Twelve films, from a single block to a working plane. Ask a question under any of them, and show me your work when you're ready. I answer here, within 24 hours.
I answer every question within 24 hours. Everyone working through this film can read the answer too.
Over the next nine films we're going to take a solid block of hard timber, cut it into three pieces, do all the work on the inside while it's open, glue it back together, and then cut and carve out our very own Krenov-style handplane.
I learned this method from Robert Van Norman at Inside Passage School of Fine Cabinetmaking, where I studied for two years. Robert studied under James Krenov at College of the Redwoods and built Inside Passage on his teaching. It came down a line, and I'm passing it on the way it was given to me — along with everything I've picked up since, from making these planes for years and from helping a lot of people build their own.
Read this whole page before you buy anything.
You have a workshop. You've been working wood for a while — long enough to have opinions about it. You can sharpen a chisel properly, you know your way around your machines, and you're not looking for someone to explain what a jointer is.
What you're looking for is depth: the reasoning behind the decisions, the tolerances that actually matter, the things that only come from making the same tool many times over.
That's what this course is. I won't be padding it with basics you already have.
If you're newer than that, you can absolutely still follow along — but expect to pause and look things up, and expect the sharpening to be the part that catches you out.
I want to be direct about this, because it affects whether you can start.
This build uses machines and hand tools together, and I don't recommend attempting it hand-tools-only.
That's not purism in reverse. It's the way Krenov worked, and the way I was taught: bring it ninety percent of the way there with the machine, and do the last ten percent by hand.
Machines are our friend. They are very good at the tedious, repetitive, dimensional work — and they're fast at it. Every hour you don't spend hogging material off a blank is an hour you get to spend on the parts that actually reward attention: the fitting, the paring, the shaping, the moment the plane starts to feel like yours. Why would you want less time for the good stuff?
That thinking runs through everything I teach at this school, not just this course. You'll see it again and again.
In this build specifically, a few operations — resawing a thick blank accurately, bringing a center block to a thickness measured against your iron, holding parallel across a block that's about to receive glued-on cheeks — are genuinely difficult to do well by hand, and getting them wrong compounds through every film that follows. That's the ninety percent. Let the machine have it.
And then the last ten percent — paring, fitting, shaping, the beaver marks, everything about how the plane finally feels in your hand — is yours.
Machines you'll want:
Hand tools:
And a flat reference surface. Granite slab, cast iron machine table, tempered glass, or a board you genuinely trust. You'll be lapping surfaces on it and checking others against it.
Abrasive: sticky-back rolls, 80 and 180 grit. I use 3M — they do a red series and a gold series, and the gold is noticeably better quality if you're not wasteful with it.
Hock. That's what I use and what I'd recommend. They're widely available, the steel is good, and they arrive flat enough to work with.
Order the iron first. Everything in this plane is dimensioned from it, and it's the long-lead item. When it arrives, measure the actual iron. Not the spec on the packet — the thing in your hand.
This is a tool, not a piece of furniture. A dense hardwood holds a tight mouth, resists wear at the sole, and will still be working in thirty years. Softer wood gives you a plane you're constantly re-flattening.
Hard exotics are what I recommend. Ekki is what I'm using here. Ipe is another good one — worth knowing that it's a known allergen, so mind the dust. Anything in that density class will serve you well.
It's harder to work, and I won't pretend otherwise. Sanding this material is not an enjoyable process. But these woods are genuinely gratifying to use as a plane, for years and years, and that's what you're building for.
This sounds backwards, so here's the reasoning.
Your blank is a thick, relatively narrow piece — the sort of thing you'd rip out of 8/4 or 10/4 stock. In flatsawn material, the growth rings run roughly parallel to the wide faces.
Those wide faces become the sides of your plane. So when the plane is assembled and standing on its sole, the rings are running vertically — from the sole up through the body, perpendicular to the sole.
Flatsawn on the bench becomes quartersawn when it stands up. Which is exactly what you want, because rings perpendicular to the sole are what keep the sole flat through seasonal movement instead of cupping across its width.
Good news for sourcing: flatsawn is the ordinary, available, less expensive material. You're not hunting for anything exotic in the way it's cut — only in what it is.
Your resaw cuts run parallel to those rings. So each cheek, sliced off on its own, is a thin flatsawn board — rings running flat across its wide face.
Thin flatsawn boards cup. That's not a risk to worry about, it's simply what they do.
This is precisely why the cheeks rest for a week or two after resawing while the center block doesn't. And it's why the cheeks have to go back exactly where they came from: when they do, the rings line up again through the assembled body and the whole plane reads quartersawn.
Flatsawn blank → flatsawn cheeks in the middle of the process → quartersawn plane at the end.
Hard, dry, and thick enough for the whole blank is a real sourcing problem. You have options, and none of them is a compromise:
A laminated plane is a perfectly good plane. Some of them look better than the solid ones. Just make sure the glue-up is sound and the stock is dry, because you're about to cut it apart and find out.
Anything still wet. Heavy figure or visible runout — save that for a wedge.
Cracks are normal in exotic timber: the outside dries much faster than the core, so the board sits under stress and cracks as it settles. Thin CA handles them, and the glue-up restores strength right across the area. Don't reject a board over drying checks.
