Eur. J. Wood Prod. (2011) 69: 183–192 DOI 10.1007/s00107-010-0415-0
O R I G I NA L S O R I G I NA L A R B E I T E N
Material properties and nondestructive evaluation of laminated veneer lumber (LVL) made from Pinus oocarpa and P. kesiya Frederico de Souza · C.H.S. Del Menezzi · Geraldo Bortoletto Júnior
Received: 26 April 2009 / Published online: 11 February 2010 © Springer-Verlag 2010
Abstract This study aimed at evaluating the mechanical, physical and biological properties of laminated veneer lumber (LVL) made from Pinus oocarpa Schiede ex Schltdl (PO) and Pinus kesiya Royle ex Gordon (PK) and at providing a nondestructive characterization thereof. Four PO and four PK LVL boards from 22 randomly selected 2-mm thickness veneers were produced according to the following characteristics: phenol-formaldehyde (190 g/m2 ), hot-pressing at 150°C for 45 min and 2.8 N/mm2 of specific pressure. After board production, nondestructive evaluation was conducted, and stress wave velocity (v0 ) and dynamic modulus of elasticity (EMd ) were determined. The following mechanical and physical properties were then evaluated: static bending modulus of elasticity (EM ), modulus of rupture (fM ), compression strength parallel to grain (fc,0 ), shear strength parallel to glue-line (fv,0 ), shear strength perpendicular to glue-line (fv,90 ), thickness swelling (TS), water absorption (WA), and permanent thickness swelling (PTS) for 2, 24, and 96-hour of water immersion. Biological property was also evaluated by measuring the weight loss by Trametes versicolor (Linnaeus ex Fries) Pilát (white-rot) and Gloeophyllum trabeum (Persoon ex Fries.) Murrill (brown-rot).
F. de Souza Institute of Agrarian Sciences, Integrated Faculties of Mineiros, FIMES, 75830-000 Mineiros, GO, Brazil C.H.S. Del Menezzi () Department of Forest Engineering, Faculty of Technology, University of Brasilia, UnB, Campus Darcy Ribeiro, PO Box 04357, 70919-970 Brasilia, DF, Brazil e-mail:
[email protected] G. Bortoletto Júnior Department of Forest Sciences, Luiz de Queiroz College of Agriculture, University of São Paulo USP, Av. Pádua Dias 11, 13418-900 Piracicaba, SP, Brazil
After hot-pressing, no bubbles, delamination nor warping were observed for both species. In general, PK boards presented higher mechanical properties: EM , EMd , fM , fc,0 whereas PO boards were dimensionally more stable, with lower values of WA, TS and PTS in the 2, 24, and 96-hour immersion periods. Board density, fv,0 , fv,90 and rot weight loss were statistically equal for PO and PK LVL. The prediction of flexural properties of consolidated LVL by the nondestructive method used was not very efficient, and the fitted models presented lower predictability. Materialeigenschaften von Furnierschichtholz aus Pinus oocarpa und Pinus kesiya und deren zerstörungsfreie Prüfung Zusammenfassung Ziel dieser Studie war die Bestimmung der mechanischen, physikalischen und biologischen Eigenschaften von Furnierschichtholz (LVL) aus Pinus oocarpa Schiede ex Schltdl (PO) und Pinus kesiya Royle ex Gordon (PK) sowie deren zerstörungsfreie Prüfung. Vier PO und vier PK Platten aus 22 zufällig ausgewählten 2 mm dicken Furnieren wurden wie folgt hergestellt: Phenolformaldehydharz (190 g/m2 ), Presstemperatur 150 °C, Pressdauer 45 Minuten, Pressdruck 2,8 N/mm2 . Nach der Herstellung der Platten wurde diese zerstörungsfrei geprüft und die Spannungswellengeschwindigkeit (v0 ) sowie der dynamische E-Modul (EMd ) wurden bestimmt. Anschließend wurden die folgenden mechanischen und physikalischen Eigenschaften bestimmt: statischer BiegeE-Modul (EM ), Biegefestigkeit (fM ), Druckfestigkeit in Faserrichtung (fc,0 ), Scherfestigkeit in Richtung der Klebstofffuge (fv,0 ), Scherfestigkeit rechtwinklig zur Klebstofffuge (fv,90 ), Dickenquellung (TS), Wasseraufnahme (WA) und permanente Dickenquellung (PTS) bei 2-, 24- und 96stündiger Wasserlagerung. Daneben wurde die natürliche
184
Dauerhaftigkeit mittels Messung des Masseverlustes durch Trametes versicolor (Linnaeus ex Fries) Pilat (Weißfäulepilz) und Gloeophyllum trabeum (Persoon ex Fries.) Murrill (Braunfäulepilz) bestimmt. Nach dem Heißpressen zeigten sich bei beiden Holzarten keinerlei Blasen, Delaminierungen oder Verwerfungen. Generell hatten die PK-Platten bessere mechanische Eigenschaften EM , EMd , fM , fc,0 , während die PO-Platten eine bessere Dimensionsstabilität mit niedrigeren WA-, TS- und PTS-Werten nach 2-, 24- und 96-stündiger Wasserlagerung aufwiesen. Die Plattendichte, fv,0 , fv,90 , sowie der Masseverlust von PO- und von PKFurnierschichtholz unterschieden sich statistisch nicht. Die Vorhersage der Biegeeigenschaften von Furnierschichtholz mittels des verwendeten zerstörungsfreien Verfahrens war nicht sehr effizient und die angepassten Modelle ergaben nur eine geringe Vorhersagbarkeit.
