with a very fine abrasive compound, and the blade is dragged backwards along it — edge trailing, never leading.

Two things happen. The compound removes a tiny amount of metal and refines the apex further than most stones reach. And the flexible surface knocks off and straightens what is left of the burr, which is why a knife that felt oddly inconsistent off the stone often feels genuinely sharp after ten light passes.

The trap is pressure. Leather compresses, wraps up around the apex, and rounds it — so heavy stropping makes a knife duller while feeling exactly like progress. Light passes, edge trailing, and stop while you are ahead.

And the honest limit: stropping maintains, it does not sharpen. A knife that has truly gone dull needs a stone. A knife that has lost its last few percent gets them back on a strop in a minute — and doing that regularly is why some people almost never need the stone.

Serrations and recurves

Serrated edges are sharpened from one side only. Each scallop gets a tapered rod, ceramic or diamond, sized to fit it, and you follow the bevel that is already there rather than inventing a new one. The flat side is left alone apart from wiping the burr off it at the end.

It is slow and fiddly, and the payoff is real: serrations keep cutting rope, webbing, and seatbelt far longer than a plain edge does, which is why rescue and marine knives have them.

Recurved blades — anything with an inward curve in the profile — cannot be sharpened on a flat stone at the curve, because the stone bridges across it and never touches the low point. A round rod or a narrow, file-shaped hone is the answer there, and it is worth knowing before you buy a blade with that profile.

The field touch-up, which matters more than any of this

The honest version of knife maintenance is that almost nobody carries a sharpening setup and almost everybody lets a blade go all the way to dull before doing anything. The fix is not a better kit at home. It is touching up early.

• A small ceramic rod or a credit-card-sized diamond plate weighs almost nothing and lives in a pack. Ten passes a side the moment the knife starts to feel lazy is a thirty-second job.
• Once a blade is genuinely dull it is a stone-and-a-session job, and you are not going to do that on a trail.
• Dry the blade before it goes away. Most corrosion on a knife happens between the last cut and the next one, and a folder's pivot holds water beautifully.
• Keep the edge off hard surfaces. Ceramic, stone, glass, and metal dull a knife faster than the cutting does. Wood and plastic are what edges tolerate.

What the steel changes

Steel choice moves two numbers in opposite directions: how long an edge lasts, and how easily you can put it back.

Harder, more wear-resistant steels — the high-carbide alloys sold as premium — hold an edge impressively and resist most abrasives. They genuinely want diamond or ceramic, and a soft water stone will barely mark them. Softer steels sharpen in minutes on almost anything, take a screaming edge, and lose it sooner.

There is no best. There is only what you want to be doing on a Tuesday night — a steel that holds its edge for a month and takes twenty minutes to bring back, or one that wants a touch-up weekly and takes two.

Testing an edge without cutting yourself

• Paper. Hold a sheet of printer paper and slice down through it. A sharp knife enters without a shove and cuts quietly; a dull one tears or skates. Receipt paper is the harder version of the same test.
• The thumbnail catch. Lay the edge lightly on a thumbnail at a shallow angle. Sharp bites and holds; dull slides. Do it along the whole blade — this is the test that finds the dull inch near the tip.
• Light. Look straight down at the edge under a bright lamp. A sharp apex is invisible; any flat or roll catches the light as a bright line. It is the most useful test there is and it costs nothing.
• Never test by running a finger along the edge. Across it, lightly, or not at all.

Where this comes from

Sharpening is not a standards-governed subject the way body armor or radio licensing is, so this page is the common ground between knife makers' own care guidance and the published angle presets of the guided-system manufacturers. That is where the degree ranges come from, and they are conventions rather than rules.

The mechanical claims here — that an edge is the meeting of two bevels, that a burr is the signal they have met, that leather compresses under pressure and rounds an apex — are mechanism rather than opinion, and you can verify every one of them with a bright lamp and ten minutes.

Spot something wrong? Tell us. Being right matters more to us than being finished.