Kumiko

Has every YouTube woodworker done an episode about kumiko, or do I just watch too much Pask Makes? The geometry appeals to me, but a project with so many small, carefully shaped pieces wasn't really on my list of things to try, despite evidence to the contrary.

But then I pulled what I thought was a piece of ash inside to dry, and it turned out to be basswood. What could I do with 20 inches of basswood 2x4? *googling* Huh, it turns out that's what lots of people use for kumiko. And they build the shaving jigs for it out of maple, which happened to be the species of the other 20 inch 2x4 drying next to the basswood. Okay then.

Both pieces of lumber milled well. I set the basswood aside and got to work on the maple. I needed at least three jigs: one at 45°, one at 67.5°, and one at 22.5°. Dividing 20 inches into three seemed like it was going to make jigs just a little too short, so I decided I would rip the 2x4 into two 2x2s, then cut each in half to make four jigs.

I stopped in my tracks after splitting the maple. The quartersawn face was gorgeous. Loads of "ray flake", plus the color variation of this "soft" maple made for quite a picture. These are going to be pretty tools!

I installed my dado stack and plowed out a 1/2 inch groove, 1/4 inch deep. Some tutorials call for 1/2 inch depth, and/or 3/4 inch width. At least one suggested that was too large to comfortably work on anything but large wall panels, though, and I'm thinking of this mostly as smaller decoration.

My magnetic digital angle gauge worked great for setting my table saw's blade angle. The only sketchy cut was the shallow angle. The jig had to go through on-end, since the blade wouldn't tilt beyond 45°. I came up with the support system you see in the image above. It was actually pretty effective. I also sneaked up on the full depth, nibbling just the corner off, then shifting the fence over, nibbling again, shifting again, etc. until I had the depth I wanted. It meant both that the blade didn't provide as much resistance to torque my piece out of true, and also if it did, I could make another nibbling pass to correct it.

With the groove and shaving faces cut, the maple work was finished, so I turned to making the depth stops. I reached for a third wood: strips of white oak left over from my adirondack chairs. Long strips of 5/32 inch thickness glued to spacers 3/16 inch thick produced a 1/2 inch slide to fit in the groove, with room for a screw to pass through the center for tightening it in place.

While glue dried on the stops, I made it back to the basswood. How best to cut a 2x4 to get a bunch of 1/8 x 1/2 inch strips? I settled on slicing through the narrow dimension, because it was the stablest running through the saw. Then I re-ripped those 1/8 x 1.5 inch strips into two stips of finished dimension. That also put the quarter-sawn edge on the wider 1/2 inch face of the strips. Maybe it doesn't matter as much in this application, but it seemed like the better choice for seasonal movement.

And now it's finally time to make some kumiko! I chose a modest 2x2 grid, with each square being about 3 inches wide. Choosing 1/8 inch as the thickness of my strips had a particular purpose: it's the same size as my saw blade's kerf. That meant I could quickly construct the grid using the box joint jig I already had. I set the peg 3 inches from the blade and threw each previously cut notch over it to get precisely repeated distances. Then I moved the peg closer and raised the blade to trim the outside ends to exactly the same length.

This is my main heresy in this project. Pull saws are all the range. I could have used my flush-trim saw, but it felt like the sort of cut meant for my dovetail saw. I am here to spread the underground secret: push saws still work. This was also only the rough dimensioning cut before taking the pieces to the shaving jig.

Now time to get out my nicest chisels, and put a fresh edge on one. From the first test parings, I could see why basswood is a popular choice. Hot knives and butter should fear their idiomatic positions.

So why four jigs for just three angles? Three reasons. First, desipte each jig having two ends on which to cut angles, I knew from watching others that it would be a mistake to try to use both ends in one project, because I would want to use them both without changing the depth stop. Second, I was confused before starting and thought I was going to need a 90° edge as well. Third, I added two more angles that weren't used in this design, but are used in others: 30° and 60°. Thus, the jigs:

As mentioned, the 90° ends were confusion. I may use them to trim the ends of the grid, or I may cut them to other angles in the future. I did use both 45° ends, because I needed two different length pieces at that angle. For a larger project, being able to keep both of these stops set, instead of having to do all of one length before the other or readjust multiple time, could be useful. I paired 67.5° with 30° instead of with 60° because it makes it impossible to choose the wrong end on the other block. I know I need one steep and one shallow, so as long as I get one right, I'll get the correct end on the other one automatically.

Alright, progression time...

Longest pieces first, then the next smallest, then the smallest. Cutting each to rough size, working with the jig to get one (or two, in the case of the middle size) to exact size, then crank the rest through the jig. The process feels like something an established craftsman might have given an apprentice to "teach" or "test" their chisel technique. It sure is satisfying to feel it all fit together just tight enough to carry it around without glue, though.

Will this show up in future projects? Well, I have the jigs and ... what would be a ton of basswood if the stuff weren't so light. So, quite possibly. But next time I need to remember that the sides of chisels often get sharp as well. 🩹 🩸

Categories: Woodworking