Parameter interactions in two-block welding and the wood nail concept in wood dowel welding

The interactions between parameters found to be determinant in wood dowel welding by high speed rotation have been evaluated. Of these, the interactions that proved to be the most significant, in descending order, were rotation rate/dowel moisture content, followed by rotation rate/ethylene glycol, and finally, at a lower level of significance, the interactions rotation rate/dowel temperature, wood grain direction/wood species and dowel temperature/wood species. Of the individual factors, once the most determinant factor already optimized in previous studies, namely the dowel/hole diameter difference, was fixed, the most significant were wood grain direction, dowel moisture content (dryness) and wood species. The optimized process yielded excellent strength results. The regression equations developed were able to predict the strength obtainable. The torque for insertion of the dowel in the substrate hole has been measured for several cases and the results are presented. In no cases the value of the torque needed for insertion was excessive and insertion was, therefore, easy. Wood joints composed of two pieces of timber held together by a dowel welded to both of them were assembled for the first time. Two further new concepts have also been advanced and tested: (i) the conical dowel, to maximize welding area and (ii) the concept of the wood nail in which a slightly conical fast-rotating hardwood dowel is inserted rapidly into a softwood substrate into which no hole has been pre-drilled. X-ray densitometry of the samples prepared with the latter approach showed some interesting mechanical interlocking features that might contribute to dowel bonding in softwoods.