Miller Dowels Mini-X When things go wrong in the shop, one of the crutches I lean on are Miller Dowels Mini-X thingies. These are stepped and ribbed dowels that you install with a super-insanely-good stepped drill bit. How to use them: First drill a hole with the bit to reinforce a questionable joint. Then tap in a Mini-X dowel with some glue.
They fit so well that the world seems like a brighter place.
I’ve long used Miller Dowels for a variety of off-label applications, especially knockdown joinery. But as I delved deeper into chairs, I have found that the Mini-X dowels are the perfect repair tool. They can go almost anywhere. The fit is perfect. And they can be easily flushed to look like nothing ever happened.
If you are Miller-curious, buy the Mini-X Dowel Joinery Kit for $35. It includes the drill bit (the most important part) and 100 birch dowels. Put the kit somewhere safe but visible. The next time something goes circling around the toilet bowl, remember this kit. It might save the day.
“The Stick Chair Book” is divided into three sections. The first section, “Thinking About Chairs,” introduces you to the world of common stick chairs, plus the tools and wood to build them.
The second section – “Chairmaking Techniques” – covers every process involved in making a chair, from cutting stout legs, to making curved arms with straight wood, to carving the seat. Plus, you’ll get a taste for the wide variety of shapes you can use. The chapter on seats shows you how to lay out 14 different seat shapes. The chapter on legs has 16 common forms that can be made with only a couple handplanes. Add those to the 11 different arm shapes, six arm-joinery options, 14 shapes for hands, seven stretcher shapes and 11 combs, and you could make stick chairs your entire life without ever making the same one twice.
The final section offers detailed plans for five stick chairs, from a basic Irish armchair to a dramatic Scottish comb-back. These five chair designs are a great jumping-off point for making stick chairs of your own design.
The arms can be the simplest part of a chair. If you’re lucky, you might find a branch in the woods that grew into the shape of a perfect arm. Or the arms can be as basic as two straight boards: one for the right hand and one for the left.
If you like, you can make a C-shaped arm that wraps around the sitter by gluing three sticks together – one for the sitter’s right hand, one for the spine and one for the left hand. On the other hand, a chair’s arms can have insanely involved joinery – mitered scarf joints or curved half-laps (for starters).
And if that’s not enough of a challenge, try steambending, where there’s a significant risk of chuck-it-in-the-trash-and-start-day-drinking failures.
With dozens of methods available, deciding how to make the arms of a chair can be daunting. So, let’s begin with some basic principles.
The Goal of the Arm The mechanical goal of every arm on every good stick chair is simple: Avoid short grain as much as possible. If you plan to build your chair with two separate, disconnected arm pieces, then things are fairly simple. You can easily find two sticks to do the job and avoid weak short grain. The troubles begin when you want your chair to have what’s called an “armbow” – a curved arm that wraps around the sitter from her right hand to the left. How in heaven’s name do you avoid short grain with a C-shaped arm? There are several strategies:
Find a curved branch that looks like a 90° bend. Saw it through its thickness (called “resawing”) to make two identical curves. Then join the two 90° curves to make an arm that curves 180°. Or get really lucky and find a curved branch that is perfectly C-shaped.
Take a straight stick and use steam to bend it over a form to make a 180° curve.
Saw up a bunch of thin (1/8″-thick) pieces of veneer. Apply glue to their faces like spreading butter on bread. Bend them over a curved “form” that represents the arm’s final shape. Let the glue dry. This is called “bent lamination.”
Purchase “cold-bend hardwood,” which is flexible when wet. You bend it over a form (similar to steambending but without the steam). When it dries, it keeps its shape.
Create a “pieced armbow.” This is where you use three or four chunks of wood that are sawn to a curved shape. You glue them up in a way that eliminates short grain, sometimes adding a piece called a “shoe” to the top to shore things up.
Typical Dimensions Arms can vary quite a bit. A steambent arm might be 1″ thick and 1-1/2″ wide. A pieced armbow might be 1″ thick and 2-1/2″ wide. A curved branch or root can be a whopping 2″ thick and 4″ wide.
