Today I carved the feet for both end pieces of the boot bench. After a lot of flatwork recently it has been really liberating to exchange the tyranny of the straight line for some flowing curves.
The feet are formed by cutting a cyma reversa detail into the bottom of each end piece. Since I prepared the full sized plans over Christmas I’ve been experimenting with a number of curves for this detail – changing the transition point between the concave and convex portions, as well as both fixed radii curves and complex curves using french curves. In the end the detail that worked best to my eye was a pair of fixed radii curves, with the transition point two fifths of the distance from the top of the detail. Placing the transition point closer to the top of the foot ensures that there is plenty of material below the bottom dado to support the shelf, and provides an elegant curve that directs the eye upwards. The foot starts 2″ from each edge of the workpiece, and finishes 3″ above the bottom edge at the centre of the workpiece. Marking on those dimensions effectively gave me two right-angle triangles back to back (one for each half of the foot detail) terminating at the lower edge of the bottom dado.
Normally I make ply templates for curved work so that the curves are easily repeatable. However, for this detail I wanted to put into practice the pre-industrial geometry that Jim Toplin and George Walker covered in By Hand and Eye, not to mention George’s excellent article in Issue 2 of Mortise & Tenon, so I laid out both feet using just my Starrett compass and the principles expounded by Walker and Toplin. Marking out repeatable curves like this is actually very straight forward once you get the hand of some simple principles. First off find the centre point for each of the two curves. To do this set your compass so that it spans the distance between the end point and the transition on the hypotenuse of the triangle marked on the workpiece, then sweep the compass in an arc from each point. Where the two arcs cross is the centre point of the circle. Lay out that curve, then repeat for the other half of the cyma reversa. Because the centre point of the convex curve fell off the end of the workpiece I butted both end pieces together to provide a continuous working surface. I’ve left the layout lines on the inside of the carcase to tell a story for any future generations who look inside. Despite the geometric terminology, this method is really intuitive and requires no numbers whatsoever (which is my kind of maths). And the Romans used it to lay out viaducts, which is pretty nifty.
With the curves marked clearly on the inside face of both end pieces, I removed the waste using a large Knew Concepts coping saw and a fresh skip-tooth blade, making sure to cut on the line. Even in 1″ thick oak the Pegus blades cut smoothly and swiftly. After using plenty of bad coping saws, the Knew Concepts saw is a revelation, particularly when loaded with Pegus blades. The coping saw removed a large whale-tail piece of waste, and left a surface that was smoother than you’d expect.
Cleaning up the curves was where the fun really started. The first priority when rasping or filing work is to elevate it so that it is close to eye level without needing to stoop. My Moxon vise is perfect for this, and the end pieces were rock solid and at a much more comfortable height. Cleaning up the saw marks was done with a 9 grain (10″) Auriou cabinet maker’s rasp, which quickly removed all blade marks and left a reasonable surface. The curves were then faired up and refined using a 13 grain (7″) Auriou modeller’s rasp which leaves an incredible surface. It is important to work from the show surface towards the back edge of the work, to avoid blowing out grain on the face of the work. For curves that will only be viewed from one side, I slightly undercut the curves from the rear, which means that I can focus on fairing the show side. Once the curves were flowing and fair I sent a photo to the client, to gauge his response. He requested a slightly deeper and dramatic curve on the convex portion of the detail, so I spent some time deepening that curve using the same rasps.
At this point the feet are probably 95% done. I’m going to leave the end pieces out so that I can live with the curves for a few days and let them seem into my subconscious. Then I will return to them mid-week and see if there is any more subtle refinement to be done. Once I’m completely happy with the feet it will be time to dovetail the carcase together.