A comparison of the different joinery methods used in the construction of wood furniture

Thumbnail Image



Journal Title

Journal ISSN

Volume Title


University of New Brunswick


Two separate series of strength tests were performed on various types of joints used in the construction of wood furniture. Six test samples of each joint type were made in a consistent manner, either by hand or with the use of machines. A gradually increasing point load was applied to each sample in order to ascertain the maximum load that the joint could withstand before experiencing a substantial drop in the load level, i.e. the point at which the joint could not recover from the stress of the load. When compared with three variations of the mortise and tenon joint, the common #20 Biscuit joint proved to be the weakest. On average, this joint was half as strong as the traditional mortise and tenon joint with the cheek and shoulder of the tenon cut from the same piece of stock. This type of joint was able to withstand over 26 kilo Newtons (kN) of applied load before elastic failure, compared to just over 12kN for the #20 Biscuit. Also included in this series of tests were two modifications of the mortise and tenon joint in which the tenon is loose or floating; one with the tenon rounded and the other with a square-edged tenon. These loose mortise and tenon joints were in the middle of the other types of joints as far as strength was concerned. The rounded floating tenon resisted a higher average maximum load of22.5kN when compared to the 20kN for the square-edged floating tenon. This may be attributed to the possibility of a more even distribution of glue around the entire surface if the tenon. The second series of tests consisted of four variations on the dovetail joint. The results were analyzed to assess how the strength of the joint was affected when modifications were made to the thickness of the dovetails as well as their angle. Again, a gradually increasing point load was applied which attempted to pull apart the dovetails and the maximum load was taken to be that at which elastic failure occurred. The results indicated that a joint having a higher number of dovetails per unit width and more allowable glue surface area on the joint were more resistant to the applied load. The thin, low-angled dovetails with three dovetails were the strongest, averaging over 8kN for maximum load, compared to the weakest load of over 6kN load for the thicker, low-angled dovetails with only two dovetails. When only the angle was changed, and not the number of dovetails or thickness of these dovetails, the 14° dovetail experienced slightly higher strength readings than the 7o dovetail. This could be attributed to the fact the higher angled joint had slightly higher available glue surface area. This would indicate that the angle of the dovetails does not greatly affect the overall strength of the joint.