Abstract:
Particleboard is a widely used composite panel product
consisting of consolidated wood particles and adhesive. On
a macro level the panel appears to be homogeneous, however,
its density varies both throughout the plane of the panel
as well as through its thickness. Of concern for this
thesis is the through-thickness density profile which
strongly influences the panel's mechanical properties such
as modulus of rupture, modulus of elasticity and internal
bond strength. Panel manufacturers realize this importance
and, as a result, have begun to use some methods to measure
the density profile in their laboratories for quality
control. However, present methods provide data for only a
small portion of the production and often hours if not days
after the panel was produced. This study proposes and
evaluates a method using a X-ray computerized tomography
technique to predict the through-thickness density profile. Three types of commercially produced particleboards
were scanned using a system consisting of an X-ray
densitometer and a computer acquisition system to collect
the attenuation data. The density profile of each sample
was then predicted by using a computer algorithm (C
language) in which two mathematical models were applied. A
two-degree polynomial regression function was used to
evaluate the accuracy of the models.
Comparisons were made of the predicted to actual
density profiles for the same sample. A good agreement, a
correlation value of r² = 0.822, between the predicted and
actual density profiles was found indicating that this
technique does provide good estimation of density profiles.
This study provides the necessary knowledge to design a
valuable nondestructive testing tool for use in monitoring
the quality of particleboard and other similar composite
products on the production line and on a real time basis.
