Hello and welcome to my blog! This post begins a short series in which I’m building a live edge tsuridana for a modern Japanese style home. A tsuridana is a single shelf or set of shelves found in specific areas of the traditional Japanese home. These shelves are normally supported by a single bamboo pole at one corner, and along the wall at the back and side.
The tsuridana are found in the tea room and adjoining service kitchen called the mizuya, above the traditional sink. In this case the tsuridana will be used in a spare kitchen above a sink. This tsuridana will float freely, with attachment to one wall. Unlike traditional Japanese building, the shelf will be attached to two steel extensions used to support it, rather than utilizing bamboo and wooden anchors.
This shelf replaces the original live edge tsuridana made for this kitchen and since removed. I’ve chosen a walnut slab, in 10/4 thickness, featuring a certain natural curve to it.
This slab is likely cut from an area near to the center of the tree, the area on the right hand side of the photo appears to be the rotted out center. Likely the slab started out at least twice this dimension and needed to be split in half. That’s a pretty handy circumstance, in this case. This tree likely had a very steep taper to the trunk and at at least one major branch splitting off from the main trunk. This creates the effect of a curve without utilizing a tree with a curved trunk. A curved trunk brings with it a lot of internal tension, which will cause unpredictable movements during seasonal humidity changes.
After sawing away most of the rotted edge and leaving just enough to feature, the slab was then placed upon my bench to be face jointed. The first step in face jointing the slab is to check for twist using winding sticks. The winding sticks accentuate any twist present, allowing for easy and accurate diagnosis.
My initial plane strokes are going crossgrain to remove cup from the slab and get a basic location of problem areas (low spots) which will need to be planed out. To remove a low spot the entire face must be taken down to that dimension.
The low spot at the near end of this slab will be partially removed when I crosscut the end of the slab, and further removed when I apply a rabbet to the exterior edges. So, while concerning, I will not take its full depth into consideration.
Next I have a low spot at the center of the crotch section, this is also of little concern because I can work to taper it into the edge with a spokeshave as I fabricate the edges.
The last concerning area is the extending portion of the center trunk which tapers away from the slab face. Rather than remove this area I will accentuate it and use it as a feature.
With the problem areas now located and a basic idea of which corners are high (twist), the next step in flattening the slab is to check for flatness along its length. I use a precision straight edge to do so and find that it is bowed toward the inside face.
I continue with crossgrain strokes, removing both the twist and bow from the slab by working problem areas until the slab begins to become flat. This worked out nicely as it allowed me to work the low spots from the face while bringing it to flat without having to remove large amounts of material.
With the slab now basically flat I next take a few strokes with the try plane to further flatten the slab and follow up with a smoothing plane until I can take full strokes along the length of the slab.
The next step in preparing the slab is to fabricate the live edge, the bark is removed and the material beneath is carved and otherwise shaved away until mostly good wood is left. Some of the areas where worms have made holes in the edge will remain.
The same approach is taken for the featured edge of the board, this is primarily done with a spoke shave. One could probably start with a draw knife at this phase, but being without one I simply go directly to the spokeshave.
Once the edge is complete I chamfer the outside edges.
The compacted wood pulp which remains in those worm holes is removed.
The final step in fabricating the edges is to trim the terminating edges. These edges are normally cut back with a chainsaw while the slab is being prepared for the sawmill, leaving a fairly coarse edge. I cut the area back at an angle which mimics that of the naturally formed edge, then plane the cuts smooth and chamfer their edges.
This slab, left on its own will move in an unpredictable manner as the seasons change. In order to support the slab, and offer some consistency in that movement, I’ve decided to create a system of dovetailed battens which will support the slab.
The battens will work in dovetailed grooves running across the slab, I did not, however want the dovetails to show on the exterior edge. I’ve marked out a large rabbet, which runs the perimeter of the edge. This rabbet will setback the front edge of the battens by some distance. My work in creating this edge detail begins in first chiseling away the waste material.
Followed by planing the exposed surface flat. I used a finish plane and shoulder plane in combination to create the inside corner.
This same approach was used along the remaining length of the slab with one addition. The inside corner was first sawn with a track saw. The remaining waste was crosscut into sections which could be more easily chiseled off.
The resulting edge was smoothed and flattened.
With the bottom now completed it was time to bring the slab to thickness. My goal at this stage was to leave the slab as thick as possible, so it was not brought down to a specific numbered dimension, but instead to a gauge line which represented the lowest level needed.
My approach again began with crossgrain work to remove bulk material from the highest sections, while keeping the slab flat across it’s width.
The slab is thicker at its ends by quite a substantial amount. I did not expect this, originally, assuming that the slab would be bellied on this face in the center. Upon further investigating I found that both were, in fact, true. The slab was bellied in the center and standing even further proud at the far end of the crotch section. The shape formed was something like a wave.
The near end of the slab was worked in long passes, which, originally further accentuated the wave.
Then reduced in the center as it was brought down along with the crotch section. The forked area of the crotch needed a great deal of material removed to bring the slab down to flat.
Finally the slab was nearly flat and the remaining work could be performed with the smoothing plane.
The resulting surface shows a reflection of the plane.
Here the straight edge now reveals that the slab is flat along its length.
Thank you for following along with the preparation of this walnut slab tsuridana, the series will be completed in the next installment, where I will detail my process for building and installing sliding dovetailed battens held in place without screws or other metal fasteners.