Nightstands v2: Finished

For the last month, I’ve felt so close to finishing the twin-nightstand project that I thought it could all come together at any moment. The final details proved as time-consuming to complete as the rest of the thing.

I’ve been varnishing for maybe a month. When I’d gotten enough onto the cabinets, I took care of two remaining steps before the final rubout. The first was how to attach the tops. I decided to use figure 8 connectors because they seemed like a pretty neat approach. Placing them into the cabinets was relatively easy. First, I marked out a box (with no particular offset):

Then I chiseled mostly down to the depth of the connector, using my Veritas mini router plane to go to the bottom:

Then I predrilled a screw hole and did a test fit:

Marking and predrilling the corresponding holes in the tops was a fairly simple matter. However, attaching the tops was more difficult due to the limited space between the shelf and the top. I used a stubby screwdriver, a socket wrench with a driver bit, and practically no vision whatsoever.

The other final task not related to finishing was to cut the legs to height. I didn’t cut the legs to length at the start because I knew there would be some adjusting to do at the end. However, I couldn’t use the ol’ “put the legs under shims, put on a level, flat surface, and mark around” method that’s become popular, because I wasn’t doing a single piece–the nightstands were identical and had to be the same height.

It was easy enough, though. I took a long, thin piece of scrap, marked the desired height to that, then flipped the pieces upside-down (without the tops) on the bench, and marked to that:

It’s a little difficult to see the scrap here because it’s also a piece of cherry, but it’s the thin thing sitting up next to the foreground leg. After making the mark, I scribed around from that mark and sawed off the ends. It worked fine.

Varnishing and sanding between coats was the main time-consumer in the final stages. The cabinets and the drawers got four coats and the tops got five coats. As I have done with the last few projects, I rubbed out the finish with a progression of lubricated #000 and #0000 steel wool only. This was a matte finish again, so I didn’t feel the need to use rottenstone. And really, #0000 does leave a very good finish if you follow the grain most of the time.

So here are the two together:

If you look closely, you’ll see that the drawer fronts are carefully arranged for continuity left-to-right as well as a balanced figure. On the left piece, the fronts have “cathedral grain” in the same orientation, and on the right, they have ellipses, also in the same orientation. But it turns out that the board used for these fronts was cut so that the front on the top left was next to the front on the top right, and the same is true for the bottom drawer fronts, so there’s a bit of continuity when you look at them left-to-right. (In the end, this probably doesn’t matter at all because they’re not going to be placed terribly close to one another and no one would notice, so don’t think of this as a sort of profound design choice.)

The handles are “halo” handles from Lee Valley.

A closeup of the right-hand cabinet and drawers:

I need to make a note about what’s going on with the woods here. In the side panel, there is a roughly 6″ wide slice of flatsawn cherry stock in the center to give it the central cathedral figure; each side panel has this. On each side of that, there are narrower strips of birch meant to provide contrast as stripes. Then, the outermost strips from the center of the panels are again cherry.

This didn’t quite work out the way I’d planned. I wanted to birch to be slightly brighter and contrast just a little more so that you could see what was going on. It was supposed to be subtle, but not this subtle. However, the grain selection worked, so I’ll take that one.

And then there’s the drawer side fiasco. I said this before, but I’ll say it again: Use something reasonable for drawer sides, not yellow birch. The sides did turn out to look interesting enough, but these were such a nightmare to thickness, surface, and cut tails into that I really don’t recommend that anyone ever do it. Yellow-poplar is a great secondary wood and I should have used that.

But what’s done is done, and none of this is terribly important in the end. What is important is that they are in use now, and the final placement looks like this:

There’s something very important to note here–These pieces do not incorporate as graceful of a design as my first nightstand. They don’t have the elegant, long legs, they don’t have the simple square footprint, and they don’t have the fancy double-arc decoration. Furthermore, they took longer to build because there are more innards.

So is it a step back from a year and a half ago? It would be if form were everything. But these were designed primarily for function, with the form fitting into a context–they flank a somewhat cheap bed that has no particularly graceful features of its own. They are a tad fancier than the bed itself. Eventually, I will build a replacement for the bed, and at that point, I can make the bed do what the nightstands cannot do on their own.

(The function should speak for itself–I had strict instructions to incorporate a large top surface, plenty of deep drawers, and easily-accessible shelves.)

