Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Properties comparison of Guadua and Moso bamboo oriented strand board with Aspen strands in the core. Zhang, Kunqian (Polo) 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_september_zhang_kunqian.pdf [ 44.47MB ]
Metadata
JSON: 24-1.0307465.json
JSON-LD: 24-1.0307465-ld.json
RDF/XML (Pretty): 24-1.0307465-rdf.xml
RDF/JSON: 24-1.0307465-rdf.json
Turtle: 24-1.0307465-turtle.txt
N-Triples: 24-1.0307465-rdf-ntriples.txt
Original Record: 24-1.0307465-source.json
Full Text
24-1.0307465-fulltext.txt
Citation
24-1.0307465.ris

Full Text

    PROPERTIES COMPARISON OF GUADUA AND MOSO BAMBOO ORIENTED STRAND BOARD WITH ASPEN STRANDS IN THE CORE by  Kunqian (Polo) Zhang  B.Sc. The University of British Columbia, 2014  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Forestry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   August 2016  © Kunqian (Polo) Zhang, 2016 ii  Abstract Bamboo (Poaceael/Graminaceae) has great potential for use in improving the properties of wood-based strand composite building materials. In previous work it has been shown that replacement of aspen surface strands with Moso bamboo (Phyllostachys pubescens Mazel) strands significantly improves the strength and water resistance of oriented strand board (OSB) of the same density made from Aspen. Guadua (Guadua andustifolia Kunth) is one of the most commercially cultivated and used timber bamboo genera in Latin America. In this study, three experiments were designed.  Six sets of 6 three-layer OSB (737 x 737 x 11.1 mm) were made with bamboo strands in the face layers and Aspen strands in the core layer. Measured board properties included internal bond, flexural properties (modulus of rupture, MOR; and modulus of elasticity, MOE), and water resistance (% thickness swell, TS; and % water absorption, WA). The 50% Guadua -50% Aspen boards (type GM) was compared with 50% Moso -50% Aspen boards (type MM) to examine the effects of bamboo species. Guadua hybrid OSB had a weaker IB strength and a higher MOE in the parallel direction. No other significant difference was found.   To examine the effect of reducing board density down to an acceptable level, three board types were compared. 1) 50% Moso - 50% Aspen boards (type MM) with target density of 760 kg/m3,  2) 25% Moso - 75% Aspen boards (type ML1) with target density of 720 kg/m3, and 3) 25% Moso - 75% Aspen boards (type ML2) with lower target density of 628 kg/m3. The lowest density group had the lowest mechanical properties and water resistance ability but met the Canadian Standards Association (CSA) standards for industrial OSB.  iii  Another two board types were designed to examine the effect of the nodes on Guadua OSB products’ properties. 50% Guadua Node – 50% Aspen boards (type GN) showed weaker IB strength and weaker flexural properties than 50% Guadua Internode – 50% Aspen boards (type GI).  iv  Preface A version of section 4.2 has been published. Zhang, K., K. Semple, and G. Smith (2015). Tailoring the addition of Moso strands to enhance the properties of OSB but reducing board density. Proceedings of the 58th International Convention of Society of Wood Society and Technology June 7-12, 2015. 788:197-205. With Kate Semple’s help on making the boards, I conducted all the testing and wrote all of the manuscript under the supervision of Greg Smith.   v  Table of Contents Abstract .................................................................................................................................... ii Preface ...................................................................................................................................... iv Table of Contents ...................................................................................................................... v List of Tables ......................................................................................................................... viii List of Figures............................................................................................................................ x List of Abbreviations ...............................................................................................................xii Glossary ................................................................................................................................. xiii Acknowledgements ................................................................................................................. xiv Chapter 1: Introduction ............................................................................................................ 1 1.1 Background .................................................................................................................1 1.2 Rationale .....................................................................................................................2 1.3 Hypotheses ..................................................................................................................4 1.4 Approach .....................................................................................................................4 1.5 Structure of Thesis .......................................................................................................5 Chapter 2: Literature Review ................................................................................................... 6 2.1 Bamboo .......................................................................................................................6 2.1.1 Moso Bamboo ....................................................................................................... 10 2.1.2 Guadua Bamboo .................................................................................................... 11 2.1.3 Differences and Comparison between Moso and Guadua ....................................... 14 2.2 Bamboo Composites Products.................................................................................... 16 2.2.1 Wood OSB Products .............................................................................................. 17 2.2.2 Bamboo OSB Products .......................................................................................... 18 vi  2.2.3 Three-layer Hybrid Bamboo OSB .......................................................................... 19 Chapter 3: Materials and Methods ........................................................................................ 20 3.1 Culm Feedstock ......................................................................................................... 20 3.2 Culm Breakdown ....................................................................................................... 21 3.3 Stranding ................................................................................................................... 23 3.4 Strands Screening ...................................................................................................... 26 3.5 Board Fabrication and Experiment Design ................................................................. 27 3.5.1 Experimental Designs ............................................................................................ 27 3.5.2 Board Fabrication Design....................................................................................... 29 3.5.3 Blending Strands with Resin .................................................................................. 30 3.5.4 Hot Pressing .......................................................................................................... 32 3.6 Specimens Cutting ..................................................................................................... 34 3.7 Test Methods ............................................................................................................. 36 3.7.1 Internal Bonding Test ............................................................................................ 36 3.7.2 Flexural Property Test ............................................................................................ 37 3.7.3 Thickness Swelling and Water Absorption ............................................................. 39 Chapter 4: Results and Discussion ......................................................................................... 41 4.1 Experiment 1: Comparison of Guadua and Moso Boards ........................................... 41 4.1.1 Thickness and Density ........................................................................................... 41 4.1.2 Internal Bonding Strength ...................................................................................... 42 4.1.3 Flexural Properties ................................................................................................. 44 4.1.4 Water Absorption................................................................................................... 45 4.2 Experiment 2: Density Effects on Mechanical Properties ........................................... 49 vii  4.2.1 Thickness and Density ........................................................................................... 50 4.2.2 Internal Bonding Strength ...................................................................................... 51 4.2.3 Flexural Properties ................................................................................................. 52 4.2.4 Water Absorption................................................................................................... 54 4.3 Experiment 3: Node Effect on Guadua-Aspen Hybrid OSB ....................................... 59 4.3.1 Thickness and Density ........................................................................................... 59 4.3.2 Internal Bonding Strength ...................................................................................... 60 4.3.3 Flexural Properties ................................................................................................. 61 4.3.4 Water Absorption................................................................................................... 62 Chapter 5: Conclusion and Future Work .............................................................................. 67 5.1 Difference between Moso and Guadua ....................................................................... 67 5.2 Lighter Bamboo OSB Met the CSA Standard ............................................................ 67 5.3 The Effects of Node on Guadua Bamboo OSB ........................................................... 68 5.4 Future Work .............................................................................................................. 68 References................................................................................................................................ 70 Appendices .............................................................................................................................. 78 Appendix A : Press Schedule................................................................................................. 78 Appendix B : Test Results ..................................................................................................... 81 Appendix C : Data Analysis in JMP .................................................................................... 129 Appendix D : Permission for reproduction for Figure 2, 3, 7................................................ 196 viii  List of Tables Table 1. Properties of Moso and Guadua polese used to produce strands ................................... 16 Table 2. Summary of experiment design per types..................................................................... 29 Table 3. Summary of experiment design per factors. ................................................................. 29 Table 4. Press sequence schedule (example of first 6 boards) .................................................... 32 Table 5. Production parameters ................................................................................................. 34 Table 6. Summary of specimens’ size and quantity.................................................................... 35 Table 7. Thickness and density for MM and GM ....................................................................... 42 Table 8. Results of flexural property test of GM and MM (perpendicular) ................................. 44 Table 9. Results of flexural property test of GM and MM (parallel)........................................... 45 Table 10. Results of thickness swelling (%) for GM and MM.................................................... 46 Table 11. Results of water absorption (w/%) for GM and MM .................................................. 47 Table 12. Experiment 2 boards types ......................................................................................... 50 Table 13. Means and standard deviation for thickness and density ............................................. 51 Table 14. Results of MOR of MM, ML1 and ML2 (perpendicular) ........................................... 52 Table 15. Results of MOE of MM, ML1 and ML2 (perpendicular) ............................................ 53 Table 16. Results of MOR of MM, ML1 and ML2 (parallel) ..................................................... 53 Table 17. Results of MOE of MM, ML1 and ML2 (parallel) ..................................................... 53 Table 18. Comparision between MLow boards and CSA standards for flexural properties......... 54 Table 19. Results of thickness swelling (%) for MM, ML1 and ML2 ........................................ 55 Table 20. Results of water absorption (w/%) for MM, ML1 and ML2 ....................................... 56 Table 21. Comparison between experiment results and CSA standard ....................................... 58 ix  Table 22. Summary of comparisons between MLow group results and CSA standards. ............. 58 Table 23. Thickness and density of GI and GN .......................................................................... 60 Table 24. Flexural property test results of GI and GN (perpendicular) ....................................... 62 Table 25. Flexural property test results of GI and GN (parallel) ................................................. 62 Table 26. Results of thickness swelling (%) for GI and GN ....................................................... 63 Table 27. Results of water absorption (w/%) for GI and GN ...................................................... 65   x  List of Figures Figure 1. Nodes on Moso Bamboo ..............................................................................................3 Figure 2. Phyllostachys sp. (Clark, 2006)  ...................................................................................7 Figure 3. Arthrostylidium sarmentosum(Clark, 2006) ..................................................................7 Figure 4. Relationship of modulus-density for materials (Wegst et al., 1993)............................. 10 Figure 5. Sympodial Bamboo (Rivière & Rivière, 1878) ........................................................... 11 Figure 6. Monopodial Bamboo (Rivière & Rivière, 1878) ......................................................... 11 Figure 7. Cured and dried crushed bamboo mats (Schroder, 2014) ............................................ 13 Figure 8. Appearance of the node plates of Moso and Guadua ................................................... 15 Figure 9. Comparison between Moso and Guadua culm(de Vos, 2010) ..................................... 15 Figure 10. OSB lay up Canadian Standards Association classifications ..................................... 18 Figure 11. Moso poles purchased cut from Canada Bamboo World ........................................... 20 Figure 12. Short billets from Guadua poles................................................................................ 21 Figure 13. Using dremel saw (left) and belt sander (right) to remove Guadua node. .................. 22 Figure 14. Top view of the strander feed box showing quarter cut culms ................................... 23 Figure 15. Top view of the strander feed box showing halve cut culms...................................... 24 Figure 16. View of disk strander showing the knife rotation direction ....................................... 25 Figure 17. Stranding culm pieces vertically ............................................................................... 25 Figure 18. Screening the strands ................................................................................................ 26 Figure 19. Blending system ....................................................................................................... 31 Figure 20. Hand orientation of bamboo strands with orienter ..................................................... 33 Figure 21. Cutting pattern with three different directions ........................................................... 34 Figure 22. Example of labeling.................................................................................................. 35 xi  Figure 23. Specimens in the condition room .............................................................................. 36 Figure 24. IB test machine......................................................................................................... 37 Figure 25. Flexural test machine ................................................................................................ 38 Figure 26. Details for thickness swelling test ............................................................................. 40 Figure 27. IB test results for GM and MM, n = 180 for each mean ............................................ 43 Figure 28. Results of 24h TS of GM and MM, n = 6 for each mean ........................................... 46 Figure 29. Results of 24h WA of GM and MM, n = 6 for each mean ......................................... 48 Figure 30. Results of IB test of MM, ML1 and ML2, n = 180 for each mean ............................. 52 Figure 31. Results of 24h TS of MM, ML1 and ML2, n = 6 for each mean ................................ 55 Figure 32. Results of 24h WA of MM, ML1 and ML2, n = 6 for each mean .............................. 57 Figure 33. IB test results of GI and GN, n = 180 for each mean ................................................. 61 Figure 34. Results of 24h TS of GI and GN, n = 6 for each mean .............................................. 64 Figure 35. Results of 24h WA of GI and GN, n = 6 for each mean ............................................ 65   xii  List of Abbreviations ANOVA – Analysis of Variance ASTM – American Society of Testing Materials COV – Coefficient of Variance CSA – Canadian Standards Association GLG – Glued Laminated Guadua bamboo  IB – Internal Bonding  MC – Moisture Content MOE – Modulus of Elasticity MOR – Modulus of Rupture PF – Phenol Formaldehyde TS – Thickness Swelling WA – Water Absorption  xiii  Glossary Bamboo Fibre refers to a distinct family of primary processed bamboo elements. Bamboo fibre usually refer to single bamboo fibre cells or an aggregation of multiple fibre cells (Liu et al., 2016). Many species of bamboo produce fibres similar in size to wood fibres and vascular wood-cells.  Bamboo Strand is a short sliver of bamboo. The thickness ranges from 0.5 mm to 0.8 mm and lit is longer than it is wide. In the case that bamboo strands are used to make Oriented Strand Board, the strands range from about 10 to 50 mm in width and 100 to 120 mm in length. Billet is the short section cut from the culm. Mostly, the culm could be cut to four billets depending on the length of the culm.  Culm, sometimes used alternately with ‘pole’ or ‘stem’, refers to the stem of the bamboo plant. The bamboo culm is further processed into smaller elements for engineered bamboo products. Halves, also known as Half-Split Culm or Half-Round Bamboo, refers to the largest form of a thick section of culm. Nodes are removed from these units using hand tools.     xiv  Acknowledgements First of all, I would like to offer my sincere gratitude to my supervisor Prof. Greg Smith for the continuous support of my M.Sc study and related research. His guidance advised me in all the time of research and writing of this thesis. Besides my supervisor, I would like to thank the rest of my committee: Prof. Simon Ellis, Prof. Thomas Tannert, for their insightful comments and those questions which encourage me to develop my research from various angles.  