All posts by gabe

Hopper Joinery

The hopper is an old form you can still see all the time in things like wheel barrows and baby’s cribs, sloped sides meeting in compound miters. Although initially intimidating, learning to determine the cut angles for butt or mitre joints is as simple as drawing a few lines with a carpenters square, opening up a whole exciting world of non-orthogonal intersecting surfaces.  Learning hopper joinery is step number one towards Japanese roof carpentry, heady stuff indeed!

Although there are western methods for deriving these cut angles, I learned it first from Chris Hall’s second volume of carpentry drawing books. They’re only available as an e-book and not cheap, but its still cheaper and faster than learning Japanese. He kind of bludgeons you (in a good way) with numerous methods of understanding this stuff, from a basic trigonometric understanding to developed drawing to my favorite and the dead simple approach with the sashigane.

Since I want to keep this simple and don’t want to rip off Hall’s hard work in teaching this stuff I’ll omit the Japanese names for the various parts of the unit triangle.  Also of import is that this method only works for regular slope, regular plan. Which is to say, the splay is the same on all four sides and they meet at ninety degrees in plan view.

Slope is expressed most commonly as a ratio, here 4/10. I apologize for the crappy surface of my work beam, the rest of it was scraped clean, but I didn’t want to erase the unit triangle until work was complete. Hopefully its clear though that I used the square to draw a triangle with right angle sides of four and ten, with the edge of the beam forming the hypotenuse. The short segment of the triangle gives the bottom bevel angle of your hopper. From there a line is extended up square to the hypotenuse and meeting at the right angle . The length of this segment that divides the unit triangle gives the face cut angle, 3.75/10. Further, the hypotenuse is now divided by this line into two segments, a short and a long. The long segment gives the edge mitre angle 9.375/10 if you want a hopper to join with 45 degree corners in plan view, and the short segment gives the butt mitre angle of 1.5/10.

I’m using butt joints, so just as I drew the original 4/10 triangle, the square is used to lay out a line on my beam for the edge butt angle and the face cut angle, and then transferred to a bevel gauge.

Its handy to have a couple of bevel gauges, each one set to the needed angles.

Apparently Shinwa makes a western framing square in stainless steel, couldn’t resist! It has the normal rafter tables and brace length table, but no Essex board foot scale. I don’t think I’ll miss that one though, never having used it.

Rather counter-intuitively, for a hopper with butted edges, the inside length is longer than the outside, which you have to watch out for when marking the edge angle.

Once you know which side of the board faces inward on the hopper you can mark the bottom bevel.

These hoppers are a real pain in the ass to nail together without some kind of joinery, so I decided to cut a simple dado. For marking the width of that dado you need to use the length of the cut line along the edge, not the thickness of the board. Its a slight difference at 4/10.

I’m using my double bladed marking knife for illustrative purposes. Marking with ink is easy on the eyes if slightly less accurate than a knife line, but my joinery was tight in any case.


With the layout almost complete I started by beveling the bottom of the hopper sides so that they will sit flat. Beveling the bottom also removes the need to cut an annoying barb where the bottom edges of the sides meet, as well as making it easier to cut a bottom panel groove with hand tools.


With the bottoms beveled I finished marking the joinery. You know something is going right if your lines meet square across the beveled bottom edge.

Cutting the dado’s with dozuki and a piece of kumiko as a guide fence to my cut line. I start the saw with a stroke or two to get a kerf all the way across the cut, then take the guide away and angle the saw. Its good practice for dozuki technique. To keep the saw from jumping the kerf you have to saw flat.

Ahhh! So much easier to assemble with a little joinery, even if this thing is getting nailed together with pneumatics.

Once the sides were together I lopped the upper corners off. The angle for that cut is a further segment of the unit triangle, important when you want to cut mortise and tenon, but for a simple hopper like this I just flush sawed using the edge of the side as a guide.

I didn’t use a bottom panel groove, just beveled the edges to fit in the hopper a little above the bottom. Surely this is not as strong as nailing the bottom on to, well, the bottom, but it looks nicer to me.

If you want to make a strong nailed box or hopper its worth studying the prototypical Japanese style crate, this one’s for my drafting tools.

So there it is! Basic hoppers are not at all intimidating once you give it a try, and you will be on the road to some really awesome woodworking possibilities once you step outside the world of square boxes. I’d really like to make a rocking crib one day that uses a hopper and splayed legs!

Rescuing an Old Dresser

Do you recall the first time being asked to restore a piece of furniture? For me it was not really that long ago. I help re-seal a customers concrete patio, lo and behold there’s an old dresser that this lady had re-finished with her grandfather when she was a little girl, and she’d like it to be fixed up.

I take a look at it, its a bit rickety. But somehow I’m thinking, fix a few drawers, strip and spray some poly, blah, blah, yah I can do that for $400. As I’m loading the dresser into the back of my pickup I tell the lady’s husband in a most sincere tone, “I’ll take good care of it.”

He replies, “Oh, that thing is a piece of junk.”

Ah, so it is, so it is.

