A sharpening routine that stuck, after three that did not

01Three routines I gave up on

The first year I sharpened the way the internet told me to, which meant a different way every fortnight. I had a coarse stone, a fine stone, a strop, a honing guide, a slab of glass with abrasive paper stuck to it and a jar of lapping compound. Each of those things works. Put together in no particular order, with no fixed angle and no clear idea of what a finished edge should feel like, they produced chisels that cut pine and stalled in oak. Twenty minutes at the stones was buying me perhaps eight minutes of good work.

What went wrong was not the equipment. I changed one variable every session and never let a routine become boring enough to judge properly. A sharpening method is only as good as your ability to repeat it in a cold shop with a headache, and mine asked for more concentration than I reliably had. The routine I use now is duller and shorter than anything from that first year, and the edges are better, mostly because I do it four times an evening instead of once a week.

02Flattening the back, once and properly

The back of a chisel or a plane iron gets flattened once in its life, and until that is done nothing else at the stones matters. A bevel is only half of an edge. If the back is hollow, or rounded from years of somebody else's honing, the two surfaces never meet in a line and the tool will not take a shaving however carefully you work the bevel. I learned this from a secondhand smoothing iron that I honed patiently for three months before it occurred to me to look at the other side.

Flattening is tedious and it is finite. I work the back on a coarse stone, pressing down close to the edge so it cannot lift, until the scratch pattern runs uninterrupted across the last centimetre. That last centimetre is the whole of it; the rest of the back can stay as it came. On a badly pitted old iron the job takes an hour and a certain amount of swearing. It never needs doing again, and the diary page listing which tools have been done has saved me repeating the work twice.

• Flatten only the last ten to fifteen millimetres of the back
• Coarse stone first; a fine stone cannot fix a hollow
• Press down near the edge so the tool cannot round over
• Write the tool down once it is done, or you will do it again

03Two stones, one angle, thirty seconds

The routine is a coarse waterstone, a fine one, and a honing guide I stopped being embarrassed about. The guide holds thirty degrees for bench chisels and twenty-five for plane irons, and because the angle is the same every time I am only ever removing metal from one narrow strip. Freehand honing is a skill I am acquiring slowly, on tools I care less about. For the irons that matter I want repeatability rather than virtue. Setting up, twelve strokes coarse, twenty fine, and the whole business is under a minute.

The other half of it is a secondary bevel a degree or two steeper than the main grind. Each honing then touches a sliver of steel rather than the whole bevel face, which is what makes the routine quick enough to do without resentment. Every fifth or sixth time that sliver has grown wide enough to be slow, and the iron goes back to the coarse stone to re-establish the main bevel. Keeping the stones flat matters as much as keeping the tools sharp, so mine are rubbed on a diamond plate before use.

04What a sharp edge actually looks like

For a long time I could not tell whether an edge was sharp, which meant I could not tell whether a routine was working. The tests I trust cost nothing. A sharp edge will not reflect light along the line where bevel meets back, because there is no flat there to catch it, while a dull one shows a bright hairline if you tilt the tool towards the window. End grain in a scrap of pine is the second test: a sharp chisel leaves a faint shine, a dull one crushes the fibres and leaves them furry.

The third test is the work itself, and it is the only one that counts. A smoothing plane with a properly sharp iron leaves oak needing no abrasive at all, and it does so with a sound, a low hiss rather than a scrape. When the sound changes I stop and hone, mid-board if necessary, because the ten minutes saved by pressing on always cost me twenty in tearout afterwards. Four quick honings in an evening is ordinary now. It felt excessive for about a month, and then it became part of the work.