I offer my enduring thanks to Dr. Kate Semple, Dr. Shayesteh Haghdan, Solace Sam-Brew, Felix Böck, and Ingrid Tsai who helped me a lot with their patience, motivation, and immense knowledge.  My special thanks go to John Hoffmann and Gordon Chow, from FPInnovation, and George Lee, Vincent Leung, Lawrence Gunther and other faculty and staff at UBC, who gave access to the laboratory and research equipment. Without their precious support, it would not be possible to conduct this research. Last but not the least, I would like to thank my mother Ms. Yi Zhang for her unconditional love and generosity through all these years!    Chapter 1: Introduction 1.1 Background Bamboo (Poaceae/Graminaceae) is a fast growing giant grass that can be tougher than wood owing to its unique microscopic structure and chemical composition (X. Li, 2004). As a renewable woody biomass plant, bamboo is used to replace wood in construction and other fields in several countries across Asia, Africa and Latin America where bamboo grows natively (Liese, 1998; Peng et al., 2010).  This Master’s study is part of the Structural Bamboo Products (SBP) project funded by the G8 Multilateral Research Initiative sponsored by the United Nations and administered by NSERC.  The SBP project aims to develop green construction materials as an alternative to the energy-intensive nature and unsustainability of conventional construction materials such as steel and concrete.  Many composite products are made of bamboo such as laminated bamboo flooring, bamboo plywood-like panels, and bamboo scrimber. However the individual manufactures are small scale and very labour intensive. The processing has low product recovery from the culm, and uses more adhesive than comparable wood composites. Bamboo processing enterprises are economically marginal due to rising competition for culm supplies and cost of labour (Semple, 2015a). According to Mr. Li (Li, 2013), who is the plant manager of Chengfeng Bamboo Industry Co. Ltd  in Anji, China, the recovery from bamboo culm to laminated bamboo lumber was about 60% to 70%. Most waste is bamboo inner and outer wall layers. No record was found for a worldwide average recoveries for industrialized bamboo products. A V-grooving method was studied in the lab with a recoveries around 77% in Malaysia (Bakar et al., 2013). This 2  method took the advantage of the cylindrical shape of bamboo culm and made the outer circumference of the cylinder the same with the inner circumference by removing parts of the outer side in a series of V-shaped grooves. 1.2 Rationale Although both the physical and mechanical properties of bamboo OSB could meet CSA industry-level standards (Lee et al.1996), the density of bamboo composites is often too high to be a practical direct substitute for commodity OSB manufactured from wood. The great concentration of vascular bundles in the culm wall makes the bamboo dense and strong (Ghavami 2005). The strength of bamboo depends largely on the number of the vascular bundles (Lo et al. 2008). The density of the bamboo fibres is 800 kg/m3 making it very difficult to nail conventionally (Li & Shen, 2011). Moso is one of the most common bamboo species used as building materials and has, for decades, produced good quality, strand-based composites (Lee, Bai, & Peralta, 1996). However, the best known commercially cultivated genus in Latin America is Guadua. Among 38 known Guadua species, Guadua andustifolia Kunth, a sympodial bamboo, is the typically species used for timber products (Schroder, 2014). But there is no literature on strand based composites made from Guadua specie. There are a few studies on Guadua composites products. Correal and Ramirez (2010) from Columbia found optimal adhesive spread rates of 300 g/m2 for glued laminated Guadua bamboo among six different rates based on glue line tests. Archila et al. (2015) formed a novel composite flat sheet material using Guadua fibre and a set range of polymers such as thermoset polymers, 3  natural latex, polystyrene and polyurethane. They called their product “plastiguadua” and assessed its physical and mechanical properties. The work reported here will attempt to fill the knowledge gap in the literature on Guadua strand based composites.  Nodes are generally the rings that appears on a bamboo pole with varying distances from one another (Mahdavi et al., 2012) and are indicated by the arrows in Figure 1. The opposite is an internode, which represents the material between nodes. The mechanical properties of the culm in the node region are lower than the culm material between nodes (Lee et al., 1996; Sulastiningsih & Nurwati, 2009). Idris et al. (1994) reported the MOR of G. apus was 502.3 kg/cm2 for parts with nodes and 1240.3 kg/cm2 for internodes; and the compression strength of G. apus was 505.3 and 521.3 kg/cm2 for parts with nodes and internodes respectively. Previous studies of Smola and Zhang found that nodes significantly reduced the bending strength of hybrid Moso-Aspen OSB (Smola, 2013; Zhang, 2013). However, there is no study on how nodes affect the strength of Guadua strands based composites products.  Figure 1. Nodes on Moso Bamboo  node internode node 4  1.3 Hypotheses Based on the literature review and previous studies (Semple et al., 2015a; Semple et al., 2015b), several questions need to be addressed. First, since Guadua has denser vascular bundles than Moso, it is expected that Guadua OSB products will have higher strength properties than Moso OSB for similar final density of products. What is the difference in properties between boards made from these two species? Second, we want to reduce the density while still retaining acceptable mechanical properties. How much would the strength be affected by reducing the density? Could we meet the requirements based on CSA standard for both the density and the quality? Third, it was found that nodes have negative effect on the bending strength of Moso OSB (P. K. Zhang, 2013). The node anatomy of each species is very different and it is not clear if the presence of nodes in Guadua will reduce board properties. The hypotheses are as follows. First, Guadua is stronger than Moso based on similar final density of products. Second, the properties will be affected by the reduced density but still meet the requirements of CSA standard. Third, the presence of node will reduce the board properties.  1.4 Approach Guadua and Moso were compared under the same manufacture condition. In previous work Aspen strands are mixed with bamboo elements to improve the compaction of the mat in the core (Semple et al., 2015b; Zhang, 2013). Thus the boards in this research used Aspen as core material. In our previous work (Semple et al. 2015b) mixed Moso bamboo (Phyllostachys pubescens Mazel) and Aspen (Populus tremuloides Michx.) wood strands together to produce a 3-layer 5  Moso surfaces/Aspen core OSB with a density around 740 kg/m3. Based on Structural Board Association design information (SBA, 1998) and TECO publication Design Capacities for OSB (TECO, 2008), the normal density is around 640 kg/m3.  1.5 Structure of Thesis The thesis consists of the following chapters: Chapter 1. Introduction: this chapter introduced basic information on bamboo and discusses the motivation of this master project. Chapter 2. Literature Review: this chapter summarized the body of previous work on bamboo and bamboo composites.  Chapter 3. Materials and Method: this chapter described the raw materials (bamboo, aspen strands, and resin). The manufacture process of hybrid three layer OSB was covered in detail in this chapter. Chapter 4. Results and Discussion: this chapter displayed the test results and discussed what could be concluded from the results. The discussion part answered the questions and examined the hypothesis stated in Chapter 1. Chapter 5. Comments and Future Work: this chapter gave a brief conclusion and list several possible directions for further research. 6  Chapter 2: Literature Review 2.1  Bamboo Bamboo has been used in construction structures for centuries, especially in South Asia and South America. The high strength and tubular form of varying diameters make bamboo different from conventional, rectangular wood materials. The hollow cylindroid form of a culm makes bamboo an optimal material in an engineering sense. Its tubular structure provides good structural stiffness per unit weight with bending strength ranging from 10.3  to 27.6 GPa (Lee et al., 1994) while its nodes behave as bulkheads and prevent buckling of the stem under compression (Amada & Lakes, 1997).  Bamboo grows faster than any other plant. Most species can reach their full height within 2-4 months while requiring about 3-8 years to reach maturity (Lee et al., 1996; Liese, 1987). Certain species grow and reach heights of 4.8 m to 28 m tall depends on species (Lewis et al., 2007) at a rate as high as 3 cm/h (Guinness World Records, 2015). With the optimized distribution of fibers and bio-matrices in resisting environmental loads in nature, bamboo is regarded as one of the most sophisticated natural materials (Low et al., 2006). In the modern world, two forms of bamboo have been cultivated: woody bamboos and herbaceous bamboos (Kelchner & Group, 2013). Different species which range in size from delicate culms smaller than a few millimeters in diameter and centimeters in height to massive culms up to 36 cm in diameter. Figure 2 shows Phyllostachys sp., one kind of lignified woody bamboo. Woody bamboos are referred as lignified bamboos, while herbaceous bamboos are non-lignified. Some species of herbaceous bamboos are used in China as an indoor ornamental plants. 7  The specie shown in Figure 3 is Arthrostylidium sarmentosum, a kind of herbaceous bamboo.  With limited vegetative branching, herbaceous bamboos are clump-forming or stoloniferous (Calderón & Soderstrom, 1980).     Figure 2. Phyllostachys sp. (Clark, 2006)           Figure 3. Arthrostylidium sarmentosum(Clark, 2006) Retrieved January 11, 2016 from http://www.eeob.iastate.edu/research/bamboo/bamboo.html . Used with permission from the photographer Dr. Clark. (See Appendix D for the permission).  In 2006, a detailed quantitative lifecycle assessment of the environmental, economic and practical performance of bamboo, van der Lugt et al. found that bamboo structures have a lower environmental impact than other more commonly used building materials, such as steel, timber, or concrete. Three years later, Nath et al. (2009) report that common bamboos of northeast India (represented by 67% Bambusa cacharensis, 18% Bambusa vulgaris and 15% Bambusa balcooa) sequestered 61 tons of above ground carbon per hectare per life span (average 2 years age), compared to 54 tons per hectare above ground carbon stocks for tropical forests and 25 tons per 8  hectare above ground carbon stocks for temperate forests during the latter part of 20th century (Gorte, 2009). Bamboo has a better rate of carbon sequestration than tropical forests, boreal forests, and temperate forests.  To be suitable for processing into similar kinds of engineered composites as small wood logs, culms of sufficient diameter, up to 150 mm, are required (Semple et al., 2015a). Among bamboo species, Moso and Guadua are both temperate woody bamboos are known as giant timber bamboo as they have diameters of 130 mm or more. The thickness of the culm wall ranges from 4 to 12mm. The wall thickness of a culm is directly proportional with the outer culm diameter (Lo et al., 2004). Similar to wood, bamboo exhibits significant anisotropy in strength. It is more than ten times stronger in tension in the longitudinal direction than in the transverse direction (Amada & Lakes, 1997). The microstructure, strength and density of nodes may affect the properties of OSB product made from bamboo. One of the differences between bamboo and wood is the outer and inner layers which cover the bamboo culm. The composites of the culm outer layer or epidermis of bamboo contains silica. Li et al. (2004)  have found the ash content of bamboo is primarily silica, calcium, and potassium. Among those, silica content is the highest in the epidermis, lower in the nodes and absent in the internodes (Li, 2004). Silica content dulls normal steel blades very fast (Shaddy, 2008). It is hard to treat bamboo with preservatives because of the  hard epidermis and the inner wax layer covering the bamboo culm prevents penetration (Lee et al., 2001; Liese, 1998). While an oil-bath treatment has proved to be successful in preventing fungal attack, this treatment severely weakens the material (Leithoff & Peek, 2001). The wax and silica contained in the inner and outer culm layers affect the wetting characteristics of the surface by making it 9  difficult to bond (Lee et al., 1998). These layers can be removed by planing or sanding but this results in significant loss at material.   To study the structural advantage of bamboo culms over other engineering materials in terms of Young’s modulus, also known as elastic modulus, E, and density, ρ, Wegst et al. (1993) developed a material selection method. The results were summarized as shown in Figure 4. To make the comparison clear, they used a line whose equation was C (a constant) = E1/2/ρ to compare the properties of bamboo with other materials. Stiffer and lighter materials fall above the line, while more flexible and heavier materials fall below the line. The ovals in the figure represents the range of the available data for a particular given material. The figure shows that only timber from palm-trees and balsas have comparable specific stiffness to bamboo, i.e. similar high MOE but low density; whereas conventional building materials, such as aluminum, concrete, and steel, have lower specific stiffness. 10   Figure 4. Relationship of modulus-density for materials (Wegst et al., 1993) Retrieved November 15, 2103 from Elsevier. Used with permission from Elsevier. 2.1.1  Moso Bamboo Of the over one thousand species of bamboo (Austin et al, 1972), Moso (Phyllostachys pubescens Mazel) is one of the few commercially used species. Bamboo can be differentiated under two categories as sympodial and monopodial (Birkeland, 2002), shown in the Figures 5 and 6. As a monopodial bamboo (intermittently spaced stems from an interconnected below-ground rhizome), Moso has been used in China for a wide range of products. China has over 5 million hectare of Moso bamboo, or 70% of China’s natural and commercial bamboo forests, amounting to over 20% of total world bamboo resources  (Jiang, 2002; Peng et al., 2010). Since Moso is easy to plant, grows straight and rapidly, and has a thick wall (Fu, 2007b), it is widely cultivated and utilized. E: Modulus of Elasticity ρ: Density 11   Figure 5. Sympodial Bamboo (Rivière & Rivière, 1878) Retrieved June 28, 2016 from Biodiversity Heritage Library (BHL). Used with permission from BHL.  Figure 6. Monopodial Bamboo (Rivière & Rivière, 1878) Retrieved June 28, 2016 from Biodiversity Heritage Library (BHL). Used with permission from BHL.  2.1.2 Guadua Bamboo Some countries in South America such as Colombia have been using Guadua plastered with mud or cement mortar in housing construction for centuries (Paudel & Lobovikov, 2003). Other simple building products made from Guadua include crushed bamboo mats (also called “esterilla” in Spanish), which is a single or multilayer plywood-like panel made from flattened, 12  thin-walled culms cut from the upper stem  (Semple et al., 2015c). Mats made from crushed Guadua bamboo are shown in Figure 7.  Colombia has made significant progress in developing engineered, glue-laminated guadua bamboo (GLG), of which the mechanical properties are better than most conventional laminated wood or bamboo species (Correal et al., 2014). Even when compared to those of the highest quality structural tropical wood products in Colombia, GLG shows equivalent performance (Voermans, 2006). It has excellent structural properties for dwellings in earthquake zones including a high shear and fastener tear resistance-to-weight ratio, high energy absorption capacity and, flexibility (Juan F Correal & Varela, 2012). Since consumers are looking for alternatives with similar appearance, density and properties to tropical timber, Guadua timber is becoming popular. Guadua stems can reach 30 m in height, 20 cm in diameter, and similar to Moso, culms are harvested around 5 years of age. 13   Figure 7. Cured and dried crushed bamboo mats (Schroder, 2014) Retrieved from March 13, 2013 from https://www.youtube.com/watch?v=RqYtEB8Lq9E  Used with permission from Youtube. Guadua has a very high storage of carbon fixed annually per hectare, with a very short growth period. Following the methodology of Riaño et al. (2002), the carbon fixation estimated for 400 clumps per hectare of Guadua angustifolia is 54.3 tons in total for a growth period of 2190 days (6 years). Another report edited by Gorte (2009) summarized the average carbon levels sequestered for several major biomes including Tropical Forests, Temperate Forests, Boreal Forests, Tundra, Croplands, Wetlands, Temperate Grasslands and etc. during the latter of part of 20th century (10 years). The weighted average carbon sequestered for all biomes is 34.6 tons of carbon per hectare in total for 10 years. In countries such as Colombia and Peru, Guadua is widely used for construction either as round culms or standardized engineered products with rectangular strips. However, about 40% of the 14  material is wasted due to natural defects or variability in dimensions (Archila et al., 2015). Researchers are interested in fully exploiting Guadua’s high fibre content and high tensile strength with the aim of creating more efficient alternatives for converting raw Guadua into standardized products.  2.1.3 Differences and Comparison between Moso and Guadua A report from Larenstein University gives a detailed comparison between Moso and Guadua (de Vos, 2010). Comparing the thickness of node plate in Figure 8 and the length of the node region in Figures 9e) and 9f) of each species, the node regions of Moso are smaller than Guadua. Also visible in Figures 9a) and 9b) is the decrease in size of the vascular bundles from the inner culm wall toward the outer wall as shown in tangential surface (de Vos, 2010). But the vascular bundles at the outer part of the culm are denser than toward the inner part (Grosser & Liese, 1971). Moso has a finer grain than Guadua because of its smaller vascular bundles. As shown in Figures 9c) and 9d), the longitudinal surface of Moso is smoother than Guadua’s surface. Both species have a higher concentration of the vascular bundles near the outside of the culm wall than toward the inside, as shown in the difference between Figures 9c) and 9e), and between Figures 9d) and 9f).  15   Figure 8. Appearance of the node plates of Moso and Guadua Bottom 1 meter portion of the culm  Figure 9. Comparison between Moso and Guadua culm(de Vos, 2010) Retrieved June 15, 2015 from Wageningen UR Library. Used with permission from Wageningen UR. a) b) c) d)  e) f )  16  Our previous study showed the average oven dry density of Moso for both internode and node plates are lower than Guadua (Semple et al., 2015c, 2015d). Guadua poles have slightly greater average diameter and wall thickness with lower frequency of nodes and longer distances between nodes than Moso poles. A summary of the basic properties comparison is given in the Table 1.  