It should come as no surprise then that once the dresser was back in my shop I realized that it needed to be disassembled and all the joints re-glued. And hey, why not French polish the thing while its in pieces? Maybe I blew my work budget right at the get go because I knew I would enjoy the work only if I could do it to my own standard of quality. When its done, its done.

Not to lie, it was a piece of junk in terms of joinery. The panels sat in the same groove as the tiny wimpy stub tenons for the rails. By the time it got to me the only thing holding it together was the nails used to get it out of clamps on the production line.

That said, taking apart a piece of furniture you didn’t make is really fun and interesting. You get to be part detective, brushing the dust off of a past place and time, exposing wood that hasn’t seen the light of a workshop in a long time.

And gosh, pulling rusty nails is just excruciating. It makes you think of all the times you’ve read “disassemble the joint with as little damage as possible”. Well, just how much damage is possible, lets not find out. The drawer stops were little pieces of dowel glued in to the front drawer dividing rails, what a joke. They got replaced with proper dovetail stops mortised into the front dividing rail.


And worn drawer bottoms.

Which are  a simple fix.

I made up some finishing samples for the customer to look over. As it turns out the fewer choices you present the better. I’ve seen a dozen samples put in from of someone that pretty well melts their decision making ability, in the end they of course pick something not even shown, i.e. “match this color from this different species of wood on the other side of my house” kind of thing, ugh.

French polished shellac is a rabbet hole, but try it, its a beautiful finish, truly. Soon you’ll be convinced that everything, including the phase of the moon needs be considered when applying shellac with a rubber and spirits.

The front faces of the drawers were warped, which had to be corrected on the newly glued runner strips. I don’t care if the drawer face doesn’t sit perfectly flush with the frame, the drawer needs to run flat on the runners.

Now lets see how many specialized tools I can use repairing a drawer.

Marking the drawer bottom for a sliding dovetail batten.

Azebiki saw for cutting the sides of the dado. As an aside, if you pay $40 dollars for a saw like this expect to get a $40 dollar saw. Or be happy and practice your saw doctoring skills!

Routering out the waste. The drawer bottoms were badly warped. Normally I might have just flipped them upside down and called it a day, but the drawer stops had carved grooves in the drawer bottoms from sagging so much. Consequently the panels had to be held flat with battens for all the marking and cutting.

White Ash tools pleasantly by hand, including a little dovetail plane.

Cut too loose and the joint is pointless, too tight a fit and you warp the panel. Maybe there’s something to be said for always cutting tapered sliding dovetails?

Whew! One drawer down, four more to go.

But something fixed with care, it adds a charm all its own.

The shellac finish convinced me to buy a proper cabinet scraper, I spent way too long sanding out tiny digs from using a hand scraping card.  The wood was red oak, filled. Stained with a brown mahogany gel, body coats of amber shellac and then clear followed by a dark paste finishing wax. And to skip over all the tedious bits, it turned out quite well. Just in time for Christmas, and passed on to its next generation of owners.


This is everyones fault but mine…

I want to rant a little bit about how a building can get mangled in the build process because of a cacophony of opinion between the home owner, architect, engineer, and carpenters. Somewhere in the mix is me, yay!, which I am most pleased about because I have a job in the mountain community where I live and don’t have to travel hours every day to work or live in some flatlander corporate paradise.

The structure pictured below is an accessory building we affectionately refer to as the Garagemahal.

Its a nice little building, right? The architect lives directly in view of the thing so he pretty well designed something he could live with looking at all the time. The homeowner wanted to have a bathroom window added, notice it drawn in sharpie directly below the point load from the front gable. Notice how the smaller transecting gable has a higher ridge height than the main gable? Seems strange but it looks nice in the drawing.

I helped frame up the grade level walls, but everything didn’t get sheathed before the main contractor went on holiday for an elk hunting trip. The truss contractors installed most of the trusses and the building leaned out of plumb because they had to pull all of the external bracing to get their crane in to lift the trusses. It leaned over more than an inch. And then they nailed the sheathing up without racking the frame plumb again. The trusses themselves were screwed up, the homeowner ordered with some kind of condensed set of plans and so they arrive without dropped gable ends that allow for lookout rafters to support the gable end overhang. But they got put up anyway, so the truss guys partially sheath the roof with a lesser gable end exposure of 18-1/2″.

Apparently I get to frame a ladder for the gable end overhang and just fasten it to the end trusses. It puts the overhanging roof load in tension, which seemed so sub-standard to me that I didn’t even know it could be done until I studied my framing books a bit. But there are corbels on the gable end that support the barge rafter so I won’t lose sleep over it.

Lets look at the truss sections for the roof.

The main gable is formed with “attic room” trusses, scissor trusses for the smaller gable, and an interior ceiling plan thats vaulted in the middle at the same pitch as the roof.

Of primary concern to me is the valley rafters. This is how the architect drew it.