Six to eight weeks, and that isn't padding.
After you resaw in Film 1, the cheeks rest for one to two weeks while the wood finishes moving. You cannot skip it — everything after depends on the pieces having settled.
Work at the pace the wood sets.
Introduce yourself. Where you're building from, what wood you found, what machines you're working with, and what you want to plane with it.
This first video does two things. It walks you through every phase of the build so the whole thing is in your head before we start — and then it takes the block to the bandsaw and cuts it into three pieces.
That order is deliberate. You'll make better decisions at the saw if you already know what the inside of this plane is going to look like.
There's more than one way to make a wooden plane. This one — one solid block, cut into two cheeks and a center block, all the internal work done while the pieces are apart, then glued back together — is what I learned from Robert Van Norman at Inside Passage.
Robert studied with James Krenov at College of the Redwoods, and built Inside Passage on his teaching. Krenov advised the school and gave the students weekly lectures over the phone until he died. I spent two years there.
So this isn't a method I read about in a book. It came down a line, and I'm passing it along the way it was given to me.
It does add work — resawing, then regluing, and everything has to go back together accurately. But it makes the inside of the plane far easier to reach and to shape, and I think that trade is worth it every time.
Go as hard as you can get. This is a tool, not a piece of furniture. A dense tropical hardwood will hold a tight mouth, resist wear at the sole, and still be doing its job in thirty years. Soft wood makes a plane you'll be re-flattening constantly and replacing eventually.
This block is ekki. Extremely hard, and it will show you its challenges — it is genuinely not comfortable to work. But that's the trade, and the finished tool justifies it.
Buy flatsawn — and here's why that isn't backwards. In flatsawn material the growth rings run roughly parallel to the wide faces. Those wide faces become the sides of your plane. So when the plane is assembled and standing on its sole, the rings run vertically, from the sole up through the body, perpendicular to the sole. Flatsawn on the bench becomes quartersawn when it stands up — and rings perpendicular to the sole are what keep it flat through seasonal movement instead of cupping.
Your cheeks will be flatsawn, and that's the point.
Your resaw cuts run parallel to those rings, so each cheek, taken on its own as a thin little board, has its rings running flat across its wide face. Two things follow, and both matter today:
Flatsawn blank, flatsawn cheeks in the middle of the process, quartersawn plane at the end.
If you can't find stock that big: glue up separate pieces to reach the width or thickness you need, or laminate several pieces before resawing and then treat the result exactly like a solid block. A laminated plane is a perfectly good plane — some of them look better than the solid ones. Just make sure the glue-up is sound and the stock is dry, because you're about to cut it apart and find out.
Expect cracks, and don't panic about them. Exotic stock dries much faster on the outside than in the core, so it sits under internal stress and it cracks and warps as it settles. I glued a few on this block with thin CA. Once everything is milled and glued back into a whole, there's plenty of strength across those areas. I'm not worried about them.
What to reject at the yard: anything still wet, and anything with heavy figure or visible runout — save that for the wedge.
I'm short on width here, so before anything gets cut I stand at the bench and add it up. Everything that eats material:
Cheek one, cut oversize~9 mmBandsaw kerf~2 mmCenter blockiron width + 2–3 mm at the sawSecond kerf~2 mmCheek two, whatever's leftneeds to be ~9 mmPlaning between cutseats a little more
If the arithmetic doesn't work, you find out now, at the bench, with a pencil. Not halfway through the second cut.
Don't resaw a cheek thinner than 9 mm. The final cheek is 6–7 mm. That surplus isn't waste — it's what lets you recover if the piece twists or winds after it's cut. Below 6 mm the cheek stops being useful.
Mark the blank before you cut it. Cabinetmaker's mark, across all three faces, while it's still one board. Once there are three pieces they go back together one way only, and the mark is the only thing that remembers which.
The center block only needs to be a little wider than the iron. About a millimetre and a half, maybe two. Roughly a sixteenth of an inch.
When I first started making planes after Inside Passage, I left much more room than that. My thinking was that more space means more lateral adjustment, and more adjustment must be better.
It isn't. Over the years I came to understand that when there's too much space in the center block, you lose your sense of where the iron is actually cutting. The gap gives you a false reading. You end up adjusting by trial and error instead of by knowing.
So: close-fitting center block, thin cheeks. The body is just a bit more than the width of the iron and no more. That's the plane that ends up feeling like part of your hand.
At the saw I won't go more than 2–2.5 mm over the iron. On this build I tried 3 mm first, looked at it, and moved the fence back in. Final measurement came out at exactly 1.5 mm.
And measure against your actual iron. Lay it on the wood. Don't trust the number on the packet.
Set the fence by eye against the blade. A carbide-tipped blade helps here for a reason that has nothing to do with cutting — the flat sides of the tips give you something visible to align to.
Then run the saw and go in about a millimetre. Stop. Look.
That millimetre tells you where you actually are, while being wrong is still free. If you're happy, take the whole cut — nice and slow.
On this block, as soon as I got into the cut the kerf started opening in a way that wasn't parallel. That's the stress inside the wood letting go. Not a surprise in this timber, but now I know it, and I know to expect movement in that cheek.
