After doing that tongue-and-groove stuff, I nailed on the bottom of the chest:
I guess I sawed off the ends with a panel saw. I don’t remember. In any case, I’ve been looking through not just the “ATC,” but also “The Woodworker” articles on traditional chests, and noted that some of the articles in the latter had said to use screws on the bottom. I went with cut nails, especially encouraged by the fact that I actually have some cut nails on hand.
Then it was on to the plinth and lower dust seal. These are just dovetailed bits and pieces and are pretty much standard fare. Pictures of those are a little later down.
The lid frame and panel presented another typical situation: determining which boards to use for the frame and panel. You normally want frames to be rift- or quartersawn because they’re more stable, and of course, those cuts are difficult to find if you’re just looking superficially (which is close to what I was doing when I bought most of the white pine that I have on hand). Fortunately, I had this one:
At first glance, this looks like a board with a lot of knots and an obnoxious cup. But slice out the center part, make some cuts here and there, and you get this:
That kind of riftsawn cut is what I was looking for.
I made the frame using through mortises (something I rarely do), did the joinery for the upper dust seal, and am now close to finished with the outer shell:
In a case of not following directions, there’s a small knot on one side here. It doesn’t show much sign of wanting to fall out, but if it does, I could repair it, or something. Because this is not a tool chest, it may not be important.
A look at the lid:
The upper dust seal is not attached or trimmed to final dimensions. This is because to really finish off the lid, I need to install the hinges, which I do not have at the moment. These are allegedly in transit.
I’ve already selected the wood for the sliding till runners and the sides of the tills themselves. I haven’t decided if I want to make some kind of divider in the bottom yet. Again, this isn’t for tools, so I don’t know if that kind of compartmentalization is appropriate here.
But it looks like I’m getting somewhere. In theory, this project shouldn’t take much longer.
In my small chest build, I started working on the bottom. This is usually done with a series of tongue-and-groove or shiplap joints. I prefer tongue-and-groove, but although I do have a set of match planes (and a Stanley 45), they are for 1/4″ grooves, and I’m working with stock where 3/16″ would be more appropriate.
Match planes are great for pumping out a lot of tongue-and-groove joints. To do it at almost the same speed, I could spend $30 on the Veritas tongue-cutter for my plow plane in that size. But none of this is necessary, and I didn’t feel like waiting around for that cutter to arrive anyway. So I set out to do it with the plow plane. Doing so requires three cuts instead of the two that you’d need to do with match planes; the tongue is nothing other than two rabbets.
You start by plowing the groove side, which is the same no matter what:
It’s best to plow all of the grooves at the same time, so that you don’t have to constantly readjust your plane.
Then, with the groove side still up, adjust the fence so that the cutter sits right on top of the “outer” side of the groove (the side opposite the fence), with the sides of the blade and groove flush:
This is where I have to say that I really like the Veritas small plow. I could do this with my Stanley #45, but moving and locking down the fence on that thing is finicky and difficult to do with precision.
Now, flip to your tongue side of a board, and (here’s the important part) clamp a scrap to the side that you’re going to cut, flush to the top. Then cut this side of the tongue:
The reason for the scrap is so that your depth stop (which is usually on the side opposite the fence) has something to hit when the cut is complete. Yes, I made the mistake of forgetting to do this on my first attempt.
Again, it’s best to cut all of your boards at the same time. Then, put a board’s groove side up again, and set the fence so that the cutter rests on the “inner” side, like you did before:
This one is more difficult to see, so feel free to use any optical method of cheating that comes to mind.
Then flip back to the tongue side, and cut this second rabbet that makes the tongue:
You probably won’t need to include the scrap this time, because the depth stop should register against the tongue itself.
It’s possible that you might feel that the fit is too tight when you’re done. In that case, just adjust the “inner” side with a rabbet plane. I’m using my miniature shoulder plane because I recently said that I seldom use it, and I feel guilty for that. Or something.
If you’ve got a lot of these to make, one possible time-saver is to reset the fence to the trimmed side after you adjust the first tongue.
