What a long and winding road to get here. It all started with this video from Michael Alm (https://youtu.be/3d9t2uBtKOI). When I first watched it, I knew then and there that I would have to make a box like that. I did not know how roundabout the path to get there would be. At the time, I was already well in to making my basement a proper shop. I knew the basics – make a mitered box and add splines for strength, but the patterned plywood immediately drew my eye. This would be a challenge that didn’t require much wood, or any new tools, and the box simply looks great. I was working on some other things and didn’t get started on it right away – but it was calling to me.
Turns out those other projects did not go as planned and I didn’t really get started on this for some time. Eventually the stars aligned and I decided to give it a try – but having had some table saw challenges on my other projects, and having never worked with baltic birch plywood before, my initial plan was to make a prototype. Turns out I had so many little squares of patterned plywood, that I made two prototypes. I posted about it back in November (https://sawdusting.com/2021/11/11/building-a-prototype/). I considered it a successful prototype – but there were areas I needed to improve. It needed to be a bit wider, I wanted the accent wood in the plywood pattern to be larger, I wanted to get rid of the “bunny ears” on the spline joints and finally, the boxes needed to have more accurate 90 degree corners as the prototype lid on one of the boxes only fit in one direction since the box was out of square. Good that I had a practice run, but there were certainly things to fix this time around.
When I found the time to start on the real thing, I began with the baltic birch plywood and walnut, to make the blanks. These get sliced up later to make little squares that are assembled to make the pattern. First I cut up strips of plywood, and glued up the right angle on the long edge. I added the walnut and glued it up to make a square. Then I cut up the slices and discovered, that the blanks were not at 90 degrees. I could not make a pattern with them that fit together properly. I posted the story about that fail at (https://sawdusting.com/2021/12/06/it-didnt-work/) if you wish to take a look.
So I returned to the very beginning. New pieces of baltic birch plywood. New zero clearance insert on the saw. I checked and re-checked the angle setting and how I stabilized the board as it went through the cut. Then when it came time for glue up, I was extra careful with setting the angled pieces on the blue tape, and how I folded them together. I checked each with a square as I put the rubber bands on. This time it worked. Well, it worked enough. Some of them still came out a little wonky, but the majority were good, and I could assemble patterns, and just set aside the ones that seemed screwy. In the end I was able to get more pieces out of the wood, based on some changes to how I cut them, and this netted me enough patterned sections to make 4 tops and 4 bottoms, with a handful of leftovers.
I already had a jig to glue up the patterned plywood from the prototype, but doing it 8 times was going to try my patience, so I quickly made a bunch more jigs – only to realize after the fourth one that I was out of clamps. Michael Alm did a later patterned plywood project where he scrapped using clamps and used wedges instead – that came out after I had done this, and so I ended up doing it in two passes. Once the glue ups were dry, I could take the sections out and trim them so they were equally sized and square. Now on to the boxes themselves.
Then came the next frustrating hurdle. I had a few days off from work, was ready to get started, and can’t seem to get the saw blade to tilt to 45 degrees. Indeed, the saw was broken as I posted at (https://sawdusting.com/2021/12/25/something-isnt-right/). At that point, I had to wait for the part to arrive before I could fix the saw, so I spent some of the time filling imperfections in the plywood so that the surface was nice and smooth. Some sawdust, CA glue and accelerator – plus sanding, and you can get a nice smooth finish.
When the saw was back in action, and I had some time again, it was on to the boxes. I struggled a bit with the sizing of the maple for the sides. 3/4″ was too much and if I tried to resaw it on the tablesaw, I would end up with two pieces about 1/4″ or so, which seemed too small – and the boards were too big to resaw on the bandsaw. So I planed them down to 1/2″ and I think the proportions work. It could have gone a bit thinner maybe….. Then I decided to try a new trick. I tried to match the grain going around the box. Since I planed them and did not resaw them, I could not get a match all the way around (too complicated to explain here), but I could get it to match around three corners, so that is what I did. It required me to cut them in sequence – short side, long side, short side, long side, and to keep the boards together for each box (I was making 4), in order, and oriented correctly. I was pleased with how that worked out, and will try matching all four sides in the future. Last step before cutting the boards into their final sizes, I added the grooves to allow the top and bottom pieces of patterned plywood to recess in to the sides of the boxes.
To cut each of the sides, I made some test pieces in scrap to work out the dimensions. Then using a stop block I could cut 1 short side from each board, for each box. Then I could move the stop block to cut the long side, but the challenge was how to get back to the setting for the short side which had to be cut after the long side to keep the grain match. I remembered seeing an ad for the “rip flip” from Woodpeckers (https://www.woodpeck.com/rip-flip-fence.html) and had the idea to “mark” the position of the first stop block setting by clamping another board adjacent to it, but one that did not extend to the bottom of my fence, so the long board would fit under it. This way I could move the stop block back and forth, and keep my settings accurate and the boards the same size. That worked out great and is a trick I will remember for the future.
Now that they were sized, it was time to cut the 45 degree angle. This was where I have had challenges before. I used a stop block to keep my lengths the same, and I also made sure to take more than one pass through the saw. It seemed that I could get a straighter, cleaner cut, if I took off only a tiny little bit. Basically I cut the angles, and then recut them again, taking off about the thickness of a playing card or less on the second pass. This seemed to help. I still think I am getting some flex in my sled and will probably build a new one before trying this again, but this got me where I needed to be.