Give the cheek a very light pass to clean the face. Light — ekki has reversing grain and standard jointer knives will tear it out if you push. Slow feed, shallow cut, more passes if you need them. A little pencil mark left behind is fine.
Then back to the saw for the second cut: the center block.
The second cheek came out thin at one corner on this build. That's fine — that corner gets cut away when the plane is shaped. You can only make that call if you already know the outline you're heading for, which is why the video starts with the overview.
Sticker them, leave air around them, and set them aside for a week or two.
They are going to move. Thin pieces respond to released stress far more than thick ones — that's exactly why I can bring the center block to thickness today and not the cheeks. If you machine them flat now, they will not be flat when you glue up, and there's no fixing that later.
Let them finish saying what they have to say.
The center block didn't warp, and I wasn't expecting it to at that thickness. So it goes straight on the thickness planer. Mine has a helical head, which makes a real difference in wood like this.
Very small adjustments. Take too much at once and the block is too thin, and the whole plan is gone.
If the bandsaw cut was good there's almost nothing to remove — one light pass took nearly all my saw marks off.
This is a hybrid build — machines and hand tools working together — and the resaw is one of the places where I'd want the machine. Resawing a thick blank of hard exotic accurately by hand is possible, but it's a great deal of work and an error here costs you the whole board.
Where the hand tools come in on this stage: cleaning the sawn face. A jointer plane will do that job perfectly well if you'd rather, and on wood this difficult a sharp, finely-set plane taking light passes may give you a better surface than jointer knives will.
Before you cut: photo of the marked-out blank with the iron sitting on it, end grain visible so I can see how it's sawn. This is the one moment where I can still save your board.
After you cut: the three pieces with the mark showing. Tell us what the wood did — did the kerf run true, or did it open up on you? did the parts immediately warp?
The cheeks now rest. work can continue on the center block.
Your cheeks are resting. The center block is at thickness. Today we prepare its surfaces and cut the two ramps that shape the whole inside of the plane.
The forward ramp is the bed — the surface your iron sits on. Everything in this build has been serving that surface. Today we make it.
The block has to stay parallel side to side. It's about to receive glued-on cheeks and a cross pin, and if it's gone even slightly wedge-shaped nothing will sit right at assembly.
So sand the surfaces before you cut the ramps.
If you cut first and then sand the two resulting pieces separately, you'll end up with different final thicknesses on them — and that's a worse problem than the one you were trying to solve, because now the two halves of your center block don't match.
Remove the minimum you can get away with. You are not reshaping this block. You are cleaning it up.
You may be about to ask why I'm not just planing this surface. Fair question, and I'll answer it honestly: hand planing this wood would not be an enjoyable process, and I'm not looking for extra work.
But it's more than my preference. This is a rowed, interlocked-grain timber, and Krenov addressed exactly this situation in The Fine Art of Cabinetmaking. Writing about difficult woods of this kind, his prescription was to use the machines as much as possible, then take the surfaces and shapes as far as you can with cutting tools, then be willing to labour with a cabinet scraper and perhaps files — and finally, to do some sanding.
That's not a fallback. On this material it's the method, and it comes from the same place everything else in this course does. You'll notice it's also the ninety-and-ten principle applied one level down: machines first, cutting tools next, abrasives last, and each one used where it's genuinely the right tool.
Take a plane to a piece of ekki end grain once and you'll stop arguing with it.
Stick abrasive down to a surface you trust — granite, cast iron, tempered glass, or a board you're confident in.
This came up in the comments on the video, and it's a good question. Any residue left behind will sit under the next sheet you stick down and ruin the flatness you're relying on.
The best answer is prevention. Clean the dust off the paper as soon as you've finished with it, take it off the surface, and store it rolled up in a closed container. Done that way I get no residue at all, and I can reuse the same sheet several times. Quality matters here too — I use 3M, and it comes off far more cleanly than cheap paper does.
The rule underneath it: the longer paper stays stuck down, the harder it is to remove cleanly.
On my granite slab I do sometimes leave a sheet on until it's worn out, depending what I'm using it for. In that case, once it's off, wipe the surface down with acetone. It's remarkably effective — like magic.
Either way, make sure you've got every last spot of adhesive off before you stick fresh paper down. A single speck under the new sheet is a high spot, and you'll be sanding against it without knowing.
Then draw thick pencil marks, or a grid, across the face of the workpiece.
This is the whole feedback system. Sanded areas turn pale from the dust. Anything still dark hasn't touched the paper yet — that's a low spot. You can see exactly where you are at every moment instead of guessing.
On the first pass here I was touching maybe 90% of the surface. That's a good start, and it means the milling was decent.
As the pencil disappears you're not just seeing where the low spots are — you're seeing a record of what your machines did to this piece. Worth learning to read, because each mark tells you something different.
Waves across the surface — that's the jointer. The face I did with the helical head doesn't leave that pattern at all. If the waves are large, your feed rate on the jointer was too fast; slow it down next time and they'll be much finer.
Snipe from the thickness planer, at the ends of the board. Normal enough, and acceptable — you'll sand it out.