That’s more or less it. And I now have everything ready for the bottom of my small chest:
My next project is a scaled-down down version of the traditional English tool chest, for use in my office as a place to store the junk that always seems to pile up on my desk. Call it “The Conformist’s Doodad Chest” or whatever. Though it will resemble the original form, let me be clear: I don’t intend to use it for tools. It’s not big enough for that purpose, at least not for woodworking. It’s half the length of the ATC as originally published, but has only 1/6th the volume. The stock I’m using isn’t quite as thick; it’s not intended to be beat around a shop.
Coincidentally, Joe just finished a tool chest about the same size, of which he will be detailing the build process soon.
As I was preparing the stock for the shell, I realized that I probably ought to make a new shooting board because I was doing stupid improvised stuff like this:
I started on the shooting board, but it will be a topic for another day, maybe.
You have to make a lot of dovetails for these chests, and I used my new bench-on-bench to make them. So far, so good, and I soon had the sides of the chest made:
I glued them up yesterday, so now I have the shell with untrimmed dovetail ends:
That scribbling reads “flat (enough).”
Now, I suppose that since I have to make a bunch of skirts and tills and stuff, it’s time to stop kicking the shooting board can down the road.
[Edit: I’ve now made a video explaining this bench.]
I’ve been interested in the idea of a so-called “Moxon” vise or bench-on-bench setup for a while, but there have been many other projects that have taken precedence. The idea of a higher sawing position for smaller cuts just seems like a good idea. Over the summer, I asked a group of woodworking pals what they thought of it, and I got some feedback including “just do it, you won’t be sorry.” Also included was a bookbinder’s finishing press treatise/rant from our friendly neighborhood bookbinder who stated that (among other things) if I really wanted to make something specialized for woodworking, I should go for bench-on-bench. Since there’s a near-zero chance of me wanting to get into bookbinding (sorry, Tom), I decided to go for bench-on-bench.
Unlike Joe’s elegantly-finished vise, I decided to go with the ever-sturdy southern yellow pine for most of the construction, and ordered the Benchcrafted hardware. As usual, the hardware sat for several months while I dealt with other projects.
There doesn’t seem to be much consensus on the construction of bench-on-bench, so I decided that I would stick to the “every one of these is different” rule and make a solid inverted “U” shape for the top and front/back, with some stretchers along the bottom. The stretchers are not for support, but to provide a way to clamp it down. Here are the parts ready for final glue-up:
This is, of course, upside-down. I waited to glue on the rear because I felt the need to do that along with the stretchers. In addition, I used a loose tongue there (with the appropriate diagonal-grain configuration) not for strength, but to keep the top aligned with the rear during glue-up. This might be the first time that I’ve used this kind of joint.
There wasn’t too much to do after that. The next step was fitting the chop:
The chop is a 4/4 piece of soft maple. The Benchcrafted instructions suggest a configuration with the vise nuts for the fixed chop recessed inside, so for no good reason (other than stupidly not doing it the quicker way on the drill press before gluing the front on), I traced the nuts on, and chopped out the waste:
Then, I chamfered the top of the front chop, added the “crubber” (the cork/rubber gasket material that BC includes with the hardware), and did a quick rubdown with some tung oil:
You can see that this is currently clamped to the bench with C-clamps on the stretchers. I suppose that I could add dog holes in the bench and use a threaded rod or something. but this seems OK for now.
I added one more component: a slide-up stop at the rear. Here’s how that looks in the “partway up” position:
It can extend higher, and is held in place with a couple of threaded levers. I’m not sure if adding this was a good idea or not, but the main reason I put it there was to try to get some bench hook functionality. For example, when doing the crosscuts to remove the waste from tenons, I thought it might be handy to take the work out of the vise and lay it along the back. Ditto for using it to stop a piece when cutting a small groove leading to a knife line with a chisel.
At worst, if the stop doesn’t prove to be useful, I can just stow or remove it.
Here’s one final view, with the vise holding a board in “dovetailing” configuration:
This seems to hold securely. I did several test cuts and nothing appeared out of the ordinary. It does seem much more comfortable than stooping down to my (low) main bench to cut stuff. The auxiliary bench that I’m using here might be a little on the top-heavy side now; I do want to add more weight to the bottom. Happily, this should be soon, with one upcoming project that will use this new tool.