Table 1. Properties of Moso and Guadua polese used to produce strands (Semple et al., 2015c, 2015d) Property Moso Guadua Internode tissue density* (kg/m3) 446.8 533.1 Node tissue density* (kg/m3) 531.8 601.6 Pole diameter (mm) 101.7 103.7 Internode length (cm) 23.98 30.67 Node Frequency (1/m) 3.80 3.30 Wall Thickness (mm) 10.9 12.0 Node Plate Thickness (mm) 2.77 7.38 Delivered MC (%) 11.7 13.3 *Oven-dry density    2.2 Bamboo Composites Products Bamboo has been planted in many places to ameliorate soil erosion and replace extensive historic forest losses. Until recently, it has been used in varies ranges of industries from slat-based laminated furniture and flooring to plywood-like panels to heavily compressed beams known as ‘scrimber’ for decades. The bamboo composites manufacture technology has a high degree of biomass recovery into product (Jyoti Nath et al., 2009). Nevertheless, technologies to convert bamboo into the same kinds of modern, engineered composite building products as wood are still in the process of development (Flander & Rovers, 2009). Bamboo’s natural hollow tube shape makes it impossible to use standard connections to connect it. Researchers have long been interested in converting bamboo from an irregular tube into shapes more suitable form for 17  structural applications (Mahdavi et al., 2012). This interest led to the development of laminated bamboo lumber (Lee et al., 1998; Nugroho & Ando, 2001; Rittironk & Elnieiri, 2008). Preliminary investigation from Mahdavi et al. (2012) showed that in order to flatten culms and create mats for bamboo composites, hammering culms can be just as, or more effective, than pressing them. After hammering, coarse sandpaper was used to smooth the inner face of the culm. These alternatives were found very effective in removing inner and outer surface layers which contained wax and silica but were very labour intensive. This process is adaptable and available to people in developing regions where heavy machinery is not accessible (Mahdavi et al., 2012). Archila et al. (2015) developed Guadua composites to protect the products against humidity, insects attack and bio-deterioration. In their research, the high strength of bamboo fibres as reinforcement was combined with polymeric matrices, which was polyester resin, to form flat sheets. The physical and mechanical properties of these sheets assessed from that research were expected to serve as a basis for further development of the engineered bamboo products.  2.2.1 Wood OSB Products OSB has been one of the fastest developing wood composite products due to its outstanding properties, ability to use logs unsuitable for veneering production, particularly in the USA (Benetto et al., 2009). OSB is a compressed mat made up of three layers of strands bonded together with a thermoset resin. Most commodity OSB is manufactured with aligned strands oriented parallel to the long edge of the panel in the surface layers, with a core of randomly oriented smaller strands and fines (shown in Figure 10b). For certain higher grades of board the 18  core strands may be oriented perpendicular to the strands direction on the surface, this structure gives the board comparatively higher mechanical strength for both directions (Figure 10a). The third type of board has both non-directional surface and core (Figure 10c). That is the original waferboard product that pre-dates OSB, and now is rarely produced.  Figure 10. OSB lay up Canadian Standards Association classifications (Structural Board Association, 1998) Retrieved November, 2013 from Structural Board Association, Used with permission from OSBGuideTM  2.2.2 Bamboo OSB Products Lee et al. (1996) had shown that the manufacture of strand boards from Moso bamboo is technically feasible. Several studies have shown that Moso bamboo in particular is a potential feedstock for OSB (Fu, 2007a, 2007b; H. Zhang et al., 2006). The OSB fabrication process represents one of the best ways for automation and mass production of bamboo-based building materials. China has been developing and promoting the use of OSB as a sustainable construction material since the mid-1970’s in order to reduce the demand for energy intensive traditional concrete and bricks (Hua, 2003). Yunnan Yung Lifa Forest Co Ltd. has spent a few years adapting OSB manufacture technology to bamboo and recently commenced production of commercial quality bamboo OSB for shipping container flooring (Anon, 2012; Grossenbacher, 2012).  a)                                            b)                                                c) 19  2.2.3 Three-layer Hybrid Bamboo OSB Because the concentration of vascular bundles in bamboo culm decreases from the outer culm wall toward the inner culm wall (Yu et al., 2008), there is a significant strength loss when the outer layer is removed as is the case for traditional bamboo products processing. Also there is lots of waste caused by removing the outer layers with up to 40% cutting waste (Archila et al., 2015; Flander & Rovers, 2009). Semple et al. (2015a) created a novel manufacturing method that uses a stranding machine for OSB to cut strands from the culms with no need to remove the outer or inner layers. Nodes were removed by hammering. With only 5% of the total culm stock chips being thrown away, recovery was about 87% (Semple et al., 2015c). In addition, a three-layer hybrid sandwich structure OSB with aspen in the core and bamboo on the surface was found to be stronger than normal uniformly mixed single layer OSB and three layers pure Aspen OSB (Semple et al., 2015b; Zhang, 2013). Thickness swell also improved by 40% compared with normal wood OSB because of the slower water absorption of the bamboo board. There is no research that compares Guadua and Moso strands based composite products. Experiments on the effect of species on the boards’ properties are required to learn the difference between Guadua and Moso. No bamboo OSB has been made with a sufficiently low density to be accepted as replacement for wood OSB. With Aspen strands in the core, the weight ratio could be adjusted to obtain the target low density. Since the nodes structure are different between Guadua and Moso, experimentation about possible node effects on Guadua OSB is required.  20  Chapter 3: Materials and Methods 3.1 Culm Feedstock For the board fabrication experiment, 20 poles of Chinese Moso culm stock in 243.8 cm lengths (Figure 11) were purchased from Canada’s Bamboo World, located in Chilliwack, BC, who imports seasoned and fumigated (methyl bromide) bamboo poles from Zhejiang Province in South East China. The Moso culms had an average diameter of 101.7 mm with an average weight of 6.6 kg. All the supplied poles were harvested at four years of age. Ten Guadua poles size 579.1 cm long were acquired from Koolbamboo, Miami, FL, USA, who import seasoned, treated (Borax) culms  harvested between four and six years of age from Colombia and Panama. Each Guadua pole was cut into 198.1 cm length before shipping. The average Guadua pole diameter was 103.7 mm with a MC at 13.3%. Other pole characteristics recorded were internode lengths (distance between nodes), number of nodes per meter, and the shape and thickness of the nodes were measured and compared between Moso and Guadua (see Table 1 in Chapter 2). These characteristics can vary greatly between bamboo species.  Figure 11. Moso poles purchased cut from Canada Bamboo World 21  All the poles were stored outside the machine lab in UBC CAWP under cover with an average temperature of 7.5 °C, and average relative humidity of 85% from March to May of the year 2014 (WeatherSpark, 2014). To calculate the moisture content and density of the raw materials, each pole was cross-cut using the Pendulum Saw (Stromab PS 50/F) into four short billets (shown in Figure 12).  Figure 12. Short billets from Guadua poles  3.2 Culm Breakdown After cross-cutting, the volume of small specimens cut from the culms was measured using the water displacement method. The oven dry density of the Moso culm is 745 ± 21 kg/m3, while the oven dry density of the Guadua culm is 806.6± 17 kg/m3. The average moisture content is 19.3 ±1.1%. 22  Semple and Smith (2014) found it is more effective to convert the billets into quarters rather than halves since the number of strands per culm round stranded is increased. Therefore all the bamboo culm rounds were cut into quarters lengthways using a band saw (type Meber SR-500). Because the maximum width and height of the feed box for the strander was only 130 mm, culm quarters were cut shorter to no more than 130 mm long using a chop saw (type Omega T55-300). In order to compare the effect of nodes in strands on the properties of boards, the bamboo culm quarters were cut to be either node free (internode) or to have a node near the middle (node). And for there to be roughly even numbers of node and node free rounds, there were about 17 to 19 pieces cut from each culm quarter. The internal plate of the node in Moso is very thin (about 2-3 mm in thickness) and easy to remove with a hammer. The internal plate of the node in Guadua is thicker (ranging from about 6 to 12 mm depending on height in the culm; and much thicker near the base). Node plate removal required a Dremel saw followed by sanding on a belt edge sander (Progress PMC-150) until flush with the inner wall (Figure 13).  Figure 13. Using dremel saw (left) and belt sander (right) to remove Guadua node. Dremel Saw 23  3.3 Stranding Before stranding, culm pieces needed to be pre-saturated with water to ease slicing and minimize damage to strands and knives during stranding. The technique was developed (Semple et al., 2014) to simulate the moisture conditions and stranding of fresh green cut culm stock. The laboratory disk strander (CAE 6/36 single-blade mounted disk) used was built by Carmanah Design and Manufacturing, Vancouver, BC. Since the effect of nodes on Moso bamboo strand boards was examined in an earlier study (Semple et al. 2015a, b), only the Guadua bamboo was converted to either node or internode pieces that were stranded separately. The Moso culms were cut into successive 130-mm-long pieces and processed together irrespective of node presence and location. The stacking and slicing configuration for the culm rounds through the radial-longitudinal plane is illustrated in Figure 14 and 15. The quarters were more efficient in terms of processing because this shape was easily fitted and securely held in place during stranding.  Figure 14. Top view of the strander feed box showing quarter cut culms Feed Box Direction 24   Figure 15. Top view of the strander feed box showing halve cut culms The pre-saturation and stranding methodology was based on earlier preliminary works designed to maximize the amounts, quality and consistency of strands from Moso bamboo. (Smola 2013, Semple et al. 2014, Semple et al. 2015c, d)  Disk rotational speed was set at 734 RPM, and knife projection was set at 0.675mm to give an average target strand thickness of 0.65 mm. Counter knife angle was 45° with a hydraulic piston-driven feed buffer rate of 0.37 m/min. The knife projection in conjunction with feed rate determines strand thickness. Sheet metal shims measuring 0.051 mm in thickness were inserted into the housing block to make tiny changes in the knife protrusion. A magnetic mounted dial gauge was used to measure the knife protrusion. From the preliminary stranding trials (Smola 2013, Semple et al. 2015a), slicing longitudinally through the culm wall (as shown in Figure 16) produced narrower smoother strands that did not curl, rather than stranding the culm horizontally as is normally done for wood logs. Figures 16 Feed Box Direction Knife Direction 25  and 17 showed the view of disk strander with knife direction in detail. Strands were oven dried at 80°C over night and left to cool for at least 4 hours before sealing them in plastic bags. The moisture content of the dried strands was approximately 2%.  Figure 16. View of disk strander showing the knife rotation direction  Figure 17. Stranding culm pieces vertically Knife Direction 26  3.4 Strands Screening The Guadua strands were dried and collected separately as either node or internode, and the Moso as mixed strands. To remove the dust, fines and chips, dried strands were sifted using a motorized screening machine designed for wood strands (Figure 18). Guadua node, Guadua internode and Moso mixed strands were screened separately. The chips were hand-picked out during the screening process. Only the 14.3 mm mesh and 3.18 mm mesh screens shown in Figure 18 were used. The fines and the dust could pass through all the screens and were collected on the bottom pan. Most unbroken large strands were collected from the top of the 14.3 mm mesh screen, while medium and smaller fragments were collected from on top of the 3.18 mm mesh screen.  Figure 18. Screening the strands  27  3.5 Board Fabrication and Experiment Design The technique of the board fabrication was based on the preliminary research (Smola, 2013;. Zhang, 2013) to produce three layers hybrid bamboo-Aspen OSB. All the process of the board fabrication were done in the Wood Composites Lab at UBC. It took two weeks to fabricate all boards and a total of 9 weeks to finish all the tests.    3.5.1 Experimental Designs There were three different experiments designed to examine the effects of bamboo species (Guadua or Moso), the effect of reducing board density to bring bamboo boards down to the expected density of commodity OSB sheathing products, and also the effect of the presence of nodes on Guadua OSB properties.  Experiment 1: Previous work (Semple et al. 2015b) found that bamboo strands were most efficiently used in the surface layers of OSB, retaining the compressible Aspen in the core layers. Therefore for the comparison between Guadua and Moso strands in the surface layers of OSB, two types of boards were produced: 1) Guadua-Aspen 3-layer hybrid OSB, and 2) Moso-Aspen 3-layer hybrid OSB with six board replicates per type. The same weight ratio of bamboo strands and Aspen strands was used to both groups. The bamboo surface strands were a mixture of node and internode strands produced with the ratio found in cutting 8 foot long Moso feedstock poles into 130 mm-long pieces. Type 1 Guadua-Mixture (GM) were 50% Guadua bamboo strands in the surface layers and 50% Aspen strands in the core. Type 2 Moso-Mixture (MM) were 50% Moso bamboo strands in the surface layers and 50% Aspen strands in the core. Both board types were produced with a target density around 760 kg/m³.  28  Experiment 2: The second experiment was aimed at retaining the mechanical properties of Moso-Aspen hybrid OSB but reducing board density. The 50%/50% MM board from Experiment 1 was used as the comparison for high density 760 kg/m³ boards. Two strategies were used to reduce board density. First the weight ratio between Moso strands in the surface and Aspen strands in the core was reduced from 50%/50% to 25%/75%. Second the overall amount of furnish used per board was reduced.  Two types of 3-layer hybrid OSB were fabricated with six board replicates per type.  Moso-Medium-density (ML1) were 25% Moso bamboo strands in the surface layers and 75% Aspen strands in the core pressed to a target density around 720 kg/m³. Moso-Low-density (ML2) were 25% Moso bamboo strands in the surface layers and 75% Aspen strands in the core pressed to a target density of around 628 kg/m³. The properties of all three board types are then compared.  Experiment 3: The third experiment was to evaluate the effect of nodes in Guadua strands on Guadua bamboo hybrid OSB. Two types of 3-layer hybrid OSB were fabricated with six board replicates per type. Type 1 Guadua-Node (GN) were 50% Guadua bamboo strands with node near the middle in the surface layers and 50% Aspen strands in the core with a target density around 760 kg/m³. Type 2 Guadua-Internode (GI) were 50% Guadua bamboo internode strands in the surface layers and 50% Aspen strands in the core with a target density around 760 kg/m³.  A summary of all six types is shown in Table 2 with details. In Table 3, the factors are listed for each experiment. 29  Table 2. Summary of experiment design per types  Board Type Target Density (kg/m³) Weight ratio of Bamboo surface - Aspen core Guadua/Aspen Hybrid  GM 760 50% - 50% Mixture of node and internodes GN 760 50% - 50% Strands with node GI 760 50% - 50% Strands without node Moso/Aspen Hybrid  MM 760 50% - 50% Mixture of node and internodes ML1 720 25% - 75% Higher target density ML2 628 25% - 75% Lower target density  Table 3. Summary of experiment design per factors.  Factors Levels Compared types Experiment 1 Species Guadua GM Guadua Mixture Moso MM Moso Mixture Experiment 2 Density 760 kg/m³ MM Moso Mixture 720 kg/m³ ML1 Moso High 628 kg/m³ ML2 Moso Low Experiment 3 Guadua Node Node GN Guadua Node Internode GI Guadua Internode  3.5.2 Board Fabrication Design Across the three experiments there was a total of six different board types, produced with six board replicates per type, for a total of 36. All boards were 740 mm x 740 mm, limited by the dimensions of the press platens. The target thickness was 11.1 mm (7/16th inch) which is a typical thickness of OSB sheathing. All 36 boards were fabricated under similar conditions (resin type, dosage, etc) during a continuous timeline. From the trial experiment on uni-directional, fully oriented boards by Smola (2013), in order for the boards to consolidate correctly, mat unevenness and delamination issues were modified by 30  making the strands in the core not oriented. In the present work, only the bamboo strands surface layers were oriented, while the aspen strands used in the core were not oriented. This produces a flatter, less voluminous mat.  3.5.3 Blending Strands with Resin Bamboo and Aspen strands were blended separately with Cascophen RBS2345, liquid Phenol Formaldehyde (PF) resin, supplied by Momentive Specialty Chemicals Canada Inc., Edmonton, Alberta, Canada. Resin content was set to 6% of the boards mass. Given the size and density of the boards, the target initial furnish mass was 4.26 kg with 0.43 kg resin mass applied per board for type GM, GN, GI, and MM, 4.04 kg furnish with 0.41 kg resin mass for type ML1, 3.52 kg furnish with 0.36 kg resin mass for type ML2. And the amount of resin was split according to the weight ratio between bamboo strands and aspen strands.   The rotating drum blender (Figure 19a), was 183 cm in diameter by 61 cm in depth, and equipped with small flights to lift and cascade strands. The required resin was applied via a compressed air-fed (30 psi) atomizer spray nozzle connected to a paint pot (Figure 19b, c). To determine that the correct weight of resin was sprayed in, the pot was placed on a tared balance. The required amount of resin was sprayed based on monitoring the drop in weight of the tared pot, and the resin supply valve shut off once the required amount had been used. 31                  Figure 19. Blending system  a) Drum blender      b) Spray paint pot      c) Spray nozzle Different types of strands (surface, core) were blended separately. Resinated strands were left in the blender and tumbled for at least five minutes to ensure the resin was evenly mixed with the strands, and after the blender was stopped the strands were left to sit for 5 min before removal to permit resin droplets to settle. Because the properties of the OSB could be affected by the humidity and temperature when pressing the boards (Zhou et al., 2009), all resinated strands were hot pressed as soon as possible after blending and mat formation. Typically, it would take 10 to 15 minutes to form the mat. Therefore the scheduling of manufacturing order of boards   a)            b)   c)   32  became important in the experiment. With six boards per type, sufficient furnish for two or three boards (depending on surface or core weight) with 10% spillage were blended per run. Table 4 shows the blending and pressing sequence for the first 6 boards from type MM. The schedule for all 36 boards, with six boards made per day is given in the appendix A.  Table 4. Press sequence schedule (example of first 6 boards) Press sequence Board Board code Day task furnish weight, kg flake type layer 1,2,3 MM1,MM2,MM3 1 blend 6.4 Moso surface 1,2 MM1,MM2 1 blend 6.4 Aspen core 1 MM1 1 form and press 4.26 Hybrid surface+core 2 MM2 1 form and press 4.26 Hybrid surface+core 3,4 MM3,MM4 1 blend 6.4 Aspen core 3 MM3 1 form and press 4.26 Hybrid surface+core 4,5,6 MM4,MM5,MM6 1 blend 6.4 Moso surface 4 MM4 1 form and press 4.26 Hybrid surface+core 5,6 MM5,MM6 1 blend 6.4 Aspen core 5 MM5 1 form and press 4.26 Hybrid surface+core 6 MM6 1 form and press 4.26 Hybrid surface+core  3.5.4 Hot Pressing After blending, the required quantity of strands for each surface layer and core layer was weighed out in plastic tubs. Bamboo strands in the bottom surface layer were evenly spread over an oiled caul plate measuring 740 x 740 x 7.11 mm, with a two inch high wooden forming box used to contain the mat and fix the position (see Figure. 20). A 30 cm high 12- vane orienter as shown in Figure 20 was used to orient the bamboo strands in the bottom and top layers of each board. Strands were manually distributed evenly into the 50 mm wide slots and dropped onto the oiled caul plate. After distributing the bamboo strands in the slots and then removing the 33  orienter, the layer was tamped down to flatten using a wooden slab. After the bottom surface layer was laid down, Aspen core strands were poured in and distributed randomly but evenly without the orienter, and tamped down. The top surface was oriented by placing the orienter on top of the core layer and arranging the top surface strands as described. The completed three-layer mat was flattened and covered with the second oiled caul plate. Figure 20 shows a 45 degree vertical view of the orienter system when making the mat.  