Exhibition Box Pt 2 archive
Thursday 2 July 2009
Having cut 275 components for his jewellery box last month, Robert Ingham now starts to put it all together
The lid was to be created by cutting through the box once it had been assembled; this is quite common practice and is straightforward if the sides of the box are made of solid wood. Because my box was made from a combination of MDF, veneers and inlayed strips I had to insert a piece of solid wood into the sides, backs and fronts through which the dividing cut could be made to separate the lid.
This was done after the top and bottom lippings had been glued in place. The receiving edges were carefully dimensioned after the strips had been inserted.
I overcame the problem of breakout on the underside of the cut by making a replacement saw-mouth plate from MDF. An insert of lignum vitae fits right up against the teeth on the rip fence side of the blade, resulting in a clean cut that is smooth enough for the lippings to be glued straight onto the edge, photo 1.
SplinesI used splines to locate the lippings accurately, knowing that the grooves that held them could be routed using the fence of my router table, photo 2. Because the grooves could be routed through from one end to the other, I was able to use my two-way feather board so that the component could be fed both ways, removing all the shavings and making sure that the cut was exactly the same as the cutter diameter. The splines were made from jelutong which is very stable and machines well. Because they were only 3mm thick I used a carrier board, thus enabling me to pass them safely through the thicknesser.
I included a strip of low-tack double-sided tape at each end because the thin strips could chatter at the start of the cut, resulting in reduced thickness. Here I must sing the praises of my Felder. The dial-controlled wheel that raises and lowers the thicknessing table is graduated in tenths of a millimetre so planing the splines to fit the grooves was an absolute joy, photo 3.
The lippings were glued and pressure was applied between two cauls held in a bench vice, photo 4.
The sides, backs and fronts were then sawn apart to receive the component that would eventually be cut to separate the lid from the lower part of the box. This component was 5mm deeper than the lippings and lattice strips to allow for the saw cut and cleaning up of the edges.
Corner postsThe corner posts were planed and thicknessed to a square section that was the sum of the thickness of the sides and the radius of the inner curve. I used one piece of wood that was long enough to make four corner posts for each box, thus ensuring that the end grain that would show on the top corners could be arranged to provide attractive visual composition. It is necessary to make these decisions with intent rather than hoping that you might notice them later on in the progress of the construction. In fact I selected the wood so that the curve of the annual rings appeared to wrap around the corners of the box.
The posts were sawn to length to be an exact sum of the height of the sides plus 60mm, which represents two lattice strips and two square spaces. This combination would enable me to cut the small mortises that hold the lattice strips with controlled accuracy on my milling machine. The inner curve was cut on my router table with a ball cutter set so that half of the diameter was contained by the fence. With a fine-depth adjuster for the vertical dimension and the use of my adapted Microfence for the horizontal dimension I was able to set up the router so that the width of wood that was left was the same as the thickness of the box sides, photo 5. I set this up using a dummy piece that had been prepared to the same dimensions as the corner posts.
The grooves were then routed with adjustable end stops fitted to the fence as they did not run the entire length of the corner posts. The same method was used to rout the grooves in the ends of the sides, fronts and backs, photo 6.
Interior partitionsThe storage potential of the box is enhanced by the inclusion of partitions which are housed into the inner surfaces. The partitions are 8mm thick and the top edges are rounded over with an 1/8in radius cutter. This meant that the housings would also have a corresponding rounded end because I did not want to cut a shoulder to cover the top edge of the joint, bearing in mind that the partitions would have to be assembled and could not be seen when the box would be glued up.
With a 1/4in-diameter TCT router cutter, some simple arithmetic and the baseboard of my milling machine I was able to cut the housing so that the partitions would slide in perfectly, photo 7.
Each box has four small rectangular trays that rest in the tops of the four corner partitions. The trays rest on small pegs located into holes drilled on the insides of the compartments. These holes were also drilled with the sides held down onto the baseboard of my milling machine.
Lattice railsThe rectangular section was planed at the same time as the top and bottom lippings were prepared. The components that made up the lattice construction were carefully cut to length on the sliding table of the dimension saw. The length of each set was transferred from my dial callipers to the sliding stop of the crosscut fence by converting the dimension to the sliding stock of an engineer's combination square, photo 8. In turn, the halving joints, mortises and tenons were cut on my milling machine. I held the pieces in my precision machine vice which also provided a register for aligning the ends with the outer faces of the jaws to ensure continuity of position before the cuts were made, photos 9 & 10.
Assembling lattices & sidesBefore this could be carried out, the grooves for the splines that held the top and bottom panels in place were cut on the router table. The same setting was used for the grooves in the edges of the tops and bottoms. Because of the small scale of the project the assembly of the lattice components was done in stages. The cross-halving joints were glued and cramped between two parcel tape-covered cauls in the vice, then the mortise-and-tenon joints were glued one at a time and held under pressure in the vice. I used Titebond which has a short pressure time, and with three bench vices I was able to keep gluing up continuously.
After assembly, the small amount of cleaning up was carried out with a finely set smoothing plane.
I mentioned earlier that the splines used to glue the lippings in place were made from jelutong. The splines between the corner posts and the ends of the sides, fronts and acks would show when the lid was cut off. For this reason I made these splines out of the same wood as the corner posts so that there was an appropriate colour match. Assembling and gluing up of the sides could now go ahead.
As always, I carried out a dry run. This enables me to work out a sequence, to set the cramps to correct distance and make sure that all cramping blocks are prepared to distribute the load evenly. The blocks for the lattice assembly were held in place with masking tape. The assembly was then glued to the bottom edges of the sides. With these in place the next stage was to glue on the corner posts.
Gluing up the boxI can remember making my first box when I was at secondary school. I was encouraged to pre-finish the inside before gluing it together, a method that has become standard practice with all carcass pieces I have since made. Because I spray finish, it was necessary to mask any surfaces that would be coated with glue later, a time-consuming task that nevertheless makes the removal of glue squeeze-out very easy later on; Titebond will not adhere to lacquer, photo 11. A temporary insert was placed into the housing slots.
When the precatalysed lacquer had dried, the surfaces were de-nibbed and lightly burnished with webrax and the box was ready to be assembled. As before, the gluing up was done in stages to enable plenty of time for the application of glue so that there was no risk of a skin forming before joints were assembled.
I had also to remember to include the partitions during this sequence.
I used a combination of a comb spreader and a small brush to apply the glue and in this way I could make sure that there was very little excess adhesive to remove later.
Although only some of the joints were being glued, the entire box was assembled to make sure that everything aligned and made contact.
A sequence of assembly was absolutely necessary and I have to say that I was glad when this phase in the making was over. If you get it wrong there is no way to reverse it, photo 12.