Several months ago, as I was starting the daybed project, I also decided to do another project in parallel: a stand for some decorative object that’s to go in the living room. I figured that it wouldn’t take much time; it would have three tapered-tenon legs, so all I needed to do was make a quick model to figure out what angles would be appropriate, and just do a little work.
I’d already made the model (leading to this). For the top, I glued up a really nice chunk of yellow pine and sawed it into a disc on the bandsaw. I made the legs and the tapered tenons on those. So all I had to do was ream the mortises and I was as good as done. Shouldn’t take more than a few days, right?
That’s when disaster occurred. For the life of me, I couldn’t keep the angle consistent when reaming with the brace. I failed immediately, reaming too far on the first one. Having ruined the top, I figured that I could learn just what I did wrong, so I tried the other two. I failed at those as well. Then I just tried a bunch more and failed at all of them by hand:
I was just having too much trouble controlling the reamer. It wobbled too much in the brace, and I couldn’t feel anything. Furthermore, because this is on the thin side, I was having trouble making adjustments without overcompensating. If I were more experienced, I might have been able to do it. Then I tried with a power drill. I was able to do some decently using that.
Sitting back, I wondered what I could do about this situation. I’d speculated before that I suspected that the reamer I was using was just meant to go into a chuck on a power tool, and this reinforced it.
Note: Chris Schwarz says in an old post that this “standard” reamer, sold by Lee Valley, is made in Canada and fits well in a brace. The former is NOT true (maybe it was back then), and as for the latter… maybe it fits better in a Spofford brace? I dunno. The shank is very soft, which caused me a lot of trouble in the brace.
What to do? I figured that I had four options:
Use a handheld power drill to do it.
Make the “banjo jig” to do it on my drill press.
See if I would have better results with a reamer with a square-taper shank that’s meant to go into a brace.
Sulk.
Door number four is never a bad choice, and I did a fair amount of that (especially considering how nicely the rings were arranged in the glue-up of the top that I ruined), so I thought, well, I’ve never really been too comfortable with my power drill; I’m don’t have the steadiest hands. Perhaps I could make the banjo jig?
But I also thought, well, I like hand work, and I don’t like the idea of another jig lying around. Maybe I could be lame and blame the tool, try a different reamer, one with a square-taper shank? I ordered the Veritas “Pro” Reamer. It was on backorder for a week or so. That became a few weeks. Then the expected date was shifted to January. This was not a cause for optimism. Much sulking ensued. But just as I was thinking, okay, well, time to build the banjo jig, mine shipped around the end of November.
To make a long story short, I am far happier with this reamer. Other than the proper shank, there are two things in particular that I like: First, there is a single blade, which seems to make it easier to control with a brace. Second, the blade is adjustable: You can set how much it protrudes (different amounts for different woods), and you can adjust the included angle slightly with set screws.
Great. My first “genius” idea was that maybe I’d try to use the thing in my T-handle in the “inline handle” configuration (or whatever it’s called; it’s the configuration on the right):
That worked, but it was really slow. So I first went at it with my largest-sweep brace:
This was quite effective. I was able to get down to near the depth I wanted quickly and at the correct resultant angle. I admit that I did switch back over to the T-handle when almost done to go more slowly when finishing up, but I think I probably didn’t need to do that.
Then it was time to chamfer the bottom edge. It’s weird to do that a curved surface. C.S. did this on a similar project in the ADB with the bandsaw, and I probably should have done that as well. Instead, I first went at it with a block plane chamfer attachment, which did not work terribly well, but got me to a certain point. Then I made this stupid thing to go the rest of the way:
It’s essentially an adjustable chamfer block with a piece of sandpaper clamped down at the bottom. This worked, but not very quickly. It loads up too much. I think it would have gone much faster with 60-grit paper instead of the 110-grit here, but eh. This might work really well with a milled-tooth file wedged down in there. I think I know how to do that, but have no need right now.