Figure 20. Hand orientation of bamboo strands with orienter The whole assembly was then placed in a hot press and pressed at 150 °C for 15 minutes. During this process, maximum mat compaction pressure was 5.8 MPa for 14 minutes and then 1 minute for the press closing and opening. All the production parameters are listed in Table 5. After pressing the board was removed, cooled, weighed and labeled with back RH(right hand) corner of press, board number (1-6) and Type (ie, GM, GN, GI, MM, ML1 or ML2). The three layer OSB was isotropic, meaning that both face layers contained the same mass of resinated bamboo strands and were oriented in the same direction. 34  Table 5. Production parameters Mat structure  Three-layer sandwich random core Resin type PF Resin solids content 57 wt% Board length 737 mm Board width 737 mm Targeted board thickness 11.1mm Board resin content 6% w/w (oven dry weight basis) Board moisture content 2%   3.6 Specimens Cutting To minimize bias due to sample position, three different cutting patterns were used and randomly assigned to each board (Figure 21). For each type with six replicate boards, two of them were cut according to one of the three different cutting patterns. In all the patterns, 30 IB specimens (51 x 51 mm) were located in different zones on the board (the small squares in Figure 21). To evaluate the effect of strands direction on bending strength (MOR), four bending test specimens (290 x 76 mm) were cut so that two had the long axis parallel-to-strands and two had the long axis perpendicular-to-strands on each board (the rectangles in Figure. 21). One thickness swelling test specimen (152 x 152 mm) was cut per board (the large squares in Figure 21).  Figure 21. Cutting pattern with three different directions 1.             2.       3. 35  Table 6 summarizes the numbers of test specimens cut for each property test. After cutting, each test specimen was labeled with board type, board (replicate) number, and test piece (observation) number. For instance, Figure 22 shows the 23rd observation of IB specimens for the 4th boards of GI type.  Figure 22. Example of labeling  All test specimens were conditioned at a relative humidity of 65±2% and a temperature of 20 ± 2°C to a constant weight and hence moisture content in accordance with ASTM D1037 (ASTM, 2012).  Specimens were kept in the conditioning room untill the weight change during 24 hours was less than 0.2%. Table 6. Summary of specimens’ size and quantity   Size of specimens mm Qty. of Specimens Qty. of Specimens Total Qty. Of Specimens per board per type Internal Bond 51x51x10 30 180 1080 Thickness Swelling 152x152x10 1 6 36 MOR/MOE Bending 240x76x10 4 24 144  36  3.7 Test Methods Tests included internal bonding test, flexural properties (bending test), and the thickness swelling/water absorption test. The process and the sample preparation were consistent with ASTM D1037 and D4442 (2012, 2014). Photographs of sample in the conditioning room are shown in Figure 23. Before all tests the required dimensions of each specimen were measured using digital calipers to 0.01 mm.  Figure 23. Specimens in the condition room 3.7.1 Internal Bonding Test The following formula gives the calculation of the IB test, which is also called the tension perpendicular to surface test. Before testing, thickness and weight were measured for each specimen. Density was calculated from the weight and the volume. Volume is the product of a, b and thickness.  ܫܤ = ୫ܲୟ୶ab        ൤ Nmmଶ൨         ୫ܲୟ୶  maximum load (N)           a   length of specimen (mm)              b                         width of specimen (mm)   37  As shown in Figure 24, the IB test machine designed by Instron according to ASTM Standards D1037  would measure the specimen’s tensile strength perpendicular to the surface of the specimen.   Figure 24. IB test machine  3.7.2 Flexural Property Test The test used the three point bending test to get the deflection and load for MOE and MOR. The formula below shows the calculation methods. 38  ܯܱܧ = ܮଷ4ܾ݀ଷ ∗ ∆ܲ∆ݕ        ൤ Nmmଶ൨      ܮ  Length of span    ܾ  Width of specimen (mm)       ݀  Thickness of specimen (mm)          ∆௉∆௬               Slope,  ∆ܲ ܽ݊݀ ∆ݕ were given by the test machine   ܯܱܴ = 3 ∗ ௠ܲ௔௫ ∗ ܮ2ܾ݀ଶ        ൤ Nmmଶ൨ ௠ܲ௔௫   Peak Load (N) ܮ    Length of span (mm) ܾ   Width of specimen (mm) ݀              Thickness of specimen (mm)  The flexural test, shown in Figure 25, shows the specimen in three-point loading with compression force in the middle of the span perpendicular to the tested surface. Two types of samples were tested; i) samples with the long edge parallel to the strand direction and ii) samples with their long edge perpendicular to the strand direction as required by the ASTM Standards.   Figure 25. Flexural test machine 39  3.7.3 Thickness Swelling and Water Absorption For the thickness swelling and water absorption test, all specimens were weighed and measured for thickness at four points 25 mm in from at the midpoint of each side prior to immersion. Weight and thickness at the same locations were measured after 2 hours soaking and again after 24 hours soaking. After 24 hours soaking the specimens were oven dried for 24 hours at 105 ℃, and re-weighed to give oven dry weight.  Figure 26 (a) is the template for marking and locating thickness measurement points in accordance with the ASTM standard (2014).  Figure 26 (b) is the tank used for thickness swelling test. Figure 26 (c) shows the caliper connected to a weight extensometer used to give a precise and consistent thickness reading. 40   Figure 26. Details for thickness swelling test (a) Marking the 4 measurement points onto a sample using a template (b) TS samples immersed in the swell tank, and (c) measurement of sample thickness at each of the points marked on the sample in (a)  (a) (b) (c) 41  Chapter 4: Results and Discussion All results were analyzed using single-factor ANOVA (i.e. board type) in JMP 10 (SAS Institute, Inc. 2012), using the 5% significance level (α=0.05). Means were compared for all pairs of means using the t-test for two treatments or using the Tukey-Kramer HSD evaluation for three or more treatments.  Board densities and thicknesses were derived from measurements of IB specimens. Flexural properties included MOR and MOE for perpendicular-to-strand direction and parallel-to-strand direction (four groups). Water resistance property was evaluated by water absorption and thickness swelling after 2 hours and 24 hours soaking (four groups). All data are recorded in Appendix B. 4.1 Experiment 1: Comparison of Guadua and Moso Boards As Guadua has thicker fibre bundles than Moso, it was expected that Guadua-Aspen hybrid OSB would have better properties than Moso-Aspen hybrid OSB. To exclude the possible effect from the presence of a node on the culm, only type GM and MM were compared in this section. Type GM was made from 50% mixed Guadua strands on the surface with 50% aspen strands in the core. Type MM was made from 50% mixed Moso strands on the surface with 50% aspen strands in the core.  All mixed bamboo strands contained both node and internode strands as the original ratio of screened strands.  4.1.1 Thickness and Density After the press was opened, the boards were allowed to cool and then cut into samples as previously described. Comparison of the pressed board thickness in Table 7, with the target board thickness, 11.1 mm, showed there was spring back. The Moso boards spring back more 42  than the Guadua boards while the density of both board types were not significantly different.  Note that the levels in Table 7 not followed by same letter are significantly different (same meaning for all the tables following). Moso mixed strands boards showed a greater spring back than the Guadua boards. We essentially hit the target pressing density, 760 kg/m3, for both types only slight higher than target. There was no significant difference between them. The full results data are given in Appendix B, while the statistical analysis are given in Appendix C. Table 7. Thickness and density for MM and GM Type Thickness  Density Mean (mm) COV (%)  Mean (kg/m3) COV (%) MM 11.55a 2.4  764.5a 8.5 GM 11.42b 2.7  770.3a 13.2 p-value <0.0001  0.5232 COV = coefficient of variation  4.1.2 Internal Bonding Strength IB testing is usually used to test the ultimate failure stress under a tensile load perpendicular to the plane of the board, which usually occurs in the weakest region of the core (Dai et al 2008, May 1983). Many studies have revealed a positive correlation between density and IB strength (Sumardi et al., 2007). For a similar final target density, the mat made from lower density material has a better consolidation because of less voids of space.  43   Figure 27. IB test results for GM and MM, n = 180 for each mean  Since Guadua has a higher density tissue than Moso (Dixon et al., 2015), it was expected that Guadua boards would have higher IB strength, but the opposite was observed. Moso hybrid boards had an IB strength at 0.769 MPa, which was 29% better than the Guadua boards that had a value of 0.598 MPa. Compared to the similar product, the pure Aspen three layer board we made in 2013 (Zhang, 2013) even had a stronger IB strength at 0.653 MPa, suggesting that the Guadua was the weak link in the IB samples. Indeed, examination of the IB samples revealed that the most common location of failure was at the interface between the surface and core layers. This was the case for both Guadua and Moso suggesting that the contact area between the bamboo and aspen strands is less than in the case of boards made with aspen surface and core. p-value <0.0001 44  The required IB strength for Strandboard and Waferboard in CSA O437.0 is 0.345 MPa (CSA, 2011). For the same applied force, Guadua compressed less than Moso. This means that the compaction ratio for Guadua is lower than for Moso. High-density species make mats with lower compaction ratios compared to low-density species (Hood, 2004).  4.1.3 Flexural Properties The MOR and MOE perpendicular to the strand direction of both board types were greater than the 12.4 MPa minimum MOR and 1.5 GPa minimum MOE required for OSB products by CSA O437.0. And there was no significant difference between the two boards types (Tables 8 and 9). Similarly, the flexural properties parallel to the strand direction of both board types were greater than the 29 MPa minimum MOR and 5.5 GPa minimum MOE required by the standard. Only MOE parallel to the strand direction test showed a significant difference between two board types. Guadua hybrid board had a 31.6% stiffer property than Moso hybrid board likely due to the different fibre properties of the two species. Since the GM and MM are both made from a mixture of nodes strands and internode strands, another possible reason could be the different nodes structure of Guadua compared with Moso. Table 8. Results of flexural property test of GM and MM (perpendicular) Type Means COV (%)  MOR (MPa) p-value = 0.4810 MM 20.25a 33.7 GM 21.93a 20.3  MOE (GPa) p-value = 0.4372 MM 2.14a 20.6 GM 1.99a 23.6 Note: the levels in the table not followed by same letter are significantly different  45  Table 9. Results of flexural property test of GM and MM (parallel) Type Means COV (%)  MOR (MPa) p-value = 0.8968 MM 64.93a 15.5 GM 64.28a 21.3  MOE (GPa) p-value <0.0001 MM 8.01b 9.5 GM  10.54a 13.7 Note: the levels in the table not followed by same letter are significantly different   4.1.4 Water Absorption All the boards were fabricated without any addition of wax (wax is normally added to OSB). Our previous experiment (Semple et al., 2015a) found that the pure Aspen three layer boards had TS values in excess 17% without wax which is above the limit 15% TS set by CSA O437.0 (CSA, 2011). Table 10 shows the TS results summary. GM boards had 55.4% more swelling than the MM boards after 2h soaked in water. This difference reduced to 22.3% for 24 h TS and became non-significant. Intuitively, these results made sense as the higher density surface Guadua strands have more void space between them and are able to absorb more water more quickly than the more compressed Moso boards. After 24 hours water is absorbed into the Aspen core and differencing decreases. Results of the 24 h TS test are also compared in Figures 29.  46  Table 10. Results of thickness swelling (%) for GM and MM  Type Means COV (%)  2h TS (%) p-value = 0.0349 MM 2.86b 36.0 GM 4.44a 27.3  24h TS (%) p-value = 0.0876 MM 9.57b 23.3 GM 11.70a 13.7 Note: the levels in the table not followed by same letter are significantly different    Figure 28. Results of 24h TS of GM and MM, n = 6 for each mean Similarly, Guadua hybrid board showed less water absorption for 2 h or 24 h WA. The Guadua boards absorbed 18.8% of its original weight water, which is 35.2% more compared to Moso p-value = 0.0876 47  boards at 2 hours (see results in Table 11). After 24 hours, this difference was reduced to 19.9%. After 2 hours Guadua boards absorbed 51.2% of the total weight of water it absorbed in 24 hours, while Moso boards absorbed 56.7% (Figure 30). Water absorbed by the boards was stored in the space in the structure. Less consolidation would permits more water to penetrate into void space between strands.  Lower density board, which has a looser structure, was expected to have more water absorption. However, in this test lower density Moso hybrid board had less water absorption. The Guadua strands have a rougher surface because of the courser grain and larger vascular bundles compared to Moso. This would result to larger void space for water to penetrate which lead to more water absorption of Guadua boards. Table 11. Results of water absorption (w/%) for GM and MM Type Means COV (%)    2h WA (w/%) p-value = 0.0355   MM 13.90b 25.4   GM 18.80a 18.3    24h WA (w/%) p-value = 0.0408   MM 32.10b 15.6   GM 38.48a 11.4   Note: the levels in the table not followed by same letter are significantly different  48    Figure 29. Results of 24h WA of GM and MM, n = 6 for each mean The IB test and the water absorption property test indicated the Guadua hybrid boards were less consolidated than the Moso hybrid board. Consistent with the findings of Dixon et al. (2015), research on their Moso imported from Bamboo Craftsman Company and Guadua from KoolBamboo showed that Guadua is stiffer than Moso, results in the Guadua hybrid board a higher MOE parallel to the strand direction. However, the positive correlation between density and strength from that finding was contrary with what was found in our study. Possible reason could be the high density of Guadua strands caused a looser structure with same target board thickness and board density. When the hot press closed to the same thickness, the pressure p-value = 0.0408 49  applied to the Moso is higher than for the Guadua due to the larger volume occupied by the Moso strands compared with the Guadua strands. . Another reason could be the material we used here for Guadua and Moso is mixed with both node and internode strands. The differences of fibre volume fractions between the species may affect the results. Guadua has a relatively coarser grain and larger vascular bundles (Dixon et al., 2015), which could cause a rougher surface after stranding. After being resinated and pressed, Guadua hybrid board is more likely to have uneven structure because of the rougher surface. With the same target density, that is a possible reason for why higher density material produces a lower compaction ratio which results in more water absorption. 4.2 Experiment 2: Density Effects on Mechanical Properties In CSA-O437.0 (2011), O-2 class OSB is recognized as being structurally equivalent to plywood when used as roof, wall and floor sheathing. Thus our results are compared to the requirements for O-2 class to determine whether low density bamboo hybrid OSB may be qualified to substitute for structural wood OSB. Furthermore, the two lower density boards will be compared with the high density MM board from Experiment 1. To make the group name corresponding to their density level, names with density level will be assigned to the three types (see Table 12). Type MM (50% Moso bamboo strands in the surface layers and 50% Aspen strands in the core with higher target density) is named MHigh to present the highest density level in this comparison. Type ML1 (25% Moso bamboo strands in the surface layers and 75% Aspen strands in the core with higher target density) is named MMed to present the medium density level in this comparison. Type ML2 (25% Moso bamboo strands in the surface layers and 75% Aspen strands in the core with a lower target density) is named MLow to present the lowest density 50  level in this comparison. According to CSA standards, no individual in the five of the panel samples shall have any property more than 20% below or above in the case of thickness swell) the listed five panel average value for that property.  Table 12. Experiment 2 boards types Assigned Name Board Type Target Density (kg/m³) Weight ratio of Bamboo surface - Aspen core  MHigh MM 760 50% - 50% Mixture of node and internodes MMed ML1 720 25% - 75% Higher target density MLow ML2 628 25% - 75% Lower target density  4.2.1 Thickness and Density Due to the spring back of the boards after the pressure was released upon press opening, board thicknesses exceeded the target thickness of 11.1 mm by an average of 0.42 mm. Between the three board types, the differences in average thickness was statistically significant (p<0.0001). 50% w/w Moso hybrid boards (MHigh) showed the greatest springback, while 25% w/w Moso hybrid low density boards (MLow) showed the lowest (Table 13). All board types hit the target pressing density within 1.2%. 51  Table 13. Means and standard deviation for thickness and density Type Assigned Name Thickness Density Mean (mm) COV (%) Mean (kg/m3) COV (%) MM MHigh 11.55a 2.4 764.5a 8.5 ML1 MMed 11.45b 2.3 728.3b 11.3 ML2 MLow 11.28c 2.7 634.8c 11.6 Note: the levels in the table not followed by same letter are significantly different  4.2.2 Internal Bonding Strength As might be expected, the low density MLow boards had the lowest IB strength (0.656 MPa). However the medium density MMed boards had the highest IB strength (0.799 MPa), while the high density MHigh boards had a lower IB strength (0.769 MPa). No significant difference was found between MMed and MHigh boards (p-value = 0.1986), whereas MLow boards were significantly lower in density (Figure 32). MMed and MHigh boards had approximately 20% better IB than MLow boards. All groups satisfied CSA-O437.0 requirements for IB strength which is 0.345 MPa (Canadian Standard Association, 2011). The lower 95% confidence interval of MLow boards was 0.632 MPa, which is above the CSA standard (details given in Appendix C).  52   Figure 30. Results of IB test of MM, ML1 and ML2, n = 180 for each mean 4.2.3 Flexural Properties For the bending test specimens (290 mm x 76 mm) tested perpendicular-to-strand direction, no significant difference between the three different board types was found (see Appendix C for details and p-value). Although both MOR and MOE of perpendicular specimens were much lower than the parallel specimens, all board types met the 12.4 MPa minimum perpendicular MOR and 1.5 GPa minimum perpendicular MOE required for O-2 class products by CSA O437.0 (2011). Results are shown in Tables 14 and 15.  Table 14. Results of MOR of MM, ML1 and ML2 (perpendicular) Type Assigned Name Means for MOR (MPa) COV(%) MM MHigh 20.25a 33.7 ML1 MMed 24.38a 25.1 MHigh – MLow, p-value < 0.0001 MMed – MLow, p-value < 0.0001 CSA Standard: 345 KPa (MM)                    (ML1)             (ML2) 53  ML2 MLow 20.39a 10.9 CSA Standard: 12.4 MPa  Table 15. Results of MOE of MM, ML1 and ML2 (perpendicular) Type Assigned Name Means for MOE (GPa) COV(%) MM MHigh 2.14b 20.6 ML1 MMed 2.72a 15.9 ML2 MLow 2.41a,b 18.0 CSA Standard: 1.5 GPa  MLow boards had significantly lower parallel MOR and MOE than MMed or MHigh types. Nevertheless they met the 29 MPa minimum parallel MOR and 5.5 GPa minimum parallel MOE required for O-2 class products by CSA O437.0 (2011). Results are shown in Tables 16 and 17. Table 16. Results of MOR of MM, ML1 and ML2 (parallel) Type Assigned Name Means for MOR (MPa) COV(%) MM MHigh 64.93a 15.5 ML1 MMed 59.09a 19.1 ML2 MLow 44.23a 19.5 CSA Standard: 29.0 MPa  Table 17. Results of MOE of MM, ML1 and ML2 (parallel) Type Assigned Name Means for MOE (GPa) COV(%) MM MHigh 8.01a 9.5 ML1 MMed 7.44a 10.1 ML2 MLow 6.05b 12.0 CSA Standard: 5.5 GPa  Compared with CSA Standards, all lower 95% confidence interval for MLow boards were above the requirements as shown in Table 18. Also, it met the requirement that no individual in the five 54  panel samples had any property more than 20% below the listed five panel average value for that property. Table 18. Comparision between MLow boards and CSA standards for flexural properties MLow (type ML2)  MOR-PD (MPa) MOE-PD (GPa) MOR-PL (MPa) MOE-PL (GPa) Lower 95% confidence interval 17.18 2.15 38.33 5.61 Means of results 20.39 2.41 44.23 6.05 CSA O437.0 Standards ≥ 12.4 ≥ 1.5 ≥ 29.0 ≥ 5.5   4.2.4 Water Absorption All three types of boards were fabricated without the addition of wax, which is normally added at about 1% w/w to wood-based OSB products (SBA 2010). Our previous study (Semple et al. 2015) found the all bamboo surface boards were below the maximum of 15% in 24 h TS required by CSA-O437.0 (2011) whereas pure Aspen boards made without wax were above 15% 24 h TS.  