Re: Federal Curve
Wiley Horne--So. Calif.
>Hi John,
Let's see if we're on the same page. I am looking in the book "The Furniture Masterworks of John & Thomas Seymour", by Robt. Mussey, Peabody Essex Museum, 2003.
There are four different bow-fronted chests of drawers shown, all constructed within the period 1800-1810. Their overall dimensions are generally 34" to 38" high x 41" to 42" wide x 21" to 23" deep. They're all 4-drawer chests, and French-footed (see Bill Tindall's recent French-footed cabinet on these pages).
The bow front in each case appears to be in the form of a slow circle. Not elliptical, though the Seymours did use elliptical fronts for some other pieces, particularly commode chests. Use of the elliptical form requires a pretty deep curve to be effective. When I say 'slow circle', I mean that the sagitta of the curve looks to be on the order of 1-1/2" to 2", compared to a 41-42" width. Now that is my guess--I'm only looking at a front view and guesstimating.
If you decide to use a circle, and if the radius is too long to draw directly, you can still proceed to construct the curve mathematically. You need to know two things: the actual width W of the chest, and an assumed radius of the curve, R. With those two things, you can use plane geometry to construct the sagitta (the degree of bulge at the center of the chest), as well as the other points defining the arc of the circle.
EXAMPLE: Assume for a chest width of 38", that the radius is 7 feet, or 84". You can work out right away, using trig tables, that an 84" radius will sweep out an arc of a bit over 26 degrees from one side of your chest to the other, the half-angle being a bit over 13 degrees. This translates to a sagitta of approx. 2". If you don't like that degree of bulge, try longer or shorter trial values of radius R. When you find one you like, you can make repeated calculations with that radius to generate the whole curve.
John, I'll stop here, cause I don't know if we're talking the same talk. If we are, I can detail the calcs. Of course, the circle form itself is based just on the Seymours. You might want to make a wider survey, just to make sure.
Wiley