1 Introduction Laminated veneer lumber (LVL) is a structural board made from lumber veneer, with a thickness ranging between 0.3 and 6.3 mm (Sellers 1985) and arranged in the same grain direction (Burdurlu et al. 2007; Gungor et al. 2006; Carvalho et al. 2004). The use of LVL boards was largely encouraged in the 1940s for manufacturing aircraft structural components, as the material is both light and resistant. Currently, LVL boards are of great structural use due to their high mechanical strength, great dimensional stability and ability to receive preservative treatment. Therefore, LVL can show strength and stiffness higher than that of the wood of origin (Kamala et al. 1999). The use of LVL boards has already been consolidated on the American, European and Japanese markets, where they are used for manufacturing furniture, door sills, window frames and stair steps (Eckelman 1993), bridges (Ritter et al. 1996), and I-beam flanges. In general, LVL boards display very good mechanical properties meaning viability for structural purposes. LVL also shows a rather uniform density throughout the board and, in general, the density ranges from 0.50 to 0.70 g/cm3 depending on the species, adhesive and other production parameters used (Shukla and Kamdem 2008; Kamala et al. 1999). The research carried out indicates that LVL boards can display modulus of elasticity in static bending above 10,000 N/mm2 , depending on the wood species, veneer thickness, number of veneers, and adhesive used (Killic et al. 2006; Aydin et al. 2004; Lee et al. 1999; Carvalho et al. 2004). On the other hand, the modulus of rupture in static bending can exceed 90 N/mm2 , and values between 48.3– 86.2 N/mm2 were observed in compression strength parallel to grain (Pio 2002; Kamala et al. 1999). The shear strength parallel to glue-line displayed values around 9 N/mm2 and
Eur. J. Wood Prod. (2011) 69: 183–192
around 10 N/mm2 in the case of perpendicular orientation (Gabriel 2007; Kamala et al. 1999). According to Pio (2002), values close to 16 N/mm2 were observed for these properties when 20 year-old Eucalyptus grandis veneers were used to manufacture LVL using the best nondestructively pre-classified veneers. LVL boards can normally be exposed to great humidity and temperature variations which usually impart susceptibility to decay. However, few studies have been performed to evaluate the durability of LVL against either fungi/termites or weathering. According to Nzokou et al. (2005), attempts were made to introduce veneers made from durable species in order to improve the natural resistance of LVL boards. Hayashi et al. (2002) studied the effect of 6-year outdoor exposure on physical properties of LVL made from seven wood species. The retentions of EM and compression strength were 78 and 77%, respectively, but heavy decay by brown-rot fungi (Pseudomerulius aureus (Fr.) Julich) took place, mainly on Abies grandis LVL samples. Three years later, Hayashi et al. (2005) observed that the decay process had achieved the Pinus radiata LVL samples. Nondestructive evaluation (NDE) has been used for determining the properties of the veneers and then enabling the production of boards according to the stiffness grades or assessing the properties of already consolidated boards. Nondestructive methods are a set of techniques intended for immediate characterization of materials without the need for taking samples and thus ensuring full subsequent use of the material (Ross et al. 1998). In Brazil, some of these techniques have been extensively adopted at scientific research level, such as the use of ultrasound (Gonçalves and Costa 2008; Miná et al. 2004; Oliveira and Sales 2002), stress wave (Ferraz et al. 2009; Souza et al. 2008; Del Menezzi et al. 2008a; Gabriel 2007) and transverse vibration (Teles 2009; Targa et al. 2005; Calil and Miná 2003). Brazil does not have an LVL industry yet, but research at laboratory scale has been conducted to evaluate the raw material from rapidly growing forests with species of the Pinus and Eucalyptus genus for production on a large scale (Gabriel 2007; Pio 2002; Matos 1997). According to Brazilian Association of Forest Plantations Producers (2008) there are 1,800,000 ha of Pinus plantation distributed mainly in the Southern, Southeastern and Central-Western regions of the country. Because of the growth potential even on low fertility soils, Pinus oocarpa and P. kesiya are very appropriate species for plantations in those regions (Kageyama et al. 1977). However, to date commercial plantations in the hottest regions of Brazil have been established for P. oocarpa alone, with a mean annual increment around 15–25 m3 /ha/year. On the other hand, for P. kesiya only provenance trials have been done, and commercial plantations have not been established yet.