As shown in Table 19, there was no significant difference between MMed and MHigh boards for 2 h or 24 h TS. Both types were at least 43.3% lower than MLow for 2 h TS. However, this difference reduced to 24.7% for 24 h TS. Nevertheless MLow boards still met the requirements of CSA-O437.0 for 24 h TS with the upper 95% confidence interval of 14.3% less than the maximum TS of 15% required by the standard. Figure 32 shows the big difference of 24h thickness swelling between ML2 and other two types. 55  Table 19. Results of thickness swelling (%) for MM, ML1 and ML2 Type Assigned Name Means COV (%) 2h TS (%) MM MHigh 2.86b 40.0 ML1 MMed 2.37b 22.6 ML2 MLow 5.05a 30.3 24h TS (%) CSA Standard: ≤15 MM MHigh 9.57b 4.0 ML1 MMed 9.84b 15.3 ML2 MLow 13.07a 14.0 Note: the levels in the table not followed by same letter are significantly different     Figure 31. Results of 24h TS of MM, ML1 and ML2, n = 6 for each mean (MM)             (ML1)      (ML2) 56  Similarly, there was no significant difference between MMed and MHigh for 2 h or 24 h WA. As shown in Table 20, they absorbed significantly less water compared to MLow at 2 hours. After 24 hours, this difference was still significant. After 2 hours MLow boards absorbed 43.4% of the total weight of water it absorbed in 24 hours, while MMed absorbed 59.0% and MHigh absorbed 56.7% of the total weight of water it absorbed. That indicated most water were absorbed in the first few hours. Figure 33 shows the big difference of 24h water absorption between ML2 and other two types.  Table 20. Results of water absorption (w/%) for MM, ML1 and ML2 Type Assigned Name Means COV (%) 2h WA (w/%) MM MHigh 13.90b 2.5 MH MMed 16.15b 44.8 ML MLow 27.81a 18.3 24h WA (w/%) MM MHigh 32.10b 15.6 MH MMed 34.64b 19.8 ML MLow 51.92a 11.5 Note: the levels in the table not followed by same letter are significantly different  57   Figure 32. Results of 24h WA of MM, ML1 and ML2, n = 6 for each mean All three groups hit the target density with a less than 1.2% difference. The MLow group, which had the lowest density, showed the lowest mechanical properties and worst water resistance. Most tests showed a positive correlation between the properties and the density, even though IB strength property and perpendicular-to-strand direction flexural properties showed that the high density group (MM) was a little weaker than the medium density group (ML1). However there was no significant difference between them in those two tests.   Yet, all properties tests showed that the 25% w/w low density bamboo hybrid board met the requirements by CSA-O437.0 (Table 21). It is noticeable that with a low density (634.8 kg/m3) and no wax addition in the fabrication, bamboo surface/Aspen core 3-layer hybrid boards still showed satisfied properties to meet the requirements for the structure OSB materials. With these (MM)     (ML1)      (ML2) 58  results, it is worth to carry on more research about how to retain the low density but improve the consolidation during the fabrication. Even though the lowest density group showed the weakest mechanical properties and water resistance ability, the lower 95% confidence interval for the group means is greater than the CSA standards (see Table 22). Table 21. Comparison between experiment results and CSA standard Type Assigned Name Means of Density (kg/m³) IB (MPa) MOR-PD (MPa) MOE-PD (GPa) MOR-PL (MPa) MOE-PL (GPa) TS (%) MM MHigh 764.5a 0.769a 20.25a 2.14a 64.93a 8.01a 9.6a ML1 MMed 728.3b 0.800a 24.38a 2.72a,b 59.09a 7.44a 9.8a ML2 MLow 634.8c 0.656b 20.39a 2.41b 44.23b 6.05b 13.1b CSA Standard ≈640* ≥ 0.345 ≥ 12.4 ≥ 1.5 ≥ 29.0 ≥ 5.5 ≤ 15 *: Preferable density, no requirement in standard for density (TECO, 2008) Table 22. Summary of comparisons between MLow group results and CSA standards. MLow (type ML2)  IB (MPa) MOR-PD (MPa) MOE-PD (GPa) MOR-PL (MPa) MOE-PL (GPa) TS (%) Lower 95% confidence interval 0.632 17.18 2.15 38.33 5.61 14.3* Means of results 0.656 20.39 2.41 44.23 6.05 13.1 CSA O437.0 Standards ≥ 0.345 ≥ 12.4 ≥ 1.5 ≥ 29.0 ≥ 5.5 ≤ 15 *: Upper 95% confidence interval  Along with IB strength, TS and WA are possibly related to consolidation of the boards. Winistorfer and Xu (1995) have found that total thickness swelling has two components: the swelling of the wood due to MC change, and a combined effect of residual stress release from the pressing and potential variance between high and low density areas in the plane of the panel. 59  With the similar final size, low density boards have more voids of space for water than the high density boards. The water absorption test result in this section is consistent with that finding.   4.3 Experiment 3: Node Effect on Guadua-Aspen Hybrid OSB As discussed in Chapter 2, the microstructure, strength and density of nodes may affect the property of OSB products made from bamboo. It was expected that internode Guadua-Aspen hybrid OSB would have better properties than node Guadua-Aspen boards. Type GI has a surface made from 50% Guadua strands without node and a core made from 50% aspen strands. Type GN has a surface made from 50% Guadua strands with a node near the middle and a core made from 50% aspen strands. The average node frequency of Guadua was 3.3 nodes per meter of pole, less than Moso poles we imported. However, the nodes of Guadua poles are 2.66 times thicker than those of Moso and are much harder which required a Dremel Saw to remove while the Moso nodes could be removed by hand or a hammer. 4.3.1 Thickness and Density Similar to the other types of board, spring back was observed with these boards. Rather than the target thickness 11.1 mm, both types of boards had a thickness exceeding 11.3 mm (see Table 23). The density of the boards also exceeded the original target by about 0.2% and 0.4%. There was no significant difference between internode Guadua board and node Guadua board for both thickness and density. This result is consistent with previous study on the effect of node on Moso-Aspen hybrid board (Semple et al., 2015b). The node has no significant effect on the spring back of thickness. In terms of density, the GI and GN boards are essentially identical. 60  Table 23. Thickness and density of GI and GN  Type Thickness Density Mean (mm) COV (%) Mean (kg/m3) COV (%) GI 11.35a 2.7 763.3a 11.6 GN 11.39a 3.0 761.8a 10.2 p-value 0.2395 0.3224  4.3.2 Internal Bonding Strength Previously, Semple et al. (2015b) found that the IB strength of boards made from internode or node strands were not significant different, and it was thought that the same result would be found for Guadua. However, IB strength tests showed the presence of nodes had a significant effect on the IB strength. Internode Guadua board (type GI) had a higher IB strength at 0.699 MPa compared to the board made with node strands (type GN) at 0.628 MPa by about 11.48%, which indicated that internode strands were compressed more than the node strands. It is noticeable that even though there was no significant difference between both types on the thickness and density, GN group which was made from nodes Guadua strands had a lower IB could be due to the presence of nodes. 61   Figure 33. IB test results of GI and GN, n = 180 for each mean 4.3.3 Flexural Properties The MOR and MOE for both directions and both types were greater than the minimum required for OSB products by CSA O437.0. And there was no significant difference between their MOE and MOR for the perpendicular to the strand direction (Table 24). The effect of nodes on the perpendicular to the strand direction flexural properties was not observed in these tests. Yet, the MOR and MOE parallel to the strand direction demonstrated higher flexural properties of internode Guadua hybrid board. The direction of vascular bundles in nodes becomes irregular compared to internode material. And as such the stiffness of the node material will be lower than 62  the internode material. The properties that one would expect to see a noticeable difference in would be on MOE. This change in structure should affect TS as well. GI group had a 26.7% higher MOR and a 21.5% higher MOE compared to GN group (Table 25). The presence of nodes on the strands significantly affected the strength of the board, which was consistent with previous study on Moso node effect (Semple et al., 2015b).   Table 24. Flexural property test results of GI and GN (perpendicular) Type Means COV (%)  MOR (MPa) p-value = 0.3224 GI 21.72a 33.1 GN 19.02a 30.4  MOE (GPa) p-value = 0.5087 GI 2.14a 28.1 GN 1.99a 25.8 Note: the levels in the table not followed by same letter are significantly different   Table 25. Flexural property test results of GI and GN (parallel) Type Means COV (%)  MOR (MPa) p-value < 0.0001 GI 74.48a 12.7 GN 58.80b 11.8  MOE (GPa) p-value = 0.0002 GI 11.48a 11.4 GN 9.45b 9.1 Note: the levels in the table not followed by same letter are significantly different  4.3.4 Water Absorption All the boards were fabricated without any addition of wax which is normally added to OSB. The previous study about the node effect on the water resistance ability showed no difference in 24 hours thickness swell and water absorption between boards made with internode or node 63  Moso strands in the surfaces. Contrary to that, the presence of nodes in Guadua strands made a difference on the results. Board made with internode Guadua strands in the surfaces had a 40.6% less thickness swell after 2 hours soaking. After 24 hours soaking in water, internode Guadua hybrid boards still had a 25.8% less thickness swell (Table 26 and Figure 35). Table 26. Results of thickness swelling (%) for GI and GN Type Means COV (%)   2h TS (%) p-value = 0.0406  GI 2.35b 32.7  GN 3.96a 37.6   24h TS (%) p-value = 0.0103  GI 9.24b 7.8  GN 12.45a 5.8  Note: the levels in the table not followed by same letter are significantly different  64   Figure 34. Results of 24h TS of GI and GN, n = 6 for each mean Also, after 2 hours immersed in water, 16.7% more water was absorbed by the boards made with node Guadua strands on the surface. This number reduced to 14.3% after 24 hours soaking in water. However, a significant difference was found between two groups for both 2 hours TS and 24 hours TS, while no significant difference was found between two groups for neither 2 hours WA and 24 hours WA (Table 27 and Figure 36). In the same duration of time, nodes did not affect the amount of water the boards absorbed. Yet, it demonstrated that with similar percent of their initial weight water absorbed, internode strands had a better ability to maintain the form less swelling. Less deformation of internode strands gave Guadua boards better deformation resistance ability. 65  Table 27. Results of water absorption (w/%) for GI and GN Type Means COV (%)   2h WA (w/%) p-value = 0.3033  GI 14.43a 22.9  GN 17.32a 32.5   24h WA (w/%) p-value = 0.0992  GI 35.51a 13.1  GN 39.10a 15.7  Note: the levels in the table not followed by same letter are significantly different   Figure 35. Results of 24h WA of GI and GN, n = 6 for each mean The internal bonding test showed the internode Guadua hybrid board had better properties than the node Guadua hybrid board. However, for the internal bonding strength both of them were lower than the Moso hybrid board which was made from mixed node and internode Moso strands (see section 4.1.2) even though Guadua had a denser vascular bundles and solid fibre 66  structure (Dixon et al., 2015). Regardless of the target thickness and density, there is a possibility for Guadua board, regardless of strand moisture, internode or node, to be compressed further.  The presence of nodes had little effect on the perpendicular to the strands direction flexural properties, but a significant negative effect on the parallel to the strands direction MOR and MOE. It showed the properties of node region could be different along the strands direction but consistent along the perpendicular direction. There was no significant difference in water absorption between the GN and GI samples. This may be due to the fact that the node strands are not as smooth as the internode strands. The node strands may pack less efficiently, leading to greater densification of the node containing part of the strands resulting in more spring back and swelling.  67  Chapter 5: Conclusion and Future Work 5.1 Difference between Moso and Guadua  In the comparison between two species of bamboo, Moso had a greater spring back after the press opened and temperature cooled down, which resulted in larger final thickness. The property test showed Guadua and Moso surface hybrid OSB had similar performance except for IB strength and parallel direction MOE. Contrary to expectation, hybrid OSB with surface made from Guadua bamboo, which is denser than Moso, had a lower IB strength. Greater compaction ratio of Moso bamboo strands is a possible reason for that result. With the same target density and weight ratio of bamboo-aspen, Moso strands filled the space better during the compression than Guadua because a lower density object has a larger volume based on the same weight. With higher density and larger vascular bundles, Guadua yet did not show better properties than Moso. Moso showed better consolidation than Guadua.    5.2 Lighter Bamboo OSB Met the CSA Standard The study on whether bamboo hybrid board could meet the density as well as property standards required by CSA O437.0 (2011) showed that the bamboo boards met the requirements. Even though the lowest density group showed the weakest mechanical properties and water resistance ability, the 95% confidence interval for the group means met the CSA standards. For high density group and medium density group, expect for the significant difference between their density and thickness because the target was set on different level, all the property test showed no significant difference between these two groups. That indicated a non-linear positive 68  correlation between the density and the performance. The density of Moso-Aspen OSB could be as low as normal wood OSB and still show compliant properties required by CSA standards.  5.3 The Effects of Node on Guadua Bamboo OSB The presence of nodes on Guadua strands affected the hybrid OSB products in terms of mechanical properties. As was with Moso, Guadua-Aspen OSB without nodes had higher IB strength, and was stiffer and stronger in the bending test than similar boards made with strands containing nodes. Nodes had negative effect on the properties of Guadua Bamboo OSB.   5.4 Future Work Possible future research could be around how to improve the consolidation of Guadua OSB products, etc. adding wood strands to fill up the space in the structure which result from the rough surfaces of Guadua strands or producing narrower Guadua strands to make hybrid OSB. The effect of the differences between the node structures of Guadua and Moso on the boards’ properties could be studied.  As the point of the project is to find the best even distribution of board density and the strength properties, there is a potential to get a better strength properties within the proper density range. How to adjust the manufacture process and technique to reinforce the board properties is a possible focus of future research.  With knowing the node have negative effect on the board properties, it is of great interest to find a solution to minimize the nodes effects on the final product. At present, the bamboo strands are produced with lab equipment and high-intensity labor involvement. If the bamboo hybrid OSB 69  products are expected to be applied in the practical manufactures field, more studies on how to improve the production flow are required. A possible solution could be removing the nodes during the process of producing strands efficiently, or changing the location of the node on the strands. A better volume ratio of nodes on Guadua culm could also be studied to show how significant of the effect of nodes on boards’ properties. In the future work, more studies could be helpful to promote the three layer hybrid products as practical construction material. 70  References Amada, S., & Lakes, R. S. (1997). Viscoelastic properties of bamboo. Journal of Materials Science, 32(10), 2693–2697. Anon. (2012). Aiming for domination. Retrieved April 10, 2013, from http://www.wbpionline.com/ Archila, H. F., Takeuchi, C. P., & Trujillo, D. J. A. (2015). Mechanical and Physical Characterization of Composite Bamboo-Guadua Products: plastiguadua. In 68th FPS-IC and WCTE. Structural Board Association (1998). OSB performance by design. SBA and US Products Standard (PS-92) (2005th ed.). Retrieved from http://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:OSB+Performance+by+Design#1\nhttp://scholar.google.com/scholar?hl=en&btnG=Search&q=intitle:OSB+performance+by+design#1 ASTM. (2012). Standard Test Methods for Direct Moisture Content Measurement of Wood and Wood-. http://doi.org/10.1520/D4442-07.These ASTM. (2014). Standard Test Method for Internal Bond Strength and Thickness Swell of Cellulosic- Based Fiber and Particle Panels After Repeated Wetting 1. http://doi.org/10.1520/D7519 Austin, R., Ueda, K., & Levy, D. (1972). Bamboo. New York: Weatherhill. Bakar, E. S., Sahrani, M. D., & Ashaari, Z. (2013). A practical method and apparatus for converting bamboo culms into flat sheets for laminated bamboo timber production. In 21st International Wood Machining Seminar (pp. 171–179). Tsukuba, Japan. Benetto, E., Becker, M., & Welfring, J. (2009). Life Cycle Assessment of Oriented Strand Board 71  (OSB): from Process Innovation to Ecodesign. Environmental Science & Technology, 43(15), 6003–6009. Birkeland, J. (2002). Design for sustainability: A sourcebook of integrated, eco-logical solutions. Earthscan. Calderón, C. E., & Soderstrom, T. R. (1980). The Genera of Bambusoideae (Poaceae) of the American Continent: Keys. Smithsonian Contributions to Botany, (44), 1–27. http://doi.org/10.5479/si.0081024X.44 Standards Council of Canada (2011). CSA-O437 Standards on OSB and Waferboard. Standard Council of Canada. Clark, L. (2006). Bamboo Biodiversity. Retrieved from http://www.eeob.iastate.edu/research/bamboo/bamboo.html Correal, J. F., Echeverry, J. S., Ramírez, F., & Yamín, L. E. (2014). Experimental evaluation of physical and mechanical properties of Glued Laminated Guadua angustifolia Kunth. Construction & Building Materials, 73, 105–112. http://doi.org/10.1016/j.conbuildmat.2014.09.056 Correal, J. F., & Ramirez, F. (2010). Adhesive bond performance in glue line shear and bending for glued laminated guadua bamboo. Journal of Tropical Forest Science, 22(4), 433–439. Correal, J. F., & Varela, S. (2012). Shear walls sheathed with glued laminated Guadua bamboo panels subjected to lateral loads. In World Congress on Timber Engineering. Auckland, NZ. de Vos, V. (2010). Bamboo for Exterior joinery - a research in material properties and market perspectives, 80. Dixon, P. G., Ahvenainen, P., Aijazi, A. N., Chen, S. H., Lin, S., & Gibson, L. J. (2015). Comparison of the structure and flexural properties of Moso, Guadua and Tre Gai bamboo. 