After cutting and recutting all the pieces, I gathered them all up for a dry fit. Three went together well, one had to be cut again to try and square up the corners. It was tedious, and it seems to have been more work than I think it should be, but in the end, it worked out. Tops and bottoms fit, the sides all came together and the miter cuts were done. Next up was some sanding on the inside, as that would not really be possible after assembly. I carefully marked all the bits and pieces and it was time for the glue up.
This is typically the most stressful part of a build – once the glue goes on the clock starts ticking, and pieces need to be quickly assembled and into the clamps before the glue sets. I have since learned that Titebond II was probably not the best choice for this, but it is what I typically have on hand. Maybe next time I will go with something that sets a bit more slowly. To help lower the stress, I used I trick I saw on Bourbon Moth Woodworking (https://youtu.be/ilyZkKBWGQY) where he attached some small mitered blocks to the corners using CA glue and then used them to get a nice square surface for the clamps. I tested it out first on some small maple off cuts, and it worked out, so I made a bunch of them and started my glue ups. It was stressful, as they always are, but in the end the boxes came together, and stayed square. I was a bit frustrated as I think some of the joints looked better when I dry fit them, than how they came out in the end, after applying glue. I think maybe the glue started to set before I got all the clamps in place. Then I let them sit overnight.
In the morning, off came the clamps and with a little sanding, the boxes were again nice and smooth on the outside. Now it was on to the splines. I used the same jig from the prototypes, but a new saw blade. This one has a “flat top grind” meaning each tooth on the blade if you were looking at it spinning towards you, would have a perfectly flat edge on top. Normally saw blades have a series of bevels on top to help the cut, but they don’t leave a flat bottom to the cut – and that leaves what are known as bunny ears, when you go to put in the splines. The jig worked fine, and the new blade was great. Somehow in the 48 cuts I had to make, I messed up one by not holding the box properly. The groove was wider than all the others. I ended up making another spline of maple, glued it up to the walnut, and slipped it in place and you can barely tell there was a mistake. That was a near miss for that box. Now that the grooves were cut, I set about making the actual walnut pieces that fit inside and then one by one glued them in place. Again, let that sit overnight.
Now I could try out my new flush cut saw. On the prototypes, I had used a japanese pull saw to cut the splines flush, but the saw was too big, and in places, it left saw marks that took a long time to sand out. The saw works fine, it was the angle I had to hold it to make the cuts on such a small box, without getting hung up on other splines that were sticking out. The new, smaller saw, worked out great. Another issue solved. After cutting, more sanding. There is always more sanding.
Moment of truth – time to separate the top from the bottom. On the prototypes, I cut this on my bandsaw, freehand by following a line. This time I decided to use the table saw. To prevent the pieces from actually separating, I set the blade so that it would cut nearly through each side, but not completely. That little bit of wood left keeps the pieces from separating as they go through the saw, helping reduce the risk of both kickback, and leaving saw marks on the piece as it came apart. This worked out well. I used my original japanese pull saw to cut the last bit by hand – but after I did that two or three times, I got smarter and used the bandsaw for the last bow or two. That actually worked out really well and is how I would do it in the future if the object fits in the saw. Next up, more sanding.
The home stretch – making a liner for the box. Since I did not plan to add hinges, and wanted the box lid to come all the way off, I needed to make a liner that woudl stick up from the bottom section, and extend in to the top of the box about 1/4″. This allows the top to “register” and fit properly, without simply sliding off the bottom. I milled down some scrap mahogany I had in the shop and then carefully cut and fit each piece, one at a time. I only put the liner on the sides, so the patterned plywood is exposed on the inside of the box bottom.
Since I planned to leave the mahogany unfinished, I needed to apply finish to the rest of the box first, otherwise there would be no easy way to keep them “finish free”. I taped the inside of the box sides with blue tape so that they would stay unfinished, and provide a good glue surface. Then I applied Watco Danish Oil to all the outside surfaces, to the inside of the lid, and interior bottom of the box. Flood the finish on, let it sit for a bit, flood it some more, let it sit, wipe it all off and let it dry for a while. Then repeat. I think I applied four coats like this. Using an ultra fine scotch brite pad, I smoothed out all the surfaces and eliminated any dust nibs.
Once dry, I could remove the blue tape from the inside, and dry fit the liner pieces again. Once everything was pressure fit in place, I took them out, gave them a quick sanding, and glued them in place. Since they pressure fit so tightly in place, I barely needed any clamps. After the glue was dry I hand sanded the top edge of the liner to make it smooth, and rounded. This allows the lid to fit comfortably, and ensures that the surfaces you touch are nice and smooth.
Just like that, and they were done. When I look back at Michael Alm’s boxes, it appeared to take a couple of afternoons at most. Mine took far longer. Some of that was the learning curve for me, some was simply the tools I was using, some was simply that I am not the fastest woodworker out there. In the end I am pleased with how they came out. I would like to try a different pattern, and thanks to a recent sale, I am once again stocked up on baltic birch. It’s calling out to me, “more patterned plywood” it keeps saying. Keep an eye on this space as I am sure there will be more.
Thanks, as always, for reading along.