A step part way along a face means either your technique on the jointer slipped, or the machine isn't set up correctly. Worth chasing down, because it'll happen on every piece until you do.
Tearout is a different order of problem. Sanding might get rid of it if it isn't too deep — but remember what we're doing here. We don't want to sand this surface much at all, and chasing tearout is exactly how the removal gets out of control and the block ends up thinner or out of parallel.
Better to fix the cause. Consider changing your planing direction on the machine — you may simply have been going against the grain. A helical cutterhead deals with tearout far better than straight knives, which is one of the real arguments for one in wood like this.
Bright, rough patches running side to side are bandsaw marks you haven't got rid of yet. Keep going.
Learning to tell these apart tells you which machine did what, where you still have work, and — more usefully — what to change next time before the marks appear at all.
Some exotics cause allergic reactions — ipe is a known one, and it's otherwise an excellent plane wood. Don't brush the dust off; that just puts it in the air. Wipe it with your hand for light work, and bring the shop vac over when the abrasive clogs or you're removing real material.
Start at 180. If the surface is rough or you're fighting bandsaw marks, drop to 80 to get the material off, then come back to 180 for the final surface.
SituationUseLight irregularities, refining180Persistent bump or bandsaw marks80Final surface after coarse work180Abrasive cloggingStop and vacuum
The start point sits somewhere between a third and the middle of the block — but not the middle. You want the back part of the sole longer than the front. That's simply how the plane is used; the weight and the pressure sit behind the iron.
From that point:
45° is right for almost everything, but it isn't the only choice. I made a 60° bed plane once — that's effectively a semi-scraper. The shaving comes off completely differently and it's harder to work with, but it has saved me a few times on difficult, wildly figured wood. You can go to 90°, or even negative, if you want to build a scraping plane.
Build this one at 45°. Then build another when you know what you want.
I have a small reference block that carries the 45° and 60° angles. Its height and thickness also determine where the cross-pin hole gets drilled — so it isn't just an angle gauge. It ties several dimensions of the plane together in one object.
This is the original one Robert gave me when I was studying at Inside Passage.
I've used it on every plane I've made since. Which is, I think, the most Krenovian thing in this entire course: a small piece of wood that carries the proportions of a tool, handed from one maker to the next.
📥 So I've drawn it up for you. There's a printable template attached to this lesson — actual size, so you can lay your stock straight on it and mark from the drawing.
Thickness11 mm (7/16")Height33 mm (1 5/16")Length~120 mm (4¾") — not critical, make it whatever suits your handEndsone bevelled at 45°, one at 60°
The thickness and the height are the numbers that matter, because those are the ones doing the cross-pin work later. The length is yours.
Make it from something hard and stable. Offcuts from your plane blank are ideal — you'll have some, and it means the block is made of the same stuff as the plane it measures.
Mark your 45 and your 60 clearly on the faces, straight away, before you set it down. The two ends don't look that different at a glance, and using the wrong one on a centre block is not a mistake you can sand out.
Make it before you lay out your ramps. Then it's yours, and it'll be on your bench for the next plane and the one after that.
Normally when you cut to a line, you keep the line. Get as close as you can, but don't cut it away — once you're into the line you start wobbling and the cut goes off.
For the forward ramp, break that rule deliberately.
That leaves the sawn face very slightly concave.
Why: a concave surface sits stable on the abrasive. A convex one — a surface with a bump in the middle — rocks and wobbles the whole time you're trying to flatten it, whether you're sanding or using a block plane. Half a minute of intention at the bandsaw saves you a fight later.
Don't overdo it. I went a little further than I meant to on this one, and flat would have been better. A hint of concave is all you want.
The surface doesn't need to be smooth or exact off the saw. Sanding establishes the final surface. You're just leaving a shape that will cooperate.
The back ramp isn't a straight 60° cut. Here's the reasoning, because it's a good example of three requirements settling against each other:
Why 60° and not less: this is a wooden plane and it needs structural integrity. Cut that ramp shallower and the piece gets thin — you'd have to make the whole body longer to get the strength back. So you keep it steep and keep the wood.
But 60° creates a problem: there's now very little space between the ramp and the cross pin for shavings to get out.
So you curve it. A slight curve — do it by eye, and go a little more than you think — opens up the passage for the shavings. It also just looks better, which is not nothing.
That curved surface is where the beaver marks go: small scooped cuts with a rounded chisel, the traditional touch on a plane made this way. We'll do those later, and they're always fun.
Fair warning — on exotic end grain your chisel needs to be genuinely sharp. It'll be fine if it is.
At the saw: a wide blade makes the curve awkward. You may need to force it slightly through the turn. Go slowly.
The plane iron bears directly on this surface. Get it right.
Flat. Or very slightly concave. Never convex.
Any bump — in either direction, along the ramp or side to side — introduces rocking in the iron. Rocking means chatter, and chatter means a plane that doesn't work, no matter what else you did well.
Checking flatness: put a light behind your square blade and look for the gap. Check edge to edge, and start to finish. You'll see light through a low spot immediately.
Checking square: to both sides of the block. Not one.