72  Construction and Building Materials, 90, 11–17. http://doi.org/10.1016/j.conbuildmat.2015.04.042 Flander, K. D., & Rovers, R. (2009). One laminated bamboo-frame house per hectare per year. Construction and Building Materials, 23(1), 210–218. http://doi.org/10.1016/j.conbuildmat.2008.01.004 Fu, W. (2007a). A study on flaking technique and manufacture of bamboo OSB. Northeast Forestry University. Fu, W. (2007b). Bamboo-A potential resource of raw material for OSB in China. China Forest Prod Ind, 34(2), 21–24. Gorte, R. W. (2009). Carbon Sequestration in Forests. Congressional Research Service, 4, 2–37. http://doi.org/10.1079/PAVSNNR20094041 Grossenbacher, M. (2012). Industrial production of bamboo OSB: A technical report. In 2nd Bieler Holzwekstoff-Workshop (pp. 28–29). Biel, Switzerland. Grosser, D., & Liese, W. (1971). On the Anatomy of Asian Bamboos, with Special Reference to their Vascular Bundles. Wood Science and Technology, 5(4), 290–312. Guinness World Records. (n.d.). Fastest Growing Plant. Retrieved March 13, 2015, from http://www.guinnessworldrecords.com/world-records/fastest-growing-plant/ Hood, J. P. (2004). Changes in Oriented Strandboard Permeability during Hot-Pressing. Virginia Polytechnic Institute and State University. Hua, Y. (2003). Projected consumption of OSB in Modern Chinese Building Systems. China Wood Industry, 17(6), 1–3. Idris, A., Firmanti, A., & Purwito. (1994). Strategi Penelitian Bambu Indonesia [Research Strategies for Indonesia Bamboo]. In E. A. Wijaya (Ed.), (pp. 73–81). Bogor: Yayasan 73  Bambu Lingkungan Lestari. Jiang, Z. (2002). Bamboo and rattan in the world. shenyang: Liaoning Science and Technology Publishing House. JMP Support. (2016). BASIC ANALYSIS - ONEWAY ANALYSIS - COMPARE MEANS. Retrieved from http://www.jmp.com/support/help/Compare_Means.shtml Jyoti Nath, A., Das, G., & Das, A. K. (2009). Above ground standing biomass and carbon storage in village bamboos in North East India. Biomass and Bioenergy, 33(9), 1188–1196. http://doi.org/10.1016/j.biombioe.2009.05.020 Kelchner, S. A., & Group, B. P. (2013). Higher level phylogenetic relationships within the bamboos ( Poaceae : Bambusoideae ) based on five plastid markers. Molecular Phylogenetics and Evolution, 67, 404–413. Lee,  A. W. C., Chen, G., & Tainter, F. H. (2001). Comparative treatability of Moso bamboo and Southern pine with CCA preservative using a commercial schedule. Bioresource Technology, 77(1), 87–88. http://doi.org/10.1016/S0960-8524(00)00145-0 Lee, A. W. C., Bai, X., & Bangi, A. P. (1998). Selected Properties of Laboratory-Made Laminated-Bamboo Lumber. Holzforschung, 52, 207–210. Lee, A. W. C., Bai, X., & Peralta, P. N. (1994). selected physical and mechanical properties of giant timber bamboo grown in south carolina. Forest Products Journal, 44(9), 14–17. http://doi.org/http://dx.doi.org/10.1108/17506200710779521 Lee, A. W. C., Bai, X. S., & Peralta, P. N. (1996). Physical and mechanical properties of strandboard made from Moso bamboo. Forest Products Journal, 46(11-12), 84–88. Leithoff, H., & Peek, R. D. (2001). Heat treatment of bamboo. In International research on wood preservation 32nd annual meeting. Nara, Japan. 74  Lewis, D., Miles, C., Hooper, A., Turtle, A., Akers, J., & Ernst, M. (2007). Farming Bamboo. Lulu Press. Li. (2013). Personal communication during a visit to Chengfeng Bamboo Industry Co. in Zhejiang province, China. Anji, Zhejiang, China. Li, H., & Shen, S. (2011). The mechanical properties of bamboo and vascular bundles. Journal of Materials Research, 26(21), 2749–2756. http://doi.org/10.1557/jmr.2011.314 Li, X. (2004). Physical, Chemical, and Mechanical Properties of Bamboo and Its Utilization Potential for Fiberboard Manufacturing. Liese, W. (1987). Research on bamboo. Wood Science and Technology, 21(3), 189–209. http://doi.org/10.1007/BF00351391 Liese, W. (1998). The anatomy of bamboo culms. INBAR Technical Report 18. Beijing. Liu, X., Smith, G. D., Jiang, Z., Bock,  maximilian C. D., Boeck, F., Frith, O., & Zhang, P. K. (2016). Nomenclature for Engineered Bamboo. BioResources, 11(1), 1141–1161. Lo, T. Y., Cui, H. Z., & Leung, H. C. (2004). The effect of fiber density on strength capacity of bamboo. Materials Letters, 58(21), 2595–2598. http://doi.org/10.1016/j.matlet.2004.03.029 Low, I. M., Che, Z. Y., Latella, B. A., & Sim, K. S. (2006). Mechanical and Fracture Properties of Bamboo. Key Engineering Materials. http://doi.org/10.4028/www.scientific.net/KEM.312.15 Mahdavi, M., Clouston, P. L., & Arwade, S. R. (2012). A low-technology approach toward fabrication of Laminated Bamboo Lumber. Construction and Building Materials, 29, 257–262. http://doi.org/10.1016/j.conbuildmat.2011.10.046 Nugroho, N., & Ando, N. (2001). Development of structural composite products made from bamboo II: Fundamental properties of laminated bamboo lumber. Journal of Wood Science, 75  47(3), 237–242. http://doi.org/10.1007/BF01171228 Paudel, S. K., & Lobovikov, M. (2003). Bamboo housing : market potential for low-income groups. Journal of Bamboo and Rattan, 2(4), 381–396. Peng, Z., Lu, T., Li, L., Liu, X., Gao, Z., Hu, T., Yang, X., Feng, Q., Guan, J., Wen, Q., Fan, D., & Jiang, Z. (2010). Genome-wide characterization of the biggest grass, bamboo, based on 10,608 putative full-length cDNA sequences. BMC Plant Biology, 10, 116. http://doi.org/10.1186/1471-2229-10-116 Riaño, N., Londoño, X., López, Y., Gómez, J., & Camayo, G. C. (2002). Plant growth and biomass distribution on Guadua angustifolia Kunth in relation to ageing in the Valle del Cauca - Colombia. Bamboo Science and Culture, 16(1), 43–51. Retrieved from http://www.bamboo.org/publications/e107_files/downloads/ABSJournal-vol16.pdf#page=3 Rittironk, S., & Elnieiri, M. (2008). Investigating laminated bamboo lumber as an alternate to wood lumber in residential construction in the United States. Taylor & Francis Group, 83–96. Rivière, A., & Rivière, C. (1878). Les bambous : végétation, culture, multiplication en Europe, en Algérie et généralement dans tout le bassin méditerranéen nord de l’Afrique, Maroc, Tunisie, Egypte / par Auguste Rivière et Charles Rivière. Paris: Au siége de la Société d’acclimatation. Schroder, S. (2014). What is Guadua angustifolia. Retrieved August 28, 2014, from http://www.guaduabamboo.com/guadua/what-is-guadua-angustifolia Semple, K. E., Smola, M., Hoffman, J., & Smith, G. D. (2014). Optimising the Stranding of Moso Bamboo ( Phyllostachys Pubescens Mazel ) Culms Using a CAE 6 / 36 Disk Flaker. In M. Barnes & V. Herian (Eds.), Proc 57th international convention of society of wood 76  science and technology (pp. 257–269). Zvolen, Slovakia. Semple, K. E., Zhang, P. K., & Smith, G. D. (2015). Hybrid oriented strand boards made from Moso bamboo (Phyllostachys pubescens Mazel) and Aspen (Populus tremuloides Michx.): species-separated three-layer boards. European Journal of Wood and Wood Products, 73(4), 527–536. http://doi.org/10.1007/s00107-015-0914-0 Semple, K. E., Zhang, P. K., & Smith, G. D. (2015a). Stranding Moso and Guadua Bamboo. Part I: Strand Production and Size Classification. BioResources, 10(3), 4048–4064. Semple, K. E., Zhang, P. K., & Smith, G. D. (2015b). Stranding Moso and Guadua Bamboo. Part II. Strand Surface Roughness and Classification. BioResources, 10(3), 4599–4612. Semple, K. E., Zhang, P. K., Smola, M., & Smith, G. D. (2015). Hybrid Oriented Strand Boards made from Moso bamboo (Phyllostachys pubescens Mazel) and Aspen (Populus tremuloides Michx.): uniformly mixed single layer uni-directional boards. European Journal of Wood and Wood Products, 73(4), 515–525. http://doi.org/10.1007/s00107-015-0913-1 Shaddy, W. (2008). The Advantages of Carbide Cutting Tools. Retrieved June 23, 2016, from http://homeguides.sfgate.com/advantages-carbide-cutting-tools-99598.html Smola, M. (2013). Producing structural elements from moso bamboo culms. Rosenheim University. Spark, W. (2014). Average Weather For Vancouver. Retrieved March 1, 2014, from https://weatherspark.com/averages/28404/Vancouver-British-Columbia-Canada Sulastiningsih, I. M., & Nurwati. (2009). Physical and mechanical properties of laminated bamboo board. Journal of Tropical Forest Science, 21(3), 246–251. Sumardi, I., Ono, K., & Suzuki, S. (2007). Effect of board density and layer structure on the 77  mechanical properties of bamboo oriented strandboard. Journal of Wood Science, 53(6), 510–515. http://doi.org/10.1007/s10086-007-0893-9 TECO. (2008). Design Capacities for Oriented Strand Board Wood Structural Panel Design Capacities Based on Span Ratings (ASD). Retrieved from http://www.tecotested.com/techtips/pdf/tt_osbdesigncapacities Voermans, J. (2006). Glued laminated bamboo: analysis of bamboo applied in laminated beams. Eindhoven University of Technology. Wegst, U. G. K., Shereliff, H. R., & Ashby, M. F. (1993). The Structure and Properties of Bamboo as an Engineering Material. Cambridge, UK: University of Cambridge. Xu, W., & Winistorfer, P. M. (1995). Layer thickness swell and layer internal bond of medium density fiberboard and oriented strandboard. Forest Products Journal, 45(10), 67–71. Yu, H. Q., Jiang, Z. H., Hse, C. Y., & Shupe, T. F. (2008). Selected physical and mechanical properties of moso bamboo (phyllostachys pubescens). Journal of Tropical Forest Science, 20(4), 258–263. Zhang, H., Du, F., Zhang, F., Liao, Z., Ye, X., Zheng, Z., & Wang, W. (2006). Research and Development of Production Technology of Bamboo Waferboard and Oriented Strand Board Based on Biological Characteristics and Timber Adaptability. Journal of Bamboo Research, 26(2), 43–48. Zhang, P. K. (2013). Improving the properties of aspen oriented strand board (OSB) by using moso bamboo strands in the surface layers. University of British Columbia. Zhou, C., Smith, G. D., & Dai, C. (2009). Characterizing hydro-thermal compression behavior of aspen wood strands. Holzforschung, 63(5), 609–617. http://doi.org/10.1515/HF.2009.111   78  Appendices Appendix A  : Press Schedule All 36 boards were pressed in 6 working days. Press sequence Board Board code Day Task Furnish weight, kg Flake type Layer 1,2,3 MM1,MM2,MM3 1 blend 6.4 Moso surface 1,2 MM1,MM2 1 blend 6.4 Aspen core 1 MM1 1 form and press 4.26 Hybrid surface+core 2 MM2 1 form and press 4.26 Hybrid surface+core 3,4 MM3,MM4 1 blend 6.4 Aspen core 3 MM3 1 form and press 4.26 Hybrid surface+core 4,5,6 MM4,MM5,MM6 1 blend 6.4 Moso surface 4 MM4 1 form and press 4.26 Hybrid surface+core 5,6 MM5,MM6 1 blend 6.4 Aspen core 5 MM5 1 form and press 4.26 Hybrid surface+core 6 MM6 1 form and press 4.26 Hybrid surface+core 1,2,3 GM1,GM2,GM3 2 blend 6.4 Guadua surface 1,2 GM1,GM2 2 blend 6.4 Aspen core 1 GM1 2 form and press 4.26 Hybrid surface+core 2 GM2 2 form and press 4.26 Hybrid surface+core 3,4 GM3,GM4 2 blend 6.4 Aspen core 3 GM3 2 form and press 4.26 Hybrid surface+core 4,5,6 GM4,GM5,GM6 2 blend 6.4 Guadua surface 4 GM4 2 form and press 4.26 Hybrid surface+core 5,6 GM5,GM6 2 blend 6.4 Aspen core 5 GM5 2 form and press 4.26 Hybrid surface+core 6 GM6 2 form and press 4.26 Hybrid surface+core 79  Press sequence – Cont. 1 Board Board code Day Task Furnish weight, kg Flake type Layer 1,2,3 GI1,GI2,GI3 3 blend 6.4 Guadua surface 1,2 GI1,GI2 3 blend 6.4 Aspen core 1 GI1 3 form and press 4.26 Hybrid surface+core 2 GI2 3 form and press 4.26 Hybrid surface+core 3,4 GI3,GI4 3 blend 6.4 Aspen core 3 GI3 3 form and press 4.26 Hybrid surface+core 4,5,6 GI4,GI5,GI6 3 blend 6.4 Guadua surface 4 GI4 3 form and press 4.26 Hybrid surface+core 5,6 GI5,GI6 3 blend 6.4 Aspen core 5 GI5 3 form and press 4.26 Hybrid surface+core 6 GI6 3 form and press 4.26 Hybrid surface+core 1,2,3 GN1,GN2,GN3 4 blend 6.4 Guadua surface 1,2 GN1,GN2 4 blend 6.4 Aspen core 1 GN1 4 form and press 4.26 Hybrid surface+core 2 GN2 4 form and press 4.26 Hybrid surface+core 3,4 GN3,GN4 4 blend 6.4 Aspen core 3 GN3 4 form and press 4.26 Hybrid surface+core 4,5,6 GN4,GN5,GN6 4 blend 6.4 Guadua surface 4 GN4 4 form and press 4.26 Hybrid surface+core 5,6 GN5,GN6 4 blend 6.4 Aspen core 5 GN5 4 form and press 4.26 Hybrid surface+core 6 GN6 4 form and press 4.26 Hybrid surface+core 1,2,3,4,5,6 ML1(1-6) 5 blend 6.05 Moso surface 1,2 ML1 (1,2) 5 blend 6.05 Aspen core 1 ML1(1) 5 form and press 4.04 Hybrid surface+core 2 ML1(2) 5 form and press 4.04 Hybrid surface+core 3,4 ML1 (3,4) 5 blend 6.05 Aspen core 80  Press sequence – Cont. 2 Board Board code Day Task Furnish weight, kg Flake type Layer 3 ML1(3) 5 form and press 4.04 Hybrid surface+core 4 ML1(4) 5 form and press 4.04 Hybrid surface+core 5,6 ML1(5, 6) 5 blend 6.05 Aspen core 5 ML1(5) 5 form and press 4.04 Hybrid surface+core 6 ML1(6) 5 form and press 4.04 Hybrid surface+core 1,2,3,4,5,6 ML2 (1-6) 6 blend 5.28 Moso surface 1,2 ML2 (1,2) 6 blend 5.28 Aspen core 1 ML2(1) 6 form and press 3.52 Hybrid surface+core 2 ML2(2) 6 form and press 3.52 Hybrid surface+core 3,4 ML2 (3,4) 6 blend 5.28 Aspen core 3 ML2(3) 6 form and press 3.52 Hybrid surface+core 4 ML2(4) 6 form and press 3.52 Hybrid surface+core 5,6 ML2(5,6) 6 blend 5.28 Aspen core 5 ML2(5) 6 form and press 3.52 Hybrid surface+core 6 ML2(6) 6 form and press 3.52 Hybrid surface+core     81  Appendix B  : Test Results Internal Bonding Test has 30 specimens per board, total 1080 specimens.  82   83   84   85   86   87   88   89   90   91   92   93   ML1 ML2 94   ML1 ML2 95   ML1 ML2 96   ML1 ML2 97   98   99   100   101   102   103   104   105   106   107   108   109   110   111   112   113   114   115   116        Flexural Properties Test has 4 specimens per board, total 144 specimens   Board type: Sample ID:GM GuaduaMixed 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)MM MosoMixed 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Mixed means internode and node mixed Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm1 240 76.27 11.24 11.34 11.31 11.24 11.28 14.2387 1.0477 384.00 33.21 198.00 6.030 110.00 3.3803 240 76.18 11.34 11.54 11.26 11.15 11.32 14.7816 1.3487 401.00 43.15 216.00 4.860 112.00 2.4501 240 75.20 11.26 11.37 11.53 11.24 11.35 23.1516 1.7749 623.00 56.47 285.00 5.070 106.00 1.9003 240 76.17 11.32 11.10 11.40 11.46 11.32 20.7282 1.9021 562.00 60.81 293.00 5.020 113.00 2.0601 240 76.35 11.32 10.97 11.25 11.73 11.32 23.7071 2.7700 644.00 88.71 333.00 3.920 113.00 1.4403 240 76.87 11.59 11.51 11.59 11.56 11.56 24.4861 2.2241 699.00 76.47 320.00 4.240 112.00 1.5201 240 76.14 11.03 11.06 11.21 11.06 11.09 16.2233 1.8340 422.00 55.11 239.00 4.330 115.00 2.0803 240 76.25 11.49 11.37 11.58 11.54 11.50 24.4042 1.9413 683.00 65.06 344.00 5.310 115.00 1.7901 240 76.25 11.41 11.36 11.65 11.53 11.49 23.4701 2.0645 656.00 69.05 256.00 3.770 111.00 1.6703 240 76.17 10.92 11.26 11.11 10.90 11.05 26.8751 2.3996 694.00 71.31 278.00 4.020 109.00 1.6501 240 76.12 11.60 11.71 11.50 11.40 11.55 24.3449 2.1345 687.00 72.49 355.00 4.960 110.00 1.5803 240 76.04 11.19 11.33 11.36 11.14 11.26 26.7971 2.4763 717.00 77.68 307.00 4.050 133.00 1.8101 240 75.36 11.03 11.26 11.10 11.01 11.10 16.0515 2.0442 414.00 60.96 226.00 3.690 112.00 1.8203 240 76.40 11.82 11.86 11.64 11.79 11.78 25.3421 2.1300 746.00 76.92 320.00 4.280 110.00 1.5501 240 76.04 11.68 11.46 11.67 11.84 11.66 12.9137 1.6346 371.00 57.05 201.00 3.480 112.00 1.9203 240 76.13 11.53 11.34 11.24 11.20 11.33 21.6330 2.0999 587.00 67.23 229.00 3.360 110.00 1.5901 240 76.26 11.30 11.47 11.27 11.44 11.37 25.5978 2.4777 701.00 80.36 331.00 4.070 110.00 1.3203 240 75.74 11.52 11.41 12.51 11.68 11.78 20.2772 2.1606 592.00 77.40 271.00 3.520 110.00 1.4401 240 76.32 11.75 11.73 11.92 11.81 11.80 21.7057 2.0819 641.00 75.59 273.00 3.650 112.00 1.5203 240 75.97 11.36 11.36 11.40 11.35 11.37 14.6320 1.6550 399.00 53.44 212.00 4.130 111.00 2.2401 240 75.95 11.66 11.75 11.75 11.71 11.72 15.5697 2.0906 451.00 73.91 247.00 3.450 111.00 1.6103 240 76.24 11.26 11.32 11.25 11.17 11.25 12.7597 1.6343 342.00 51.33 188.00 3.640 111.00 2.1401 240 76.13 11.68 11.53 11.44 11.55 11.55 36.7943 3.2122 1038.00 109.03 406.00 3.940 104.00 1.1703 240 75.92 11.25 11.56 11.30 11.39 11.38 19.6797 2.4713 537.00 79.90 278.00 3.600 115.00 1.560ThicknessAverage thickness (mm)ResultsSample IDLength of spanGM2GM3GM4GM5GM6Board type Board No.Data Sheet for MOR/MOE - GM/MM perpendicularMMMM1MM5MM6GMGM1MM2MM3MM4118   Board type: Sample ID:GM GuaduaMixed 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)MM MosoMixed 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Mixed means internode and node mixed Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm2 240 76.29 11.40 11.28 11.52 11.37 11.39 53.7367 9.8687 1478.00 322.12 1053.00 3.470 383.00 1.3904 240 76.65 11.64 11.63 11.75 11.52 11.64 38.4412 7.8071 1108.00 272.73 450.00 1.760 120.00 0.5502 240 75.93 11.17 11.10 11.21 11.05 11.13 61.3250 10.7473 1603.00 325.77 850.00 2.720 534.00 1.7504 240 76.29 11.83 11.67 11.43 11.77 11.68 80.4904 11.0799 2325.00 389.22 1469.00 3.860 566.00 1.5402 240 75.38 11.56 11.34 11.50 11.71 11.53 51.1783 9.3462 1424.00 312.26 441.00 1.450 110.00 0.3904 240 77.24 11.07 11.04 11.10 11.13 11.09 64.5198 12.3375 1701.00 375.58 440.00 1.210 117.00 0.3502 240 76.07 11.43 11.58 11.55 11.62 11.55 53.8980 9.5040 1518.00 321.90 432.00 1.360 94.00 0.3104 240 76.12 11.11 11.05 11.12 10.98 11.07 72.4273 13.0595 1875.00 389.68 1134.00 2.890 530.00 1.3402 240 76.15 11.06 11.09 11.04 11.00 11.05 68.4836 10.7075 1768.00 318.11 830.00 2.790 426.00 1.5204 240 76.31 11.73 11.69 11.90 11.78 11.78 66.2468 9.5149 1947.00 343.00 452.00 1.440 109.00 0.4402 240 76.35 11.60 11.50 11.51 11.45 11.52 72.2230 10.7897 2031.00 363.95 953.00 2.630 418.00 1.1604 240 76.75 11.60 11.18 11.28 11.36 11.36 88.4008 11.7055 2430.00 380.59 1264.00 3.370 617.00 1.6702 240 75.97 11.76 11.78 11.84 11.52 11.73 77.8665 8.2477 2259.00 292.24 444.00 1.620 105.00 0.4604 240 76.07 11.44 11.29 11.36 11.16 11.31 74.1088 9.2466 2004.00 294.64 442.00 1.620 112.00 0.5002 240 75.95 11.83 12.04 11.95 11.76 11.90 61.7741 8.9938 1844.00 332.65 443.00 1.340 117.00 0.3604 240 76.04 11.42 11.32 11.46 11.46 11.42 64.9282 7.6160 1787.00 249.24 433.00 1.820 104.00 0.5002 240 76.12 11.52 11.52 11.70 11.55 11.57 47.4623 7.2902 1344.00 248.85 445.00 1.800 119.00 0.4904 240 75.73 11.94 11.92 12.03 11.96 11.96 65.7078 8.0204 1978.00 300.85 461.00 1.750 109.00 0.5802 240 75.87 11.73 11.47 11.68 11.82 11.68 62.9734 7.6026 1809.00 265.60 436.00 1.750 104.00 0.5004 240 76.55 11.78 11.51 11.76 11.70 11.69 53.6047 7.2059 1557.00 254.81 446.00 1.790 102.00 0.4402 240 75.97 11.71 11.83 11.86 11.58 11.75 63.7920 8.5001 1857.00 302.73 448.00 1.600 115.00 0.5004 240 76.24 11.48 11.25 11.54 11.73 11.50 57.3773 6.6772 1607.00 224.03 447.00 2.100 102.00 0.5602 240 76.19 11.50 11.36 11.56 11.48 11.48 65.5957 8.0862 1828.00 269.35 446.00 1.750 112.00 0.5104 240 75.91 11.64 11.71 11.73 11.69 11.69 83.9121 8.5825 2419.00 301.34 825.00 2.850 376.00 1.360MM2MM3GMGM1GM2GM3GM4GM5GM6MMMM1MM4MM5MM6Data Sheet for MOR/MOE - GM/MM parallelBoard type Board No. Sample IDLength of spanThicknessAverage thickness (mm)Results119   Board type: Sample ID:MH Moso High Density 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)ML Moso Low Density 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm1 240 75.95 11.64 11.43 11.38 11.40 11.46 29.8708 3.1384 828.00 103.87 399.00 4.020 104.00 1.1803 240 76.06 11.35 11.37 11.42 11.76 11.48 26.7432 2.9102 744.00 96.77 381.00 4.460 111.00 1.6701 240 75.94 11.39 11.30 11.37 11.37 11.36 19.5147 2.4070 531.00 77.49 259.00 3.380 111.00 1.4703 240 76.17 11.49 11.51 11.54 11.69 11.56 20.8404 2.4959 589.00 84.92 281.00 3.450 112.00 1.4601 240 76.38 11.86 12.58 11.76 11.83 12.01 36.8746 3.6179 1128.00 138.43 434.00 3.160 117.00 0.8703 240 76.15 11.03 10.98 11.15 11.06 11.06 24.1766 2.6336 625.00 78.40 302.00 3.860 106.00 1.3601 240 76.58 11.32 11.50 11.48 11.46 11.44 16.8105 2.3614 468.00 78.34 281.00 3.690 111.00 1.5203 240 75.98 11.35 11.27 11.56 11.14 11.33 18.6395 2.5856 505.00 82.68 267.00 3.371 109.00 1.4601 240 76.24 11.20 11.22 11.27 11.67 11.34 28.8980 2.7887 787.00 89.71 326.00 3.630 108.00 1.2003 240 75.95 11.50 11.98 11.55 11.71 11.69 29.5078 3.2442 850.00 113.75 387.00 3.390 114.00 0.9901 240 76.03 11.21 11.15 11.36 11.36 11.27 23.1505 2.3490 621.00 73.97 273.00 3.570 111.00 1.3803 240 76.11 11.42 11.43 11.47 11.29 11.40 17.5351 2.1480 482.00 70.13 219.00 3.050 111.00 1.5101 240 76.03 11.42 11.36 11.33 11.38 11.37 21.5636 2.8309 589.00 91.60 233.00 2.540 113.00 1.2303 240 76.00 11.15 11.18 11.08 10.94 11.09 22.0788 2.5176 573.00 75.46 273.00 3.540 110.00 1.3801 240 76.25 11.65 11.48 11.47 11.67 11.57 19.4065 2.4429 550.00 83.42 264.00 3.140 108.00 1.2703 240 75.01 11.28 11.29 11.37 11.22 11.29 21.0102 2.5640 558.00 80.08 299.00 3.820 110.00 1.4601 240 76.22 11.38 11.52 11.37 11.62 11.47 20.2035 2.0019 563.00 66.67 224.00 3.160 112.00 1.4803 240 75.94 11.34 11.42 11.56 11.60 11.48 17.1220 1.6407 476.00 54.55 191.00 3.240 113.00 1.8101 240 76.33 10.88 10.90 10.93 11.19 10.98 20.2828 2.9700 518.00 86.71 238.00 2.700 114.00 1.2703 240 76.37 11.38 11.90 11.24 11.38 11.48 21.5154 2.5724 601.00 85.89 324.00 3.710 117.00 1.3001 240 76.15 11.31 11.22 11.23 11.48 11.31 23.0987 3.0006 625.00 95.65 287.00 3.070 111.00 1.2303 240 76.35 11.09 11.34 11.06 10.90 11.10 19.7174 2.0175 515.00 60.91 227.00 3.630 107.00 1.6601 240 76.24 11.09 10.93 11.10 11.21 11.08 22.9519 2.4316 