Now, have I lost my mind, or did the architect? Vaulted ceiling means the center square of this building is stick framed. Tripled girder trusses support both the roof and floor load of the smaller gable and so all that load goes onto the perimeter load bearing wall (remember that sharpie drawn bathroom window?).  But the ridges are at different heights, and the valley rafters have to hang on the lower ridge and carry down to the girder trusses. So why did the architect draw them as all intersecting in the middle?  Not to mention that there is no top plate for the valley rafter to land on, its going to hit the side of the girder truss. The top chord of the girder truss is 2×6, not nearly enough depth to nail an LVL sized to carry the load so it requires some kind of metal hanger. With my luck that connection point will be smack in the middle of a metal gusset plate. And shouldn’t that connection point be specified by the truss company per IRC? Having to move the valleys to where they need to be will change the interior ceiling plan as well, fun times!

So basically the architect drew it wrong, the engineer took their money but said nothing, the homeowner provided bad info to the truss designers who delivered the wrong trusses which then proceeded to fuck the building when the contractors installed them. Or maybe I just missed something.

Now I’m apparently the guy that can handle roof geometry and will get up on a 12/12 pitch roof in January in the mountains. Thank god for fall protection equipment. But I don’t get paid enough for shit like this.

So now I’ll leave you with a cute picture of a little girl learning to play chess and we can all feel better.

Summer’s warmth and providence

The days here are getting longer again, looking forward to summers warmth.  Lets take a look back at last summer.

With the warmth and sunlight grass grows, matures, and can be cut for hay. For the past couple of years I’ve been using a scythe, which is indescribably enjoyable work, very honest. Properly peening a scythe blade takes a bit of practice. But anyway I got it into my head that I’d like to see how much work making hay is on a small scale.

Making hay by hand is hard work, lots of it, in a constant rush because summer thunderstorms might ruin all your work. But you have a good reason to make a wooden hay rake. Riving your pegs is a necessity, hopefully from a stout wood like hickory.

Loose hay takes up a ridiculous amount of space. Still, there’s good arguments for leaving it loose in a haystack. Most importantly, baling hay is a lot of work, but you get to make a baler. And if you’re like me, and do a bit of blacksmithing and metalwork, you’ll be right at home in the making.  The plans for this baler are free online from Tillers International.

The attraction of the continuous baler is production, but it needs skill to load the hay into the baler evenly. The tendency is to push too much to the bottom of the infeed box and your bales will want to banana apart.

Good compound leverage, very effective. As I recall there was an error in the plans with the placement of the bearing rod for the plunger. The plunger face has to push fully past the spring loaded dogs that hold the hay from pushing back into the infeed box. Isn’t there always a sneaky error in a set of plans though? You’re clever people and will figure it out.

Do you love small unexpected touches? I bought these pens which shipped from Japan and came with a little folded paper crane. It got me wondering if there’s  a machine that can fold these mass production style, it wouldn’t surprise me.

Back from reminiscing about summer’s warmth, baby its cold outside. But maybe time to make some snowshoes?

Gabe Dwiggins

I read Sebastian’s final blog post this morning:

I too haven’t written much lately, mainly the struggle to express the ways in which life changes and how it changes us in turn.

And I must say thank you so much for what you have been willing to share, I greatly respect the honesty in your work and writing. For what the future holds let us favour optimism and give thanks for all things crafted with love.


A Japanese Saw Sharpening Vise

Its long overdue that I make a better saw sharpening vise. Jason posted this great diagram of ChoMasaru’s vise quite a while ago when he put one together:

The image quality is not so good, but it looks to be in shaku scale.

Odate has a small diagram of a saw vise in his woodworking tools book, which I had used as the basis for the last one I made, but I was never happy with it.

I started by making a rule with sun (寸) increments (30.3mm) using a piece of kumiko I had laying around.

And then joined up some white oak, wide enough to fit the full length of my 350mm rip saw which tends to see the most sharpening of all my saws.

From the diagram a wooden pattern was made to trace the shape on the edges of the joined white oak.

I started on the inside of the pattern, slaying the waste with a router.

The rest of the inside shaping was with hand planes.

Before starting the shaping on the outside I clamped both pieces together to match drill for the small bolts I used to hold the two pieces together.

I had a newcomer to the shop as well, making a fidget spinner out of wood. The irony of someone getting completely absorbed and spending hours and hours making a small toy meant for people with attention problems was not lost on me. Young people need tools in their hands, not fidget toys. I could do without my tools being dropped on the concrete floor though, best to have a student set of stuff on hand for such occasions . Oh, and a first aid kit for when they cut themselves, which they always do, no matter how carefully you explain the whole “don’t cut into your hand with the chisel” thing.

After drilling for the bolts the rest of the outside profile was cut. I sped things up a bit with a power planer before refining with a low angle block plane.  Too much time lost to repairing a chipped plane blade dropped on the ground, haha.

Fitting the wedge took a while, I started off with an angle too obtuse to stick in there.  Now I just need the little box to lay it up against and collect the filings. Having a good vise for filing makes all the difference in getting nice teeth geometry, and the ChoMasaru pattern is beautiful!

My shop is located at the other end of this rainbow if you want to visit.