Toshi-nuki Shelves – Build 1

It’s time to start building some knockdown shelving. A few weeks have passes since the last post where I talked about different joinery options for this project, and after giving it some time I’ve decided to go with the watari-ago joint. With one exception. The wedges have continued to bother me. When they are installed on the top side of the shelf stretchers they interfere with both the useable shelf space and also the installation of the shelves themselves. So I kept thinking about different joinery, or different possibilities for orienting the wedges, and finally it hit me, why not just flip the whole joint upside down? Here’s how that would look.

I like this a lot. It solves the problems described above, and visually I think it actually looks better; the wedges now look like they are supporting the shelves kind of like the bracket structure you see on temples here in Japan. A few people have expressed concern that the wedges are now going to be more prone to falling out, and I’ve been pondering that possibility too. But I’m not convinced that it’s going to be a major problem. If the wedges do seem to come loose, one possible solution is to actually use a light dab of glue when installing them, particularly with a weak glue like rice paste glue, or even nori glue used to attach paper on shoji screens. A weak glue would still allow the wedges to be easily knocked out with a light tap from a hammer.

So with the construction details finally decided, I bought some wood. I got a combination of Japanese ash (tamo タモ) and white ash. I used the white ash for the posts, and tamo for all the remaining parts (stretchers and shelves). The two species are really similar both in color and grain, but the white ash does have a bit more yellow-ish sap wood, and in general the tamo is more uniformly light brown in color.

After sawing and milling the parts to size I laid out the joinery on the posts for the watari-ago mortises. If you look closely at the layout on the front piece, you can see an angled line designating the slope of the wedge.

Pretty quickly I ran into some issues cutting the mortises. Not to get too far off track, but the past three years I’ve been splitting my time between Tokyo and Okayama, Japan. I’ve been working every other month on my own work in Tokyo and with Somacosha in Okayama. After three years of living a split lifestyle I decided it was time to be in Tokyo full-time with my wife and our pup, and put my effort fully into my own business. So, long story short, I’m currently working out of a shared shop here in Tokyo. It ridiculously close to my home, which is a huge plus, but the machinery is pretty lacking. Including a machine I desperately wanted to use in this case, a mortiser.

They actually have a mortiser at the shop, but it’s one of these Hitachi drill/mortiser hybrids. Surprisingly it’s a pretty capable machine for light-duty work, and I used it for the mortises on the western red cedar shojis. I like it better than the cheap benchtop mortisers commonly available in the US, since it has a x-y table allowing easy adjustment of the work. Overall the quality of machining, movements, and functionality is pretty good. So I had hoped to use it for cutting the watari-ago joints. But when I put in a fresh 15mm chisel and tried a test cut in a piece of ash it became immediately clear that the machine didn’t have the strength to cut mortises of that size. I was able to start cutting, but it took so much pressure that I felt like I’d break the machine in no time if I proceeded to try and cut all 18 of the mortises.

So after feeling a bit defeated and realizing that I had overestimated the capabilities of the little Hitachi mortiser, I resigned myself to doing it the old-fashioned way. Well not so old-fashioned, I used lots of machines, but the overall process required a lot more work than if I had access to a quality mortiser.

The challenge of cutting square holes is a big one. There aren’t many machines that can do it well, and unfortunately the options available to those of us who still want to cut square holes are incredibly limited. You have a handful of companies selling essentially the same rehashed benchtop mortisers that leave a ton to be desired. I bought a little benchtop Grizzly unit many years back and returned it almost immediately. The quality of the machining, accuracy of the moving parts, etc. are so poor that it feels like you paying hundreds for a hunk of scrap iron. I’ve used Jet, Powermatic, and the Grizzly benchtop machines and they all felt more or less the same. Those companies seem to put the minimum amount of effort into making these machines. Meaning there’s little to no improvement, and if anything the machines just keep getting worse in favor of saving cost. I don’t think there are any manufacturers in the US still making a production level mortiser. To get a quality mortiser you are basically left with needing to buy an old industrial machine. Usually massive and in varying states of use/abuse, but still a huge step up from what’s available commercially. The only other option is to import machines from Europe or Japan, where professional grade mortisers are still made.

Sorry for the rant, but I really wish there were better options for cutting mortises. Lately I’ve been dreaming of what it would be like to have a tool that could cut mortises like a Festool Domino but with some kind of additional cutter to leave the corners square. A portable, lightweight, fully adjustable machine for cutting square holes….. that’s the dream.

Anyways, back to building the shelving. My approach for cutting the 18 mortises for the knockdown shelf was as follows:

First I drilled out the waste at the drill press

Then using chisels I squared up the holes, cutting to my layout lines. I went at this pretty quick. A few taps with a hammer were enough to establish flat edges deep enough for the next step.

From here I used a router with bearing bit to clear out the waste, running the bearing along the flat surface cut by the chisel.

Here are the mortises after routing. Since these are all through-mortises, I worked off of both sides to clear out the entire mortise. From there I used a couple of chisels to square up the interior of the mortises on the side with the watari-ago joint. I didn’t bother squaring up the other side, since I would soon be making an angled cut on that side to accommodate the wedges.

Next I used the router to cut the 6mm deep watari-ago notches, and free handed the cut as close to my layout lines as possible. From there I cleaned the corners up with a chisel.

The final step for these mortises was to cut the angle to accommodate the wedges. I made a paring block with an 85 degree angle from a piece of scrap wood, to guide my chisel when making these cuts.

Here are the finished mortises. 6 posts with 3 mortises each makes 18 in total. The whole process pictured here took me about 8 hours. Despite my complaining above about not having a mortiser to do this work, once I got started I realized that for intricate mortises like these, with angles and notches, a more hands-on approach leaves a much better finish and final result. I have never used a mortiser that can leave as clean of a cut on end grain as a sharp chisel. And while the insides of the joint are not super visible, it’s hard to accept a joint that looks crusty and full of tear out. I do still want to get my hands on a quality mortiser though. Even if a bit of hand work is required, rough cutting a mortise with a hollow chisel mortiser can really save a lot of time. When drilling or routing mortises like I did here, cleaning up the round corners takes a lot of time. If I had been able to rough cut these mortises with a square chisel, I think I could have cut the time down considerably.

Finally, here’s a sneak peak of the joint after mortising and cutting one of the stretchers. It’s pretty solid, and I’m really glad I decided to flip the joint, putting the wedge on the bottom. There’s plenty of friction on the wedge that I feel pretty confident it won’t be falling out anytime soon. Not to mention that with weight on the shelves, more pressure will be place on the wedges, which I think will further work to hold them in place.

That’s all for now. Next week I’ll share the last few construction details as well as some photos of the finished shelves. So stay tuned, and thanks for reading.