597.00 73.02 249.00 3.580 111.00 1.6903 240 76.12 11.39 11.51 11.25 11.09 11.31 15.6763 1.9014 424.00 60.59 216.00 3.450 113.00 1.750ML2ML3ML4MLML1ML5ML6MHMH1MH2MH3MH4MH5MH6Data Sheet for MOR/MOE - MH/ML perpendicularBoard type Board No. Sample IDLength of spanThicknessAverage thickness (mm)Results120   Board type: Sample ID:MH Moso High Density 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)ML Moso Low Density 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm2 240 75.23 12.03 11.96 12.11 12.26 12.09 37.9112 6.5960 1158.00 253.73 445.00 1.780 105.00 0.4404 240 75.73 11.15 11.08 11.37 11.26 11.22 57.7888 8.0419 1529.00 248.57 453.00 1.830 105.00 0.4302 240 76.34 11.56 11.77 11.65 11.58 11.64 53.8437 6.7247 1547.00 234.27 443.00 2.000 108.00 0.5704 240 76.32 11.54 11.73 11.52 11.57 11.59 40.6636 6.0111 1158.00 206.67 451.00 2.270 110.00 0.6202 240 76.32 11.41 11.88 11.68 11.36 11.58 63.9576 7.9437 1819.00 272.58 446.00 1.720 108.00 0.4804 240 75.55 11.80 11.53 12.03 11.75 11.78 50.6360 6.6996 1474.00 239.26 434.00 1.930 111.00 0.5802 240 76.31 11.72 12.00 11.97 11.68 11.84 63.0382 7.7188 1874.00 283.06 454.00 1.620 103.00 0.3804 240 76.16 11.26 11.11 11.30 11.22 11.22 69.7711 8.0631 1859.00 251.15 435.00 1.910 106.00 0.6002 240 76.07 11.81 12.00 11.76 11.72 11.82 70.1891 8.5446 2073.00 310.78 432.00 1.480 115.00 0.4604 240 76.25 11.42 11.37 11.59 11.57 11.49 61.7880 7.6086 1727.00 254.48 455.00 1.870 114.00 0.5302 240 75.88 11.55 11.55 11.75 11.63 11.62 69.6764 7.6938 1983.00 265.04 434.00 1.670 108.00 0.4404 240 76.19 11.35 11.50 11.51 12.05 11.60 69.8130 7.5741 1989.00 260.80 433.00 1.670 107.00 0.4202 240 76.19 11.64 11.42 11.54 11.66 11.57 49.9885 7.0851 1415.00 241.61 444.00 1.900 113.00 0.5304 240 75.28 11.33 11.30 11.47 11.29 11.35 57.5644 7.1905 1550.00 228.86 445.00 1.990 104.00 0.5002 240 76.35 11.52 11.50 11.60 11.69 11.58 42.1777 5.0088 1199.00 171.72 447.00 2.680 107.00 0.7004 240 75.13 11.85 11.53 11.37 11.63 11.60 46.9033 6.0806 1316.00 206.06 444.00 2.180 104.00 0.5302 240 76.66 11.73 11.71 11.62 11.91 11.74 27.4163 4.9998 805.00 179.57 443.00 2.550 109.00 0.6904 240 76.13 10.89 11.22 10.94 10.98 11.01 43.5545 6.2146 1116.00 182.58 434.00 2.420 109.00 0.6402 240 76.35 11.27 11.08 10.98 11.03 11.09 43.8971 5.9702 1145.00 179.89 453.00 2.610 113.00 0.7204 240 76.31 11.69 11.67 11.80 11.60 11.69 47.0185 6.1762 1362.00 217.86 441.00 2.200 136.00 0.8002 240 76.12 10.91 10.65 10.99 11.13 10.92 40.1364 6.6246 1012.00 190.00 447.00 2.470 105.00 0.6704 240 76.01 11.72 11.79 11.77 11.85 11.78 57.2124 6.3314 1677.00 227.78 438.00 2.010 110.00 0.5702 240 76.08 11.82 12.00 11.89 11.82 11.88 33.9154 5.2379 1012.00 193.45 439.00 2.350 114.00 0.6704 240 76.18 11.08 11.04 11.09 11.10 11.08 41.0136 5.6414 1065.00 169.04 446.00 2.750 113.00 0.780ML2ML3ML4MLML1ML5ML6MHMH1MH2MH3MH4MH5MH6Data Sheet for MOR/MOE - MH/ML parallelBoard type Board No. Sample IDLength of spanThicknessAverage thickness (mm)Results121   Board type: Sample ID:GN Guadua Node 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)GI Guadua Internode 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm1 240 76.01 11.16 11.33 11.20 11.15 11.21 19.9378 2.0237 529.00 62.70 269.00 4.290 111.00 1.7703 240 75.77 11.14 11.11 11.07 11.58 11.23 23.9445 2.6386 635.00 81.82 297.00 3.860 108.00 1.5501 240 75.99 11.27 11.73 11.36 11.18 11.39 28.3258 2.7512 775.00 89.27 321.00 3.850 113.00 1.5203 240 76.47 10.98 10.92 10.90 11.14 10.99 9.9874 1.2247 256.00 35.92 177.00 4.240 140.00 3.2101 240 76.12 11.74 11.86 11.73 11.69 11.76 14.3750 1.4301 420.00 51.16 207.00 3.810 141.00 2.5203 240 76.02 11.56 11.31 11.49 11.51 11.47 19.6981 1.9540 547.00 64.81 252.00 3.980 112.00 1.8201 240 75.94 11.28 11.30 11.36 11.36 11.33 23.3230 2.3931 631.00 76.38 300.00 4.060 106.00 1.5203 240 76.23 11.77 11.94 11.61 11.61 11.73 23.5696 2.4416 687.00 86.98 298.00 3.430 111.00 1.2801 240 76.03 10.87 10.90 10.71 10.81 10.82 11.8853 1.3547 294.00 37.78 164.00 4.500 113.00 3.1503 240 76.26 11.56 11.45 11.72 11.67 11.60 23.7508 2.2013 677.00 75.82 347.00 4.660 115.00 1.6001 240 75.95 11.66 11.32 11.59 11.86 11.61 13.4740 1.5644 383.00 53.77 223.00 4.370 116.00 2.3803 240 76.00 11.35 11.32 11.40 11.53 11.40 15.9644 1.8934 438.00 61.69 208.00 3.590 113.00 2.0501 240 76.37 11.51 11.42 11.04 11.26 11.31 27.7245 2.9667 752.00 94.78 336.00 3.680 118.00 1.3803 240 76.09 11.10 11.41 11.08 11.37 11.24 32.7680 3.0915 875.00 96.65 372.00 3.920 112.00 1.2301 240 75.82 11.01 10.99 11.22 11.00 11.06 25.8359 2.2903 665.00 67.89 278.00 4.370 111.00 1.9103 240 76.21 11.12 11.31 11.26 11.40 11.27 19.8142 2.0310 533.00 64.15 248.00 3.850 112.00 1.7301 240 75.97 11.38 11.58 11.37 11.21 11.39 32.3182 2.8907 884.00 93.77 396.00 4.170 110.00 1.1203 240 76.06 11.25 11.19 11.37 11.30 11.28 13.5836 1.4658 365.00 46.27 176.00 3.650 114.00 2.3101 240 75.57 11.38 11.50 11.41 11.22 11.38 25.6871 2.4235 698.00 78.05 358.00 4.840 102.00 1.5603 240 75.99 11.13 11.02 10.95 11.07 11.04 18.4935 2.0147 476.00 59.65 252.00 4.360 116.00 2.0801 240 76.21 11.37 11.61 11.32 11.48 11.45 12.2253 1.3898 339.00 45.95 181.00 4.330 113.00 2.8503 240 76.45 11.79 11.81 11.68 11.66 11.74 19.9355 1.9134 583.00 68.40 282.00 4.210 111.00 1.7101 240 76.00 11.52 11.77 11.81 11.40 11.63 12.2329 1.4039 349.00 48.50 193.00 4.190 112.00 2.5203 240 76.25 11.35 11.41 11.49 11.38 11.41 19.9909 1.8320 551.00 60.00 233.00 4.160 113.00 2.160Data Sheet for MOR/MOE - GN/GI perpendicularBoard type Board No. Sample IDLength of spanThicknessAverage thickness (mm)ResultsGNGN1GN2GN3GN4GN5GN6GIGI1GI5GI6GI2GI3GI4122   Board type: Sample ID:GN Guadua Node 1,3 perpendicular MOR=3*PeakLoad*LengthOfSpan/(2*width*thickness2)GI Guadua Internode 2,4 parallel MOE=LengthOfSpan3/Slope/(4*width*thickness3)Slope = (Pmax-Pmin)/(Ymax-Ymin)Width 1 2 3 4 MOR MOE PEAK LOADSlope Pmax ymax Pmin ymin(mm) (mm) (mm) (mm) (mm) (mm) MPa GPa N N/mm N mm N mm2 240 76.22 11.62 11.70 11.60 11.59 11.63 52.9265 8.5509 1515.00 296.46 453.00 1.670 118.00 0.5404 240 76.24 11.31 11.20 11.09 11.06 11.17 69.0540 10.6563 1823.00 327.18 449.00 1.430 112.00 0.4002 240 76.10 11.03 10.99 11.00 10.91 10.98 61.4196 9.7176 1566.00 283.45 918.00 3.260 507.00 1.8104 240 76.23 12.01 12.04 12.15 11.99 12.05 55.1183 8.2732 1694.00 319.09 464.00 1.540 113.00 0.4402 240 76.14 11.66 11.68 11.67 11.73 11.69 52.3235 8.8773 1511.00 312.04 452.00 1.560 115.00 0.4804 240 76.19 11.89 11.70 11.95 11.89 11.86 51.6526 8.5731 1537.00 315.09 449.00 1.570 115.00 0.5102 240 76.00 11.75 11.67 11.77 11.60 11.70 57.3621 8.8818 1657.00 312.62 441.00 1.550 119.00 0.5204 240 76.41 11.60 11.73 11.59 11.61 11.63 55.6743 9.8036 1599.00 341.18 448.00 1.580 158.00 0.7302 240 76.35 11.45 11.05 11.77 11.36 11.41 50.8359 9.2032 1403.00 301.82 453.00 1.900 121.00 0.8004 240 76.40 11.48 11.52 11.33 11.35 11.42 63.4095 10.0440 1755.00 330.69 444.00 1.710 110.00 0.7002 240 75.67 11.24 11.11 11.18 11.36 11.22 70.6007 10.9932 1869.00 340.21 448.00 1.370 118.00 0.4004 240 76.18 11.77 11.75 11.74 11.78 11.76 65.1966 9.8229 1908.00 352.15 1197.00 3.560 542.00 1.7002 240 75.24 11.05 11.05 10.78 11.00 10.97 64.5694 11.0249 1624.00 316.86 836.00 2.790 291.00 1.0704 240 76.42 11.91 11.57 11.75 11.58 11.70 81.7650 11.6824 2377.00 414.00 1304.00 3.360 476.00 1.3602 240 76.37 11.20 11.17 11.15 11.38 11.23 86.1591 12.0800 2303.00 377.55 1296.00 3.630 556.00 1.6704 240 75.74 11.48 11.57 11.53 11.28 11.47 79.9860 11.0120 2212.00 363.70 1164.00 3.490 633.00 2.0302 240 75.42 11.32 11.25 11.00 11.35 11.23 68.7720 11.1447 1817.00 344.44 902.00 2.900 406.00 1.4604 240 76.51 11.47 11.53 11.81 11.55 11.59 83.8574 12.9386 2394.00 445.95 446.00 1.020 116.00 0.2802 240 76.14 11.39 11.29 11.46 11.60 11.44 83.7806 12.8341 2317.00 422.78 1327.00 3.640 566.00 1.8404 240 76.00 11.25 10.99 11.12 11.24 11.15 58.2949 9.8028 1530.00 298.82 985.00 3.370 477.00 1.6702 240 76.27 12.10 12.24 12.24 11.99 12.14 80.8980 12.9082 2527.00 510.00 1281.00 2.580 618.00 1.2804 240 76.20 11.54 11.43 11.63 11.40 11.50 63.5518 10.7910 1779.00 361.86 456.00 1.740 105.00 0.7702 240 76.19 11.63 11.41 11.79 11.60 11.61 69.3320 8.8087 1977.00 303.70 440.00 1.350 112.00 0.2704 240 76.37 11.87 11.76 11.99 11.90 11.88 72.7453 12.7200 2178.00 471.29 1053.00 2.560 577.00 1.550GI2GI3Data Sheet for MOR/MOE - GN/GI parallelBoard type Board No. Sample IDLength of spanThicknessAverage thickness (mm)ResultsGNGN1GN2GN3GN4GN5GN6GIGI1GI4GI5GI6123  Thickness Swelling Test has 1 specimens per boards, total 36 specimens.   124   125   126   127   128       Appendix C  : Data Analysis in JMP Following pages show the data analysis results using JMP 10. They are in the order of density, thickness, IB, MOR – perpendicular, MOR – parallel, MOE – perpendicular, MOE – parallel, thickness swelling in 2hrs, water absorption in 2hrs, thickness swelling in 24hrs, water absorption in 24hrs.                   Oneway Analysis of density By type20040060080010001200GM MMtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.001139-0.0016585.36834767.3973360t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-5.7508.99911.947-23.4470.95t RatioDFProb > |t|Prob > tProb < t-0.638983580.52320.73840.2616-30 -20 -10 0 10 20 30Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares2975.52609015.52611991.0Mean Square2975.517287.75F Ratio0.4083Prob > F0.5232 130Oneway Analysis of density By typeOneway AnovaMeans for Oneway AnovaLevelGMMMNumber180180Mean770.272764.522Std Error6.36306.3630Lower 95%757.76752.01Upper 95%782.79777.04Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber180180Mean770.272764.522Std Dev101.82064.870Std Err Mean7.58934.8351Lower 95%755.30754.98Upper 95%785.25774.06Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixGMMM-17.697-11.947-11.947-17.697Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 131Oneway Analysis of thichness By type1111.51212.5GM MMtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.0451820.0425150.29285511.48636360t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence0.1270560.0308700.1877640.0663470.95t RatioDFProb > |t|Prob > tProb < t4.115881358<.0001*<.0001*1.0000-0.15 -0.10 -0.05 0.00 0.05 0.10 0.15Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares1.45288030.70345332.156333Mean Square1.452880.08576F Ratio16.9405Prob > F<.0001*Means for Oneway AnovaLevelGMMMNumber180180Mean11.422811.5499Std Error0.021830.02183Lower 95%11.38011.507Upper 95%11.46611.593Std Error uses a pooled estimate of error variance 132Oneway Analysis of thichness By typeMeans and Std DeviationsLevelGMMMNumber180180Mean11.422811.5499Std Dev0.3102410.274369Std Err Mean0.023120.02045Lower 95%11.37711.510Upper 95%11.46811.590Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixMMGM-0.060710.066350.06635-0.06071Abs(Dif)-LSDMM GMPositive values show pairs of means that are significantly different. 133Oneway Analysis of IB By type200400600800100012001400GM MMtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.1938380.191586175.5018683.4628360t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence171.63718.500208.018135.2550.95t RatioDFProb > |t|Prob > tProb < t9.277902358<.0001*<.0001*1.0000-200 -150 -100 -50 0 50 100 150 200Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares26513231102671513678038Mean Square265132330801F Ratio86.0795Prob > F<.0001*Means for Oneway AnovaLevelGMMMNumber180180Mean597.644769.281Std Error13.08113.081Lower 95%571.92743.56Upper 95%623.37795.01Std Error uses a pooled estimate of error variance 134Oneway Analysis of IB By typeMeans and Std DeviationsLevelGMMMNumber180180Mean597.644769.281Std Dev193.956154.863Std Err Mean14.45711.543Lower 95%569.12746.50Upper 95%626.17792.06Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixMMGM-36.38135.26135.26-36.38Abs(Dif)-LSDMM GMPositive values show pairs of means that are significantly different. 135Oneway Analysis of MOR_PD By Type1520253035GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.022827-0.021595.76643321.0901824t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-1.68762.35413.1945-6.56980.95t RatioDFProb > |t|Prob > tProb < t-0.71688220.48100.75950.2405-8 -6 -4 -2 0 2 4 6 8Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares17.08864731.53862748.62725Mean Square17.088633.2518F Ratio0.5139Prob > F0.4810Means for Oneway AnovaLevelGMMMNumber1212Mean21.934020.2464Std Error1.66461.6646Lower 95%18.48216.794Upper 95%25.38623.699Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber1212Mean21.934020.2464Std Dev4.450936.83321Std Err Mean1.28491.9726Lower 95%19.10615.905Upper 95%24.76224.588 136Oneway Analysis of MOR_PD By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGMMM-4.8822-3.1945-3.1945-4.8822Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 137Oneway Analysis of MOR_PAR By Type30405060708090GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.000782-0.0446412.0248164.6030824t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence0.6444.90910.825-9.5370.95t RatioDFProb > |t|Prob > tProb < t0.131253220.89680.44840.5516-15 -10 -5 0 5 10 15Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares2.49103181.11383183.6048Mean Square2.491144.596F Ratio0.0172Prob > F0.8968Means for Oneway AnovaLevelGMMMNumber1212Mean64.280964.9252Std Error3.47133.4713Lower 95%57.08257.726Upper 95%71.48072.124Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber1212Mean64.280964.9252Std Dev13.693510.0837Std Err Mean3.95302.9109Lower 95%55.58058.518Upper 95%72.98171.332Means ComparisonsComparisons for each pair using Student’s t 138Oneway Analysis of MOR_PAR By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixMMGM-10.181-9.537-9.537-10.181Abs(Dif)-LSDMM GMPositive values show pairs of means that are significantly different. 139Oneway Analysis of MOE_PD By Type11.522.53GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.027672-0.016520.457792.06708324t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence0.147880.186890.53547-0.239710.95t RatioDFProb > |t|Prob > tProb < t0.791276220.43720.21860.7814-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares0.13121694.61058574.7418025Mean Square0.1312170.209572F Ratio0.6261Prob > F0.4372Means for Oneway AnovaLevelGMMMNumber1212Mean1.993142.14103Std Error0.132150.13215Lower 95%1.71911.8670Upper 95%2.26722.4151Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber1212Mean1.993142.14103Std Dev0.4742140.440755Std Err Mean0.136890.12723Lower 95%1.69181.8610Upper 95%2.29442.4211 140Oneway Analysis of MOE_PD By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixMMGM-0.38759-0.23971-0.23971-0.38759Abs(Dif)-LSDMM GMPositive values show pairs of means that are significantly different. 141Oneway Analysis of MOE_PAR By Type67891011121314GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.5685350.5489231.1524739.27237524t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-2.53320.4705-1.5575-3.50900.95t RatioDFProb > |t|Prob > tProb < t-5.3841522<.0001*1.0000<.0001*-3 -2 -1 0 1 2 3Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares38.50312029.22025967.723379Mean Square38.50311.3282F Ratio28.9891Prob > F<.0001*Means for Oneway AnovaLevelGMMMNumber1212Mean10.53908.0058Std Error0.332690.33269Lower 95%9.84907.3158Upper 95%11.2298.696Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber1212Mean10.53908.0058Std Dev1.440670.76214Std Err Mean0.415890.22001Lower 95%9.62367.5215Upper 95%11.4548.490 142Oneway Analysis of MOE_PAR By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGMMM-0.97571.55751.5575-0.9757Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 143Oneway Analysis of 2h TS By Type123456GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.3730090.310311.1249493.65271712t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-1.58420.6495-0.1370-3.03130.95t RatioDFProb > |t|Prob > tProb < t-2.43909100.0349*0.98250.0175*-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares7.52875212.65510220.183854Mean Square7.528751.26551F Ratio5.9492Prob > F0.0349*Means for Oneway AnovaLevelGMMMNumber66Mean4.444802.86063Std Error0.459260.45926Lower 95%3.42151.8373Upper 95%5.46813.8839Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber66Mean4.444802.86063Std Dev1.213731.02854Std Err Mean0.495500.41990Lower 95%3.17111.7812Upper 95%5.71853.9400 144Oneway Analysis of 2h TS By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGMMM-1.44720.13700.1370-1.4472Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 145Oneway Analysis of 2h WA By Type1012141618202224GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.3710990.3082083.49146616.3451212t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-4.89672.0158-0.4052-9.38810.95t RatioDFProb > |t|Prob > tProb < t-2.42914100.0355*0.98220.0178*-8 -6 -4 -2 0 2 4 6 8Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares71.93203121.90334193.83537Mean Square71.932012.1903F Ratio5.9007Prob > F0.0355*Means for Oneway AnovaLevelGMMMNumber66Mean18.793513.8968Std Error1.42541.4254Lower 95%15.61710.721Upper 95%21.96917.073Std Error uses a pooled estimate of error variance 146Oneway Analysis of 2h WA By TypeMeans and Std DeviationsLevelGMMMNumber66Mean18.793513.8968Std Dev3.446843.53553Std Err Mean1.40721.4434Lower 95%15.17610.186Upper 95%22.41117.607Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGMMM-4.49150.40520.4052-4.4915Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 147Oneway Analysis of 24h TS By Type89101112131415GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.2638770.1902641.94845910.6380812t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-2.12991.12490.3766-4.63640.95t RatioDFProb > |t|Prob > tProb < t-1.89333100.08760.95620.0438*-4 -3 -2 -1 0 1 2 3 4 148Oneway Analysis of 24h TS By TypeOneway AnovaAnalysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares13.60920937.96490751.574117Mean Square13.60923.7965F Ratio3.5847Prob > F0.0876Means for Oneway AnovaLevelGMMMNumber66Mean11.70309.5731Std Error0.795450.79545Lower 95%9.93067.8007Upper 95%13.47511.346Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber66Mean11.70309.5731Std Dev2.729260.37965Std Err Mean1.11420.1550Lower 95%8.83889.1747Upper 95%14.5679.972Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGMMM-2.5065-0.3766-0.3766-2.5065Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 149Oneway Analysis of 24h WA By Type2530354045GM MMTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.3552940.2908234.71165435.2882312t TestMM-GMAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-6.3862.720-0.325-12.4470.95t RatioDFProb > |t|Prob > tProb < t-2.34754100.0408*0.97960.0204*-10 -5 0 5 10Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares122.34107221.99683344.33790Mean Square122.34122.200F Ratio5.5109Prob > F0.0408*Means for Oneway AnovaLevelGMMMNumber66Mean38.481232.0953Std Error1.92351.9235Lower 95%34.19527.809Upper 95%42.76736.381Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGMMMNumber66Mean38.481232.0953Std Dev4.385885.01632Std Err Mean1.79052.0479Lower 95%33.87926.831Upper 95%43.08437.360Means Comparisons 150Oneway Analysis of 24h WA By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGMMM-6.06110.32480.3248-6.0611Abs(Dif)-LSDGM MMPositive values show pairs of means that are significantly different. 