And there’s one more final step forward with these pieces. I kept them in the same room as the first nightstand while I waited for any remaining solvents to evaporate, so I was able to compare them. The craftsmanship is definitely a step up from back then.

As a postscript, I again need to thank Kirk Eppler, who helped out in a couple of ways when I was building these things.

Nightstands v2: First Mouldings

The second nightstand project is lumbering and stuttering forward, this time with the tops. I glued up the tops a few weeks ago, smoothed them off, and cut the left sides even and square. I’d been debating what to do about the edges for a while, but then I got the order that they ought not to be “so plain.” Emboldened by a somewhat recent episode of The Woodwright’s Shop with St. Roy hosting Bill Anderson, I thought that perhaps I’d be able to do a simple single ellipsoidal curve.

After fooling around a bit on some scrap, I got a profile that I liked, and was satisfied that I sort of had the idea. So I moved to one of the tops. I started by marking the profile into the sides, and then marking the lines for a rabbet to remove a good portion of the waste:

I should note here that I used a surprisingly large number of marking gauges for this–four (two for the rabbet, two for the edges). Because I wanted to use the same profile on six sides, and all six sides were not ready for marking from the very start, I kept the gauges on their settings so that I’d be ready to mark when able and not have to figure out everything again.

Then I started on the cross-grain side and cut the rabbet:

This plane doesn’t have a depth stop and I didn’t need one (as seen on the show). Then, also like the show, I used a chisel to knock off a bit more of the waste from the edges of the rabbet:

Later on, I switched to a narrow rabbet plane; for this small amount, it worked a bit faster than the chisel.

Finally, I hit it with a simple Taiwanese round and completed the shaping:

So there you have it, my first moulding. I followed this with the moulding on the opposite side, doing the same cross-grain cut. Then I did the one on the front. The cuts on that one went with the grain and was consequently much easier.

Nightstands v2: Making Drawer Bottoms

It’s been so long since I talked about the second nightstand project that I sometimes wonder if I’m making any progress. So I looked back at that last post and realized that since then, I’ve done the following:

  • Made the rest of the cabinet components.
  • Glued up the cabinets.
  • Made all of the drawer sides, fronts, and backs.
  • Resawed and milled the pieces for the drawer bottoms.
  • Milled half of the tops and roughed out the other half.

The drawer sides and backs were a pain because for some bizarre reason, I chose birch to be my secondary wood. Don’t do this. Use yellow-poplar, pine, or something that people who should know better would use. The stock I had a ridiculous number of grain reversals, leading to a lot of tearout when planing, so milling this stuff took forever. It also dulled my plane blades quickly, so I was constantly resharpening. But even after I milled it, the dovetails took longer than they should have because it seemed like I needed to sharpen my chisels after every couple of swipes.

So it took forever, but I finished, and that left the drawer bottoms. I haven’t really talked much about how I’ve been making drawer bottoms, though I did one particular post that kind of touched on panels in the tool cabinet. So I figure I could post something on that.

First, I mill the wood to thickness, preferably a little thicker than the grooves that they’ll fit in. I always need to glue them together.

Next, I saw the glued-up panel to rough depth–about a half-inch wider than it will eventually be. In the following photo, I ran out of wood in one board of western redcedar and had to use another very differently-colored board for the last little bit at the end. The rip panel saw that I use is on top.

Now I plane the top of the panel to remove excess glue and get a finished surface. I do this after trimming the depth because I often use the cutoff somewhere else, so I want that cutoff to be as thick as possible to start.

I used a Taiwanese plane for this because it was sharp and the blade is nice and wide.

With the surface planing done, I trim one of the sides square to the front with a plane. I used a Milllers Falls #11 (this is like a Stanley 5 1/4) because it’s easy to control (and its blade happened to be sharp).

Then I take the drawer front and mark off the width:

I mark this side square to the front as well, and after verifying that it is in fact square, I trim it off as well.

As I mentioned earlier, the panels are typically a little thicker than the grooves that will house them. I don’t measure them because I care only about the face side being reasonably flat. Now it’s time to get three of the edges to fit into the grooves.

To do this, I make a rabbet. I was thrilled to be able to use my new Taiwanese rabbet plane so soon:

I start with the sides, going across the grain. Every so often, I check the thickness by seeing if the rabbeted edge fits in the groove that I’ve plowed in the drawer side. When going across the grain, it definitely helps to take out a little bit of the end of the rabbet with a chisel so that you don’t blow out the grain.

