Verification of Test Conditions to Determine the Compression Modulus of Elasticity of Wood

The obtaining the modulus of elasticity in compression of the wood with the use of dial gauges, fixed on opposite faces of the specimens, may lead to deformation values and consequently to different elastic modulus as a function of the faces chosen for its attachment, being the timber an anisotropic material. This study aimed to evaluate the influence of two distinct positions for setting the dial gauges (A and B) in wood specimens tested in compression, using the assumptions of the test methods and calculation of the Brazilian standard ABNT NBR 7190: 1997. The woods evaluated in trials were the Pinus elliottii and Corymbia citriodora, being used seven specimens per species. A specimen was taken to the rupture, obtained the values of the maximum stress and strain (references) needed to obtain the elastic moduli of the six remaining specimens per species, certain non-destructively (two tests per piece). The results of analysis of variance revealed the equivalence between modules elasticity in compression parallel to the grain for both wood species investigated, resulting not significantly arrangement of dial gaugesto determine the properties of stiffness. However, the anisotropy of wood, these results cannot be extrapolated to other woods of the same or d ifferent species, justifying the setting of dial gauges in two different positions, allowing for judging whether or not the equivalence between the modulus of elasticity.


Introduction
The wood itself as one of the oldest building materials, being used mainly because of its availability in nature, ease of handling, manufacturing and excellent relationship weight/strength [1][2][3].
The timber was presented as a cellular material, produced by a mechanism of continuous growth of plants. There are several species of trees throughout the world, but with all common features such as a cellular structure with an arrangement in the form of concentric rings, which ensures orthotropic mechanical propert ies of wood, directly related to its orientation relative to the main axis [4].
Chemical and mechanical propert ies can d iffer for the same species of wood acco rding to the locat ion o f their extraction . Other parameters such as climate and so il cond itions can affect th e g ro wth o f th e tree, d irectly influencingtheir propert ies. Moreover, factors such as the presence of us, opening cracks during drying and inclination of the fibers cause the strength of the woods have great variations [5][6][7].
According to [1], the mechanical properties of wood are dependent on the density, the percentage of juvenile wood, the width of the rings, the angle of the microfibrils, the amount of extract ives, moisture content, the intensity of insect attack, the type and location and number of nodes, among other factors, making it difficult to obtain need all their elastic parameters to be used in structural projects [8,9].
In order to enable the rational use of wood in structures mechanical tests are performed to obtain the equivalent properties, obtained from experiments and calculation procedures of standardized normat ive docu ments, such as the standard ABNT 7190 [10], widely used by engineers, architects and designers for material characterizat ion due to mechanical stresses and also for proper and safe design of structural elements.
Among the mechanical p roperties of materials used in the design of a structure highlights the modulus of elasticity (MOE), enabling the setting to provide displaced and deformations in structural components subjected to the action of the imposed loads (limit state).
Be of g reat interest for the knowledge of the modulus of elasticity in co mpression wood, allo wing the design of structural elements subject to compressive stresses, several studies have been conducted [11][12][13][14][15][16][17][18][19], in order to verify experimentally and numerically the influence of co mposition anatomical tissue timber (anisotropy) in physical, chemical and mechanical properties, as well as to characterize wood species not yet known.
This study aimed to investigate the influence of positioning of dial gauges (positions A and B) to determine the modulus of elasticity of wood in co mpression parallel to the grain, enabling determine possible differences between then.

Materials and Methods
The wood species used in this study were the Corymbia citriodora (Strength class C40) and Pinus elliottii (Strength class C30), made seven specimens per type of timber to perform in co mpression test [10], extracted fro m different parts of a batch considered homogeneous, with mo isture content near 12%, as established by the Brazilian standard [10].
The specimens were manufactured with square cross section of 5.0cm and 15cm of length [10], and are free of defects. The dimensions of the sides of the specimens were performed with a caliper accurate to 0.1 mm.
The dial gaugeswere fixed in two different positions, A and B, as illustrated in Figure 1, t wo bending tests were performed in the same specimen perwood species. Of each species, one of seven specimens was taken to rupture, allo wing discover tensions (σ) and strain (ε) for the 10% and 50% of the ma ximu m stresses and strains, used to determine the modulus of elasticity (Equation 1) in the other specimens, as required by the Brazilian standard [10]. To check the statistical equivalence between the modulus of elasticity for the two species of wood was used analysis of variance (A NOVA), performed as an aid software Min itab ® version 14.

Results
Tables 1 and 2 present the descriptive statistics related to the modulus of elasticity (MOE-A,-B MOE) in co mpression parallel to the grain o fCorymbia citriodora and Pinus elliottiiwood respectively, obtained with the use of dial gauges positioned on the faces A and B (Figure 1), X m is the arith metic mean, SD the standard deviation and CV the variation coefficient of specimens.  Figure 2 shows the normality graphs of the modulus of elasticity ofCorymbia citriodora and Pinus elliottiiwood respectively. The P-values of normality tests of Anderson-Darling ( Figure 2) on the modulus of elasticity for the Corymbia citriodora (0.161) and Pinus elliottii (0.513) woods were both higher than 0.05, p roving to be normal the data distribution [21]. Table 3 shows the results of the ANOVA factor, position of the dial gauge to obtain the modulus of elasticity (MOE-A; EOM B).

Conclusions
By the results of the analysis of variance was verified statistical equivalence between the modulus of elasticity of both wood species, revealing, for the specimens tested, not significant position of dial gauges in the calculation of the modulus of elasticity. As the wood an anisotropic material (orthotropic), the results obtained in this study cannot be extrapolated to the same wood species or different species, imply ing the use of d ial gauges in two different positions in the specimen, enabling assess equivalence or not of the elastic moduli obtained.