Correcting an angle: if you're out of square, you can shift your pressure slightly left or right. But be careful. Too much pressure on one side creates a bump or a drop on the other, and then you're sanding more to fix that, and you've entered a loop that's hard to get out of.
This is the one place where sanding is genuinely worse than planing. With a block plane you can take partial passes and walk an angle back into square. On the abrasive you can't be that precise, so even pressure is safer than correction. Don't let a small error grow into one you have to chase.
On the angle itself: you might be wondering how I know it's still exactly 45° after all that sanding. I don't, and it doesn't matter. 44½°, 45½° — no meaningful difference to how the plane works. Whatever it is, it is. The flatness and the squareness are what you protect.
A tiny bandsaw mark that doesn't affect the mechanics can technically stay. But you'll be holding this plane for the rest of your life, so I'd rather it were clean.
Don't compromise on this step.
Post your bed. A photo of the forward ramp with your square held against it and a light behind — the same check you just did. That backlit gap is the single most useful thing you can show me at this stage, and it's far easier to fix now than after glue-up.
Also: show us your magic block. First plane, and already your first tool made for making planes.
Your cheeks are still resting. The center block has its ramps cut and sanded. Today we finish the inside of it: carve the curved back ramp, deal with any cracks that have shown up, and shoot the two chamfers that decide how tight your mouth will be for the life of this plane.
That last one deserves your attention. The chamfer work in this film is what stands between you and a plane that opens up and stops smoothing after its first winter.
Before anything else, the gouge gets sharpened. I'll be using a small one — smaller diameter means more work, but I like the personality it leaves. The surface is distinct without being obvious, and it's lovely under the hand.
First, make yourself a strop. Carve a small block of wood to match the curve of your gouge, and put a bit of honing compound on it. That's what removes the burr afterwards. It takes two minutes and you'll use it every time from now on.
On the 1,000 stone. Much the same as a straight chisel, except you turn the tool as you go — a rocking motion so the curved edge contacts the stone across its full width. It takes a bit of getting the hang of, but it isn't complicated.
Two things to watch:
What you're looking for is contact on the leading edge across the whole width, and a burr. The burr can be hard to feel on a curved edge — take your time and be sure.
On the 8,000. Here's the thing people get wrong: a nice shiny bevel is not the goal. You can have a beautiful polished surface and still not have your leading edge in contact. The leading edge is what you're checking. If it still looks a bit bright and grey from the 1,000, keep going.
Clean towards the edge — not sideways, and definitely not backwards off it. You don't want to break the burr off yet.
Stropping. On the compound block, pull backwards. If you push forward even slightly, unlike on a stone, the edge digs into the wood.
A note on strops: I use compound on a flat piece of wood rather than leather. Leather compresses under the tool, which rounds the edge slightly — it changes your attack angle without you deciding to. Flat wood doesn't.
That's what Robert called them, and it's what I've called them ever since. Small overlapping gouge cuts across the curved back ramp, replacing the bandsaw texture with something deliberate.
Clamp the piece where the gouge can enter comfortably. This sounds like a small thing and it isn't. If the work sits too low or at an awkward angle, you'll find yourself making strange wrist motions just to start each cut — and you'll fight the tool for the whole surface.
Adjust the height and angle as you go, too. The grain changes along the curve; some stretches feel harder, some easier, and a small change in setup makes a real difference.
A tail vice is lovely for this. Any vice will do.
Start slightly inside the edge, not right at it. Starting from the very edge is awkward, and you don't need to — that whole edge gets cut away later when the plane is shaped.
Work side to side, in rows. You can make them short, like little dimples, or longer for a more flowing pattern. Whatever you prefer, but keep it reasonably consistent rather than chaotic.
They won't be perfect. That's the point — this is meant to look organic, not machined. I'm not chasing symmetry.
Rest your hand on the workpiece. It stabilises where each cut begins and it's most of the control you have.
Bandsaw marks. If you skip spots, you'll see them as brighter, paler patches — that's rough sawn surface still showing through. What you want is a consistently darker, textured surface all the way across, cut by the gouge.
Keep looking for those bright spots as you go. And check against the light at the end.
Coming to the edges, you're in end grain. A straight, careless motion here will lever a piece right off, and you won't just lose it — you'll see the breakout after glue-up, sitting there permanently.
Tilt the gouge slightly toward the safe direction as you approach the edges. Otherwise cut straight, and cut with the grain. You'll find out very quickly if you're going against it.
You could skip all of this. Scrape the curved area, or sand it, and move on. You could even leave it bandsawn — the plane would work.
And that would be fine, as long as it's an honest choice.
Here's the distinction I'd ask you to hold onto. Choosing a simpler finish because you've considered it and that's what you want is a decision. Choosing it because you don't feel like sharpening, or because you're not confident with the tool and would rather not find out — that's something else, and only you will ever know which one you did.
If it's the confidence, practise on some scrap first. That's the honest route through.
I'll tell you what you get for the work. The surface has far more character. It feels good when you put your hand in there to clear shavings. And every time I use this plane, I remember the afternoon I carved that ramp.
That's what you're buying with the extra hour.
A crack turned up in this ramp that I'm fairly sure wasn't there when I cut the piece. That happens; the wood is still settling, and it will keep telling you things for a while yet.
