A CURVED DRAWER

I want my furniture to be fun. Useful and elegant is nice too, but I love lots of delightful surprises — discovering a secret drawer or some sparkling detail. A tall cabinet I have been working on has many drawers, all of them fun, some secret, some not so secret, three swing out and are a combination lock for a door, and one is curved. 

The curved drawer arcs to the left as you pull it out. It’s surprising and unusual. The shape is amusing too — a drawer with sides that appear elastic enough to bend into a curve. Otherwise my curved drawer doesn’t look that complicated— but it turned out to be. 

I usually start building a drawer with the pocket made, guides in place, and carefully fit the face first, and then build the drawer to it. Because of the way this cabinet is constructed the interior around the drawer is not accessible later, so I had to build from the inside out. My drawer also had some tight tolerances, fitting within 1/16” of a tambour door that slides by it. An accurate paper pattern seemed the best way to get the tolerances right, and to see how the curve of the drawer felt. I also used it to set the left guide; the right one I could fit after the drawer was built. 

For a drawer to work smoothly it has to be a consistent width — parallel sides —so it glides in and out of its pocket without any looseness or snugness. The guides that control it must be parallel too, and in this case the same curve as the drawer sides. I knew that by laminating the sides I could make them thin, strong, and smoothly curved. I cut a pine laminating form to an arc of 20” radius and laminated the two sides on it. With 4 thin laminates, for sides just under 1/4” thick, I got very little springback.   

The problem is that these sides were only parallel when right next to one another, and as I unhappily discovered not at all when I built them into a drawer. For my next drawer only 2” wide, parallel sides would have to be arcs of 20” radius and 18” radius. That would require two pretty exact bending forms, and even then I wasn’t sure they would yield truly parallel sides. I worked on other parts of my cabinet while I thought about my next move. 

The solution turned out to be easy and accurate. I made a single form to the smaller radius and laminated the “inner” side on it. And then I laminated the “outer” side on top of the first, only spaced with blocks the exact width of the inside of the drawer. Since my drawer was so small I actually did both laminations at the same time. These sides are perfectly parallel, even if they are not exact arcs of a circle (due to springback).  Before unclamping my laminations I marked the sides with a line perpendicular to the form, so I could keep them in the same orientation within the drawer. 

A simple solution to bending two parallel drawer sides — separate them with blocks equal to the inside width of the drawer.

A simple solution to bending two parallel drawer sides — separate them with blocks equal to the inside width of the drawer.

My sides were wider than I needed. Some of that extra width became the guides. The rest was cut away jointing the top and bottom edges. I use either of two methods (or both) to hold curved parts to joint them. One way is to hold an end in a wooden jawed clamp fastened to my bench. Another is to jam the far end into a “stop” or V slot in a piece of scrap clamped firmly. I occasionally checked the side on my jointer table to see if it stood square.  

For my first curved drawer, the one with the same radius sides,  I cut half blind dovetails to join the face and sides. It wasn’t easy, but if you lay out the parts together (on that paper pattern) and get the angles accurate for the ends of the sides and the face, those are the surfaces that guide cutting the joint. For my second much smaller drawer I cut a locking rabbet joint by hand. The form was a useful place to support the sides while laying out and cutting the joinery. 

The curved bottom and curved groove were further challenges. I’ve used a router with a three wing cutter (cutting a groove horizontally) housed in a rounded “fence” to cut accurate grooves in curved parts. I’ve also made a simple scratchstock cutter to scrape a groove. Then there was the issue of how to slide the bottom in on a curve.

My solution was again simple. In Japan I saw many drawers with a thin paulownia bottom merely glued (and sometimes pegged) to the underside of the sides and face. The drawer rides on the bottom, and since paulownia is very stable, it somehow works.  Since I am using as many native woods as possible for this cabinet, I chose curly maple for the bottom, reasoning it was hard, fine grained, and would slide sweetly. Since I can ignore seasonal movement in such a small bottom, for simplicity and strength I ran the grain down the length of my drawer. 

I was going to add a fancy face to my drawer, in this case birdseye maple with some inlaid dots, so I hung it down to cover the thickness of the bottom. 

The last step was fitting the outside guide. I slid the drawer in, laid a piece of paper as a shim along the outside, moved the guide into position, and screwed it down lightly. After trying my drawer a few times, and a light tapping here and there on the guide to position it perfectly, I secured it with a few glue blocks on the outside.  A pine block at the rear is the stop for the drawer.

 

What’s next? A secret drawer hiding in the back of a larger drawer could be fun. 

 

 

 
 

Urban Wood Conference, California

Last week I was in California speaking at a conference of the Western Chapter International Society of Arboriculture, including urban foresters, botanists, wood workers, and lovers of wood. I was speaking about my experiences harvesting wood on our farm and how that wood ends up in the furniture I design and build. Following the conference I led 20 people in building small tables, all from local urban wood.

We learned about the nitty gritty of how trees grow, and the different cell structures of hard and soft woods. Finally I understand the scientific basis for why quarter sawn lumber is more stable than flat sawn and moves half as much — because of the ray cells. They run radially from the heart of the tree to the outside and act like steel reinforcing. It's the exposed ray cells that shimmer on quatersawn faces . 

Naturally figured woods were talked about, and the question came up about how birdseye maple comes about. In my experience the birdseyes can be dense in one part of a perfect log, and then dissipate away. No one yet understands why they form.  One theory is the tree gets damaged and lots of dormant buds form but never sprout. But birdseye logs rarely show such damage, or any bark imperfections at all! 

This is the first half of the talk I gave.