The arm has to be strong enough that it won’t crack during assembly or in service. And this challenge is made more difficult by all the holes you drill in the arms for sticks. Make the arm too bulky, however, and it might look ugly. It’s a balancing act.
In the world of stick chairs, a typical arm is about 1″ thick, give or take. In a strong material, such as oak, I’ll accept 7/8″ thick. For width, I like 1-3/8″ wide for arms that I’ve bent. And about 2-1/4″ wide for pieced armbows.
If the arm has a shoe, I usually shoot for 1″ thick for that component, though I have seen much thinner ones on historical chairs.
When assembled, the armbow is typically wider than the seat. If my seat is 20″ wide, then my armbow will be 23″ to 26″ wide overall. The depth of the armbow varies according to the design of the chair. If the back of the chair leans a lot, you might have to make the armbow deeper (or not, depending on where you want the hands to end up). Sometimes the hands hang over the front edge of the seat. Sometimes they are in line with the front edge of the seat. Sometimes they are a few inches back. Here’s a good starting point: My armbows are typically about 16″ deep, and that works for most of my chair designs.
All that said, the arms can vary a lot in a stick chair. Don’t be afraid to stray from these guidelines when copying an old chair.
Arms in the Hedge Among stick chairs from Wales, it’s fairly common to find a chair where the arm’s shape was determined in part by the tree. A tree branch grew in a graceful curve, and it was harvested by a cunning chairmaker. I first learned about this bit of cleverness from chairmaker Chris Williams and Emyr Davies, a conservator at St Fagans. They planted the following idea into my brain: “Chairmaking begins with a walk in the woods.”
That is, you can find a chair’s arms in the branches, and the chair’s design begins there.
During my visits to the forests in Wales, these simple words became real. I looked up into the branches of these craggy Welsh trees and saw the arms of chairs waving back at me. Curved branches are quite common in trees that are part of the intertwined ecosystem of hedgerows and sunken lanes.
When I returned to the United States, I went to the forest to look for arms, but above me I found only legs, sticks and stretchers – straight stuff. The North American forest tends to produce arrow-straight tree trunks as the leaves stretch upward for sunlight.
Of course, naturally bent wood is out there in American forests and towns, but it’s not nearly as common as it is in Wales, where the landscape nurtures these curves.
If you do find curved material for arms, harvesting it, sawing it, drying it and shaping can be a challenge.
Naturally bent wood can possess significant internal stresses. The reward, however, is an armbow with no short grain.
There are two typical ways to use the curve of a branch in an arm.
With a branch that possesses a shallow curve, use the curve as-is, like in a root-back Welsh chair. These arms act more like a backrest, really. Sometimes they have a shoe (aka doubler) that is carved from the solid arm. Sometimes a shoe is applied.
With a branch that bends 90°, saw it through its thickness and join the two pieces into an armbow. The joint can be a scarf or a half-lap.
While I have looked for arms during many walks in the woods, most of my success has come from “walks by the stream.” Trees that grow adjacent to a stream can have roots that bend from the bank then plunge down. Sometimes erosion can expose these bent roots. They are ideal for arms. (Thanks to chairmaker John Porritt for showing this trick to me.)
Steambent Arms Steambending is challenging, time-consuming and there’s always the risk of failure. Despite this, I have loved it since I bent my first comb in 2003. You need a steambox, a way to make steam (I use a wallpaper steamer), a bending strap and a form. The biggest challenge, however, is getting the right wood. The grain has to be dead-straight along its length, or it is likely to split while being bent. Air-dried or green wood bends the easiest – it still has lots of moisture in it, which helps carry the heat into the stick. If the wood has been kiln-dried, it needs to be rehydrated before bending. Cut the stick to shape then soak it in water for a week or two.
But even when you do everything right, sometimes steambending goes wrong.