Use thin CA, not wood glue. The point is penetration — you need it to travel the full depth of a narrow crack, and ordinary wood glue is far too viscous to get in there. Thin CA wicks straight through; on this one it came out the other side almost immediately, which told me it was all the way in.
You don't need much. Then clamp it — a vice is fine — for real pressure across the crack while it cures.
Both ramps get a chamfer. They are referenced differently and they exist for different reasons, so take them one at a time.
Not square to the sole — square to the ramp.
The reason is simple. Without it you're left with a thin edge at the front of the mouth, and a thin edge bends and catches on things. The chamfer removes it.
The exact size isn't critical. What matters is that after you've removed material from the sole and got the iron projecting properly, there's still a significant chamfer left there — so the end of it isn't thin and flexible.
Don't overdo it either, or you lose support for the front of the iron.
Here's the whole reasoning, because it's the most important thing in this film.
This is going to be a smoothing plane, which means you want a tight mouth. After glue-up you'll carefully remove material from the sole until the iron projects exactly right, and that gives you your tight opening.
But a wooden plane moves. It can take a whole year for one to properly stabilise to your workshop, and during that time it'll warp a little and you'll need to flatten the sole again.
Think of the chamfer as a little vertical wall. That wall is what holds the mouth relatively tight. As long as it's there, flattening the sole costs you very little. But once the wall is gone, you're back to that shallow angled surface — and from then on the mouth opens up fast, every single time you flatten.
So the wall has to be tall enough to survive years of flattening. And it can't be too tall, because a high wall means shavings jamming in that tight space between the iron and the wood, and a plane that clogs.
The numbers: start at about 3 to 3½ mm. What you're aiming for is a final wall — after all the flattening that gets the blade properly protruding — of around a millimetre and a half. That's the sweet spot: enough to last a long time, not enough to jam.
Look at a plane that's been in use for years and you'll still see a bit of chamfer there. That's the wall, still doing its job.
A shooting board does this work — and don't forget what you're planing here. This is end grain, which is why the setup matters as much as the technique.
Mine has a fence that adjusts both for square and sideways. Bringing it in close gives near-zero clearance right where the cut exits, and on end grain that's the difference between a clean chamfer and blowing a chunk off the far corner.
Sharpen your iron to a higher angle before you start. On ekki it'll take some edge damage regardless, but a higher angle survives it.
Check square against a light. I thought I was square, held it up, and could see light on the right-hand side. Checked from the other side — same story.
When that happens, adjust the fence, don't keep planing. A small tap sideways with a hammer shifts where the material comes off. Trying to correct by planing harder on one side just makes a worse angle.
Pressure goes exactly at the corner you're cutting. Push too much one way and you change the chamfer angle; too much the other way and you change it differently. It's about finding the right amount, and it's the part that takes practice.
Go slow at first. Be careful, and check repeatedly that you're still square — stopping to look costs you seconds, and a chamfer planed out of square costs you the piece. Once you've done this a few times and your technique with the shooting board settles, things will speed up on their own. Let that happen rather than pushing for it.
Take lighter passes as you get close to your number. I went slightly past 3 mm on purpose — a little over is where I want to be.
Don't trust the measurement. Simulate the assembly.
Take the iron out of its chipbreaker, lay something straight along the bottom to stand in for the sole, and set your two parts in position as they'll actually sit once the cheeks are on.
The edge of the blade should meet roughly the middle of the chamfer.
Middle. Then mark your line and plane to it.
Second part of the check: the base of the sharpening bevel should still have wood underneath it. Mine's at about 27° at the moment — and if I later sharpen at 30°, there's still support there. That margin is what you're confirming.
These are the things I actually reach for. The links are Amazon affiliate links — they cost you nothing extra and they help support the work.
Norton sharpening stones. Nothing like them, and the two grits in this film are the two I use for almost everything.
CA glue, various viscosities — you want the thin one for cracks like the one in this video: https://amzn.to/4w8wcVb
You'll also need: a small gouge, a shooting board with an adjustable fence, honing compound, and a scrap of wood to carve into a strop block.
Show us your beaver marks. Photograph the back ramp with a light raking across it — that's how you see the texture, and how you spot the bright patches you missed.
And post your iron check. The blade in position against the chamfer, so we can see where it lands. This one is worth getting a second opinion on before glue-up, because the mouth you end up with for the next thirty years is decided right here.
The screw that connects the blade to the chip breaker runs in a slot in the center block. The slot has to be wide enough for lateral adjustment, deep enough to clear the screw head, and long enough for the blade to protrude through the sole.
I once cut this slot without enough length. When I started working with the plane, the screw head bumped into the end of the slot and I couldn't move the blade any further.
It's one cut on the router table, but it's three separate setups — depth, centering, and stopping point — and each one gets checked before the real piece goes near the bit.
What you'll do in this film:
The slot carries the screw that joins the blade to the chip breaker. Two jobs:
This is why a ¾" bit for a screw a fraction of that size isn't sloppiness. It's the design.
The cut is made on a flat, straight, unwarped face — you checked and prepared that surface earlier in the build. Everything below assumes it.