Eur. J. Wood Prod. (2011) 69: 183–192
In this context, the present work aimed at evaluating the material properties of LVL as well as evaluating nondestructively LVL made from P. oocarpa and P. kesiya plantation species which have demonstrated very good silvicultural development in Brazil.
2 Material and methods 2.1 Board manufacturing Wood logs of the Pinus oocarpa Schiede ex Schltdl (PO) and Pinus kesiya Royle ex Gordon (PK) species were obtained at ages 28 and 30, respectively, from the experimental plantation site in the Monte Olimpo region, at the Luiz de Queiroz College of Agriculture—ESALQ/USP located in the municipality of Piracicaba-SP/Brazil. The logs were rotary peeled into 2 mm veneers as described by Bortoletto Júnior (2008). An overall number of 176 two mm thick and 1000 mm long veneers were obtained with 88 for each species. The veneers were selected on the basis of visual quality grades according to the parameters established by the Brazilian Association of Processed Wood and Plywood and presented by Santos (2008). Only veneers classified as A, B, C, and C+ were used for LVL manufacturing. Therefore, veneers with too many knots (D class) or cracks were discarded as well as those showing excessive roughness. Veneers were cut parallel to grain into 230 mm sections and subsequently dried to reach 6–8% moisture content in a press at 110°C and 0.7 N/mm2 for 15 min. The adhesive composition was: phenol-formaldehyde resin (100 parts), wheat flour (5 parts), coconut husk flour (8 parts), and water (5 parts). The LVL boards were consolidated according to the methodology suggested by Carvalho et al. (2004), as follows: 190 g/m2 adhesive spread by a roller gluing machine in 22 randomly selected veneers (230 × 1000 mm2 ), hot pressing at 150°C for 45 min and 2.8 N/mm2 specific pressure. After cooling, the boards were ripped into five 40 × 40 × 960 mm3 beams (width × thickness × length), resulting in 20 PO and 20 PK beams. Finally, the beams were placed in a conditioned room (20°C; 65% RH) until they reached equilibrium moisture content. 2.2 Nondestructive evaluation All 20 beams of each species measuring 40 × 40 × 960 mm3 (width × thickness × length) were evaluated lengthwise by the Metriguard Stress Wave Timer model 239A, in both flatwise and edgewise positions. The equipment has an impact pendulum that generates a stress wave which propagates through the beam. Two accelerometers are connected to the beam to measure the stress wave transit time (t), which is the time required for the wave to travel between them. When
185
the beam is hit, the first accelerometer triggers, whereas the second one stops the timer of the equipment when the wave reached it and the transit time is displayed. Each beam was hit three times. The mean value of the transit time was used to determine the stress wave velocity (v0 ) and then the dynamic modulus of elasticity (EMd ) according to (1) and (2) adapted to the international unit system (Del Menezzi et al. 2008a). L , t × 10−6 v2 × D EMd = o × 10−5 , g vo =
(1) (2)
where: vo : stress wave velocity, m/s; EMd : dynamic modulus of elasticity, N/mm2 ; D: density, kg/m3 ; L: length of the beam, m; t: transit time, µs; g: acceleration of gravity, 9.8 m/s2 . 2.3 Mechanical properties After nondestructive evaluation, the mechanical properties were determined using a universal testing machine at the facilities of the Forest Products Laboratory (LPF), Brazilian Forest Service (SFB). Both the modulus of elasticity (EM ) and the modulus of rupture (fM ) were obtained according to ASTM D 5456/06 (ASTM 2006) and ASTM D 198/99 (ASTM 1999a) standards for third-point test. Fifteen beams of each species were tested in flatwise position (loading perpendicular to glue line) and 5 beams in edgewise position (loading parallel to glue line). The greatest span/depth ratio allowed (21 times) was used, which meant a total span of 840 mm. The