Tahiana Ramananantoandro

Effects of machining parameters on raised grain occurring after the application of water-based finishes

The application of water-based finishes to wood surfaces is known to increase surface roughness by raising the grain. This phenomenon is detrimental in the appearance of the finished wood surface. To eliminate it, light denibbing is needed prior to applying the second coat of finish. The objective of this study is to optimize wood machining to minimize raised grain caused by water-based coatings and to avoid, or at least, to minimize the sanding operation between each coat of finish. Experiments were done on beech samples. Complete experimental design was defined to generate several surface qualities from “smooth” to “rough” which are linked to the average thickness of the removed chips. The planing parameters were the cutting mode, cutting direction, feed speed, and rotation speed. To compare with existing practices in joinery industries, specimens were also sanded. Two layers of water based coating were applied. Half of the samples were sanded after the first coat and the other half were not. Following each coat, surface roughness was measured and raised grain was evaluated through tactile tests. Results show that sanding operation is responsible for more raised grain than planing. Whereas for planing, high chip removal thickness gives more raised grain after coating application. The effects of the other parameters on raised grain occurence are not found to be statistically significant. The average roughness (Sa), quadratic mean roughness (Sq), reduced peak height (Spk) and the material volume of the peak section (Vmp) are the roughness parameters best correlated with tactile evaluation of grain raising, but the correlations are weak.

Adaptation et variabilité génétique de la croissance de Liquidambar styraciflua L. de l'essai de provenances à Mandraka, Madagascar

Liquidambar styraciflua , or the American sweet-gum tree, is a deciduous species origi- nating in the south-eastern United States and Central America. Its behaviour in plantations is outstanding and its timber is used for mul- tiple purposes. In 1986, under an interna- tional provenance trial, the Commonwealth Forestry Institute (CFI) in Oxford distributed Liquidambar styraciflua seeds of thirteen dif- ferent provenances, mainly harvested in the natural range of distribution of the species extending from the eastern United States to Mexico and Central America. In Madagascar, they were planted by the forestry and fish- eries research department (DRFP-FOFIFA) at Mandraka. Data on annual growth were col- lected for 25 years, but the results, unlike those for Brazil and Mexico, have not yet been published. The aim of our study was to assess adaptation of the species in Man- draka and to investigate intra-species growth variability. Up to 25 years of age, the trees grew steadily to reach, on average, a height of 26 m, a diameter of 38 cm and a basal area of 0.12 m² per tree. These average val- ues differ significantly according to prove- nance. At 25 years of age, heritability in the broad sense is 0.76, 0.60 and 0.63 for height, diameter and basal area respectively, indicating genetic control over growth. Under similar ecological conditions and at equiva- lent age, the growth of L. styraciflua is equiv- alent to that of species of the genus Eucalyp- tus and better than in Pinus . In comparison with the international CFI trials, the prove- nance trial for L. styraciflua at the Mandraka site shows better adaptation and a high growth rate for the species.