151 152 153Oneway Analysis of thichness By type10.51111.51212.5MHigh MMed MLowtypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.1360830.1328650.28231911.4282540Analysis of VarianceSourcetypeErrorC. TotalDF2537539Sum of Squares6.74198042.80117749.543158Mean Square3.370990.07970F Ratio42.2937Prob > F<.0001*Means for Oneway AnovaLevelMHighMMedMLowNumber180180180Mean11.549911.454711.2801Std Error0.021040.021040.02104Lower 95%11.50911.41311.239Upper 95%11.59111.49611.321Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber180180180Mean11.549911.454711.2801Std Dev0.2743690.2621770.308379Std Err Mean0.020450.019540.02299Lower 95%11.51011.41611.235Upper 95%11.59011.49311.325Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.35023Alpha0.05 154Oneway Analysis of thichness By typeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMHighMMedMLow-0.069940.025280.199890.02528-0.069940.104670.199890.10467-0.06994Abs(Dif)-HSDMHigh MMed MLowPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMHighMMedMLowABCMean11.54988911.45466711.280056Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMHighMMedMHigh- LevelMLowMLowMMedDifference0.26983330.17461110.0952222Std Err Dif0.02975910.02975910.0297591Lower CL0.19989270.10467050.0252816Upper CL0.33977400.24455170.1651628p-Value<.0001*<.0001*0.0042* 155 156 157Oneway Analysis of MOR_PD By Type1520253035MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.1187820.0653755.45276521.670736Analysis of VarianceSourceTypeErrorC. TotalDF23335Sum of Squares132.2556981.17731113.4330Mean Square66.127829.7326F Ratio2.2241Prob > F0.1241Means for Oneway AnovaLevelMHighMMedMLowNumber121212Mean20.246424.380120.3856Std Error1.57411.57411.5741Lower 95%17.04421.17817.183Upper 95%23.44927.58323.588Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber121212Mean20.246424.380120.3856Std Dev6.833216.130492.21863Std Err Mean1.97261.76970.6405Lower 95%15.90520.48518.976Upper 95%24.58828.27521.795Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.45379Alpha0.05 158Oneway Analysis of MOR_PD By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMMedMLowMHigh-5.4623-1.4678-1.3286-1.4678-5.4623-5.3231-1.3286-5.3231-5.4623Abs(Dif)-HSDMMed MLow MHighPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMMedMLowMHighAAAMean24.38014220.38559220.246367Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMMedMMedMLow- LevelMHighMLowMHighDifference4.1337753.9945500.139225Std Err Dif2.2260822.2260822.226082Lower CL-1.32857-1.46780-5.32312Upper CL9.5961239.4568985.601573p-Value0.16730.18710.9978 159Oneway Analysis of MOR_PAR By Type2030405060708090MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.4503890.41707910.0506956.0827136Analysis of VarianceSourceTypeErrorC. TotalDF23335Sum of Squares2731.72783333.53766065.2654Mean Square1365.86101.02F Ratio13.5212Prob > F<.0001*Means for Oneway AnovaLevelMHighMMedMLowNumber121212Mean64.925259.089744.2332Std Error2.90142.90142.9014Lower 95%59.02253.18738.330Upper 95%70.82864.99350.136Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber121212Mean64.925259.089744.2332Std Dev10.083711.28018.6097Std Err Mean2.91093.25632.4854Lower 95%58.51851.92338.763Upper 95%71.33266.25749.704Means ComparisonsComparisons for all pairs using Tukey-Kramer HSD 160Oneway Analysis of MOR_PAR By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.45379Alpha0.05HSD Threshold MatrixMHighMMedMLow-10.068-4.23310.624-4.233-10.0684.78810.6244.788-10.068Abs(Dif)-HSDMHigh MMed MLowPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMHighMMedMLowAABMean64.92524259.08972544.233175Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMHighMMedMHigh- LevelMLowMLowMMedDifference20.6920714.856555.83552Std Err Dif4.1031754.1031754.103175Lower CL10.62374.7882-4.2328Upper CL30.7604224.9249015.90387p-Value<.0001*0.0027*0.3414 161Oneway Analysis of MOE_PD By Type1.522.533.5MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.2452820.1995410.4360532.42399236Analysis of VarianceSourceTypeErrorC. TotalDF23335Sum of Squares2.03926146.27469498.3139562Mean Square1.019630.19014F Ratio5.3625Prob > F0.0096*Means for Oneway AnovaLevelMHighMMedMLowNumber121212Mean2.141032.723332.40762Std Error0.125880.125880.12588Lower 95%1.88492.46722.1515Upper 95%2.39712.97942.6637Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber121212Mean2.141032.723332.40762Std Dev0.4407550.4331670.434199Std Err Mean0.127230.125040.12534Lower 95%1.86102.44812.1317Upper 95%2.42112.99852.6835Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.45379Alpha0.05 162Oneway Analysis of MOE_PD By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMMedMLowMHigh-0.43682-0.121120.14548-0.12112-0.43682-0.170220.14548-0.17022-0.43682Abs(Dif)-HSDMMed MLow MHighPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMMedMLowMHighAA BBMean2.72332502.40762502.1410250Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMMedMMedMLow- LevelMHighMLowMHighDifference0.58230000.31570000.2666000Std Err Dif0.17801790.17801790.1780179Lower CL0.145481-0.121119-0.170219Upper CL1.0191190.7525190.703419p-Value0.0069*0.19410.3051 163Oneway Analysis of MOE_PAR By Type56789MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.5689050.5427780.7479517.16250836Analysis of VarianceSourceTypeErrorC. TotalDF23335Sum of Squares24.36281318.46123442.824047Mean Square12.18140.5594F Ratio21.7746Prob > F<.0001*Means for Oneway AnovaLevelMHighMMedMLowNumber121212Mean8.005777.435006.04676Std Error0.215910.215910.21591Lower 95%7.56656.99575.6075Upper 95%8.44507.87436.4860Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber121212Mean8.005777.435006.04676Std Dev0.7621410.7523650.728959Std Err Mean0.220010.217190.21043Lower 95%7.52156.95705.5836Upper 95%8.49007.91306.5099Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.45379Alpha0.05 164Oneway Analysis of MOE_PAR By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMHighMMedMLow-0.7493-0.17851.2097-0.1785-0.74930.63901.20970.6390-0.7493Abs(Dif)-HSDMHigh MMed MLowPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMHighMMedMLowAABMean8.00576677.43500006.0467583Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMHighMMedMHigh- LevelMLowMLowMMedDifference1.9590081.3882420.570767Std Err Dif0.30534990.30534990.3053499Lower CL1.209740.63898-0.17850Upper CL2.7082742.1375081.320033p-Value<.0001*0.0002*0.1636 165Oneway Analysis of 2h TS By Type1234567MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.5697290.512361.1104313.42930618Analysis of VarianceSourceTypeErrorC. TotalDF21517Sum of Squares24.49066918.49584642.986516Mean Square12.24531.2331F Ratio9.9309Prob > F0.0018*Means for Oneway AnovaLevelMHighMMedMLowNumber666Mean2.860632.372625.05467Std Error0.453330.453330.45333Lower 95%1.89441.40644.0884Upper 95%3.82693.33896.0209Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber666Mean2.860632.372625.05467Std Dev1.028540.535321.53451Std Err Mean0.419900.218540.62646Lower 95%1.78121.81083.4443Upper 95%3.94002.93446.6650Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.59747Alpha0.05 166Oneway Analysis of 2h TS By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMLowMHighMMed-1.66530.52881.01680.5288-1.6653-1.17721.0168-1.1772-1.6653Abs(Dif)-HSDMLow MHigh MMedPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMLowMHighMMedABBMean5.05466672.86063332.3726167Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMLowMLowMHigh- LevelMMedMHighMMedDifference2.6820502.1940330.488017Std Err Dif0.64110750.64110750.6411075Lower CL1.016790.52878-1.17724Upper CL4.3473083.8592912.153274p-Value0.0022*0.0099*0.7317 167Oneway Analysis of 2h WA By Type101520253035MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.5957980.5419045.50352519.2868718Analysis of VarianceSourceTypeErrorC. TotalDF21517Sum of Squares669.6895454.33181124.0213Mean Square334.84530.289F Ratio11.0551Prob > F0.0011*Means for Oneway AnovaLevelMHighMMedMLowNumber666Mean13.896816.149527.8144Std Error2.24682.24682.2468Lower 95%9.10811.36023.025Upper 95%18.68620.93832.603Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber666Mean13.896816.149527.8144Std Dev3.535537.238615.09597Std Err Mean1.44342.95522.0804Lower 95%10.1868.55322.466Upper 95%17.60723.74633.162Means Comparisons 168Oneway Analysis of 2h WA By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.59747Alpha0.05HSD Threshold MatrixMLowMMedMHigh-8.25343.41165.66423.4116-8.2534-6.00075.6642-6.0007-8.2534Abs(Dif)-HSDMLow MMed MHighPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMLowMMedMHighABBMean27.81438316.14945013.896783Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMLowMLowMMed- LevelMHighMMedMHighDifference13.9176011.664932.25267Std Err Dif3.1774623.1774623.177462Lower CL5.664243.41157-6.00070Upper CL22.1709619.9183010.50603p-Value0.0015*0.0060*0.7620 169Oneway Analysis of 24h TS By Type789101112131415MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.6122350.5605331.38644810.8298318Analysis of VarianceSourceTypeErrorC. TotalDF21517Sum of Squares45.52482728.83356774.358394Mean Square22.76241.9222F Ratio11.8416Prob > F0.0008*Means for Oneway AnovaLevelMHighMMedMLowNumber666Mean9.57319.842913.0735Std Error0.566010.566010.56601Lower 95%8.3678.63611.867Upper 95%10.78011.04914.280Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber666Mean9.57319.842913.0735Std Dev0.379651.502031.83480Std Err Mean0.154990.613200.74905Lower 95%9.1758.26711.148Upper 95%9.97211.41914.999Means ComparisonsComparisons for all pairs using Tukey-Kramer HSD 170Oneway Analysis of 24h TS By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.59747Alpha0.05HSD Threshold MatrixMLowMMedMHigh-2.07921.15151.42121.1515-2.0792-1.80951.4212-1.8095-2.0792Abs(Dif)-HSDMLow MMed MHighPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMLowMMedMHighABBMean13.0735009.8428509.573133Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMLowMLowMMed- LevelMHighMMedMHighDifference3.5003673.2306500.269717Std Err Dif0.80046610.80046610.8004661Lower CL1.421181.15146-1.80947Upper CL5.5795545.3098372.348904p-Value0.0015*0.0029*0.9396 171Oneway Analysis of 24h WA By Type2530354045505560MHigh MMed MLowTypeAll PairsTukey-Kramer0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.720750.6835166.00655839.5532618Analysis of VarianceSourceTypeErrorC. TotalDF21517Sum of Squares1396.7979541.18121937.9791Mean Square698.39936.079F Ratio19.3576Prob > F<.0001*Means for Oneway AnovaLevelMHighMMedMLowNumber666Mean32.095334.640351.9242Std Error2.45222.45222.4522Lower 95%26.86929.41446.698Upper 95%37.32239.86757.151Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelMHighMMedMLowNumber666Mean32.095334.640351.9242Std Dev5.016326.874455.98454Std Err Mean2.04792.80652.4432Lower 95%26.83127.42645.644Upper 95%37.36041.85558.205Means ComparisonsComparisons for all pairs using Tukey-Kramer HSDConfidence Quantileq*2.59747Alpha0.05 172Oneway Analysis of 24h WA By TypeMeans ComparisonsComparisons for all pairs using Tukey-Kramer HSDHSD Threshold MatrixMLowMMedMHigh-9.0088.27610.8218.276-9.008-6.46310.821-6.463-9.008Abs(Dif)-HSDMLow MMed MHighPositive values show pairs of means that are significantly different.Connecting Letters ReportLevelMLowMMedMHighABBMean51.92421734.64030032.095250Levels not connected by same letter are significantly different.Ordered Differences ReportLevelMLowMLowMMed- LevelMHighMMedMHighDifference19.8289717.283922.54505Std Err Dif3.4678883.4678883.467888Lower CL10.82128.2762-6.4627Upper CL28.8367026.2916511.55279p-Value0.0001*0.0005*0.7476 173Oneway Analysis of density By type5006007008009001000110012001300GI GNtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)7.87e-5-0.0027183.34964762.5301360t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-1.4758.78615.804-18.7530.95t RatioDFProb > |t|Prob > tProb < t-0.167863580.86680.56660.4334-30 -20 -10 0 10 20 30Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares195.72487083.92487279.6Mean Square195.756947.16F Ratio0.0282Prob > F0.8668Means for Oneway AnovaLevelGIGNNumber180180Mean763.267761.793Std Error6.21256.2125Lower 95%751.05749.58Upper 95%775.49774.01Std Error uses a pooled estimate of error variance 174Oneway Analysis of density By typeMeans and Std DeviationsLevelGIGNNumber180180Mean763.267761.793Std Dev88.801477.5154Std Err Mean6.61895.7777Lower 95%750.21750.39Upper 95%776.33773.19Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixGIGN-17.278-15.804-15.804-17.278Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 175Oneway Analysis of thichness By type1010.51111.512GI GNtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.0038630.0010810.32608111.37236360t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence0.040500.034370.10810-0.027100.95t RatioDFProb > |t|Prob > tProb < t1.1782873580.23950.11970.8803-0.10 -0.05 0.00 0.05 0.10Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares0.14762338.06567138.213293Mean Square0.1476230.106329F Ratio1.3884Prob > F0.2395Means for Oneway AnovaLevelGIGNNumber180180Mean11.352111.3926Std Error0.024300.02430Lower 95%11.30411.345Upper 95%11.40011.440Std Error uses a pooled estimate of error variance 176Oneway Analysis of thichness By typeMeans and Std DeviationsLevelGIGNNumber180180Mean11.352111.3926Std Dev0.3057710.345197Std Err Mean0.022790.02573Lower 95%11.30711.342Upper 95%11.39711.443Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixGNGI-0.06760-0.02710-0.02710-0.06760Abs(Dif)-LSDGN GIPositive values show pairs of means that are significantly different. 177Oneway Analysis of IB By type20030040050060070080090010001100GI GNtypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.0486550.045997159.8519663.9342360t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-72.1016.85-38.96-105.240.95t RatioDFProb > |t|Prob > tProb < t-4.27893358<.0001*1.0000<.0001*-100 -50 0 50 100Analysis of VarianceSourcetypeErrorC. TotalDF1358359Sum of Squares467849.79147839.89615689.5Mean Square46785025553F Ratio18.3093Prob > F<.0001*Means for Oneway AnovaLevelGIGNNumber180180Mean699.984627.884Std Error11.91511.915Lower 95%676.55604.45Upper 95%723.42651.32Std Error uses a pooled estimate of error variance 178Oneway Analysis of IB By typeMeans and Std DeviationsLevelGIGNNumber180180Mean699.984627.884Std Dev154.682164.860Std Err Mean11.52912.288Lower 95%677.23603.64Upper 95%722.73652.13Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet1.96661Alpha0.05LSD Threshold MatrixGIGN-33.13738.96238.962-33.137Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 179Oneway Analysis of MOR_PD By Type101520253035GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.0445030.0010726.52823620.3685524t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-2.69782.66512.8293-8.22500.95t RatioDFProb > |t|Prob > tProb < t-1.01226220.32240.83880.1612-10 -5 0 5 10Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares43.66956937.59317981.26273Mean Square43.669642.6179F Ratio1.0247Prob > F0.3224Means for Oneway AnovaLevelGIGNNumber1212Mean21.717519.0196Std Error1.88451.8845Lower 95%17.80915.111Upper 95%25.62622.928Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber1212Mean21.717519.0196Std Dev7.196495.78327Std Err Mean2.07741.6695Lower 95%17.14515.345Upper 95%26.29022.694Means Comparisons 180Oneway Analysis of MOR_PD By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGIGN-5.5272-2.8293-2.8293-5.5272Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 181Oneway Analysis of MOR_PAR By Type505560657075808590GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.4946240.4716538.27616166.6368824t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-15.6783.379-8.671-22.6850.95t RatioDFProb > |t|Prob > tProb < t-4.64025220.0001*0.9999<.0001*-20 -15 -10 -5 0 5 10 15 20Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares1474.82791506.88632981.7142Mean Square1474.8368.49F Ratio21.5320Prob > F0.0001*Means for Oneway AnovaLevelGIGNNumber1212Mean74.476058.7978Std Error2.38912.3891Lower 95%69.52153.843Upper 95%79.43163.753Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber1212Mean74.476058.7978Std Dev9.424886.93984Std Err Mean2.72072.0034Lower 95%68.48854.388Upper 95%80.46463.207Means Comparisons 182Oneway Analysis of MOR_PAR By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGIGN-7.00718.67118.6711-7.0071Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 183Oneway Analysis of MOE_PD By Type11.522.53GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.020102-0.024440.5598312.06600424t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-0.153540.228550.32044-0.627530.95t RatioDFProb > |t|Prob > tProb < t-0.67181220.50870.74560.2544-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares0.14145036.89504417.0364944Mean Square0.1414500.313411F Ratio0.4513Prob > F0.5087Means for Oneway AnovaLevelGIGNNumber1212Mean2.142781.98923Std Error0.161610.16161Lower 95%1.80761.6541Upper 95%2.47792.3244Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber1212Mean2.142781.98923Std Dev0.6021320.514062Std Err Mean0.173820.14840Lower 95%1.76021.6626Upper 95%2.52542.3159 184Oneway Analysis of MOE_PD By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGIGN-0.47398-0.32044-0.32044-0.47398Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 185Oneway Analysis of MOE_PAR By Type8910111213GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.4768860.4531081.10988610.4643524t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence-2.02920.4531-1.0895-2.96890.95t RatioDFProb > |t|Prob > tProb < t-4.47837220.0002*0.9999<.0001*-2 -1 0 1 2Analysis of VarianceSourceTypeErrorC. TotalDF12223Sum of Squares24.70571327.10063751.806350Mean Square24.70571.2318F Ratio20.0558Prob > F0.0002*Means for Oneway AnovaLevelGIGNNumber1212Mean11.47909.4498Std Error0.320400.32040Lower 95%10.8148.785Upper 95%12.14310.114Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber1212Mean11.47909.4498Std Dev1.310850.86335Std Err Mean0.378410.24923Lower 95%10.6468.901Upper 95%12.3129.998Means Comparisons 186Oneway Analysis of MOE_PAR By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.07387Alpha0.05LSD Threshold MatrixGIGN-0.93971.08951.0895-0.9397Abs(Dif)-LSDGI GNPositive values show pairs of means that are significantly different. 187Oneway Analysis of 2h TS By Type123456GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.3559150.2915061.1856363.15543312t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence1.609130.684533.134350.083910.95t RatioDFProb > |t|Prob > tProb < t2.350723100.0406*0.0203*0.9797-2 -1 0 1 2Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares7.76793014.05731821.825248Mean Square7.767931.40573F Ratio5.5259Prob > F0.0406*Means for Oneway AnovaLevelGIGNNumber66Mean2.350873.96000Std Error0.484030.48403Lower 95%1.27242.8815Upper 95%3.42945.0385Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber66Mean2.350873.96000Std Dev0.769861.48956Std Err Mean0.314290.60811Lower 95%1.54292.3968Upper 95%3.15885.5232 188Oneway Analysis of 2h TS By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGNGI-1.52520.08390.0839-1.5252Abs(Dif)-LSDGN GIPositive values show pairs of means that are significantly different. 189Oneway Analysis of 2h WA By Type1015202530GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.1053430.0158774.61575215.8752812t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence2.89172.66498.8295-3.04610.95t RatioDFProb > |t|Prob > tProb < t1.08511100.30330.15170.8483-10 -5 0 5 10Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares25.08608213.05166238.13774Mean Square25.086121.3052F Ratio1.1775Prob > F0.3033Means for Oneway AnovaLevelGIGNNumber66Mean14.429417.3211Std Error1.88441.8844Lower 95%10.23113.122Upper 95%18.62821.520Std Error uses a pooled estimate of error variance 190Oneway Analysis of 2h WA By TypeMeans and Std DeviationsLevelGIGNNumber66Mean14.429417.3211Std Dev3.310805.62574Std Err Mean1.35162.2967Lower 95%10.95511.417Upper 95%17.90423.225Means ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGNGI-5.9378-3.0461-3.0461-5.9378Abs(Dif)-LSDGN GIPositive values show pairs of means that are significantly different. 191Oneway Analysis of 24h TS By Type789101112131415GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.4981720.4479891.76215610.8439312t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence3.205501.017385.472370.938630.95t RatioDFProb > |t|Prob > tProb < t3.150737100.0103*0.0052*0.9948-4 -3 -2 -1 0 1 2 3 4Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares30.82569131.05192461.877614Mean Square30.82573.1052F Ratio9.9271Prob > F0.0103*Means for Oneway AnovaLevelGIGNNumber66Mean9.241212.4467Std Error0.719400.71940Lower 95%7.63810.844Upper 95%10.84414.050Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber66Mean9.241212.4467Std Dev1.683591.83736Std Err Mean0.687320.75010Lower 95%7.47410.518Upper 95%11.00814.375 192Oneway Analysis of 24h TS By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGNGI-2.26690.93860.9386-2.2669Abs(Dif)-LSDGN GIPositive values show pairs of means that are significantly different. 193Oneway Analysis of 24h WA By Type2530354045GI GNTypeEach PairStudent’s t0.05Oneway AnovaSummary of FitRsquareAdj RsquareRoot Mean Square ErrorMean of ResponseObservations (or Sum Wgts)0.2482840.1731125.33003236.3073712t TestGN-GIAssuming equal variancesDifferenceStd Err DifUpper CL DifLower CL DifConfidence5.5933.07712.449-1.2640.95t RatioDFProb > |t|Prob > tProb < t1.817386100.09920.0496*0.9504-10 -5 0 5 10Analysis of VarianceSourceTypeErrorC. TotalDF11011Sum of Squares93.83264284.09243377.92507Mean Square93.832628.4092F Ratio3.3029Prob > F0.0992Means for Oneway AnovaLevelGIGNNumber66Mean33.511139.1037Std Error2.17602.1760Lower 95%28.66334.255Upper 95%38.35943.952Std Error uses a pooled estimate of error varianceMeans and Std DeviationsLevelGIGNNumber66Mean33.511139.1037Std Dev4.388976.12825Std Err Mean1.79182.5018Lower 95%28.90532.672Upper 95%38.11745.535 194Oneway Analysis of 24h WA By TypeMeans ComparisonsComparisons for each pair using Student’s tConfidence Quantilet2.22814Alpha0.05LSD Threshold MatrixGNGI-6.8566-1.2640-1.2640-6.8566Abs(Dif)-LSDGN GIPositive values show pairs of means that are significantly different. 195196  Appendix D  : Permission for reproduction for Figure 2, 3, 7   197    

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0307465/manifest

Comment

Related Items