The Chair – The arm-backrest joint (part 2)

This is the joint that is creating controversy on Wood Central :)

Ellis wrote, “ .. I would try to preserve the rounded points on the finger joints (no rhyme intended), primarily to preserve the look of the original. There are a few things that I regard as inalterable, one of which is a design detail in a reproduction of a classic piece. You may have the best intentions and the most well-rationalized reasons for taking liberties with it, but it isn't your design. An informed viewer will see -- consciously or unconsciously -- that there's something different about this joint with the pointy fingers, and the overall effect will be diminished”.

The thing is, I agree with Ellis. I asked myself, which would I prefer – pointed fingers or rounded. The answer is rounded.

I started a reply, and then paused when I found myself saying … “the problem is the difficulty in making the round fingers. The fingers do not dominate the design and are hardly likely to be noticed. The pointed fingers are less risky …”

When I caught myself writing that I knew I had to find a way to do the rounded fingers. This build is all about risk, and not about being dominated by safety-first decisions.

I have an idea, but that is for a little later. For now I thought I would show you how the factor did the fingers and joined the arms and backrest. There is an irony here: the factory developed a machine method to build a chair that was probably first built with power and handtools … and here we are seeking to reverse the process, to find a handtool method to do what is done with machines! :)

I managed to extract some shots of work process from the video offered by the Moebler factory.

At the factory the arm is cut from a slab of 5” thick oak using a template that resembles the one I built from an actual arm …

In the picture above the grain direction looks straight, however the photo below indicates that there is considerable run out ..

Below is the arm in the sanding copier just prior to the shaping that will take place …

Below is the shaped backrest . This is ready for the fingers …

What I find interesting is that a saw blade cuts the fingers, and that the reason for the rounded ends is that this is the limitation of the method used! The thickness of the blade cannot saw to a point.

The finished cuts …

Arms and back ready for gluing up …

Being glued up in a jig …

And the backrest is shaped with a CNC machine.

This leaves roughish marks that are removed with spokeshave and sander.

Now let me take you back the template of the fingers I made …

Here is a second template, one of the backrest as it joins to these fingers …

My idea is to use the points to position holes (the round ends), and then to connect these with saw cuts to the outer edges of the holes.

In theory that sounds easy, however getting the holes lined up is the tricky bit.

Why holes and not simply file the ends with a chainsaw file (of which I have a bunch), as suggested by Ellis. Well, the file is just not reliable enough. The fit of finger to finger must be spot on – the shaping angles deeply into the fingers and any deviation from a tight fit will show up as a gap or an irregular shape. A round, drilled hole on the other hand is predictable. I think that sawing to the edge of a hole is far more complicated that sawing to a point, but what the hell …

The need now is to find a way to drill the holes. By hand (so a drill press is out).

The obvious method is to use a dowel guide. I have a fine one in the Stanley #59. The problem with this is that, while it will fit across the width of the arms, the backrest is far too wide for it …

Consequently, my idea will be to build a jig similar to the #59 (just larger).

Any other ideas?

Regards from Perth

Derek

January 2014