distance from the reaction point to nearest load point was 280 mm. The parallel compression strength (fc,0 ) was performed according to ASTM D 5456/06 (ASTM 2006) and ASTM D 198/99 (ASTM 1999a). Twenty samples measuring 40 × 40 × 180 mm3 (width × thickness × length) were obtained for each species. The length of the fc,0 sample was such that the ratio length/radius of gyration was 17, the maximum allowed. The shear strength parallel (fv,0 ) and perpendicular (fv,90 ) to glue line were conducted according to ASTM D 5456/06 (ASTM 2006) and ASTM D 1037/99 (ASTM 1999b). Three samples per board were obtained, which resulted in 12 samples per species measuring 40 × 40 × 63.5 mm3 (width × thickness × length). For these tests, sample sizes were adapted from the sizes provided in ASTM D 1037/99 (ASTM 1999b), due to dimensional limitations of the available material. 2.4 Physical properties The sample for determining density at equilibrium moisture content (D), water absorption (WA), thickness swelling (TS) and permanent thickness swelling (PTS) measured
186
Eur. J. Wood Prod. (2011) 69: 183–192
Table 1 Dynamic modulus of elasticity and stress wave velocity of Pinus oocarpa and P. kesiya LVL boards Tab. 1 Dynamischer E-Modul und Spannungswellengeschwindigkeit von Pinus oocarpa und Pinus kesiya Furnierschichtholzplatten Property
Position
Pinus oocarpa
Pinus kesiya
Dynamic modulus of elasticity (N/mm2 )
Flatwise**
16,512
17,476
(0.721)
(0.724)
15,895
16,872
(0.752)
(0.708)
Flatwise**
4775
4946
Edgewise**
4687
4860
Edgewise**
Stress wave velocity (m/s)
** Difference
between species statistically significant at 1% level; values between parenthesis are the mean density (g/cm3 ) of the samples tested
40 × 40 × 40 mm3 (width × thickness × length). Three samples per board were obtained, which resulted in 12 samples per species. WA, TS and PTS tests were conducted according to ASTM D 5456/06 (ASTM 2006) and ASTM D 1037/99 (ASTM 1999b). WA and TS testing were done for 2, 24, and 96-hour periods with the same samples. PTS was assessed based on the percentage difference between the measurement taken at the 24-hour period of water immersion and measurement after sample reconditioning (20°C; 65% RH). Afterwards, the WA/TS samples were oven-dried (103 ± 2°C) in order to determine the equilibrium moisture content (EMC) of the material. 2.5 Biological properties The biological properties of LVL boards were determined according to ASTM D 2017 (ASTM 2005) (soil block-test). Samples measuring 25 × 25 × 9 mm3 (width × length × thickness) were taken from the ends of the beams used for the static bending test. Six samples per board were obtained totaling 24 samples for each species. The samples were obtained in flatwise position, meaning that the glue lines were parallel to the sample surface. Inoculate of Trametes versicolor (Linnaeus ex Fries), Pilát (white-rot), and Gloeophyllum trabeum (Persoon ex Fries.) Murrill (brown-rot) fungi previously cultivated in liquid Malt-Distilled Water medium were used. A supporting plate made of Cecropia spp. and Pinus spp. wood was used to promote initial fungus growth. Glass flasks were autoclaved at 121°C for 60 min and placed in an incubating chamber with humidity at around 70 ± 4% and temperature ranging from 26 ± 2°C. Weight loss (WL) percentage was assessed for both fungi types after a 12-week period of incubation. 2.6 Statistic analysis To evaluate the effect of the wood species on mechanical properties (EM , fM , fc,0 , fv,0 , fv,90 ), physical properties (D, WA, TS, PTS), biological properties (WL-brown and
white-rot) and nondestructive evaluation (v0 , EMd ) a one way ANOVA was run. The results of the nondestructive analysis were used for predicting the flexural properties in flatwise position by using the simple linear regression model (y = a + bx); with EMd or v0 as an independent variable (x) and EM or fM as dependent variables (y).
