After the work on the legs and stretchers in the previous post, I moved on to working on the arms and back rest. The first step was to use the bandsaw to rough cut the shape out of the chunky glue-up from the previous post. From there I moved to hand tools to creep up on the finished profile. Leading up to this phase of the project I was continuously thinking about how I would do the carving and shaping of the arms and back rest. I have a range of carving gouges and a few small kanna for cutting curves, but what I really needed was a nankin-ganna (Japanese spokeshave). After a bit of hemming and hawing over whether I should just buy one or make my own, I decided to just bite the bullet and cut up a block of nice oak that I had been saving for another plane, and made a quick nankin-ganna.
Below is the finished product. I won’t go into details about making the nankin-ganna in this post, but I will say that the second I started using it I was really glad I took the time to make it. This tool makes shaving wood on curved surfaces ridiculously easy. It’s also much easier on your hands to hold for long periods of time, compared to mame-ganna (small planes) that are usually used for tasks like these. I pretty quickly found myself wishing I had a few more for different curvatures and profiles, but for this project managed to do all the shaping with this one nankin-ganna.
I spent time making sure the blade fit really well, but did a very quick and dirty job of shaping the actual handle itself. I figure as I use this tool more, I will get a better idea of what kind of shape makes most sense for comfortable holding.
On the bottom side I inserted a couple pieces of brass right in front of and behind the blade. These areas wear quickly on wood planes, particularly when planing round surfaces with narrow points of contact between the plane and workpiece. So I inset a couple pieces of brass to ensure a consistent surface in these crucial areas.
So with the nankin-ganna in hand I moved forward with shaping the arms, beginning with the bottom surface. Below you can see one arm roughly carved, and temporarily joined to the tenon on one of the front legs.
After a bit more planing with the nankin-ganna I arrived at this shape.
Back when I wrote about the design for this chair I ended by saying I had two design styles in mind for the arms. While I visually prefer the design below on the left, I was worried that the narrow round arms would be uncomfortable in use. So I created the arm profile in the right image as an alternative. My plan going into this chair was to start out by shaping the arms somewhat flat on top but round on the bottom, similar to the style in the second image. But I wanted to angle the arms/back like they are in the first image. Essentially I decided to merge elements of these two designs.
But after doing the rough shaping on the arms, when I took a step back I really wasn’t too happy with the design. I really found myself wanting to round out the top side of the arms too. So that’s exactly what I did. I started by taking a router and cutting a chamfer to speed up the process of rounding out the shape.
And then using the nankin-ganna I smoothed things out. I didn’t form a perfectly round shape, rather the top side maintained a slightly wide round profile. But instead of a sharp transition like the design in the rendering above, the whole profile is much more uniform. A lot of the chair designs I’ve been looking at lately by other furniture makers are full of bold lines and crisp edges, but my gut in this instance was telling me to go round. Aesthetic decisions like this are tricky. But the more furniture I build the more comfortable I’m becoming with following my gut. Overall I think the softness of this profile fits with the other elements of the chair better; the round legs, the soft round profiles of the woven seat, etc..
Here are the arms/back after rounding out the whole upper surface. You can also see how I started to carve the round profile onto the end of the arms as well, which I mainly did with a chisel.
In the next image you can see the arms and back assembly full carved. At this point I also sanded out the planing marks left from the nankin-ganna. I really wanted to find a way to leave the surface with a planed finish, but it was incredibly difficult to get a clean surface straight off the nankin-ganna. The grain direction of the wood changes dramatically at the joints between the back and arms and in those locations a certain amount of planing had to be done against the grain. Usually the main problem with that would be tearout, but I actually managed to get a clean surface with little to no torn grain. The bigger problem I had was the getting a really clean transition in the spots where I would start and stop planing. The nankin-ganna almost has to hit the surface like a plane landing and taking off, otherwise you end up with really distinct lines where the plane starts and stops.
So I ended up sanding out the surface left by the nankin-ganna. I did however leave the carved ends on the arms in their raw faceted state. I like the textured surface and it’s contrast with the smoothly sanded arms.
With the arms and back assembly shaped the final task to complete before assembly was to scribe the legs to the arms. I used a similar process and the same scribing tool as pictured in the previous post. In the two pictures below you can see the legs after a couple of rough scribes. You can also see the pencil lines marking the next scribe.
After a few rounds of scribing and chiseling, I gradually moved closer and closer to a clean joint at each intersection. This process took a while but the results are worth it. There’s something really satisfying about two round pieces cleanly joining in this way.
Why not just drill a big hole into the arm and insert the whole leg you say? There are a few downsides to that method, but the biggest in my opinion is the fact that when the two parts are nearly the same size as they are in this case, if you were to drill a hole big hole in the arm, there would be very little material left on the sides of the hole. And what material does exist will be really weak grain that has a tendency to break an crumble just by looking at it. By cutting and scribing into the endgrain of the joining piece (in this case the legs) you are working with a grain direction that is much stronger and won’t crumble as easily as the previously described method.
With the scribing done I moved ahead with gluing up and assembling the whole chair. I first glued and clamped the legs and stretchers, and after an hour or so removed the clamps and proceeded with gluing on the arms.
And with that the main structure was assembled, and the most difficult work finished. In the next post I’ll share a few shots from weaving the seat as well as some final images of the completed chair.
Thanks for reading,
Jon
