Started 12/1/2017, completed 6/29/2018
Finish: Minwax gloss wipe on polyurethane, cut 1:1 with mineral spirits, one coat wipe on wipe off.
Some notes regarding what I was trying to achieve with this chair build, and things I might choose to do differently in future builds.
First, the chair as a whole: still very much a Hal Taylor design, with the exception of the shape of the headrest and the extension of the rear legs above the headrest joint (the “horns). The latter are a characteristic though not invariable feature of a Sam Maloof rocking chair, although Sam built his chair with the headrest grain running horizontally, while I continue to build using Hal’s coopered vertical grain headrest. This not only is a better design from the perspective of wood joinery between leg and headrest, but for a strongly striped wood like this african mahogany it achieves a much better harmony with the wood figure throughout the chair. See below.
The “view from the top” shows the typical paddle-shaped arms with the cove cut forming a hollow, a nice wide surface for resting the forearms, as well as the figure and symmetry of the seat and arms.
And here’s a head on view. Note that the curves for the headrest top and bottom continue into the rear legs. More on the headrest and the horns further down the blog post.
Symmetric back braces
One of the new things I tried with this build is mirror image symmetry of back braces. Not my original idea, but I’ve forgotten who on the Hal Taylor chair group discussed this previously. The left/right symmetry is achieved by flipping some of the lams over. The symmetry is then close to perfect. When I say “close to”, I mean that the figure changes a little from lam to lam in the flitch, but using a thin kerf (ca 0.60″) circular saw blade in the table saw to cut the lams minimizes the figure change from one lam to the next. The photo shows 8 consecutive laminations cut from a board, the flitch order from left-to-right 1-3-5-7-8-6-4-2, with lams 2,4,6, and 8 flipped over. Lams 7 and 8 will be glued edge to edge and the center 1-3/8″ cut out to form the central back brace face lam.
I could have changed the design to make a back rest using an even number of braces (no central brace), which would avoid the edge gluing, but that would require making a new back brace hole template, which I didn’t want to attempt for now.
The following photos shows the fixture I made to facilitate edge gluing the thin lams. The wedges keep the joint tight, and the long bow clamp keeps things flat while drying. Wax paper (actually Cheerios bags I save) keeps glue squeeze out from sticking to clamps or fixture.
Next is a photo of the back of the chair, showing not only the symmetry of the seat and legs, but also the rear side of the back braces.
More on the shaping of the headrest and horns.
The back of the headrest is contoured to a convex “belly” shape, which is a Maloof thing. This presents opportunities and challenges in the transition to the horns.
My first idea was to create a concave hollow in the horn, and I roughed out the hollows using a bent gouge (a #7 or #8 I think). I just couldn’t resolve all the divergent edges that presented to me. I believe Maloof did put this type of hollow in his horns, but his rear leg was much narrower back to front at the headrest joint, and I think that the hollow works out better in that scenario. So I used my trusty rasp to remove the hollow and transition to a more or less flat surface. Hopefully the following three photos show this.
Next is a purposely somewhat unflattering view of the horn from the side, emphasizing a few things I’m not satisfied with:
1. should the profile of the horn from the side be squared off, as I ended up doing, or curved in some way. I still can’t decide, but this photo makes it look less than beautiful I think. An unresolved problem.
2. I made a measurement error laying out the three holes for the screws bringing the headrest/rear leg joint together, and as a result the plugs are not even. Measure twice, drill once, right? Not a big deal but once it’s pointed out it’s an obvious slightly odd thing.
3. The “adder piece” for the rear leg that makes the leg/arm transition more graceful still gives me a lot of trouble. For one thing, depending on the figure of the wood, the adder piece can look a lot different from the wood it’s glued to, even though it comes from the opposite side of the leg billet. In this case the adder has strong vertical figure but the outside of the leg doesn’t, which is less than ideal. Also there’s significant color variation. It actually doesn’t look nearly as bad as it appears in the photo. I know that the mahogany darkens with age, so the differences will gradually fade.
There’s also the issue of difficulty getting a really thin glue joint when attaching the adder piece to the leg: if it isn’t perfect, there will be a wide bit of glue visible when feathering out the adder piece, as often happens in my hands.
All of these difficulties have me wondering whether the problem the adder solves, which is achieving a smooth flow between the leg and arm, is at the expense of creating several new problems. Is it worth the trouble? I don’t know what Hal thinks but it’s been bugging me for several chairs now and I’m trying to come up with another design that doesn’t require the adder.
New front leg build
Here’s a view of the front leg including the leg-to-seat transition. This is the “bandy legged” style front leg (there’s also a straight version as viewed from the front). This is the first time I changed the build of the front leg to move the glue seam from the middle of the leg to the joinery area. I like this because the seam in the middle of the leg sometimes leads to figure or grain mismatch. It also wastes less wood, as instead of using two full-length billets glued together, you just glue a small block where the joinery is cut. However, in designing this new method, I didn’t make this adder block long enough top-to-bottom, which forced me to make the radius of the leg-to-seat transition smaller than intended. It’s OK, but I’ll use a 5″ long adder in the future, not a 3″ adder.
The next photo shows an intermediate step in shaping the front legs. The temporary blocks at the end are to create a stable platform when cutting the joinery. Why four front legs? I was building two chairs simultaneously, the legs for the small chair described in this blog on the left. The adder blocks should have been 5″ long, not 3″ as shown, to get a gentler transition from the horizontal seat to the vertical leg.
Next is, more or less, the same view of the other front leg, but showing a little more of the front leg-to-arm transition. Some folks are grinding off much more of the bottom of the arms to remove almost all of the (flat) side edges. I do remove a lot of material but haven’t quite gotten to the point of thinning the arms to an edge. Maybe next chair?
A view showing that there are still side edges on the arms, though much wood has been removed through freehand shaping with angle grinder.
The final photo is a top view of the right arm and seat, for no particular reason, except perhaps to remind myself that I could grind off more of the inside edge of the rear leg/arm transition to straighten out the transition – there’s a bit of a convex curve there:
I hope you enjoyed this tour through the build of my latest Hal Taylor design rocking chair (with a hat tip to Sam Maloof). Please feel free to comment (constructive criticism only please).