After steaming the stick for an hour or so, you bend the stick around the form, secure it with clamps and let the stick dry. You can let it air-dry for a couple weeks, or you can build a primitive kiln using some insulating board, duct tape and a light bulb. You want the bulb to heat the kiln to 115°-125° (F). After a few days in the kiln, the arm will be dry enough to keep its shape.
People have written entire books about steambending. The chapter on the comb-back with a bent arm goes into detail on this technique.
Bent Laminations I’m not a fan of using bent laminations in stick chairs. Laminations usually look wrong to my eye. Basically, making a bent lamination involves sawing multiple thin strips of wood from a board in sequence. You apply glue to their faces, bend the wet mess over a form and let the glue dry. Then you machine the glue-encrusted part to shape.
I am happy to use bent laminations when making contemporary pieces, but in a vernacular stick chair, I’m going to opt for something else because it can look a bit like fancy plywood.
Cold-bend Hardwood Surprisingly, the easiest way to bend an armbow or comb is using a high-tech material called “cold-bend hardwood” or “comp wood” (“comp” is short for compression). This material has been heated with steam and compressed along its length. When it arrives in your shop, it is wrapped in plastic and is pretty wet (about 25 percent moisture). It also is flexible. You cut it with a band saw and bend it around a form. It’s like steambending without the steam, strap or failure. I’ve had only one failure in 10 years of working with it.
What’s the downside? It’s expensive. A stick of comp wood that will get me three armbows might cost $150. When I sell a chair for $1,400, a $50 armbow isn’t all that big a deal. If fact, it might be cheaper than steambending because there is almost zero risk of failure when bending an arm. However, if you are a hobbyist, your time is your own and you can make these decisions without worrying about the clock.
After you bend the comp wood, you clamp it to the form. Then you can let it air-dry for a week or put it in the kiln overnight. When its moisture drops below 15 percent or so you can take it out of the form. I have found it quickly acclimates to your shop’s equilibrium moisture content.
The comp wood is indistinguishable from wood that has been steambent, so it looks fine in a stick chair. And I go into detail on using comp wood in the chapter on the comb-back with a bent arm.
Pieced Armbows My favorite way to make an armbow is the easiest method overall.
A pieced armbow is made from two to four bits of wood that are sawn and glued to avoid short grain. A pieced armbow allows you to use flat boards from the lumberyard (or sticks from your backyard) and, with a bit of cleverness in selecting the grain, create a sturdy armbow.
The joinery can be as simple as butt joints and glue, or as showy as mitered lap joints or long scarf joints.
Making a pieced armbow begins with choosing the shape of the arm, choosing the joinery, then making patterns for the parts.
It will come as no surprise to those who know me that I got distracted by other work…and that I forgot it takes me five times longer (at least) to do things with a camera pointed at me than when no one is watching. And that is why the video I promised a few weeks back on kitting out the interior of an “Anarchist’s Tool Chest” is taking a bit longer than expected. But with Wally’s help, we’re nearing the finish line. (A special thanks to its soon-to-be owner for bearing with the delays.)
The video will feature: • installing the till runners • a brief look at dovetailing the three tills (goodness knows if you’ve built this chest, you know how to cut dovetails!) • fitting the till bottoms to the runners • installing a moulding plane till at the back • making and installing a hanging hole-y rack with slots behind it for backsaws • making and installing a saw till for larger handsaws • notes on finishing the interior • tips on fitting the lid • hardware installation, including hinges, chest lifts, ring pulls and a “crab lock” • caster installation • surprisingly few cats (unless we add then in post production).
Chris plans to start editing at the end of next week, and he’s fast – so it should be available soon thereafter. Below are a few pictures I snapped during the process.
The following is excerpted from “The Anarchist’s Design Book,” by Christopher Schwarz, an exploration of furniture forms that have persisted outside of the high styles that dominate every museum exhibit, scholarly text and woodworking magazine of the last 200 years.
There are historic furniture forms out there that have been around for almost 1,000 years that don’t get written about much. They are simple to make. They have clean lines. And they can be shockingly modern.