3 Results and discussion 3.1 Nondestructive properties The results of the nondestructive properties are presented in Table 1. It is clear that PK board presented higher values of EMd and v0 than PO in both positions studied. Several studies have suggested that an increase in material continuity, i.e., a decrease in empty spaces, increases stress wave velocity on the board (Han et al. 2006; Brashaw et al. 2004). Han et al. (2006) evaluated several types of wood based boards and the following stress wave velocities were obtained: ≈1870 m/s (particleboard), ≈2770 m/s (OSB) and ≈4300 m/s (plywood). According to these data, it can be inferred that the higher the homogeneity or continuity of the material, the higher the stress wave velocity. In this context, probably the anatomical arrangement of the PK wood provided greater tissue continuity inside the veneers or less empty spaces, which determined a greater v0 in these boards. Gabriel (2007) evaluated the wave velocity of LVL board made from Pinus caribaea (var. caribaea, bahamensis and hondurensis) and Pinus oocarpa. The values observed were between 4470–5158 m/s. It can also be observed that EMd and v0 were higher in flatwise position for both species. This phenomenon occurred probably due to the LVL production process, where high temperatures and pressures tend to lead to densification of surface layers, thus improving the stress wave velocity in flatwise position (Souza et al. 2008; Carvalho et al. 2004). In general, flexural properties of LVL board obtained nondestructively by acoustic methods have been reported to present higher values than those obtained in static bending testing (Lee et al.
Eur. J. Wood Prod. (2011) 69: 183–192
187
Table 2 Mechanical properties of Pinus oocarpa and P. kesiya LVL boards Tab. 2 Mechanische Eigenschaften von Pinus oocarpa und Pinus kesiya Furnierschichtholzplatten Mechanical properties Static bending
Modulus of elasticity (N/mm2 )
Position
Pinus oocarpa
Pinus kesiya
Flatwise**
16,199
17,087
Edgewise** Modulus of rupture (N/mm2 )
(0.721)
(0.724)
16,160
15,062
(0.752)
(0.708)
Flatwise**
85
94
Edgewisen.s
93
92 57
Compression parallel to grain (N/mm2 )**
–
55
Shear parallel to glue line (N/mm2 )n.s
–
8.7
9.8
Shear perpendicular to glue line (N/mm2 )n.s
–
12.4
12.5
** Difference
between species statistically significant at 1% level and density (g/cm3 ) of the samples tested
2001; Pu and Tang 1997). In the present work, the same also happened as will be shown later. 3.2 Mechanical and physical properties Table 2 shows the results observed for the mechanical properties of PO and PK LVL boards. In general, PK boards presented higher flexural and compressive strength properties than PO boards, but no effect was observed on shear properties. As argued above, the presence of less empty spaces in the PK veneer might explain the higher EM and fM results in PK boards, since the density of the boards of the two species was virtually the same. The comparison between loading positions (edgewise and flatwise) for each species did not reveal a clear trend. For PK boards the EM values in flatwise position were higher than those in edgewise position, while for fM the difference was not significant. On the other hand, for PO boards EM values were slightly higher in flatwise position, but the opposite behavior was observed for fM values. During the manufacturing process the LVL billet is pressed perpendicular to glue line, i.e., in the flatwise position. This way, because of the pressure and temperature, some densification of the surface veneers is expected. According to Carvalho et al. (2004), this phenomenon can improve the flexural properties of the board when tested in flatwise position. They noticed that EM in LVL boards manufactured from Eucalyptus urophylla/E. grandis hybrid showed a 13,792 N/mm2 value in flatwise position and 12,917 N/mm2 in edgewise position. As for fM , the values observed were 59 and 55 N/mm2 , respectively. Gabriel (2007) observed that LVL boards manufactured from Pinus caribeae (var. caribeae, bahamensis and hondurensis) and Pinus oocarpa veneers presented values for EM in flatwise position between 13,512–14,555 N/mm2 and between 80–114 N/mm2 for fM . As for edgewise position, values were between 11,555–18,144 N/mm2
n.s statistically
non-significant; values between parenthesis are the mean
