Variabilidade e amostragem em painéis aglomerados

• Ano de defesa: 2010
• Autor(a) principal: Guimarães Júnior, José Benedito
• Orientador(a): Mendes, Lourival Marin
• Banca de defesa: Carvalho, Ana Márcia Macedo Ladeira, Mori, Fábio Akira, Osajima, Josy Anteveli, Trugilho, Paulo Fernando
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: UNIVERSIDADE FEDERAL DE LAVRAS
• Programa de Pós-Graduação: DCF - Departamento de Ciências Florestais
• País: BRASIL
• Palavras-chave em Português: Painéis de madeira aglomerada; Chapas de madeira aglomerada; Propriedade; Avaliação; Coeficiente de variação
• Área do conhecimento CNPq: CNPQ_NÃO_INFORMADO
• Link de acesso: https://repositorio.ufla.br/handle/1/4770

At first, particleboards are considered elements of great homogeneity in their physical and mechanical properties. However, it is possible to notice high values for the dispersive measures of the same ones. This factor, many times occurs due to incorrect sampling, reduced intensity and sampling area, the no effective control of all processing variables or intrinsic characteristic of the variation coefficient for such experiments, which is often high. In this sense, the aim of this work was to study the possible causes of variability that exists in particleboards. In the first stage of the work, 4 boards from industries were used, from which the specimens were obtained for evaluation of properties. These ones were obtained in several sizes. From the best results, (higher property value and lower variability) it was possible to option for the following dimensions: water absorption and thickness swelling: 7,5 x 7,5 cm; 2) static bending: 5,0 x 27,0 cm and 3) perpendicular tension: 4,0 x 4,0 cm. During the second stage, boards were produced in laboratory. Each board was totally reduced to specimens obtained in lines for the determination of certain property. Boards were also produce in order to make the sampling according to Mendes (2001). Taking the best results as a base, it was suggested a new sampling for particleboards. It was observed that this new method evaluated water absorption, thickness swelling and parallel elasticity modulus better if compared to the sampling suggested by Mendes (2001). In the third stage, the generated database was obtained by information collected of variation coefficient from scientific journals. From this point, the normality test was applied (SHAPIRO; WILK, 1965). When the data presented normality, the methodology described by Garcia (1985) was applied. In the case the distribution did present itself as normal, the proposition of Costa, Seraphin and Zimmermann (2002) was used. At the end, a classification for the variation coefficient for each property of particleboards was generated.