This book explores 18 of these forms – a bed, dining tables, chairs, chests, desks, shelving, stools – and offers a deep exploration into the two construction techniques (staked and boarded) used to make these pieces that have been forgotten, neglected or rejected.
Here’s a simple stool design that represents a lot of false starts, research and prototyping. It’s not perfect, but it is a nice, stout stool. The stool is based on 18th-century low stools from American homes. This stool is also an excellent introduction into building the seat and undercarriage of a full-blown chair.
This stool has a pine seat and hardwood legs – ash in this case. The seat is 13″ in diameter and the legs hold the seat about 16″ off the ground – a good height for a low stool. The H-stretcher is a bit of overkill. But I think you should include it. It will teach you how to add stretchers to any of the chairs in this book – or from other people’s books. So let’s go.
Make the Seat The seat is a softwood that is about 1-1/2″ to 1-5/8″ thick. You can glue up the seat from two bits of wood (that’s what I did) and put the seam in the dead center of the seat. Keep the leg joints away from this seam; you don’t want the legs levering the seat apart. (Yes, a long-grain-to-long-grain joint is stronger than the wood itself in a perfect world. But that is not where we live.)
With the seat blank glued up, use a compass or trammel points to lay out the 13″-diameter seat. Cut the seat to shape. Then cut a 1/2″ x 1/2″ bevel on its underside. This bevel lightens the look (and the weight) of the stool. You can do this on the band saw or do it with a block plane or spokeshave. Now you can mark out the location of the joints. Here’s the easy way. On the underside of the seat, draw a line through the centerpoint of the circle. Make this line parallel or perpendicular to the glue seam in the seat (if you have a seam).
Place a protractor on your pencil line and mark the seat at 45° on both the left and right sides of 90°. Connect the marks with the centerpoint and you will have a perfect “X” on the underside. Now take a ruler and mark out the location of the four leg mortises 1-1/2″ in from the edge of the seat’s bevel. (It’s all shown in the photo [below] if you look closely.)
Make the Legs The legs are 1-1/2″ x 1-1/2″ x 18″ and are made from a dead-straight hardwood such as oak, maple, ash or hickory. Knock off the edges until the legs are octagons. Then taper the legs so they taper to 1″ at the top.
For this project, I decided to use cylindrical tenons. While I prefer tapered tenons, cylindrical tenons are far more common in the historical record and are easier to make, especially if you own a lathe.
The tenons for this stool are 1″ in diameter and 2-1/4″ long. Turning them on the lathe is straightforward. If you want your tenons to be dead-on accurate, I recommend you purchase a bedan tool with a sizing attachment. A bedan tool is basically a wide parting tool with its sides relieved (like a traditional mortise chisel) to allow it to maneuver in the cut without binding. The sizing tool is an attachment that clamps to the tool and allows you to set the diameter of the cut.
To use the bedan tool and sizing attachment, first drill a test mortise and gauge the exact diameter of the bit that will drill your mortises. Set a dial caliper to that measurement (lock it) and use the caliper to set the bedan tool and its sizing attachment.
Bore the Mortises The leg mortises are bored at an 21° resultant angle. The sightline is 0° and runs directly into the centerpoint of the seat. Set a sliding bevel to 21° and tape it to the sightline. Clamp a backing board below the seat to reduce (but probably not eliminate) any splintering.
With this stool, I’m using a 24 mm bit from WoodOwl that is supposed to leave a clean exit hole without splintering. It does a pretty good job, though no bit is perfect (hence my backing board). These particular bits work best in an electric drill.
Drill the four mortises. Then put the legs into their mortises and have a gander at how accurate you were.
Rotate the legs in their mortises and orient them so their attractive surfaces face out. Then meaningfully mark the legs and the seat so you can get the legs back into this ideal arrangement.
Now it’s time to bore the mortises for the side stretchers. These are positioned about 4″ to 4-1/2″ up from the floor. Here’s how to mark them out. First, level the stool like you are preparing to cut the legs to length. Shim the feet until the seat is level all around. Then cut a 4×4 block of wood to 4-1/2″ long and place it on the bench. Fetch the half-pencil (it’s a pencil planed to half its thickness). Mark the location of the mortises for the side stretchers on the legs.
