| Summary: | 碩士 === 國立中興大學 === 森林學系 === 84 === The stress-wave nondestructive test (NDT) method was applied in
this research to test the joint strength for four kinds of
longitudinal joint and three kinds of T-type joint of Selangan
batu, Red oak, Douglas fir and American hemlock, and try to
predict the joint defect. The relationship of joint strength
predicted by NDT and tested traditional destructive method was
correlated.For the longitudinal joint , the wave velocity of all
samples would become obviously faster after jointed, but no
constant tendency in vibration frequency was observed. Bending
test showed that the MOR among the wood species, Selangan batu
had higher strength value, and Douglas fir was the lowest.
Among the joint methods, the best one was finer-joint, and the
worst one was butt joint. MOE of tested wood species, showed
same tendency as MOR showed, but the variation of joint was not
significant. The relationship between MOEd and MOE was tested by
linear regression analysis, the Coefficients of correlation
determination showed that the highest value was Douglas fir
(r2>0.85), and the lowest for American hemlock (r2<0.50).For
the T-type component side, only the stress-wave timer and tap
tone analysis system were applied. The vibration frequency
tester showed unstable result, therefore this method was
excluded in this study. The wave velocity did not have
significant difference before and after the T-type component was
assembled, but it showed slower velocity after the components
were assembled. While the longitudinal frequency was measured by
tap tone analysis system, it was found that the value of
softwood was lower than that of hardwood. The vibration
frequency was positively correlated with tenon length, and the
value of withdrawal component of non-spread glue (3401 Hz) was
lower than the double-spread one (3824 Hz). The withdrawal
strength showed that the best was Red oak, and the worst was
Douglas fir. For joint methods, it showed that the best was
oval-end mortise and tenon joint, followed by square-end mortise
and tenon joint, and dowel joint. The withdrawal strength was
positively correlated with tenon length. Among the all
components, the highest withdrawal strength was 2003 kgf for 40
mm tenon length , and the lowest was 137 kgf for the case
without glue. The withdrawal strength of tenon width with +0.5
mm fitness would be obviously decreased.For the bending moment
of the T-type components, the best was Red oak, and the second
was Selangan batu with a little difference. Among the joint
methods, the best was square-end mortise and tenon joint, the
second was oval-end mortise and tenon joint, and the worst was
dowel joint. The bending moment of the tenon thickness with +0.5
mm fitness was better than ±0 one, and there was significant
variation among three kinds of glues used in this study. The
correlation of withdrawal strength and bending moment measured
by destructive test and predicted by NDT was not significant
tested by linear regression analysis. The different joint type
which r2 value of regression analysis by withdrawal strength and
vibration frequency was 0.80. It could be concluded that the tap
tone frequency analysis system could be applied in prediction
the variation of different joint components of oval-end mortise
and tenon joint.
|
|---|