In any case, it got done. I glued and wedged in the legs next. This left me with the final cutting task: getting the legs to length. I marked them with the “slide block of wood with pencil taped on” method, but then I needed a good way to clamp the octagonal legs to the bench to actually saw them. I was thinking about making some kind of custom caul, and then I just grabbed the above stupid tool and clamped with that:
It was far more effective at this task, so the futility was diminished somewhat.
After sanding, I finished it with tung oil. Here it is in its intended location, without the object that will sit on top:
Aside from the top being southern yellow pine, the legs are ash; one of them was seen in this post. A project that I thought would get me a few quick XP instead dragged on without much action for a couple of months. At least I no longer need to sulk over this one.
Additional: I have a plan for that initial ruined top. I do not know when I’ll be able to act on it.
The only remaining task after gluing in the daybed legs was to glue in the railing, which brought us to this point:
It’s hard to notice, but the rail has a slight concavity on top, to echo off the front of the platform.
Well, that’s all fine and good, but almost no one wants to lounge around on a slab of wood, right? I haven’t really mentioned anything about the cushions, except that I’d finished them. I don’t have any photos of that process. It’s not exactly woodworking, and I don’t approach sewing projects in the same way that I do woodworking. So you’ll just have to settle for me sewing together two scraps of the cushion fabric:
Grayscale to make it look more “old-timey-looking,” when in reality, I was just too lazy to deal with the white balance down there.
Levity aside, I’d like to thank The Funky Little Chair for making a great video showing how to make cushions. I’ve never done anything like that before.
I could have made my life easier if I’d done square cushions first instead of a curved thing with weird little cutouts in the back. But I managed, so here’s the final piece:
Strangely, even though I’d only made a very rough drawing of this before starting, it came out almost exactly as I envisioned it. That doesn’t seem like it should happen very often.
To recap, this is southern yellow pine with red oak legs and railing supports. The raw material cost wasn’t terribly high; I think the foam might have been the biggest single-ticket item. The main problem with materials is finding what you want. The wood is one thing, but if you’ve never shopped for fabric before (while trying to get your spouse to agree on a selection), there’s a whole new world of fun waiting.
I’d originally intended the daybed project to have tapered-tenon legs and railings, but after I thought about the design a bit and what I wanted to accomplish (and with the help of a spectacular failure with the tapered reamer), I decided to use normal mortise-and-tenon joints. Well, sort of normal–I decided to use angled twin tenons.
I don’t have many photos of the process of making these, but the biggest difference is that, because the mortises are the weirdest part of this, I cut the tenons first. I marked the tenon thicknesses using the width of my mortise chisel–something that I almost never do. Then I marked one side of the “outer” mortise” and use a block cut at an angle (10 degrees) to guide the chisel. Here is a simulation of how that worked:
With the first mortise made, I then inserted the tip of one of the “outer” twin tenon into the mortise, and gave the end of a leg a good whack. This put an indentation of the other tenon into the platform, and then I knew where to line up the block to chop the second mortise. Surprisingly, this worked; I was not terribly optimistic about it.
In any case, with the legs made, it was time to shape them. I cut out the rough shape on my bandsaw because when you have one, that’s what you do. Then it was time to do the medium-grade shaping. Doing the “front sides” was easy because those are straight; I just used a jack plane like this:
I followed that up with a smoothing plane. Hmm, look at that, you can actually see the twin tenons in this photo.
In any case, the rear sides of the legs were to have concave curves, which excluded the jack plane. However, I did have a small compass plane that I got in Taiwan several years ago (Japanese blade, Taiwanese body), and was able to use that for much of the work:
But once that reached its limit, it was rasp time. The shaped legs looked like this:
The next steps were to sand the legs smooth and finish them. I don’t have any photos of that. I used a tung oil/varnish blend.
After that, they sat for a long time until today, when I glued them into the platform of the bed:
This project is almost done. The other wooden component is the railing, which is done but not glued on. Otherwise, there are three cushions that go on the top, which I’ve also completed but won’t address in this post.
