Burr Elm Box archive

Thursday 18 September 2008

Robert Ingham creates this box out of burr elm and bog oak


I am sure you have heard fellow craftsmen talk about the inspiration that lies in a piece of wood that they have in stock, waiting for the right moment to be released and expressed as an object. I was given some pieces that had been cut off a square block of burr elm from which a disc had been sawn for a turned bowl. They lay on a shelf until the day after I saw a programmme about Islamic Art, in which a chest was featured, decorated with geometric patterns.

The strong natural pattern of the burr elm seemed to ask for a frame to define it so I decided to use some bog oak to do this. The offcuts were quite small, which limited the area that could be created without edge jointing. Also, I was concerned about the stability of the elm, which even with clean straight grain, is very unreliable. While the question of radial and tangential shrinkage has very little application to burrs, which consist of the growth of many small shoots produced by the tree to counteract irritation usually caused by diseases of the bark, there can be problems with distortion from stress relief, as is evident when they are cut into veneers. So I decided to create the front, sides and top of my box from several small squares with strips of bog oak to define them - in effect, several small panels framed with black rails and stiles. I could have achieved the same visual result with marquetry and veneers - I chose not to because I was stimulated by the challenge of using the very strongly patterned burr elm that I already had.


The thickness of the sides was to be 12mm (1/2in) and the burr elm "panels" were 45mm (1 3/4in) square. I did not have enough material to make the squares out of solid wood and I was also concerned about the long term stability of the elm, so I decided to make a sandwich of two layers of 3mm (1/8in) glued to both sides of a 6mm (1/4in) core of MDF. I needed 124 squares, so the economic use of the amount of burr elm I had was of paramount importance. Careful planning was therefore necessary and when I had decided on a strategy, I sawed the wood up into pieces from which I could further convert them into strips.

The bandsaw in partnership with a surface planer, was the ideal combination that produced 3.5mm thick strips with one planed surface and one sawn surface. The strips were long enough to produce three squares, so I set about gluing the sandwiches using Titebond and clamping them together in the two vices on my bench. The short pressure time meant that I could glue-up two sets every fifteen minutes while I was working on another project.

Rails & stiles

The next process was to dimension the squares to size. The first box I made using this construction, I planed the edges square using my upright shooting board in the vice, after roughly cutting the strip into pieces. With the adjustable depth stop set to 1mm (3/64in) over-size, I planed two edges to get them square and then moved the stop to the finished size and repeated the process. This was stressful and time consuming. The next attempt was carried out using the cross slide on my router table, once again working to an adjustable stop so that two edges could be established to ensure squareness, and then the remaining two sides could be cut with the stop reset to the final dimension. Now, with a reliable sliding table, I would cut the squares on my tablesaw.

The bog oak for the rails and stiles were prepared using a planer and thicknesser. The squares were located into the construction with loose tongues in grooves. I used 1.5mm (1/16in) aeroply for the tongues, and cut the grooves with a slotting cutter mounted in my router table. Because of the short length of contact between the edges of the squares and the router fence, which would have been a very dangerous process, I made a supplementary fence from a strip of extruded aluminium, which was held onto the main fence with slot screws. This had a window slot cut into it through which the slitting blade projected, providing support above and below the groove for the entire length of the cut.


The main challenge was that of assembling the combination of components. For this, I made a jig, which consisted of a base plate, two fixed fences and moveable pressure bars. The fixed fences were glued and screwed in place to resist the force from the pressure bars. These bars were made from hard maple and were held down onto the base plate with coach screws and wing nuts. I chose coach screws because the square section below the head could be located in a corresponding slot in the underside of the base plate, which prevented it from rotating when the wing nut was tightened.

The clamping force was applied by a series of screws located parallel to the surface of the base plate. Hexagon head screws driven through tee nuts, distributed the pressure through a maple strip faced on one side with aluminium to resist damage, and a glue resistant plastic strip on the other. The pressure bars were held down onto the base plate through a grid of holes that gave a permutation of squares, rails and stiles to accommodate different dimensions.


Unlike a large frame and panel in which the panel has to be free to expand and contract, the squares were small enough to be glued in place as they were made stable with the addition of the MDF core. Because of the number of components that made up each surface of the box, it was not possible to assemble more than one horizontal line of squares and stiles at a time, due to the short open assembly time of the glue. On the other hand, the short pressure time meant that I did not have to wait long before the next line and rail could be assembled. Once each surface had been assembled, the horizontal and vertical corner rails that held the box together could be glued in place to form the assembled box. The lid was then created by sawing through the horizontal rail that had been made deeper to allow for this process.

Woodworkers Institute

Tagged In:

Robert Ingham , Burr Elm , Box

"Error reading XSLT file: cwsTerminology.xsltcwsTerminology.xslt

"The main challenge was that of assembling the combination of components"

Diagrams Click an image to enlarge