Date of Award

12-2010

Degree Name

MS in Engineering - Materials Engineering

Department

Materials Engineering

Advisor

Charles Chadwell

Abstract

Understanding how different processing variables influence wood welded bonds is vital if the technique will ever be used to create engineered lumber without using adhesives. A variation of vibration welding, wood welding uses pressure and friction to bond materials together. During welding, heat causes a softening in the wood, a naturally occurring composite material. This softening leads to fiber entanglement and a bond forms upon cooling.

The goal of this research was to investigate several processing aspects of the wood welding procedure. A prototype wood welding machine, designed and fabricated from the ground up, was used to investigate the effects of various welding parameters using birch wood. Wood welds were evaluated on the basis of bond coverage and ultimate shear strength.

Four experiments were performed: welding frequency and duration interaction, grain orientation effects, alternative welding completion metrics, and strength development over time. During the wood welding process, three distinct phenomena were repeatedly observed: smoke creation, welding residue formation, and an audible pitch change. The presence of each was recorded for every wood welded specimen and used later in additional data analysis. Investigating each of the welding phenomena was done in an attempt to better characterize when fusion was achieved at the weld interface.

ImageTool, an image analysis software package, was used to investigate and quantify the often irregular bonds exposed after shear fracture. The results of the various welding variables were analyzed on the basis of shear strength and bond uniformity.

From the birch samples, it was shown that better bonds result from lower welding frequencies and longer welding durations. The grain orientation analysis demonstrated that welding orientation marginally affects the average shear strength of the wood weld. The data from the alternative welding metrics suggests that welding time is not a quality indicator of welding completion (bond coverage). The strength development trials confirmed previous research; wood welds obtain most of their strength in a relatively short period of time.

Douglas fir and poplar both proved to be weldable for the first time, but they were sufficiently weaker than birch. When welding was attempted with Douglas fir under similar pressures used for birch, Douglas fir samples would commonly “washboard.” With reduced welding pressure, Douglas fir formed wood welds more easily.