All of that fussing around with making those clamps has finally gotten me ready to get started in earnest on my next project: a daybed/”couch”-like thing for an area of the house that currently has zero furniture. I had a sort of “vision” about how I wanted it to appear one day, and knew that I had to build it before someone else in the house bought one, because what I imagined seemed pretty neat, at least at the time. And when I read what Roubo had to say about them (yes, it’s in there), I became more convinced.
I wanted this to go a lot more quickly, but getting the large platform glued up took much longer than I had wanted–I had to make a whole bunch of stuff, including the clamps, before I could get to where I am now.
The first operation on the glued-up platform (after planing it off) was to shape it. I started with the ends, which are to be semicircular. I don’t have any trammel points to make a beam compass for such a large radius, so I just drove a couple of small nails into a board at the appropriate distance and used that to scribe it out:
As made, the radius of this improvised beam compass was a little too much, but to correct it, I just tapped one of the nails to bend it inward. I marked over the scribe line with a pencil.
Then there was a curve to lay out on the front. I guess I’ve been watching too much “Tally Ho,” because I tried to imitate Leo’s curve marking with a batten tacked into place:
I wasn’t sure if this would work, but it did. There were a few options for curve shape, mostly dealing with the tangent angle at the point of inflection in the middle. I chose a fairly mild one.
So I had one side done, and to do the other, I traced that side’s lines onto some of the paper that I use for making patterns, then cut along the lines on the paper, flipped it over, and transferred the lines to the other side of the platform.
Then it was time to start removing wood. I always get a little nervous the first time I cut into a panel (especially one that I’ve spent so much effort on gluing up), but it had to be done:
I removed as much as I could with a saw, working my way around the ends and trying to get as close as possible to the line. The cuts on the long grain were more challenging, so I used a drawknife to get rid of most of that material:
I finished the initial shaping with my Shinto saw rasp, and had this:
The front curve was somewhat tricky because it’s concave; I sawed a relief cut in the center and took out most of the waste with the drawknife and a chisel. (Note: Southern yellow pine is not particularly easy to use a drawknife on, but there’s worse.)
It came out as planned. Pay no attention to the ugly knots in the middle; those will not be visible with a cushion on top. I would have preferred clear wood, but due to multiple factors (the proximity of the nearest source of large southern yellow pine, the inconveniences of the source at the time that I went to buy it, not to mention the obnoxious process of picking through the pile alone on a steamy hot day in the middle of a pandemic), I settled for straight, mostly-quartersawn boards that would yield a clear perimeter, which is the only part of the wood that will be visible.
There isn’t much remaining work on the wood to do. I have to make the legs and the railing, do the final profile of the edge, and finish it. During finishing, I’ll make the cushion (hopefully I’ll have some sort of an idea about what I’m doing there).
After I made my panel clamps, I’ve now used them several times and have been able to work around some of the issues that came up. Here’s a partial shot of the latest project getting its final glue-up (this is big–roughly 7×2 feet (2130x610mm) in area:
I have four clamps–three long ones and a shorter one that I made earlier as sort of a prototype. I used all of them on this project.
Let’s address the the two comments that on speculation of “stuff that might go wrong” that I got last time.
First, Lionel asked if glue squeeze-out is a problem; can you glue the work to the clamps? The answer is yes. I said that if this were a problem, I’d just wax the insides, and that’s exactly what I did. The wax eliminated the problem.
Then, Matt asked if I used a single wedge, would it tend to wear/dent stuff into the pegs? I tried it, and it does indeed mush into the pegs (especially when they are southern yellow pine).
On that note, I wasn’t terribly happy with the way that the single or double wedges were working. In particular, all sorts of things can happen when you try drive them in:
The force of the mallet blow can shift and tilt the entire clamp over just a bit.
On double wedges, it can be tricky to hit one wedge without loosening and dislodging the other wedge.
On double wedges, it’s pretty easy to hit them so that they go askew of each other, and then they can slip out.
If the clamps are too close together, it can be difficult to find enough room to get your mallet strike started.
Fussing around with the wedges during glue-up (especially when you don’t have anyone to help you) takes precious time. The liquid hide glue that I use gives me a little more time, but still.