and 68–101 N/mm2 , respectively. Burdurlu et al. (2007) evaluated the effect of the veneer organization and the loading direction on the flexural properties of LVL made from eight veneer combinations of Fagus orientalis and Populus nigra. According to the results, the fM values were slightly higher in edgewise position, while for EM an opposite behavior was observed. Information concerning flexural properties of LVL board made from Pinus wood species is not so common. Özçifçi (2007) studied the flexural properties of phenol-formaldehyde (PF) bonded LVL made from P. brutia. The 5-veneer LVL presented fM value of 110.1 N/mm2 and EM value of 14,100 N/mm2 . Çolak et al. (2004) produced LVL from P. sylvestris bonded with four types of adhesive. For PF bonded LVL they observed the following values for flexural properties: 100.5 N/mm2 (fM ) and 6,264 N/mm2 (EM ). For LVL made from P. taeda, Pu and Tang (1997) obtained EM values between 9,340–12,300 N/mm2 depending on the veneer grade. Biblis (1996) observed that P. taeda LVL manufactured from 20-year-old plantation trees presented values of EM ranging from 8,874 N/mm2 to 9,956 N/mm2 and fM values ranging from 34.4 N/mm2 to 38.1 N/mm2 depending on the number of veneers. As seen, the values of the flexural properties observed in this present work are within the range usually obtained in other researches for LVL made from Pinus wood species. In this context, it can be stated that both species presented suitable mechanical properties, which means that they can be used for manufacturing engineered wood products. Recently, Santos et al. (2009) showed that either OSB or plywood webbed I-beam flanged with P. kesiya LVL board presented appropriate bending strength and stiffness. Table 3 shows the results observed for the physical properties of PO and PK LVL boards. It can be seen that PO boards presented better dimensional stability than PK. It is known that high thickness swelling is expected from high
188
Eur. J. Wood Prod. (2011) 69: 183–192
Table 3 Physical properties of Pinus oocarpa and P. kesiya LVL boards Tab. 3 Physikalische Eigenschaften von Pinus oocarpa und Pinus kesiya Furnierschichtholzplatten Physical properties
Immersion period (h)
Density (g/cm3 )n.s
–
Equilibrium moisture content (%)n.s
–
11.7
11.5
Water absorption (%)
2**
12.45
29.34
(0.723)
(0.718)
24**
24.44
40.25
96**
34.50
45.52
2**
1.54
3.53
24**
4.05
4.86
96n.s
5.30
5.01
24**
1.96
3.08
Thickness swelling (%)
Permanent thickness swelling (%) ** Difference
between species statistically significant at 1% level and density (g/cm3 ) of the samples tested
water absorption. This behavior was observed in relation to PK boards. It can also be seen that thickness swelling in PK boards occurred faster than in PO: the 2h TS value for PK boards was more than two times the one for PO. It probably happened due to the faster release of the compression stress for PK board in comparison to PO board, which can be inferred by observing the permanent thickness swelling (PTS) values in Table 3. The PTS value of PK board was 57% higher than that observed in PO board, which indicates that more compression stresses were required during hot-pressing of PK board. According to Del Menezzi et al. (2009), when the boards are soaked in water the compression stresses are released and, in conjunction with natural wood swelling, determines the TS intensity. If the board is dried, the wood tends to return to its original thickness but the TS relative to the compression stress does not, and thickness of the board remains higher than before the test. Shukla and Kamdem (2008) observed that water absorption after 24 hours of immersion of PVAc bonded LVL was 60% for silver maple LVL, 40% for yellow poplar, and 55% for aspen. After the same time period, thickness swelling varied between 3–5% in the LVL of the three species studied. Recently, the same authors (Shukla and Kamdem 2009) observed a direct linear relationship between water absorption and thickness swelling for LVL made from Liriodendron tulipifera. Pio (2002) noticed that in Eucalyptus grandis LVL boards made with veneers from 15 year-old trees, the 2-hour TS ranged from 4.42 to 11.33% while for 20 year-old trees it ranged from 6.59 to 14.70%. As for the 24-hour immersion period, the TS observed was between 6.31–13.60% and between 6.95–12.44% for the LVL manufactured with veneers from 15 and 20 year-old trees, respectively.
Pinus oocarpa 0.717
n.s statistically
Pinus kesiya 0.712
non-significant; values between parenthesis are the mean
3.3 Prediction of flexural properties Table 4 presents the linear models fitted for predicting flexural properties of the LVL boards. For PO board it was not possible to fit any statically significant model to predict EM and fM . On the other hand, for PK board fitted models were able to predict the EM , but not the fM . Figure 1 presents the data dispersion for these models. The models presented lowmedium coefficient of determination (
