Full text
| |
| Author(s): |
Maristela Gava
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | Piracicaba. |
| Institution: | Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC) |
| Defense date: | 2010-03-05 |
| Examining board members: |
Jose Nivaldo Garcia;
Luciano da Fontoura Costa;
Nilson Franco;
Israel Luiz de Lima;
Laura Alves Martirani
|
| Advisor: | Jose Nivaldo Garcia |
| Abstract | |
The present research work had the aim to contribute for the development of testing procedures and evaluation of displacements to determine shrinkage of Eucalyptus urophylla, E. saligna and a hybrid E. urograndis in the radial and tangential directions using the Thin-plate Spline TPS method. It was carried out a theoretical and practical study in order to contribute to reviewing of Brazilian Standard NBR 7190/97 looking forward to optimize the dimensional variation testes mainly on the specification of specimen and method of assessment of shrinkage. Thirty trees of those species from the Forest Sciences Experimental Station of Anhembi were utilized for the study. The statistical analysis were made according full randomly experimental design with independent sampling and five replications through factorial scheme 3 x 2 x 4 (species x longitudinal location x radial location) for the study on variables obtained from the specimens which had the wedge shape; 3 x 2 (species x longitudinal location) for the study on variable obtained from the NBR 7190 standardized specimen and for the study on natural moisture content in trees and 3 x 2 x 3 (species x longitudinal location x method) to compare the methods of assessment of wood shrinkage. The proposed method showed to be sensible to samples shrinkage and make possible a detailed analysis of strains in the whole specimen surface. Shrinkages variations from pith to bark were studied observing that the TPS method applied on the wedge shape specimen produced shrinkage values smaller than those obtained through standardized procedures. When applied on cubic specimen produced values statistically the same as those obtained through NBR 7190/97. The natural moisture content in alive tree and the basic density of wood were not enough to explain the verified shrinkage variation but it was verified a tendency of the radial shrinkage to decrease as those variables increase and a tendency of tangential shrinkage to increase as the natural moisture and basic density also increase. (AU) | |