In the shop I use "deep" for two different things, and it's worth separating them here so the notes stay clear:
If the slot isn't deep enough, the screw head scrapes the top of the slot, or gets pushed up by it. That pressure lifts the assembly and the blade no longer sits snugly on the ramp — which is the whole point of the bedding. So be sure the slot is cut deep enough. Keep the two ideas separate in your head and both setups get simpler.
Rule: screw-head thickness + about 1.5 mm.
Why the extra 1.5 mm rather than a tight fit:
Verify, don't eyeball. Set the height by measurement, then run a piece of scrap and measure the actual cut. The height scale on a router lift is a starting point, not a result.
This is the part people guess at. Don't guess — derive it.
Because the flat material beside and in front of the slot is structural. The blade beds on the ramp; the chip breaker and wedge press down on the blade and drive it against the ramp. That pressure needs a solid, flat reference area to push against. Running the slot too far forward removes the reference and weakens the block exactly where the load is.
Put the pencil mark on the face you'll be able to see while feeding — you will not see the bit during the cut.
¾" spiral solid carbide, upcut. (The one I use — affiliate link.)
The slot should sit roughly centered across the block. Not laboratory-centered — but centered.
Process:
The doubling rule — this is the one that catches people:
Moving the fence changes both sides at once. Take 1 mm off one side and you add 1 mm to the other. The visible imbalance changes by twice your fence movement.
So when your square tells you you're ⅛" out, move the fence half of that. And expect to overshoot at least once anyway — I did, on camera. It happens faster than your hands expect.
Also expect to go the wrong way once. These adjustments are genuinely confusing in the moment. Make a small move, take a shallow test cut, read it, and if it got worse you now know the direction. Two or three iterations is normal.
Go deeper by only a couple of millimeters per test cut. Just enough to get a readable edge. If the previous test cut is in the way of reading the new one, going slightly deeper wipes it out and gives you a fresh reference.
Lock the fence. Then check it's locked. Then check again.
You can't see the bit. So bring the reference up to where you can see it — the table surface.
Marking the stop line on the table:
Ways to stop the cut:
MethodNotesPhysical stop, 45° faceMost dependable. Use an offcut from your center block. Lower the bit first, position the stop against the exact cutting location, then raise the bit. The angled face gives a much better contact and reference than a square stop, which tends to feel vague as you reach it.Visual stop at the table lineWhat I do here. Watch carefully, stop when the marks meet, travel back out.Stop and switch offReach the mark, kill the router, wait for the bit to come to a complete stop, then withdraw.
On withdrawing: travel back out, don't lift.
Lifting the workpiece vertically off a spinning bit is sketchy for no gain. If you don't clear the cutter completely on the way up, you tear the piece. Pull straight back the way you came, with pressure held against the fence.
Assemble the blade, chip breaker and screw and offer them into the slot. You're confirming three things:
If all three read right, you're done shaping these parts. Ramps, chamfers and slot are complete.
MistakeConsequenceFixSlot too narrow / close to screw diameterNo lateral adjustment; blade can't be squared to the mouthUse the ¾" bit as specifiedSlot not deep enough (height)Screw head scrapes or gets pushed up, lifting the blade off the rampScrew-head thickness + 1.5 mm, verified on scrapSlot too short (length)Blade can't project — plane won't work. The mistake I made.Derive the endpoint from blade position, add 3 mmSlot too longWeakened block, lost bedding reference under the wedgeStop at the derived point + 3 mm, not moreCorrecting the fence by the full errorOvershoot past center, repeatedlyMove half the error; every move counts doubleTesting on a finished surfacePermanent test cut on a face that survives shapingTest in waste that comes off during shapingLifting the piece off a running bitTorn workpieceTravel back out, or stop the router firstDull or resinous bitVibration, burning, less controlClean it or replace it — spirals are cheap now
Previously: ramps, chamfers and the center-block geometry (Films 2–3), while the resawn cheeks rested.
After this film: the center block is finished. Ramps, chamfers and slot are all complete — no further shaping needed on these parts.
Next: preparing the cheeks, assembly, and drilling for the cross pin.
Post a photo of your slot with the blade, breaker and screw assembled in it — and tell us how far past your calculated point you stopped. Did you find yourself wanting to add "just a bit more" insurance? Where did you draw the line, and why?
A good follow-up thread: what's a mistake in your own work you only discovered once you started using the piece?
The cheeks have been resting since the resaw. In this film they come back to the bench, get milled flat and thin, and then the plane's most important hole gets located and drilled.
Everything so far has been the centre block: the ramps, the beaver marks, the chamfers, the screw slot. That work was deliberately scheduled while the cheeks sat and finished moving.
By the end of this film the three pieces are pinned together in their final positions, and the cross-pin hole is drilled through both cheeks in alignment.
Machines: jointer, thickness planer, drill press
Hand tools: a plane you trust in this wood, hand drill, flush-cut saw, chisel, file, square, sliding bevel, marking point
Bits: 6 mm for the dowels, 8 mm for the cross pin
Consumables: penetrating superglue (maybe), paraffin wax, sticky back sandpaper on a flat surface (180 grit, here's my amazon link: https://amzn.to/4n33VuZ), 6 mm dowels (or rod you can cut)
From earlier films: the magic block (11 mm × 33 mm, 45° and 60° bevels), the blade and chip breaker, push blocks, clamps, zero clearance inset \ board for the bandsaw
Many people don't know this. When a blank is warped, the instinct is to take it straight to the jointer and let the machine sort it out. But a warped board rocks, and a rocking board gives the cutterhead nothing stable to reference.