One of the speculative ideas that I had last time was I could add a hook to the upper wedge, so that I’d only need to hit one end of it. I did this, with one other change:
I planed a groove into the top and a matching tongue into the bottom with some old wooden match planes I picked up in Alameda many years ago. (Note to self: regrind those plane blades, they’re in awful shape.) This eliminates the slipping askew and generally makes it much easier to get the wedges set up.
With that in place, it was much easier, but I still had some difficulty finding the room to set the wedges at times, and didn’t really have a feel for how much I was tightening the joints. So I came up with this dumb hack to do the final tightening:
Even though using a little F-clamp like this might look like it could back the hook off of the peg, it hasn’t happened to me in practice; it just shoves the lower one into place. In any case, this made the job a lot easier.
In addition, I made new pegs out of ash to discourage dents from forming in them. I suppose that I could do that with the upper wedges as well, but that hasn’t caused any problems (after all, the upper wedges aren’t supposed to slide).
So now what do I think of using them?
Overall, I’m pretty happy. One of the things that I like the most is that you perform glue-up against a wall with the work sitting on its side. This makes it a lot easier to get things in place, and it’s also relatively easy to do a dry run of the glue-up and clamping, or at least get a good idea of what’s going to happen. Once you have everything in place, you can easily check both sides by just tilting it one way or the other. And because it’s already against a wall and not on a bench or anything, you don’t have to move it afterward to reclaim your space.
My plans today were to get the legs onto the stand that I’m building. Things did not quite go according to plan. The first impediment was that I haven’t been too content with the way you use the tenon cutter, and it came to a head today. I prefer to twist the stock into the cutter rather than twist the cutter around because it’s easier to keep it centered that way. Unfortunately, it’s not really easy to put the tenon cutter into a vise or something to keep it put. Not like it’s easy to twist the cutter, either, mind you.
So I picked up a squared-up ash cutoff, put a rabbet into one edge, and screwed the tenon cutter into the opposite edge (it has holes and little half-moonish registration standoffs to help with the attachment):
Then I can just put it into the vise, with the rabbet registering on the benchtop. The result is that (in theory, at least), you can hold the stock parallel to the benchtop and bench front to keep it in the right place:
With this system in place, I was still having problems. The tenon cutter isn’t large enough to engulf the whole octagonal profile of these legs. I was doing a rough shaping with the drawknife and saw rasp to get a cylinder to shove into the tenon cutter, with a little transition area to the outer surfaces. I determined that you need to be fairly precise about the size of that cylinder. I’d made it too small, which resulted in tapered tenons with a long rough unshaped area between the tapered part and the transition to the octagon:
This happens because the transition area eventually meets the end of the tenon cutter and cannot advance any more. This wouldn’t happen if the tenon cutter were larger than the stock being worked (which would be the case for most cylindrical legs), or if I wanted to do a more gradual transition (assuming that I were actually good enough with a drawknife to do that). Even though this 9/16″ version isn’t the largest tenon cutter that Lee Valley offers–the 5/8″ keeps selling out too quickly–it wouldn’t make much difference here.
I decided to cut new transitions farther back in the stock and make newly-shaped cylinders bigger in order to bring the tapered part closer. Something tells me that this would be a lot easier with a lathe, but I don’t have one of those. I’d been putting it into my tail vise and just whacking away with a chisel, but the end would keep sliding around too much. So I interrupted myself again and made a stop with a mouth on one end that I could secure into two dog holes (to keep it from swiveling around):
This worked. As a bonus, you don’t need to secure it with the tail vise this way, making it really easy to turn the stock around so that you can work on all of the facets.
With this aid, I was able to shape everything a bit better and use the tenon cutter again. This made for a much nicer result:
I’m not worried about the 1/4″ between the transitions and the tapered part (this is the distance between the tenon cutter’s blade and its end).
Unfortunately, I spent so much time fooling around with these little shop aids and trying to figure out the best way to use them that I didn’t have any time to ream the holes for these things today. Oh well.
Galoototron 