ABSTRACT
This study, investigated the creep behaviour of Black Afara (Terminalia ivorensis) timber in accordance with EN 408 (2004) and EN 13153 (2002). The data were analysed based on the Eurocode 5 (2004), Eurocode 0 (2002), JCSS (2006) and EN 384 (2004) specifications using Easyfit statistical package. The statistics of the reference material properties (density, bending strength and modulus of elasticity) were calculated. The mean value of the density was found to be 449.24kg/m3. The corresponding coefficient of variation is 15%.The mean values of the modulus of elasticity is 16171N/mm2, and the corresponding coefficient of variation is 21%. For the bending strength, the mean value was given by 81.05N/mm2. The corresponding coefficients of variation is 16%. The mean values were adjusted to 12% and 18% moisture content, to agree with the European and Nigerian reference moisture contents. Five theoretical distribution models (normal, lognormal and gumbel, weibull and frechet) were fitted to the reference material properties using kolmogorov smirnov statistical distribution fitting technique. The best fit theoretical distribution models were found to be normal distribution for density, Lognormal distribution for modulus of elasticity and Weibull for bending strength. The Terminalia ivorensis timber specie was graded in accordance with the requirements of EN 338 (2009). This yielded the appropriate strength of D18. One-year creep test was measured from the laboratory experiments on Terminalia ivorensis timber samples. The creep data were fitted to linear, exponential and logarithmic regression models. The coefficients of determination for each of the model was determined. The logarithmic model has the highest correlation with coefficient of determination of 0.941. This implied that, logarithmic regression model is the most appropriate to model the creep behaviour of the timber. The selection of the logarithm models was buttressed by the values obtain from regression modelling error analysis. Microstructure analysis was conducted on the Terminalia ivorensis timber specie using scanning electron microscope on the specimen. The analysis was conducted before and after creep deformation. The microstructure analysis revealed that, the cellular structure contained empty vessels which was found to have closed after creep deformation. This implied that, the application of the long term loading lead to the closure of the empty vessels within the cells microstructure, leading to the densification of the timber.
TABLE OF CONTENTS
Cover page
Title page ii
Declaration iii
Certification iv
Dedication v
Acknowledgement vi
Abstract viii
Table of Contents ix
List of Figures xiii
List of Tables xvi
List of Plates xviii
List of appendices xix
Notations xx
CHAPTER ONE: INTRODUCTION
1.1 Background 1
1.2 Problem Statement 4
1.3 Justification of Research 5
1.4 Aim and Objectives 6
1.4.1 Aim of Study 6
1.4.2 Objectives of Study 6
1.5 Scope of the Research 7
1.6 Limitation of the Research 7
CHAPTER TWO: LITERATURE REVIEW
x
2.1 Preamble 9
2.2 Structure of Wood 12
2.3 Types of Timber 15
2.3.1 Softwoods 15
2.3.2 Hardwoods 16
2.4 Timber as a Structural Material 17
2.5 Properties of Timber 18
2.5.1 Physical Strength Properties 19
2.5.1.1 Moisture content (MC) 19
2.5.1.2 Fiber Saturation Point 19
2.5.1.3 Equilibrium Moisture Content 20
2.5.1.4 Dimensional instability 20
2.6 Some Common Nigerian Timber Species 20
2.7 The Material Properties of Timber 21
2.8 Codes of Practice for Design of Timber Structures 24
2.9 Background on the Tested Timber Specie 25
2.10 Strength Classes of Timber 26
2.10.1 The Timber Strength Classification System 27
2.10.2 Timber Grading System in the Nigerian Code of Practice (NCP 2) 30
2.11 Effect of Load Duration on Timber Structures 31
2.11.1 The Eurocode 5 Strength Modification Factor for Load Duration Kmod 31
2.12 Creep 32
2.12.1 Creep of Wood 34
xi
2.12.1.1 Effect of Environment on Wood 36
2.12.1.2 Effects of temperature 37
2.12.1.3 Effects of moisture 37
2.12.1.4 Mechano-sorptive behavior 39
2.12.1.5 Mathematical Models for Creep in Wood39
2.12.2 Creep Rupture Modelling 40
2.12.3 Eurocode 5 Specification on Timber Deformation 44
12.13 Distributions Fitting and Tests of Goodness of Fit of the Measured 45
Parameters
2.13.1 Probability Density Function 47
2.13.2 Distribution Models 47
2.13.2.1 Normal Distribution 47
2.13.2.2 Lognormal Distribution 48
2.13.2.3 Gumbel Distribution 49
2.13.2.4 Frechet Distribution 50
2.13.2.5 Weibull Distribution 51
2.13.3 Allocation of Strength Classes 52
2.13.4 Moisture Adjustment Factors 53
2.14 Microstructure and Chemistry of Wood 55
2.14.1 Microstructural Analysis 56
2.14.2 Chemical Components of the Wood 60
2.14.2.1 Carbohydrates 60
2.14.2.2 Lignin 61
xii
CHAPTER THREE: MATERIALS AND METHODS
3.1 Preamble 63
3.2 Materials 64
3.3 Methods of Test 64
3.3.1 Determination of Density and Moisture Content 65
3.3.2 The Bending Test 69
3.3.3 Creep Deformation Tests 76
3.3.4 Microstructural Analysis of Terminalia Ivorensis 81
CHAPTER FOUR: ANALYSIS OF RESULTS AND DISCUSSIONS
4.1 Preamble 81
4.2 Model Distribution Test and Goodness of Fit 81
4.2.1 Moisture Content of Terminalia Ivorensis 82
4.2.2 Density of Terminalia Ivorensis 89
4.2.3 Bending Strength of Terminalia Ivorensis 95
4.2.4 Modulus of Elasticity of Terminalia Ivorensis 100
4.3 Stochastic Models of Material Properties 105
4.4 Moisture Adjusted Material Properties 107
4.4.1 Adjustment Factors for Bending Strength 108
4.5 Other Material Properties 112
4.6 The Measured Stochastic Models 119
4.7 Strength Allocation Class for Terminalia ivorensis 119
4.8 Developed Creep Deformation Models 120
xiii
4.9 Microstructure of Terminalia ivorensis Timber 125
CHAPTER FIVE: CONCLUSIONS AND RECOMMENDATIONS
5.1 Conclusions 129
5.2 Recommendations 130
REFERENCE 132
APPENDICES
CHAPTER ONE
INTRODUCTION
1.1 Background
Wood is one of the oldest and most used materials known. From a simple log lying over a creek acting as a footbridge to large multi-storey timber buildings. It is usable for a wide range of structures (Trygve, 2015).
The term timber is frequently used to referto wood that is suitable for building or structural usage. It is one of the most frequently used building materials in both ancient and modern engineering constructions. Recently, the use of timber structures has increased in the construction industry due to attributed advantages such as environmentally friendly nature of timber, fully renewable potential and low handling costs (Afolayan, 2005; Hassani et al., 2014). Timber is widely available natural resources throughout the world, which with properly managed wood plantation, there is potential for continuous and sustainable supply of raw timber materials.
In addition, it exhibits unique material properties; it is a light weight material and, compared to its weight, the strength is high; the strength/weight ratio is even higher than for steel that is why it is used widely as a structural material for roofing system and pedestrian or bicycle bridges. Also it has low modulus of elasticity compared to concrete and steel (Shuaibu, 2010), this implies low stiffness capacity and consequent poor resistance to deflection in service.
The above mentioned attributes among others, gave rise to the acceptance of timber worldwide as building material that can compete stiffly with the conventional/popular
2
concrete and steel in the arena of the building industry with considerable advantages of low embodied energy, low carbon impact, and sustainability.
Unfortunately, itsstrong hygroscopic character basically affects all related mechanical properties leading to degradation of material stiffness and strength over the service life resulting in the loss of capacity and consequently structural integrity even after being in use for decades (Hassani et al., 2014).
In addition to the hygroscopic character, time-dependent phenomena like long-term visco-elastic creep (Liu, 1993) and mechano-sorption under changing environmental conditions (Houska and Bucar, 1995; Hanhijarvi, 1995) also affect strength and stiffness of timber in service.
The time-dependent behaviours of wood has been widely investigated, and the following measures of time-dependent response are commonly used in experimental investigation: creep, the increase of the deformation with time, under a sustained applied action; relaxation, the stress decay under constant deformation; stiffness variation in dynamic mechanical analysis (Ferry, 1980); and rate of loading (or straining) effects (Boyd and Jayne, 1982).
These basic techniques can also be generalized to include recovery (the recovery of strain once a creep load is removed) and more general loading histories (Barrett, 1982). In order to investigate the time-dependent behaviours of Terminalia ivorensis in this study, the creep technique was employed.
The prognosis of the long-term behaviours of timber structures under service loads and environmental influence is one of the most discussed issues in research and design of timber structures(Foschi and Barrett, 1982). Since deformations developed are due to
3
very complicated physical and mechanical time-dependent properties, a wide range of factors, such as variable moisture and temperature conditions, stress level, span-depth ratio, the structure of wood and other factors affect the wood and should be taken into consideration to enable the prediction of deformation in timber structures more accurately for design purpose.
There have been serious investigations devoted to load duration effects in wood carried out by researchers in different countries up to now, and more extensive reviews on the topic have been documented by Morlier (1994), Hunt (1999) and Dinwoodie, (2000). Nevertheless, an establishment of a plan mathematical model and/or definition of numerical parameters for prediction of final deformation for Nigerian timber species are not available.
The serviceability limit state of timber structures is seriously influenced by the increase of deformation due to creep of material. The creep process lead to a time-dependent increase of deformation of structural elements that can cause inadmissible deformation and even collapse of an entire construction. Lots of rheological models have been designed by various researchers (Bodig, 1993; Toratti, 1992; Hanhijarvi, 2000; Martensson, 1992; Dubois et al., 2005; Chassagne et al., 2006; Dinwoodie et al., 1990; Pierce et al., 1985) during the last decades with the aim of describing and simulating the time dependent behaviour of a natural viscoelastic material-wood. However, the mathematical models derived contain a great variety of constants to be determined by large sample tests.
Wood loaded under a long period of time will experience an instant deformation right after load is applied. With time, creep deformations will develop in the loaded specimen.
4
Part of the deformation will be elastic and disappear right after the load is removed. The other part is a plastic deformation that is due to viscous flow within the molecules that leaves a permanent deformation (Hoffmeyer, 1990).
Timber which is a product from wood differs from other capillary-porous material, for it is living and is composed of cells. These cells can swell and shrink to a relatively great extent as they absorb or release moisture (Krus and Vik, 1999), resulting in variability within and between timber species. The inherent variability of timber which is unique in its structural and mode of growth results in characteristics properties which are distinct and more complex than other common structural materials such as steel, concrete and brickwork (Shuaibu, 2010). Its hygroscopic character however, affects all related mechanical properties leading to degradation of material stiffness and strength over the service life resulting in the loss of capacity and consequently structural integrity even after being in use for decades (Hassani et al., 2014).
The focus of this research is on Nigerian grown Terminalia ivorensis (black Afara) timber species which is commonly used within the construction industry in Nigeria. The rheological properties studied established relationships between stress, strain, rate of strain, and time.
1.2 Problem Statement
The serviceability limit state of timber structures is greatly influenced by the increase of deformation due to creep of material. The creep process leads to a time-dependent increase of deformation of structural elements that can cause inadmissible deformations and even collapse of an entire structure. A lot of different rheological models have been designed by various researchers (Bodig, 2000; Toratti, 1992;
5
Hanhijarvi, 2000; Mårtensson, 1992; Dubois et al. 2005; Chassagne et al. 2006; Dinwoodie et al. 1990; Pierce et al. 1985) during the last decades with the aim of describing and simulating the timedependent behaviours of a natural viscoelastic material-wood. However, the mathematical models derived contain a great variety of constants to be determined by large sample tests.
Traditionally, structural design in Nigeria is based on British Standard. Eurocode 5, which is the current design code for timber structures in Britain. This code is based on limit state design approach that requires the treatment of the ultimate and serviceability limit states separately. A deterministic constant, kdep is used in Eurocode 5 to accommodate creep of timber. There is the need to develop a model based on creep test data that is capable of adequately but simply predicting timber creep, in order to fully accommodate timber rheology in design.
1.3 Justification of Research
As reliability based design replaces traditional deterministic design practices, a long-term creep law for wood becomes necessary to account for the time-dependent behavior of wood.
Improvements must be sought in our knowledge of the long-term behavior of non-lumber materials. This information could help improve structural modeling, and also clarify questions of the linkages between strength degradation and creep, as well as the interaction between environmental effects, threshold levels for strength, upper limits of creep deformations, etc.
6
Furthermore, Leichti et al., (1990) cited the lack of information about the long-term behavior of structural wood composites as the main reason why designers are reluctant to use them in primary structural applications, such as beams and columns.
Nigeria has very vast timber resources. However, the resources are not fully utilized to their maximum capacity in the area of design and construction. The area it gained wide applicability is in roof truss systems which is based on prescriptive design. To effectively design timber structures based on the current designed philosophy of the Eurocode 5, material properties, strength classification as well as time dependent behaviour of timber must be established.
1.4 Aim and Objectives
1.4.1 Aim of Study
The aim of this study is to investigate the creep behaviour of black afara (Terminalia ivorensis) timber.
1.4.2 Objectives of Study
The objectives of the study to achieve the aim are as follows;
1. To determine the reference material properties; density, modulus of elasticity and bending strength of Terminalia ivorensis (black afara) timber specie according to EN 338 and Eurocode 5 as well as the other material properties; tension strengths parallel and perpendicular to grain, compression strength parallel and perpendicular to grain, shear strength and shear modulus.
7
2. To fit the requisite properties of the Terminalia Ivorensis (Black Afara) timber to adequate stochastic distribution models using Kolmogorov-Smirinov test in easyfit statistical software package.
3. To determine deformation from one-year creep test on Terminalia ivorensis timber specie and performregression analysis to develop time-dependent creep prediction models for the timber under investigation.
4. To investigate the microstructure of Nigerian grown Terminalia ivorensis timber before and after creep in order to find out the effect of creep at the micro level.
1.5 Scope of the Research
This study is focused on Nigerian grownBlack Afara(Terminalia ivorensis) timber specie. The strength and stiffness properties test carried out are those reference properties specified by EN 384 (2004) and Eurocode 5. Creep behaviour of Terminalia ivorensis timber species was measured and modelled over a period of three hundred and sixty days (one-year) according to DIN 50, 119 (1978). The developed statistical model considered only two variables; creep deformation and time. Effect of moisture content and temperature were implicitly accommodated.
1.6 Limitation of the Research
The following are the limitation of the research:
i. The timber samples were not obtained directly from the parent trees, but from open market already converted. The actual age of the timber could not be established.
8
ii. The results obtained from this thesis is based on standard controlled laboratory condition. There may be variation from what is obtained in the field.
iii. Three creep testing machines were used for the long term creep tests and each machine can accommodate only one sample. Therefore, only three samples were used for the creep tests.
iv. The mathematical models developed are only for one-year creep tests. Creep prediction with the developed models may only be approximate if used for more than one-year creep.
IF YOU CAN'T FIND YOUR TOPIC, CLICK HERE TO HIRE A WRITER»
