Monday 9 July 2018
I experienced a pleasant feeling of achievement when the cramps were removed from the assembled box. Although the components had been put together previously, this was the first time that it matched my expectation of the design that had been in my mind from early in the progress of its evolution. There are many different ways of describing the design process but at this point the word creative was uppermost in my mind.
I removed the small amount of glue that had oozed out of the joints during the glue-up with a combination of a smoothing plane and a chisel. Very little wood had to be removed because the junctions between adjoining surfaces were precise.
The surfaces were then sanded to a finish that would require a light re-sanding later on when the basecoat would be applied.
Although I was anxious to cut through the box, I sat down and thought about any processes that could be best carried out before separating the lid. The most important was to mark out the positions of the hinges which were going to be morticed into the adjoining rails.
I also checked to see if there was any distortion by placing the box upside down on the table of my planer-thicknesser. It is absolutely flat and works well as an engineers surface plate.
I then turned the box over so that it stood on its feet. I was pleased to see that there was no distortion and this meant that I could use the two aspects that had been tested as references to saw through the box and to clean up the adjoining edges.
Separating box & lid
Sawing through the box was carried out using my Hitachi band resaw. The 75mm-wide stellite-tipped blade has a 1.5mm kerf, photo 1. Although the cut is very clean I made an allowance of 5mm for the saw cut and cleaning up to achieve a good fit between the two adjoining edges. This was carried out with a trapezoidal-tooth blade on my dimension saw,photo 2.
I removed the saw guard and riving knife and lowered the height of the blade to correspond with the thickness of the corner posts. The rip fence was set so that it was flat on the saw table.
For extra support on the underside of the cut I fixed a board made of thin MDF down onto the table with double-sided tape. It was carefully positioned so that the edge touched the inner side of the saw teeth. It also provided a smooth clean surface that reduced the risk of the box getting scratched and gave me much more control as I fed the box forward past the saw blade. The fence was then raised into the vertical position and the edges of the lid were cleaned up.
The adjoining edges made perfect contact and all that was necessary was light sanding which was carried out after the hinges and lock were installed.
The dimensions of the hinges were determined by the width of the lattice strips and by the diameter of half inch router cutter.
Strips of prepared wood were glued together with newspaper between the edges so that they could be separated after the profiles had been formed. Although the paper is very thin, it delaminates when a force is applied at 90 degrees, thus separating the pieces without removing any material which would be the case if a saw cut was used.
The hinge profile was formed in stages on my router table. In addition to the round section of the hinge I routed a groove which would result in a small tenon when the individual pieces were cut.
The slots, corresponding cylindrical recesses and holes were formed on my milling machine, photo 4.
My precision machine vice and stops held in place with G-cramps ensured accurate positioning of the very small hinge sections.
The next stage was to rout the mortice slots to accept the small tenons that projected from the hinges. I had already marked out their positions, using scalpel and try-square, before the lid was cut off.
With the slots routed, I carefully pared back to the knife line to square the ends.
A dry assembly of the two parts of the three pairs of hinges assured me that the fit would be accurate when they would be finally glued in place and the pins were inserted, photo 5.
In the past I have used conventional metal stays to hold the lid in the vertical position when it is opened, but for this box I wanted to avoid a mechanism that would take up space inside the piece when the lid was lowered.
I borrowed the idea of a stay that was built into a hinge and was first used in Victorian times. Two flanges with 45 degree faces were incorporated into the hinge.
When the box was opened the flanges made contact at a point that held the lid just beyond the vertical position. I achieved this by creating the 45 degree faces on two pairs of projections that were made from my wooden hinge components.
The angled facets were formed by removing half the profile of a hinge on the bandsaw. To do this, the hinge blank was held in a jig which was also used to produce the angle on my disc sander, photos 6 & 7.
The partitions at the ends of the box were straight components that were let into housings and installed during the main glue-up of the box.
The area in the centre of the box was also divided up into partitions with a cylinder in the middle; in fact all the partitions were echoes of linear details that appeared on the lid.
The straight pieces were inserted into housings that were cut before the box was assembled and the cylinder, made from four quadrants held together with splines, slid downwards onto the projecting ends.
An extra piece of wood was glued onto the outer face so that it could be cut to create square ends that would provide accurate reference surfaces, photo 11. The grooves to accept the splines were cut on my milling machine with the components held on the baseboard, photo 12.
The quadrants were then glued together and held in a vice with two additional F-cramps while the glue set, photo 13.
This assembly was then glued to an MDF backing board with newspaper in between so that the cylinder could be split off after it had been finished.
The assembly was then set up in the 4-jaw self-centring chuck of my engineering lathe and the inside bored out. To make sure that all three cylinders had exactly the same internal diameter, I zeroed the final setting of the first cylinder and then turned the other two in succession to the same measurements. This process was then repeated for the external diameter, photo 14.
The baseboard was then fixed to the rotary table which was mounted in the vertical position on the milling machine. Using a quarter inch-diameter router cutter, with two passes I was able to cut a housing with one-eighth of an inch-radius corners that matched the rounded top edges of the straight partitions that engaged the cylinder.
After the first housing had been cut, I wound the rotary table through 90 degrees and cut the second and so on until all four housings had been created, photo 15.
The handle had to be inserted after the cylinder had been turned. After preparing the rectangular component and holding it between the inside faces of the chuck jaws, I measured the internal diameter of the cylinder with dial callipers and turned a radius on the ends of the handle strip to match the calliper setting, photo 16.
The curved profile and finger depressions were formed on my milling machine and disc sander and the top edges were rounded off on my router table using a safety jig to hold the component, photo 17.
After the cylinder had been lacquered the handle was pushed down inside and bonded in place with super glue.
So far many aspects of the making of this box have relied on machines and engineering principles, so I felt that I had to include the joint that I still make entirely by hand: the dovetail. The rectangular trays at the ends of the box fitted the bill even though one corner was rounded to match the insides of the corner posts, photos 18-20. This had to be made on my milling machine, photo 21.
I spray-finished the outside of the box with a basecoat and topcoat of Morrells pre-catalysed lacquer. After denibbing very lightly with 320-grit paper, the surface was burnished with a grey webrax disc on the random orbit sander to produce a gentle matt appearance.
The bottoms of all the trays and partitions were then lined with crushed velvet. The fabric was wrapped around a thin cardboard carrier so that the edges would be protected.
I coated both sides of the card with double-sided tape so that one side held the fabric and the other side held a secondary piece of card which was 6mm narrower all round.
The overlapping edge of the fabric was then folded over to form a hem held in place by the 6mm strip of adhesive.
The second layer of card was also coated with tape. When the backing was removed the combination of velvet and card could be lowered into the area to be lined, photo 22.
Finally, small discs of suede were fixed to the undersides of all the trays. I coated a strip of suede with double-sided tape and cut the discs with my disc punch set, photo 23.
It is my hope that these three boxes will be acquired by people who will respect them as heirlooms. To add to their historic value I provide a provenance which includes information about the piece and advice on care and maintenance. This information will also be useful if they need restoration in the future.
Taking advantage of the convenience of digital photography I now include a CD of the progress of the piece in the workshop.