To bore the mortises, flip the chair upside down so the seat is on the benchtop. Place a couple sticks between the seat and benchtop to let the tenons poke through the seat. Then take an awl and mark the centerpoint of each mortise on each leg. I do this by eye. Measuring always seems to make it worse.
Chuck a 5/8″ Forstner bit into a cordless drill. I drill the 7/8″-deep mortises in the legs entirely freehand, using the seat and the marks on the legs as a guide. Rotate the leg in its mortise so you can get the drill and the bit in position in line with the leg. The drill and bit should be aligned with the mortise on the opposite leg. The photo [“On the tip of disaster”] shows how this works.
If you lack confidence because this is your first rodeo, have a spotter give you some directions. They should be able to tell you if your drill bit is in line with both mortise locations in the legs. Drill the blind mortises, stopping before the bit explodes out the backside.
Make the Stretchers The stretchers are 1″ x 1″ material that has been planed octagonal. After preparing the overlong stock for the stretchers, you need to determine how long they should be for your stool. To do this, fetch two skinny scraps. Pinch them together and press the ends into the bottoms of the mortises. Make a pencil mark across the two scraps. Remove them from the mortises. Reassemble them with the marks aligned. Measure the overall length, and that’s the finished length of the stretcher.
Do this for both stretchers. Mine were slightly different lengths. If you are cutting the tenons on the lathe, then add 2″ to the calculated length to give you some room to work without running your tools into the headstock and tailstock of the machine. (I wrote this sentence to remind myself to do this next time.)
Cut the 5/8″-diameter tenons on the side stretchers using the same techniques outlined for the legs. Yup to the bedan tool and the sizing attachment. After turning the tenons, saw the stretchers to their final length and install them in their mortises.
The medial stretcher is easy. Mark the centerpoint on each side stretcher. Use the same 5/8″ Forstner bit to drill a 5/8″-deep mortise in each side stretcher. Once again, I drill these freehand. Keep the bit 90° to the stretcher and parallel to one of the facets of the octagonal stretcher.
You know what to do next. Get the skinny scraps and use them to determine the finished length of the medial stretcher. Cut the stretcher 2″ overlong. Turn the tenons on the ends with the bedan tool. Cut the medial stretcher to finished size and fit everything. If the stool doesn’t explode, you are ready to glue it up.
Assembly Before you disassemble the dry-fit stool, mark where the wedges should go in the legs’ tenons. I use a Sharpie for this to avoid confusion. Disassemble all the parts and mark them up so you can assemble them in the same orientation with glue in the equation.
Kerf the legs to receive wedges. Use a band saw or a handsaw for this. You want the kerf to be of significant thickness. Make some 1″-wide wedges for the legs.
Right before assembly, clean up all the tool marks left on the legs, stretchers and seat with planes and spokeshaves. This is quick work with sharp tools.
Here is the sequence for assembly. Learn this and you’ll be ready for a full-on chair in your future. Glue the medial stretcher to the side stretchers. Twist the parts until the assembly sits flat.
Put glue in the mortises in the legs. Wipe off any excess and put the stretchers’ tenons into the legs in the mortises. This will be an ungainly thing, like a baby goat. Rotate the legs until the assembly is stable. Set it on the bench.
Paint the interior of the mortises in the seat with glue. Do not skimp or get in a hurry. Take a deep breath.
Navigate the legs into their mortises. This might require some grabbing and bending. That’s OK as long as the seat doesn’t split. The goal is to get the tip of each tenon into its mortise.
Tap the legs down, working around the stool’s four legs until the legs are seated. Small taps are better than big ones.
Flip the assembled stool over. Paint the wedges with glue and drive them in with a hammer.
Let the glue dry overnight. The next morning, saw the tenons flush to the seat. There are (at least) 50 ways to level your tenons. When your seat is flat and not saddled, the fastest way is with a Japanese ryoba saw. We took a hardware-store saw and stoned the sides of its teeth with a diamond plate to remove the set of the teeth. It now barely scratches the seat in use.