-----------------------------------------------------

Nearly every aspect of my life connects to trees and wood. Primarily I am a furniture designer and maker and have been for over 40 years. I’m also a teacher, teaching all over the world about hand tools and techniques. For almost 25 years I have been a contributing editor at Fine Woodworking magazine writing about design and technique, along with 2 books about indispensable hand tools.

We live on a small farm in Vermont with about 13 acres of productive forest. From it has come much of the wood to build our house, and nearly all the wood for 2 barns, and my brick shop (it has a heavy wooden structure and wood roof, floors and ceilings). Our forest supplies us with the fuelwood that keeps us warm and cooks some of our food, fenceposts, habitat for wild turkeys, bobcats, and bear, and maybe most importantly tranquility during my almost daily walk among the trees. I once imagined I knew every tree in our forest. 

Vermont is about 95% forested. Our farm is surrounded by deep forest.

Vermont is about 95% forested. Our farm is surrounded by deep forest.

For 37 years I have been harvesting, milling, and drying wood for my furniture. While not all of what I use comes directly from our forest, a great deal does, and much of the rest is local. We grow over 20 species of good hard and soft woods — oak, ashes, maples (some of the best hard or sugar maple in the world I am told), cherry, birches, aspen, butternut, beech, basswood, white pine.  

In an average year I harvest 2 - 3000 board feet of logs during the winter months. Some years it has been closer to 5,000, which is a lot of wood — a good start on a modest sized barn. I fell trees in the winter when the woods are more open and accessible.  Snow cushions the fall of good logs (that can suffer damaging shock cracks if dropped too hard) and makes moving them easier and cleaner.  Clean logs make clean lumber, and keeps my sawyer Don Lawrence happy. He arrives every spring with his portable Woodmiser sawmill. He can pull right up alongsidea log, roll it on, and saw it with a blade so thin hardly any wood is wasted. We’ve sawn 36” diameter butt logs that yielded board after board 21+ inches wide, but more often logs 14 - 24” in diameter. We saw a lot of 5/4 thick hardwood for furniture. It’s a useful dimension to mill to 1”, or resaw into 7/16” panels.  We cut 6/4 for legs or heavier structural parts, and 8/4 for still heavier parts or for resewing into thinner parts. A variety of thicknesses to choose from is efficient and cuts down on waste. Building materials are anything from 1” boards to 2X4, 3X7, 6X6. We always saw for quality, for clear straight grained wood over wide flat sawn boards. And when it is worth it we turn and turn the log to get maximum quarter sawn boards, for their alluring fleck pattern, linear grain, and better stability. 

Sawing a butternut log about 24" diameter

Sawing a butternut log about 24" diameter

Making your own lumber is not quick, nor is it easy. Green wet lumber is heavy. Each board needs to be carefully stacked and separated with 1”X1” stickers, in a place with lots of airflow. Spring is the ideal time for this with the air cool and dry, so there is less chance of molds that can start growing almost overnight.  I air dry in stacks as long as I can, but always at least a year to the inch of thickness. After that lumber moves into the top of my shop or one of the barns and can be stacked without stickers. Typically I dry to somewhere around 11% moisture content outside, and to about 7 or 8% in my shop. 

So let’s deal with the question of air dried vs kiln dried. Is one better? Kiln drying might slightly improve a board’s stability and does hasten the time from sawing to using it. It will set the sap on softwoods like white pine that can bleed resins. Still, I much prefer air dried wood. Putting a plane to it I can immediately feel the difference. The fibers are more elastic and flexible, so my shavings are fluffy and continuous. End grain cuts cleanly, not into tiny bits but real shavings. Even shavings and sawdust from my machines are bigger, less dusty, so healthier for me.  Air dried wood steam bends with far more flexibility than kiln dried, and thin laminates flex better for cold bending — a process I use to produce curved elements for my furniture. Air dried lumber has richer color. Steaming walnut for instance, leaches out the subtle reds, purples, and browns (darkening the light sapwood). 

Chopped end grain cherry, air dried on the right and kiln dried on the left.

Chopped end grain cherry, air dried on the right and kiln dried on the left.

And lastly air dried wood is stable. What exactly do I mean by “stable”? A stable part doesn't warp or change shape unexpectedly. Far more important than the inherent stability of the species — and they are all different — is using my experience to choose the best wood species for what I am making, and the best grain for my parts. Even grain, quatersawn or rift sawn, no knots or wild grain — this is what I look for. The dried boards tell you a lot, by how flat they are.   

Careful selection of straight quatersawn door and case parts, that are  attractive and highly stable.  

Careful selection of straight quatersawn door and case parts, that are  attractive and highly stable.  

Some species have issues. These are the ones I avoid sawing and using because they can warp unpredictably. Beech is one that I love but don’t trust. Aspen is another, where some trees are exceptional and the rest yield boards that warp into shapes you might use to plank a boat. Warp is the result of uneven shrinkage or stress, which all goes back to the tree and the way it grew. Each board mirrors that tree, from curving grain lines due to a gentle bend of the trunk, to knots, uneven growth, or some other anomaly. 

Wood will likely warp in some way as it dries from green (wet) to equilibrium to where it lives (8%?), be it a dry climate or a wet one. Wood never stops absorbing or giving up moisture, and changing shape. Some woods have cell structures that give them high stability, such as mahogany or walnut. It’s no mystery why so much of the highest style antique furniture made here and in England was made from mahogany, and to a lesser extent walnut, only some from cherry, and little out of curly maple, a less stable species. Of course mahogany and walnut are handsome, but they are also exceptionally stable, and in mahogany’s case, once came in boards 2 - 5’ wide. 

PART 2 further explains the advantages of sawing your own lumber, and woods that I especially love and why.