I then make an identical rabbet on the other side, and then do it for the front edge, this time checking against the groove in the front drawer.

Now it’s time to trim the panel/bottom to final depth. I assemble the drawer front and sides without the back, and slip the panel in. (This is also a good time to verify that the panel isn’t too long.) I make sure that the panel goes all the way into the groove on the drawer front. Then, I mark off where I think I should cut the panel:

I make marks on both sides, draw a line between them, and then saw and/or plane to that line. Then I test-fit the panel into the 3/4 assembly again, and measure the depth with my double square:

With the square set, I drop this end into the groove on the rear of the drawer to see how close I am to the bottom:

You can see the gap; you want a gap so that the panel has room to expand in the groove. I usually go for somewhere around 1/16″-3/32″, but it’s never exact, just the kind of thing you know when you see it.

With the final depth trimmed with a plane, I rabbet the underside of the rear panel edge like I did earlier for the sides and front.

Then, the final test-fit comes; here’s the view inside showing the face surface:

and here’s the underside that no one sees, with the rabbets and the saw marks that I’m too lazy to clean up:

So there are the drawer bottoms. I have two done and I’ll do the other two when I get back from the ski trip I’m on now.

I’m also making some progress with the tops. Here are the pieces that will form the tops. I had to do a lot of sawing around defects to get to this point:


Goofs Illustrated: Repairing Grooves

I thought I would be posting about the glue-up on the current twin nightstand project at this point, but there was one remaining thing that was really bugging me.

It turns out that I’d carelessly torn out the sides of a couple of grooves when I was cutting them, and although it wasn’t severe, with the panels fit, the shadow there would stick out like a sore thumb whenever I saw it. Unfortunately, I did it in a prominent area, and at the exact same place on each of the nightstands. So I wouldn’t be able to hide from it, and since I plan to use these things for a long time, I’d have a constant reminder of when I messed that up. Something needed to be done.

I started by marking out the repair area with a cutting gauge. In the following photo, the tearout isn’t easy to spot at first (it’s on the near edge of the groove), but if you look closely, it will be apparent.

Note also that I’m taking all of these photos with a macro lens so that you can see everything a bit better. For reference, that’s a 1/4″ groove.

Then I clamped a flat piece of stock right up against the mark I’d just made and started wasting out the repair area. You have to do this really carefully–use a small chisel diagonally (or a gouge) to avoid tearing out your repair area when you’re starting.

The photo below shows the last part of the wasting step–registering the chisel back against the scrap and paring out the last little bit.

Your chisel has to be very sharp and the back must be flat for this to work properly. For most chisel operations, the back may not need to be flat, but this is an exception.

I also gratuitously used my new Veritas miniature router plane in the repair area. It probably wasn’t necessary.

Now at this point, I was lucky enough to have not made it look any worse than it did at the beginning, so I carefully milled a small piece of wood two-square and pared down the length until it was the length of the repair area. I also made sure that this piece would have the same grain direction as the wood in the original. If I’d been smarter, I would have also matched the vector of the medullary rays, but I wasn’t, so I didn’t.

I glued it in place and banged in a couple of pine wedges to “clamp” it:

After letting it dry overnight, I pulled out the wedges and removed most of the material on the repair that was proud of the surface and groove (again, carefully).

For the final touches, I used planes. This was the very first time I got to use the side rabbet plane that I got from Veritas when they first released it:

It was very handy for this, but as they say, it’s a plane you don’t want to have to use.

After doing the sides, I planed off the top and the repair was complete:

There’s a tiny little nick at the top of this one where I accidentally dented the edge of the repair area, but I can live with that. I’ll only notice it every now and then and it’s something that I should notice occasionally. The important part is how it looks with its panel in place, and that seems to be OK:

The moral of the story? Don’t ever have to do this. I had thought that I was careful enough when I made the grooves; I’d marked the grain with a mortise gauge and worked the area with a chisel beforehand. But I still tore it up with my router plane because I was just too ham-fisted. If I’d spent an extra 30 seconds total being just a bit more careful, this would have never happened and I would have saved the hour that it took to do these repairs.


Nightstands v2: Rear Panels and Alignment Guide

Since my last update, I smoothed and fitted the side panels to the main cabinets, which left just the back panels before I could glue up the cabinets. I’d resawed the western redcedar for these panels, but I hadn’t cut them to size to fit them to the grooves that I also did in the last week or so.

The top section (backing the shelf) was easy–I just sliced the panel to width, cut the rabbets, and test-fit. Then I started on the backing panel for the first drawer. That opening is 5.5″; add about 1/8″ on each side for the housing grooves, and you’ve got about 5.75″. Because the redcedar I resawed was from a 1×6, that’s perfect, right?

Well, no.

I always forget how crazy they go when they surface these softwoods. What I had was a total width of about 5.5″–a quarter-inch short. Oh, so annoying. (Note to lumberyards: I really, really want my softwood unsurfaced, like my hardwood.)

So I jointed an edge from the offcut from the previous panel, jointed an edge on the slightly-too-small 5.5″ piece, glued it up (sprung-joint style), and rather than getting fancy with clamping as in my previous episode, I just slammed it into the front vise on my bench:

That vise sometimes makes some good arguments for why it should stick around.

Then after the glue dried, the most annoying part came: I sawed off all but about 3/8″ of the smaller piece that I had just glued on so that the panel was the right size. It’s a really good-looking, almost seamless joint when planed down, but no one will ever see it because it’s on the back of the piece. That means that, in the future, I get to spin the piece around, point at a practically invisible line, and say, “this here was really annoying.” I’ll also be suspected of having gone off the deep end at some point.

So right now, I have these rear panels finished on one of the cabinets. To help align the cross-members of the piece, I made myself a little guide out of scraps:

And here it is in use on the rear frame of the cabinet:

The main goal of making this guide was for glue-up, because you have a lot less time to go measuring stuff during that process. And hopefully by the next time I post, I should have the cabinets glued up.

Nightstands v2: Panels, Decorations

Work on the new nightstand projects has been excruciatingly slow this month, but has not stopped. In part, I’ve needed to do a lot of stock preparation and a lot of resawing (that’s the bandsaw calling me with its siren song again). The other problem is a lack of time–external stress causing most of it. But who cares about that? Let’s get into the woodworking.

I’ve been primarily working on the panels for the sides and backs of the cabinets. The sides are similar to the ones I’ve used before–1/4″ thick wood from the same stock (or similar) as the frame. There’s a slight difference, though: I decided that because I didn’t have any single piece of cherry wide enough to cover an entire side, I’d make a decorative touch with another species (birch) when gluing them up to make the piece.

The birch I picked was particularly annoying to work with, mainly because in many places, the grain reverses halfway through the width. Oh yeah, and it adores tearout.

This time, I used the double-wedge method to secure the panel stock against dogs (my low-profile versions) as I was planing it:

The panel jointing/glue-ups gave me a chance to use my new Veritas bevel-up jointer with the fence. I’d already used it in this piece back with the shelves, but this operation was tricker because the stock is so thin (it tends to bend when you press it).

I cut pieces out of the stock as I needed them, first jointing one edge with a slight hollow, then sawing off the desired stock, then jointing again. This was not as simple as it sounds, because when you saw off a small strip, the strip often slightly changes in geometry, because it’s under tension. So the hollow that you had before on that one edge may now be a bit convex, and you have to redo that edge.

But the most annoying thing, by far and away, was that I couldn’t use the jointer fence on the now-thin strip as I did before, because the fence is too deep and I don’t have a workbench trick for that yet (though I might in the future). So I had to use the old “clamp the plane upside-down in the vise and pull the strip through” trick.

It works, but I’m not terribly fond of it. I’m always afraid of planing a knuckle, and you tend to get sweat on the plane sole, leading to rust if you’re not careful. Still, it was the only option I had at the time.

With that done, it was time for laying out the panel components and choosing which pieces would go on which panel. Some people use stroke marks or triangles to mark which pieces go where, but because I had 20 pieces, I decided to go with a more detailed system that told me which part went where, and which panel each piece ultimately belonged to.

Now the fun part began: glue-up. There’s been a bit written about gluing up panels lately, and wouldn’t you know that The SchwarzThe Chris would post something just as I was doing mine. However, that method is for thicker stock. Plus, I was itching to try a variation on something I saw on page 20 of Toshio Odate’s shoji book.

Essentially, that method calls for you to sandwich the panel that you’re gluing between two boards, wrap a rope around the panel/boards, and put some blocks of wood underneath the rope to tension it and smash the panels between the boards to keep everything flat.

I modified it a little, making it sort of a hybrid with go-bars:

  • I used my workbench top instead of the board.
  • I didn’t use a board on top, but rather, just put the blocks directly onto the panel.
  • I used two sets of rope instead of one (one for each side of the panel)
  • Instead of using the rope to tension both on the top and the sides, I used it on the top only. For the sides, I used wedges, dog holes, and a stop on my workbench.
  • I put wax paper between the panel and the workbench to keep the panel from getting stuck to the benchtop.

It’s much easier to show this in a photo:

It took me a while to finalize the setup. I actually used a single piece of rope (twine, really) that I clamped to the bench in strategic places, and didn’t even cut it from its spool. I didn’t have scrap blocks handy, so I had to search all around the place for one of my boxes of scrap. And since I’d never done it before, I didn’t have a feeling for what the tension should be like and how everything fit together.

The good news is that once I figured it out, the actual glue-up process was a snap and took only a couple of minutes to execute. That’s important, because I had to do eight repetitions. (There are four panels, and each panel has five pieces, and hence four joints. I glued only two joints at a time to reduce complexity.)

Additional good news: it worked like a charm. Those panels came out really flat and seamless on the faces, and the wax paper did a perfect job at preventing the panel from sticking to the bench (and the paper). You have to replace the wax paper now and then because the wax coating comes off because it’s stuck to the glue.

I’m really happy with the results of the method and I’ll be trying it again. I’d struggled doing this sort of glue-up earlier because the panels are so thin. Being able to put the interim work into a spot where it won’t shift around is nice. I just hope that my experience with this time reduces the time it takes to set up next time (although part of that time was spent looking through boxes to find the rope).

I also acquired and resawed the stock for the backs of the nightstands. I picked up a bunch of “vertical grain” (quartersawn) western redcedar. It was so easy to resaw and plane–what a relief.

I guess there’s also this obligatory shot of this board’s grain (sorry to you folks on Google+, you’ve already seen these shots):

The other thing I’ve been doing on this project has been the decorations. I changed the design to be a little closer to the one I used on the first nightstand, to give it a little less transparency and eliminate the need for plinth on the bottoms of the sides:

With all of these components made, I’m nearly ready to glue up the main cabinets.

Nightstands v2: Frames Complete, Shelves Joined

It’s been a little while since I posted anything on the twin-nightstand project, so it’s time for a quick checkpoint (one of the reasons I write this blog is so that I can record when I’ve finished various phases of a project).

Both frames, exterior and interior, are complete. The interior frames are to support the shelf and drawers in each piece. In the first nightstand project, I shaped some of the exterior pieces to provide drawer supports in strange ways, and I said to myself that this method was too complicated to be worth the effort next time. So in the new one, I’ve been making the supports as separate pieces, in a secondary wood (yellow-poplar salvaged from an old bed frame).

I was finished with the exterior frame several weeks ago, and I finished with the interior frame about two weeks ago. This week, I’ve been working on the shelves that go above the midsection of each piece. Each shelf is made from two roughly 1/2″ panels edge-glued; I did the glue-ups yesterday and today.

So now the pile of components looks like this:

You can clearly see the secondary yellow-poplar members here. The components that I have yet to make are:

  • panels for the sides and back
  • drawers
  • drawer bottoms
  • tops
  • decorations (these will line the base of the pieces)

In theory, none of this should take too long, but it’s still a fair amount of work to do. There will be a non-trivial amount of resawing for the panels. I can’t really put a time estimate on how long this is going to take, because I’ve encountered a lot of diversions when making this project (though I’m starting to feel like I really need to finish this and move on to the next thing).

There is a new tool in the preceding photo that I may feature sometime in the future, depending on how just I like it. You could guess which one it is, though the excitement factor may not be entirely present in doing so.


Nightstands v2: Exterior Frame A

The new twin-nightstand project is coming along. After some quality time with saws and planes, I had all of the external frame pieces milled for each piece:

In the center are eight sticks of yellow-poplar that I salvaged from an old bed frame. They’ll be used for the rear of the nightstands. Looking at the timestamps of the photos, it looks like it took (gasp) almost two weeks to produce this stack. That’s just the speed you’ll experience when you’re working by hand, have at most an hour and a half to work each day, and maybe aren’t the quickest person around. (Would a bandsaw greatly facilitate the speed? You bet it would.)

The next step was to come up with a frame map, as I did for the previous nightstand. Because there are so many components in this piece, it’s best to keep close track of them. And again, as I did in the previous version, I made a really ugly but still-workable and accurate drawing:

There are two kinds of labels here. One begins with a P, meaning “piece,” so P1 is the far left leg in this case. The others, which are simply numbers, label joints. The joints are in four vertical levels in the piece, so I decided to have the lower-level joints start with 1, the next with 2, and so on. Then, the second number denotes the position. In this drawing, the x1 joints are in off the right side of the front left leg. So joint 11 is the lowest one of these, and 21, 31, and 41 are directly above. Naturally, in this system, there will be plenty of holes–for example, there are no joints 25 or 35.

Finally, because there are to be two identical nightstands, I decided to add a letter A or B to the end to identify each one. So you’ll have all sorts of labels like P3A and 43B.

Whew. But that wasn’t the hard part. That honor went to planning which piece would go where:

If I recall correctly, the components for piece A are on the right and piece B’s are on the left. In any case, it was tricky. Because the components came from different offsets into a single flatsawn board, the cut profiles were different. The ones from the center were more or less quartersawn in flavor, but most were riftsawn in one way or the other. The worst part was that two of the legs had very tangential grain on their sides because they came from the center of the board. To produce a consistent look, I decided to put both of those legs on one side, and use matching rails between them. The reason? You can’t look at both sides at once, and in the end, I can always choose to hide that side against the side of the bed if it looks too ridiculous. In any case, that will be on nightstand A, and I’ll try to show that when all of this is done.

I wasn’t just trying to match the grain on each side, though. I decided to try out the arrangement of shoji described in Toshio Odate’s book, where you arrange the pieces so that the core of the tree is pointing inward (bark side out). Because this is a 3-D piece, I also tried to place the legs and rails so that the cores would face the interior of the nightstand. It didn’t look so bad to me, so maybe this traditional arrangement has a bit of merit.

With everything arranged, I labeled each piece and its orientation, put half of the pieces away, and started work on the actual joinery. One week later, and I’ve finished the 24 mortise-and-tenon joints of the exterior frame of nightstand A:

Hey, how about that, it doesn’t seem to look too bad and it took a lot less time than I thought it would. Now it’s B’s turn.

Or, it will be B’s turn when the Thagomizer recovers from its run-in with a holdfast. I broke it in a different spot this time, a clean shear across the top:

Of course, it had to wait until I was working on the next-to-last mortise. I finished that one and the final one with it still broken (it hadn’t split down the middle). Then I removed the head, gunked on and spread the hide glue, and clamped it together. Yeah, I’ve thought about making a new mallet one of these days, but I just don’t have the time right now to be dorking around with tools (sigh). Maybe when this project is near an end and in the finishing process.

Hand Resawing Notes

When you’re resawing by hand, there is a big consideration in determining how difficult it’s going to be: Under how much stress is the wood? There are two big factors that I’ve noticed:

  • The cut of the wood
  • Moisture content

Let’s start with the moisture content, because that’s perhaps a little easier to understand. Green wood is usually under a lot less internal stress than dry stuff because that’s more or less its natural state before being cut down. As R. Bruce Hoadley’s book explained, you can cut slices off of a fresh log and it won’t crack apart until it starts to lose moisture. Sure, there might be pitch, resin, sap, or whatever else inside to make it more difficult to cut, but it does mean that the wood will be less likely to close up on you.

Because I rarely get the chance to work with green wood, the cut of dry wood is what causes me the most grief. Dry wood is always under stress of some sort; it’s unavoidable because you’ve taken away the water that was maintaining equilibrium when it was (probably) a stable, living organism. What you want is another state of equilibrium where the lignin is strong enough to hold everything together without cracking. Quartersawn wood is great because it’s under the least stress–because it’s got a consistent density pattern across the cut, it dries evenly and doesn’t tend to cup. Just a little more shrinkage at one end is pretty easy to handle.

Well, that’s nice, except that quartersawn wood is more expensive and it’s a lot harder to find than flatsawn stuff (not to mention that the figure may not be what you’re looking for). So this is the sort of thing I usually end up with:

This is a cross-section the 8/4 stock for the new nightstands that I wrote about earlier. The wood on the bottom that’s closest to the pith (center) is under a lot of stress. You can see a minor check here–this was due to that knot and its proximity to the end of the board.

You can see that this board was fed through a planer a little after drying because the cross-section across the cut is flat. But below is what happens when you resaw it:

That pencil line (and the sawcut) was straight when I started, but as I got further into the cut, the stress along that cross-section was relieved slightly, and this cupping was the new state of equilibrium.

The problem is that this stress relief is a major pain in the tuchus when you’re resawing dry flatsawn stock by hand. Basically, some part of the wood inevitably closes up around your saw as you are sawing, and you find your saw incredibly hard to push or pull through the cut, especially if you’re using a deep-bladed affair such as a Disston handsaw. It’s less severe with a frame saw, but it’s still there.

To work around it, break out your wedges and keep tapping them in as you’re working:

They’ll keep the kerf from closing in around you and it will be easier to saw. You really want to put them in as soon as you can, because the easier you saw, the straighter your cut will be. If your blade is pinched by the wood, it is being bowed out of a straight line, and your cut will wander.

Wedges are easy, except for one minor detail. You mostly want to be working like a bandsaw, perpendicular to the grain. However, when working with a fairly thin blade such as this, you need to be able to establish a shallow kerf in both sides so that your blade can track inside. To get these small kerfs, you need to tilt the saw at a diagonal, and this is precisely where the wedges and/or stress-relieved wood will get in your way. If you pop out a wedge or if you’re not able to drive them in deep enough, the wood will close up around your blade.

So what you often end up doing is tilting the saw just a little and establishing a small kerf until you can get deeper.

A bandsaw arguably does this better, because it always cuts perpendicular to the grain because you’ve got a fence, and the blade will always be cutting next to some (hopefully) straight wood. But that’s not to say that the wood isn’t under stress and isn’t closing around the wood–it is. That’s why you need a fairly beefy bandsaw to resaw wide flatsawn stock. You can cheat a little with wedges by turning the machine off in mid-cut, banging some wedges in, and resuming.

As far as the current project goes, I’ve chosen to cut the thinner slice from the more stressed parts of the wood, the “pith” side. The result is that the thicker slice, having a more consistent density, mostly retains its straight profile, but the thin slice (shown above) cups like crazy. My plan is to make those thin slices into panels, where I can use their more interesting figure to my advantage, and not care about their increased instability, because they will be flexible and slide in grooves.

That said, here is what became of that 8/4 piece of cherry in an intermediate state (the chunk on the right is what remained at the end and I probably won’t use it):

It’s hard to believe how large that board used to seem, and now it’s so much more compact. The crazy thing is though it’s not much different in volume (I was actually efficient at slicing this for a change), it seems like it’s a reasonable weight now, whereas before, it was really unwieldy. Of course, there will be a significant change in volume when I whittle those panels down to desired thickness.

Eh, whatever. What it boils down to is that I’ve got my new project frames right there! There are a few more notes I have for how I chose to cut this, but I’ll save them for a later post.

Nightstands v2: Wood Acquisition

Normally I wouldn’t write about the mundane process of purchasing and transporting some wood for a project, but after some thought, I noticed that I really haven’t talked much about how I’ve gone about that process. Then, I realized that I actually hadn’t done any halfway serious wood shopping in a few years. I’d gotten the beech for the stool and first nightstand projects through a group purchase off of craigslist, and a fellow BAG dropped it off at my house (thanks again, Kirk!), and the rest of it was either really cheap home center softwood or some stuff I’d gotten long ago in this initial purchase (wow, is that ever a blast from the past–I was still living in my Inner Sunset apartment then!).

I’d also gotten another piece of beech somewhere between those two, which marked the first time that I’d used my roof rack for that purpose. We’ll get back to that rack in a second–it goes along with this post. I’ve seen a few posts from other folks regarding wood transport (including some crazy stuff like bicycles), but not many, which probably indicates that a lot of people have trucks. And for those of you wondering, yes, I may do all of my work by hand, but when I need to move some wood around, you can believe that I’m not going to do it by wheelbarrow or drag it on the sidewalk.

Let’s go back and start with the wood selection. My design was for cherry and called for 1.25″ thick legs and 1″ frame pieces, meaning that I’d some 6/4 or bigger stock . Then each nightstand will have drawers, panels, a top, and a shelf, which calls for the usual 4/4 stuff. I had the amount of board-feet and linear feet necessary for 6″ wide stock computed before I went.

But when you go to the lumberyard, you really don’t know what they’ll have. That especially holds true for cherry, where the stock on hand has all sorts of variations in in color, grain straightness, and especially defects. So you have to have a basic plan at what stock you’ll need for your cutting list, but you need to be able come up with a plan B.

And plan B it was. I really didn’t like any of the 6/4 cherry boards that I saw today. However, the 8/4 stock was only 16 cents more expensive per board-foot, and far better-looking. So even though it will mean a bit of exhausting resawing, I bagged a decent-sized one of those for the legs and the frames, and a 4/4 piece for panels and drawer fronts. I may get another piece for the top later, if necessary.

I also got a piece of 4/4 birch for my secondary wood. I don’t know if that was a great idea or not, but I like birch, and it doesn’t cost much.

Now we get to transporting the wood, and it’s where we get back to the story about the rack. Here’s the load of three boards back at my place (those are ACF and Penn State decals in the window, in case you’re wondering what sort of pointless stuff that is):

The rack is a Yakima system that I’d gotten off ebay when I had my CRX. Now, if you know anything about these rack systems, you can switch them from one kind of car to another, as long as you’re willing to pay the king’s ransom that Yakima wants for the little clips that will fit your car (essentially, very expensive powdercoated pieces of pressed steel). As a special surprise, I also had to get new crossbars, because the new Civic’s roof is much wider than the old one. Thankfully, I was able to dodge Yakima’s cash vacuum on that one by just getting some 3/4″ galvanized pipe from the hardware store.

Now here’s how the choice of car gets a little tricky. I like my 2-door, mostly to annoy any passengers that may end up in the back seat, but this is one instance where having the 4-door is better. You see, on the 2-door, you place the bars only 18″ apart (unless you pay another arm and a leg for Yakima’s extender kit), but on the 4-door, you put them 32″ apart. A longer spread helps keep the load from rotation forces and keeps boards from springing around so much. To avoid that altogether, you can tie the load down in the front and the back. Because this was a short trip (and wood has a much smaller aerodynamic profile than, say, a canoe), I didn’t need to do that.

So, yeah, had I thought about this when deciding between the coupe and sedan, it would have been a consideration. However, the 4-door had a big problem that I don’t think I could have overcome: You can’t get it in the color you see above. [Edit: Also, new for the 2012 Civic 4-door, you can’t get the EX trim with a manual transmission. This is far worse than the wrong color. But you can still get it with the EX coupe.]

[Note: The Schwarz also has a what is essentially a Civic, and but he’s basically gone the infill route by getting the old Acura version. I have to admit that I miss the hatch of the CRX.]

Okay, back to the load. Let’s see how I fasten the boards to the roof. Here is the driver’s side, which has the 8/4 cherry:

I put pieces of foam pipe insulator around the bars, then put the wood on top, and fastened things together tightly with a ratchet tie-down–which is essentially a band clamp. The ratcheting ones let you get the grip good and tight, compressing and locking the board into the foam. This not only takes a lot of slop out of the fastening, but also adds some friction to the surfaces; everything is always in contact with something else, even if you hit a bump in the road (and California roads have plenty of those). The boards were shorter than the length of the car, so I did not need to put a red flag on the end to indicate an overhang (as you would need to do if you let your boards hang out the back of a pickup or van).

These tie-down clamps have really long cords, so you have to do something with the excess. On the driver side, I’ve wrapped them around the board several times, then tied the loose end to the bar. It doesn’t need to be a strong knot because the ratchet mechanism is doing all of the hard work.

On the passenger side, I put the two thinner boards together, but this time, I got lazy and put the loose end of the straps into the car. I wouldn’t do this on a long trip because it makes too much noise:

Once I was finished fastening everything, I gave them a bunch of pretty hard shoves to make sure that they wouldn’t budge.

As far as the capacity of the rack goes, I’m pretty comfortable with these two boards on top of each other, but I don’t know about three. In any case, there isn’t room for much more. The weight limit on the rack and the roof itself is 125 pounds, which translates to around 35 board-feet of this density of wood. There were 23 board-feet on there today. This is just fine for a small-time woodworker like me; it’s just very unlikely that I’d ever need to buy much more wood than this at once, given what I currently do.

But what if I did need to haul more? Would I get a truck? A significantly cheaper option would be to get a trailer. My car’s towing capacity is 1000 pounds.