After sawing off the tenons, plane the seat to remove any toolmarks.
Level the Feet As shown in other sections of this book, there are lots of ways to level the feet. Picking one method depends on how your head works. Here’s how I did it for this stool. I first leveled the seat using wedges underneath the four legs.
Then I determined the final seat height (16″) and made a block of wood to guide a pencil. The height of the block represented the amount of leg I needed to saw off to achieve the final seat height. In this case, the block was about 1″ high. I placed this block on the benchtop and used a half-pencil. Then I sawed off the legs to their finished lengths. I then chamfered the feet to prevent the feet from splintering out when the stool is dragged across a floor.
And then you are done with construction. Finishing these stools can be as simple as a coat of linseed oil and wax. Or you can dive into milk paint, soap finishes or the Wild World of Wiping Varnishes. Do your best work – you don’t want to be accused of polishing a turd. (And you thought you’d get away without a single stool joke.)
I have a small flower garden in front of my house, and after years of being annoyed by having to sit down on my porch to don an old pair of running shoes for weeding (and checking them for spiders first, because they lived on the porch), I finally broke down and bought myself some garden clogs. They’re easy to slip on and off as I walk in and out the door, which is great, but I don’t want to leave them on the porch (they are much more likely to take a walk than my nasty old running shoes). But then I got annoyed by the amount of dirt I was tracking into the house, and bought myself a galvanized steel boot tray. But…just inside my front door is an HVAC return, so the boot tray had to go across the hall, and well…that was a good excuse to build a piece of furniture* to hold it – something that could fit just inside the door to make it easier to stow my clogs, and provide enough space underneath to allow for air flow.
Before all I built was tool chests, my M.O. was to find a Shaker piece that I like and modify it to suit my space, needs or both. So I went back to my old ways and spent a few days leafing through the various Shaker books in the Covington Mechanical Library.
I had almost settled on a washstand when Will Myers dug up a photograph of a piece illustrated in “The Encyclopedia of Shaker Furniture” by Timothy D. Rieman and Jean M. Burks (Schiffer, 2003), and in volume 2 of Ejner Handberg’s “Shop Drawings of Shaker Furniture and Woodenware” (Berkshire Traveller, 1975) – a “Bake Room Table” that was in the North Family Dwelling House at Mount Lebanon. I liked the drawings, but didn’t fall in love with the form until I saw the (unique) table in a photo.
The shelf that cuts across the side cutouts was, according to Handberg, probably added later. And what are those round cutouts at the back? Maybe it fit around a pipe of some kind?
The original, at 66″ long and just more than 28″ high, looks lighter and more graceful than what I made, but I didn’t have the space to copy its size. And I chose to make the cutout at the feet a bit shorter so that I could fit the shelf (my entire reason for building the piece!) above the void. I also skipped the support at the front between the drawers; the overhang on my top wasn’t enough to require it. Still, it was the starting point I needed – that, and the size of the boot tray it was destined to hold.
I also changed the construction, with a 3/4″-thick solid dust board (or perhaps it should be called a drawer support, given there are no drawers below it?), dadoed below the drawers instead of the nailed-on 2-1/4″-wide rails front and back, and nailed-on runners (if Handberg is correct). That last decision was a tactical error; the wider board provides more protection against racking…so I ended up pocket screwing (go ahead – come at me; there are period pocket screws in plenty of Shaker and other period work) a rail behind the drawers and under the back of the shelf (likely overkill, as I am wont to do).
And instead of classic Shaker wooden pulls, I used iron ring pulls, to match the iron nails that attach the top. As a result of its almost-square form, size and metal hardware, I think my result skews a bit Arts & Crafts.
It’s a simple build for a simple customer. Were I building this for a more discerning end user (i.e. one that is paying me), I’d probably use sliding dovetails to attach the dust board/drawer support (and possibly use a web frame instead of solid wood there) as well as the drawer divider, and inset the back rails in grooves. And my drawer dovetails would be better.
The Build I started by gluing up the side panels and drawer support. All the wood that shows is cherry, but I glued up the drawer support from a 5″-wide or so piece of cherry on the front, and poplar behind. (I guess it does show at the back to the cats, but they don’t care).
Then I marked out the dado locations, and played around with the curved cutouts at the bottom – and I ended up with curves that are slightly higher than what I drew. I cut them out on the band saw, then cleaned up the cuts with a combination of the spindle sander and sandpaper (#120 and #180) wrapped over a piece of the cutout (a sanding “fid”) to clean up the rough spindle-sander scratches.
Then, I sawed the walls of the dados (which are a 1/4″ deep), knocked out most of the waste with a chisel and cleaned up the dado bottoms with a router plane (there’s a video here of this process, should you care to watch).
With one side arranged dados-up on the bench, smear a bit of glue in the dados, then put the shelf and support in place. Brush glue in the mating dados then put the other side in place on top. If the shelf and support are held tightly in well-fit dados and not moving around once they’re in their housings, it makes glue-up a lot easier to handle by yourself. If your fit is too loose, knock a wedge or two in on the underside to push out any gap and tighten things up. After the glue is dry, you can use a chisel so cut off the protruding end of a wedge (or, if it’s below the shelf at the bottom where it won’t show, just leave it).
Then, lift the assembly to get clamps across it front and back at the dado location and check it for square … then foolishly move the now-quite-heavy assembly off your bench and onto the floor by yourself because it will look better in a picture that way. Be sure to then complain that your back hurts.
After I took the clamps off, I realized I was being boneheaded; the piece needed some wider horizontal members to keep it from racking. So, I pocket-screwed 3″-wide rails between the sides at the back above the drawer support (poplar) and below the shelf (cherry). Then, the right side developed a slight cup at the top front – and I was afraid it would get worse. So, I pocket-screwed a 3/4″-thick x 3″-wide piece of cherry about 1/2″ back under the front of the drawer support. Uh … it’ll create a nice shadow line.
Tangent: I should have known things would go at least a little bit wrong. Some of the cherry I used for this piece is cursed. I’d bought the 5/4 stuff in mid-2017, with plans to build a Stickley 808 server for a Popular Woodworking article. I was seduced by the wood’s curly grain and remarkably low price … even though I knew curly cherry could be a right royal pain in the butt at the best of times, and that the low price indicated it was already misbehaving. But … so pretty!
I commissioned reproduction hardware from John Switzer at Black Bear Forge, and stickered the wood to acclimate for a month or so in the PW shop. Then I surfaced it to 7/8″ and glued up my panels. It all looked fabulous. For about a week. Then all the panels developed a gentle cup. OK – I could flatten it again, and build the piece out of 3/4″ instead of 7/8″. It happened again. So I put the panels aside and decided to build the server out of white oak, just like the originals. I bought the oak … then I was no longer employed at PW.
Those cherry panels left PW with me, and they’ve been in the basement at Lost Art Press ever since. (John’s gorgeous hardware has been in my basement ever since.) Chris recently used one of the panels as a desktop across trestles. And after flattening it and putting heavy stuff atop it, that panel has remained flat; I thought the curse had been broken. Chris says he’s protected from it there is no joinery involved in his setup.
The curse was not broken – but the addition of a front rail seems to have at least overcome it. For now. But never forget that wood hates you.
After the clamps were off, I marked centerlines on the drawer support and 5″-wide drawer divider, then clamped the divider firmly in place, and countersunk screws to hold it in place from underneath. It is simply butted tight to the underside of the top.
I had a beautiful wide piece of cherry from C.R. Muterspaw for the top that I sure wish I could have used at full length. Not only was cutting it painful, but I think a longer overhang to either side added lightness – but rendered it not fit for purpose.
I cut a small roundover on each end of the top’s back rail, sanded it, then glued it in place to the back of the top. I could have gotten away with leaving it off from a functional POV, but I wanted the extra overhang that offered for the front edge.
I then dithered over best to attach the top, and after considering the use of traditional buttons (which would require 1″ shorter drawer sides and backs to accommodate their attachment), had decided on figure-8 fasteners (which would require drawers sides and backs only 1/8′ or so narrower than the fronts), when Chris talked me into using blacksmith-made nails. I liked the way they looked on the cupboard in “American Peasant,” so…
The only great-looking nails we had (they were made by Mark Kelly, a blacksmith at Mt. Vernon) were 2-3/4″ long; I really should have used 2″ nails. It was a bit scary to drill so deeply into the cherry sides. But I got away with it… or so I thought for about 14 hours. I did a test drilling setup for tapered pilot holes, and after successful tests drilled my pilots and hammered the nails in place. They looked great. Whew! By the next morning, a small split developed at one location. But you can’t really see it without bending down and looking closely. And I’m not showing you.
The customer is dismayed but accepting of the flaws. (I blame the curse.)
Further adding to the flaws count: It turns out I didn’t get the divider perfectly centered; the left opening is 1/16″ smaller than the right. So I fit each drawer front and back to its opening with a No. 51 shooting plane. For those who don’t already know, a tightly fit drawer is key to smooth movement. These have maybe a 1/32″ reveal side to side. (And wood movement won’t be an issue, as that will be top to bottom, where I left about 1/16″.)
Another tangent: Years ago, I erred on the side of too loose, and those drawers bug the bejeezus out of me to this day. It’s this piece – the drawers, which have just under a 1/8″ reveal on both sides, rack every time I slide them in. Yes, I have some thin UHMW tape…no, I have not yet applied it even though I know it would likely solve the problem.
The drawers are half-blind dovetails at the front and through dovetails at the back (the tailboards are on the side, so that the drawers can’t be pulled apart in use). The only advice I have on cutting drawer dovetails is to run the grooves for the drawer bottoms before transferring the tails to the pin boards. That way, you can stick a shim of the right size in the grooves to help align the pieces (a trick I’m pretty sure I learned from a Chris Becksvoort article in Fine Woodworking).
The drawer back is 3/4″ narrower than the front, to allow the bottoms to slide in underneath it. The bottoms are 1/2″-thick paint-grade plywood, rabbetted to fit the 1/4″ wide x 1/4″ deep grooves. I cut a slot in the center back, then nailed it to the underside of the back to keep it from moving. (A more discerning client would get solid wood, sized to allow for expansion and contraction – but that’s not an issue with the plywood.)
I cut half-blinds so rarely that I can’t remember if I prefer to secure the board vertically in a twin-screw vise or flat on the bench to make the cuts. These were cut in the vise – but I think my overcuts end up longer – thereby making the waste in the corners slightly easier to remove – when I clamp it flat to the bench. (Yep, I know lots of folks pooh-pooh overcutting. “Whatev,” as the kids haven’t said for some time, now. There are plenty of period drawers that employ overcuts; I’m in good company.)
And to finish things off, instead of leaving blue-tape pulls in place for years, I decided on traditional iron ring pulls (32 mm). Simply drill a hole where desired to fit the staple legs, then bend each leg back with a pair of pliers, and hammer the staple legs in place. I decided to locate the pulls slightly above center. I taped off the location, marked the hole, then drilled it over a backer board so as to avoid blowout on the backside.
The finish is soft wax 2.0 – easy to make, safe to use and simple to re-apply if it proves necessary. You’ll find directions for making and using it in Chris’s latest book, “American Peasant” – a free PDF download (see pages 65-7).
Now that my new piece is at home and in use, I’m confident the Curse of the Curly Cherry is finally broken … as long as the drawers and tray stay in place, covering up that cursed wood. (But I remain leery of using the two panels still in our basement!)
And if I ever get to make this again, well, I’ve identified all the problems! If I call this one a prototype, maybe the mistakes won’t bother me as much. (They will. That is my curse.)
– Fitz
*aka a procrastination technique to put off building a pantry cabinet or the bookcase for the bottom of my staircase.