So the first job is flattening — but very selectively. You're taking material off two specific corners and nowhere else. Neither a jointer nor a sanding board can do that: both reference the whole surface. They will flatten it eventually — but the blank tips onto one of its low corners, and the machine keeps cutting until the entire face has come down to meet that corner. You get a flat surface by removing far more material than the warp actually required. On a piece that's heading for 8 mm, that can be the difference between a usable cheek and one that's too thin. A hand plane is the only tool here that lets you choose exactly where material comes off.
Find the diagonal it rocks on, hand plane the two high corners down, and stop as soon as it stops rocking. Half a millimetre of lift is fine. You're not trying to finish the surface here — you're making it stable enough for the machine to do its job properly.
Nothing at this stage has to be perfect, because the rest of the sequence completes what the hand plane starts. The hand planing is what does the initial flattening. The jointer then cleans that face up enough to reference from, and the thickness planer takes care of the other side.
This is the school philosophy in miniature: the machine takes the work ninety percent of the way, and the hands do the ten percent the machine can't. Here the order is inverted — the hands go first — but the principle is the same. Each tool does the part it's actually good at.
Rounding is a real risk with a sanding block, but not with good sticky-back paper stuck down to a flat surface — that stays flat. The problem is the one above: a flat abrasive is still referencing the whole face, so it brings the piece down to its lowest corner instead of taking off the two high ones.
The other reason is residue. Sanding leaves abrasive grit sitting in the pores of the wood, and that then goes across your jointer knives.
I take these down to just under 8 mm, and I go as thin as the hardness of the wood will allow. This is not about weight or looks. Thin cheeks let you feel where the blade is while you're planing. Thick cheeks put a wall between your hand and the cutting edge, and you lose that feedback.
That's a design decision, not a rule. But it's the kind of decision that only shows up years later, in use.
The machined faces are already good. I still sand them, by hand, on a flat surface — because the goal isn't a strong joint, it's a joint you can't see. On a finished plane, a visible glue line down the side is the thing your eye keeps returning to.
The discipline here is stopping. Every extra pass is a chance to put a new warp or a taper into the thickness, and then you've traded a cosmetic problem for a structural one. If a low spot sits in an area the shaping will cut away, leave it alone.
The cross-pin hole needs to sit in the right place relative to the blade, and in the same place in both cheeks.
That's why it goes to the drill press rather than being drilled by hand, and why the two cheeks get drilled separately rather than stacked and drilled through.
Mine came from Robert at Inside Passage, and it does two jobs in this build. The two bevelled ends carry the angles — 45° at one end, 60° at the other — so the ramps can be marked off it directly. And its thickness and height are what set the cross-pin position: 11 mm thick and just over 33 mm high, or roughly 7/16" by 1 5/16". Length is about 12 cm, which doesn't matter much — it just needs to be comfortable to hold against a line.
The height is the part doing the quiet work. It already accounts for the 1.5–2 mm you'll lose off the sole later, so when you add it to your layout you end up with the right gap between cross pin and blade once the plane is finished, not just as it sits now.
If you don't have one, make one. It's a small offcut of stable hardwood, thicknessed accurately and with the two bevels cut on the ends. Once you've made it, the layout in this film becomes a matter of stacking references rather than measuring.
Notice how much of the layout is copied rather than measured. The ramp angle gets read off the ramp you actually built, not set to 45° from a drawing. The blade position gets set by eye to the middle of the chamfer. The cross-pin height comes from the magic block, which already carries the blade stack, the wedge clearance and the 1.5–2 mm you'll lose off the sole later.
Every one of those is a reference taken from the object in front of you. That's what keeps the finished plane consistent with itself, rather than consistent with a number you wrote down three films ago.
A cross-pin hole that's too small. I've drilled 7 mm in the past. The pin takes too much stress over years of use and eventually breaks — and then you're drilling it out and fitting a brass rod. That's why I recommend 8 mm.
Drilling the cross pin by hand. Don't.
Drilling straight through the assembled sandwich. With both cheeks pinned to the centre block on the locator dowels, there's nothing underneath the upper cheek where the bit comes out — so it blows out on the inside face. You can fit a backing piece in there, but it's a hassle to make, it puts extra stress on the bit, and it does nothing about the bit flexing on the way through. That's why the first cheek gets drilled on its own, the assembly goes back together so the drill can locate the second cheek with the machine off, and then it comes apart again to drill that one separately.
Siting the locator dowels badly. Too far in and they land inside the finished shape. Too close to the edge and the corner breaks out during assembly and disassembly. If you're unsure where your shape will fall, sketch it on now and drill outside the line.
Assuming the pin clearance doesn't matter. Too close to the blade and there's no room for a wedge at all. Too far and the wedge becomes so thick the action between pin and blade goes soft.
Two things worth sharing under this lesson: