UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental evaluation of manufacturing parameters on the structural performance of rounded dovetail… Anastas, Hiba 2006

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

Item Metadata

Download

Media
831-ubc_2007-0309.pdf [ 5.09MB ]
Metadata
JSON: 831-1.0074932.json
JSON-LD: 831-1.0074932-ld.json
RDF/XML (Pretty): 831-1.0074932-rdf.xml
RDF/JSON: 831-1.0074932-rdf.json
Turtle: 831-1.0074932-turtle.txt
N-Triples: 831-1.0074932-rdf-ntriples.txt
Original Record: 831-1.0074932-source.json
Full Text
831-1.0074932-fulltext.txt
Citation
831-1.0074932.ris

Full Text

EXPERIMENTAL EVALUATION OF MANUFACTURING PARAMETERS ON THE STRUCTURAL PERFORMANCE OF ROUNDED DOVETAIL CONNECTIONS by H I B A A N A S T A S B . S c , The University of Brit ish Columbia, 2003 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E STUDIES (Forestry) T H E U N I V E R S I T Y OF BRIT ISH C O L U M B I A August 2007 © Hiba Anastas, 2006 ABSTRACT The revival of timber framing in recent years and the development of computer numerically controlled (CNC) machines have led to a number of research projects and related studies on traditional timber joinery. This paper examines the experimental evaluation o f the structural performance of rounded dovetail connections. The study investigates four different climatic conditions to evaluate their influence on the load carrying capacity, stiffness, and failure modes o f single and double dovetail joints under static shear loading. The study also examines the effect of manufacturing tolerances such as machining speed and joint tightness (tight joint versus loose joint) on the overall performance of the dovetail joints. It was found that the specimens manufactured and tested in the dry condition outperformed the specimens evaluated under the other climatic conditions. In the study of different manufacturing tolerances, it was determined that the joints produced at low speed and without a gap had a higher capacity than those produced at higher speed and with a gap. TABLE OF CONTENTS A B S T R A C T i i T A B L E OF C O N T E N T S i i i L IST OF T A B L E S v L IST OF F I G U R E S v i i A C K N O W L E D G E M E N T ix C H A P T E R 1. I N T R O D U C T I O N 1 1.1 General 1 1.2 Objective and Scope 7 C H A P T E R 2. L I T E R A T U R E R E V I E W 9 2.1 Introduction 9 2.2 Testing of Traditional Timber Connections 12 2.3 Preliminary Investigations of Rounded Dovetail Connections 17 C H A P T E R 3. E F F E C T OF C L I M A T I C C O N D I T I O N S 21 3.1 Effect of Moisture on Timber Behaviour 21 3.2 Statistical Experimental Design 23 3.3 Materials and Joint Configuration 24 3.4 Joint Test Procedure 26 3.5 Results and Analysis 31 C H A P T E R 4. E F F E C T OF M A N U F A C T U R I N G T O L E R A N C E S 48 4.1 Introduction 48 4.2 Statistical Experimental Design 49 4.3 Material and Joint Configuration 49 4.4 Joint Test Procedure 51 4.5 Results and Analysis 51 i i i C H A P T E R 5. C O N C L U S I O N S A N D R E C O M M E N D A T I O N S 64 5.1 Conclusions 64 5.2 Recommendations for Future Work 65 B I B L I O G R A P H Y 67 A P P E N D I X 74 iv LIST OF TABLES Table 3- 1 Summary of material properties of main beams used in the climatic condition tests 26 Table 3- 2 Summary of material properties of joists used in the climatic condition tests 26 Table 3-3 A N O V A results for climatic conditions effect on the connection ultimate capacity 40 Table 3-4 A N O V A results for climatic conditions effect on the connection capacity at 5mm deformation 41 Table 3-5 A N O V A results for climatic conditions effect on the connection capacity at 3 mm deformation 42 Table 3 -6 A N O V A results for climatic conditions effect on the connection shear stiffness 43 Table 3 -7 A N O V A results for climatic conditions effect on the connection adjusted shear stiffness 44 Table 3-8 A N O V A results for climatic conditions effect on the connection maximum deformation 46 Table 3-9 Summary of test results of the climatic conditioning tests 47 Table 4- 1 Summary of material properties of main beams used in the tolerances tests.. 50 Table 4- 2 Summary of material properties of joists used in the tolerances tests 50 Table 4- 3 A N O V A results for production parameters effect on the connection ultimate capacity 57 Table 4- 4 A N O V A results for production parameters effect on the connection capacity at 5mm deformation 58 Table 4- 5 A N O V A results for production parameters effect on the connection capacity at 3 mm deformation 59 Table 4- 6 A N O V A results for production parameter effect on the connection stiffness 60 Table 4- 7 A N O V A results for production parameter effect on the connection adjusted stiffness 61 Table 4- 8 A N O V A results for production parameters effect on the connection maximum deformation 62 Table 4- 9 Summary of test results of the production parameters tests 63 Table A 1 Material properties and test results for control (DD) specimens 74 Table A 2 Material properties for D W D series specimens before and after conditioning 75 Table A 3 Material properties and test results for D W D series specimens before testing 76 Table A 4 Material properties for W D specimens after conditioning 77 Table A 5 Material properties and test results for W D series specimens before testing.. 78 Table A 6 Material properties and test results for W W series specimens 79 Table A 7 Material properties and test results for S5 and S9 series specimens 80 Table A 8 Material properties and test results for G l and G2 series specimens 81 v i LIST OF FIGURES Figure 1-1 Typical timber frame structure with traditional timber connections (Brungraber 1985) 1 Figure 1-2 Common traditional timber joinery; mortise and tenon (a) and dovetail joint (b) .....2 Figure 1- 3 The Hundegger K2-5 machine at the University o f Brit ish Columbia 6 Figure 2- 1 Rounded Dovetail Joint (Technische Universitat Miinchen 1999) 11 Figure 2- 2 Comparison between (a) single and (b) double dovetail connections 11 Figure 2- 3 Geometric parameters o f rounded dovetail connection 20 Figure 3-1 Joint Configuration of single (left) and double (right) dovetail connection .. 24 Figure 3 -2 Dovetail joint set-up test 28 Figure 3- 3 Load application regime according to E N 26891 29 Figure 3-4 Crack formation in test specimen 31 Figure 3-5 Load-deformation curves for 10 single DD-dovetail specimens 33 Figure 3 -6 Load-deformation curves for 10 double DD-dovetail specimens 34 Figure 3 -7 Load-deformation curves for 5 single DWD-dovetai l specimens 35 Figure 3-8 Load-deformation curves for 5 double DWD-dovetai l specimens 35 Figure 3 -9 Load-deformation curves for 5 single WD-dovetail specimens 36 Figure 3-10 Load-deformation curves for 5 double WD-dovetai l specimens 37 Figure 3-11 Load-deformation curves for 5 single WW-dovetai l specimens 38 Figure 3-12 Load-deformation curves for 5 double WW-dovetai l specimens 38 Figure 3-13 Comparison for ultimate capacities for different climatic conditions 39 Figure 3-14 Capacities at 5 mm deformation for different climatic conditions 41 Figure 3-15 Capacities at 3 mm deformation for different climatic conditions 42 Figure 3- 16 Shear stiffness for different climatic conditions 43 Figure 3-17 Adjusted shear stiffness for different climatic conditions 44 Figure 3-18 Max imum deformation for different climatic conditions 45 Figure 4-1 Load-deformation curves for 10 single dovetail specimens produced at low speed and without a gap 52 Figure 4- 2 Load-deformation curves for 5 single dovetail specimens produced at medium speed 53 Figure 4- 3 Load-deformation curves for 5 single dovetail specimens produced at high speed 53 Figure 4- 4 Load-deformation curves for 5 single dovetail specimens produced with a 1-mm gap 55 Figure 4- 5 Load-deformation curves for 5 single dovetail specimens produced with a 2-mm gap 55 Figure 4- 6 Comparison for ultimate capacities for different production parameters 56 Figure 4- 7 Comparison for capacities at 5 mm deformation for different production parameters 57 Figure 4- 8 Comparison for capacities at 3 mm deformation for different production parameters Figure 4- 9 Shear stiffness for different production parameters 60 Figure 4- 10 Adjusted shear stiffness for different production parameters 61 Figure 4- 11 Max imum deformation for different production parameters 62 v i i i ACKNOWLEDGEMENT I would like to thank Dr. David J. Barrett and Dr. Frank Lam for their guidance throughout this project and Thomas Tannert for the help provided during testing. I would like to thank the Department of Wood Science at the University of Brit ish Columbia and Markus Steiniger for his contribution in developing the double dovetail shape and conducting initial research that provided valuable guidance. I would also like to thank all my friends and family for their help in the past years. CHAPTER 1. INTRODUCTION 1.1 General Traditional timber framing is basically a post and beam structure that incorporates a complex system of joinery where the timber connections employ pegged mortise and tenons, wedges, splines, and simple bearing joinery. Some of the most common timber joinery found in timber framing is the mortise and tenon, the dovetail and the knee brace joint. In the past, primeval forests existed in some countries such as Northern Europe, Japan, China and India, and so the techniques of joining large timbers into a structural skeleton were developed as a craft. The timbers were joined together to form timber-framed buildings without the use of metal fasteners. Instead, these large timbers were connected using mortise and tenon and dovetail notches fastened with thick wooden pegs. This kind of post and beam structure was designed to be a more stable dwell ing intended to last for centuries (Benson and Gruber 1980). Figure 1-1 was removed because a letter of copyright of permission could not be obtained. The figure illustrates a typical North American timber-frame with traditional carpenter's j oinery. Figure 1-1 Typical timber frame structure with traditional timber connections (Brungraber 1985) Figure 1-2 shows some of the most common traditional timber joinery. The mortise and tenon and its variations are very common joints used in timber framing. They 1 are used as tension joints to fasten posts and beams, connectors, knee braces, wall plates, struts and collar ties. They are tightly secured with hardwood pegs or dowels to transfer loads under service conditions. The dovetail joint is another notch used in timber framing to connect purlins and joists to rafters and girts. The interlocking of the joint is dependent upon the weight of the timber and the wedging effect of the dovetail joint thus eliminating the use of wooden pegs (Hewett 1980). Figure 1- 2 Common traditional timber joinery; mortise and tenon (a) and dovetail joint (b) The history of timber framing evolved for more than a thousand of years beginning around the 3 r d century A D (Chappell 1998). To this day, many of the surviving European and M i d and Far Eastern timber framing structures still stand tall, and they include stave churches, bridges, temples, palaces and cathedrals. The success of these 2 structures is attributed to the intrinsic toughness in the connections (Brungraber 1992a). Since North America was under the Brit ish colonization, they were influenced to adopt methods and techniques of building based on the European model. Timber framing prevailed and became a dominant method of construction in North America from 1620s until 1870s (Sobon and Schroeder 1984). In the past two hundred years, stud framing (wood frame) replaced timber framing because this method of building is faster and cheaper. Timber joinery has lost its pre-eminence because of the development of inexpensive mechanical fasteners and adhesives that now have almost entirely replaced timber joinery in timber structures (Price 1996), and also because of the intricacy and the variations in the individual craftsman's techniques and skills, making the behaviour of timber framing somewhat difficult to predict (Goetz el al. 1989). In the middle 1800s, the techniques used for timber framing started to rapidly vanish. This was attributed to the rapid growth of sawmills producing standardized sizes of dimension lumber accompanied by mass production of nails and metal fasteners (Elliot and Wallas 1977). It was also contributed to the desire of enclosing a building that required less skills and time, which was in demand for the rampant growth of the American population. B y the early twentieth century, studs replaced timber framing in residential buildings, and steel fittings and cast iron in some commercial buildings. Timber framing was more or less in abeyance for about 120 years in America, whereas in the Far East and Europe, residential and commercial structures were still built using heavy timber and the traditional methods of connection (Charles 1984). To this day in Japan, it is common for 3 "traditional post and beam" single family houses to be constructed using mortise and tenon joints to frame members (Lam et al. 2001) Techniques used in timber-framed structures differ from practices used to build conventional wood-framed structures. Timber framing commonly uses fewer and larger timbers with dimensions in the 15 cm to30 cm (6 in to 12 in) range as opposed to conventional wood framing which employs more wooden members with dimensions ranging between 5 cm and 25 cm (2 in and 10 in). The methods of fastening the frame members are different between timber framing and conventional wood framing, in which the latter uses nails and mechanical fasteners to jo in the members, whereas timber framing employs more complex joinery such as mortise and tenon fastened using wooden pegs (Lewandoski 1992b). Techniques used in timber framing make an efficient use of forests and timber, by using members of all lengths, sizes and shapes, whereas in conventional framing, labour and time constraints are a factor, therefore it uses straight and long beams which require less joinery work. The activities and interests in timber framing have seen a remarkable revival and have undergone resurgence in popularity in the past two decades. The revival is contributed in large to the development of high speed wood machining and the methods of joining distinct pieces of machined timber. This rebirth was also stimulated by the increased awareness of existing information on the trade and the advent of more energy efficient materials, making timber frame skeletons viable, energy efficient structures that are durable and aesthetically pleasing (O'Connel l and Smith 1999). 4 While traditional timber-framing construction has a longer track record in construction over conventional heavy timber framing, it lacks the specifications, standards and codes set for today's building techniques and requirements (Chappell 1984). Where heavy timber construction use metal fasteners and hybrid joining systems that are governed by codes, traditional timber framing does not have developed techniques and codes that wi l l satisfy both the traditionalists and designers/ regulators. Accordingly, practical and tested design guidelines for traditional timber joinery are sought (Schmidt et al. 1996). The design of timber-frame connections is currently beyond the scope of the National Design Specification for Wood Construction (NDS) (Schmidt and M a c K a y 1997). The introduction of C N C machines radically changed the timber frame manufacturing industry. Curves are as easy to cut as straight lines, complex 3-D structures are easily produced, and the number of machining steps requiring operator intervention has been dramatically reduced. Timber frame buildings have always had visual appeal but today's structures have become more complex while also embracing building science that relies on tight construction and energy conservation. This had been made possible because of better building envelopes, introduction of sophisticated steel connectors, C A D software and C N C cutting technology. C N C machines, interfaced with 3-D modeling software, are the most recent innovation in adding significant capabilities to timber operations. These types of processing machinery, migrating from their base in Europe, provide the means to produce 5 advanced joinery details that exhibit a high degree of accuracy and repeatability with tolerances of 0.3 mm (12/1000 of an inch). The market leader in North America with over 50 machines in use is Hundegger Maschinenbau GmbH. The Hundegger K 2 (Figure 1- 3) is a fully "automated timber cutting machine" (Brungraber 1999) that excels in all types of wood construction; frame construction; prefabricated construction; log home joinery and glue-laminated joinery. Figure 1- 3 The Hundegger K2-5 machine at the University of Brit ish Columbia In the 1980s, Kessel (1985, 1989) and Natterer (1985) actively promoted the application of C A D / C A M technology in timber construction, initiating CAD-software developments in heavy-timber construction. Machine manufacturers, such as Hundegger, have developed fully automated timber-processors, continuously improving the manufacture of timber structures. Bamford (2003) investigated the acceptance within the architectural design and structural engineering society in British Columbia for an increased application of CNC-t imber processing, comparing CNC-cut joinery with steel 6 fastener systems used for similar structural applications, and concluding that most of the all-wood connections provide a better cost-effectiveness. There is a large potential for the use of CNC-processed wood-to-wood connections in North America. 1.2 Objective and Scope There are many variables that affect the design of timber joints. These variables include the load-bearing performance, the type of fasteners used to interlock the joint, moisture content, the quality of the manufacturing, the wood species, and the existence of knots. A l l timber joints rely on interlocking capacity by the fasteners used or by their own form. Therefore, it is essential to have accurately cut timber components that provide interlocking surfaces, which in turn provide stiffness in the joint (Erman 2 0 0 2 ) . Traditional timber joints were designed to provide resistance against different forces. Timber is used as posts since it has good compressive strength parallel to the grain, and used as joists and beams when loaded perpendicular to the grain since it has good resistance against bending moments. Timber joints are classified into two groups: orthogonal, in which the joining angle is a right angle, and diagonal in which the joining angle is other than a right angle. Much of the knowledge about the performance of most traditional timber connections is lost either because of modern mechanical fastener replacements or because of the excessive craftsmanship required. Three primary objectives exist for this research. 1. To study the maximum load carrying capacity, stiffness, and failure modes of the R D C under static shear loading. 2. To determine the effects of seasoning and dovetail type on rounded dovetail joints under shear. 3. To determine the effect of manufacturing parameters on the overall performance of rounded dovetail joints. Chapter 2 consists of a literature review of the experimental and numerical testing conducted on dovetail connections and various other traditional timber joinery. It also includes the preliminary investigations of rounded dovetail connections at the University of British Columbia. Chapter 3 examines the effects of different climatic conditions on the structural performance of rounded dovetail connections. Chapter 4 includes experimental tests conducted on rounded dovetail connections to examine the effect of manufacturing tolerances, such as machine processing speed and gap between mortise and tenon, on the connections strength, stiffness and deformation. Chapter 5 presents the conclusions and recommendations for future studies. 8 CHAPTER 2. LITERATURE REVIEW 2.1 Introduction Traditional wood-to-wood connections are believed to have huge potential in commercial projects and residential timber frame homes. However, there is reluctance among architects and engineers to use such connections because strength and design values are unknown and because of the difficulty of fabricating the joints by traditional handcrafts methods. Studies concerning the structural performance of Rounded Dovetail Connections (RDC) under different loading configurations have been carried out, but information on their behaviour under long-term loading and in different climatic conditions is still needed. A structure is " a series of connections joined to one another by members" (Brungraber 1992b). It is essential to establish the connections capacity against applied shear, axial and rotational loads in order to assess the capacity of a structure. Timber joinery has lost its pre-eminence because mechanical fasteners and adhesives have almost entirely replaced it in modern timber structures. A lso because of its intricacy and the variations in the individual craftsman's techniques and skil ls, the structural performance of timber joinery is sometimes considered to be more unpredictable compared to modern wood frame construction. Various factors come into play that make timber joinery less desirable as opposed to mechanical fasteners. High labour costs, high degree of workmanship, greater amount of time and ski l l required for measuring, fitting, and cutting, increased member size due to the significant removal of wood fibre in the area of the connections. In addition, wood engineering codes do not provide any specific design guidelines for timber joinery due to lack of data and the 9 sensitivity of quality of workmanship and material quality. However, the activity and interest in timber framing has seen a remarkable revival and has undergone resurgence in popularity in the past two decades (Schmidt et al. 1996). Unt i l the mid 20th century, wood-to-wood connections were commonly used in construction where their design and manufacturing was based on the experience of carpenters (Sobon 2002). High labour costs, high degree of workmanship required, assembly difficulties and inefficient use due to over dimensioning of members made the timber frame system too expensive and timber less competitive as construction material. Over the last years, important developments in wood processing machines has taken place that have created the possibility of producing wood-to-wood connections cost effectively (Bamford 2003; Bobacz 2003). The wooden dovetail is a centuries old connection that has been extensively used throughout Europe and Japan. The R D C is a new design that has a dovetail-similar shape in longitudinal direction of the secondary beam and is specifically adapted to be processed semi-manually or completely automatically with a CNC-t imber processor. The newly developed R D C wi l l not be widely adopted in North America until more technical data are developed including performance data under long-term loading conditions. Load duration effects relating to wood-to-wood connections are of concern in two aspects of design: joint strength (long-term design load) and serviceability (e.g. deflection under typical sustained loading and whether this value is acceptable for the structure and the structure's components). 10 Rounded dovetail connections are commonly used to connect purlins to rafters in classic timber-frame bent structures, floor joists to r im or suspended girder beams, jack rafters to valley or hip rafters, or rafters to ridges and top wall purlins in European purlin roofs. Figure 2- 1 was removed because a letter of copyright permission could not be obtained. The figure shows a diagram of a type of a rounded dovetail connection that was developed at the Technical University of Munich. Figure 2- 1 Rounded Dovetail Joint (Technische Universitat Miinchen 1999) Different configurations of the dovetail connection have been used over the centuries, primarily governed by practical considerations such as the carpenter's skills and available tools. The development in C N C wood-processing machinery has drastically changed these constraints and almost any desired geometry is possible at little added cost. One particular shape of interest is the Double Rounded Dovetail Connection ( D R D C ) that was developed at the University of British Columbia. Figure 2- 2 shows the single (a) and the double (b) R D C geometries. Figure 2- 2 Comparison between (a) single and (b) double dovetail connections 11 2.2 Testing of Traditional Timber Connections Although timber framing was the dominant method of building in the past, applicable research conducted on timber framing and traditional timber joinery is very limited in Europe and North America. One of the first prominent research projects on timber framing was carried out by Brungraber (1985). The research was broad in scope, and it examined the experimental and numerical results of a full timber frame structure and individual joints as well. Kessel and Augustin's work in Germany focused on the capacity of mortise and tenon joints loaded in tension and stipulated recommended design values and allowable loads (Kessel and Augustin 1995, 1996). A number of other studies focused on the mechanical characteristics of newly developed engineered configurations of traditional timber connections (King et al. 1996; Bulleit et al. 1999; Seo et al. 1999). Since most timber frame structures are erected using green timber, Schmidt and Scholl (2000) investigated the effect of load duration and seasoning on full-sized mortise and tenon joints loaded in tension. It was concluded that the "design of mortise and tenon joints for long-term loading is a serviceability concern rather than a strength issue." The test results from Schmidt and Scholl research were used to revise the detailing requirements developed by Schmidt and Daniels (1999) to ensure ductile failure of the peg rather than failure of the joint first. Other research focused on the behaviour of the frame structure as a whole, not isolating the behaviour of the individual joints (Erikson 2003). Erikson and Schmidt developed frame stiffness parameters by conducting tests on both single- and multi-storey bents subjected to lateral loads (Erikson and Schmidt 2001, 2002a). Mi l le r (2004) tested the capacity of mortise and tenon joints under tension, shear 12 and direct bearing. A three-dimensional model was developed to validate the numerical results against the physical testing results. Bulleit et al. (1996) developed a semi-rigid connection model to analyze the performance of frames connected with traditional mortise and tenon joints. Sandberg et al. (1996) tested different variations of mortise and tenon joints under simulated gravity loads. The results indicated that these joints have a significant load capacity and resistance to catastrophic failure. Parisi el al. (2000) performed monotonic tests on unreinforced timber connection to obtain information on the characteristics ofthe joint and its sensitivity to several parameters, and to help engineers, architects, and timber framers understand the behaviour of these connections. In the United Kingdom, recent renewal of traditional timber framing with green oak has promoted the use of pegged mortise and tenon connections. Shanks and Walker (2004) conducted tests on machine processed mortise and tenon joints loaded in tension, shear and bending. It was concluded that the "fit of the tenon within the mortise has a very important influence on both stiffness and strength of the joints in tension, shear and bending." Shanks and Walker (2003) also conducted tension testing on eleven different pegged connections using green oak. A current development in analyzing timber structures has led to research on traditional timber frames in Taiwan. Since timber connections are considered as semi-rigid joints, it is important to examine their rotational performance (Chang and Hsu 2005). K ing and his colleagues (1996) conducted studies on the rotational stiffness of one-third scale timber joints. Chang el al. (2006) developed a theoretical model to 13 estimate the rotational stiffness of traditional Nuk i joints with gaps. Nuk i joints are used in Japan and Taiwan to connect beam and column. It was established that the gap in the joint alters the rotational performance of the joints and increases the complexity of the analysis. A number of experimental studies on R D C provided insight into their structural performance (Bobacz 2002; Hochstrate and Kessel 2000). The observed failure modes were tension perpendicular to grain failure at the mortise base or failure of the tenon with splitting of the joist member, depending on support conditions and geometry parameters. It was shown that the load-carrying capacity of the connection depends crucially on the load carrying mechanism between the connected members. It is to be differentiated between cases with properly matching geometry and cases with a groove in the fillet of the connection. The tenon tapering in vertical direction transfers the load evenly over the connection height; it distributes straddling and friction forces, and reduces problematic tension stress concentrations that occur in conventional notches or mortise and tenon joints. Hinkes (1987) studied the strength of handcrafted end-notched joint configuration, a type of mortise and tenon joint, using Spruce of Grade II material, and considering different joint geometry and variable moisture contents. Tests were carried out according to German D I N 4074. The characteristic strength values were used to develop design equations for end-notched configurations, and to introduce it into the 14 German timber code 1052-1988. The tests were used to understand the bearing mechanisms of rounded dovetail joints. Heimeshoff and Kohler (1988) studied the strength, stiffness and performance of handcrafted dovetail lap joints using softwood of Grade II material. Tests were performed according to German D I N 4074 and was concluded that the joints are very sensitive to tolerances - shrinkage due to drying of wood - causing substantial slip. It was suggested that the joint is suitable for short-term loading but not recommended for sustained loading in tension. Gorlacher and Kromer (1993) examined the failure mechanisms of dovetail lap joints under tensile loading. They used softwood handcrafted specimens usually found in historic timber structures and were processed at 30% M C then conditioned to 15% M C . The tests included eleven different lap joint set-ups considering a tight and a loose fit. Long-term loading was studied in which the specimens were subjected to a constant load for 100 days under different climatic conditions, and followed by short-term tests until failure. Wood dowels were employed as reinforcement for some of the specimens. Gorlacher and Kromer noticed that dovetail lap joints with no wooden dowels displayed relatively large displacement under tensile loading; and the changing of climatic conditions caused wood shrinkage, which in turn led to loose fit and slip. It was recommended that dovetail lap joints are not suitable for long-term loading in tension. From the tests conducted on the specimens with dowel reinforcement, it was deduced that the dowel acted as a load-bearing element. 15 Kreuzinger and Spengler (1999) studied the load bearing performance of C N C -processed dovetail joints in beam-to-beam configurations. Different parameters were investigated, such as the size and the cone angle of the dovetail tenon and the support condition of the main beam. The support condition of the dovetail tenon was also examined (the mortise and the tenon was either in full contact or had a small recess). The tests were conducted according to D I N 4074 on kiln-dried softwood specimens. It was found that the load bearing behaviour of dovetail joints is better compared to that of simple end-notched lap joints. A lso, varying the moisture content of the specimens had a substantial impact on the joinery fit. Further studies were required to investigate the long-term performance of dovetail joints under sustained loading condition. Holzner and Spengler (1999) developed a simple connection model and design equations for specially processed dovetail joints. A study on optimizing the connection geometry was carried out. The results indicated that the load bearing performance of the joint is very sensitive to changes in moisture content and tolerances. Hochstrate and Kessel (2000) conducted an experimental study on the load bearing performance of CNC-processed rounded dovetail joints. The tests included different parameters by varying the dimensions of the main beam and the joist, the size of the tenon, and the distance from the dovetail tenon to the loaded bottom edge of the main beam. Tests were conducted according to D I N 4074 on kiln-dried softwood specimens, and according to D I N 1052-1 on Glulam specimens. The results of the characteristic strength values derived from the test series were used as recommendations for design strength considering L imi t State approach of the European timber code and the new German D I N 1052 (Draft). Harada et al. (2005) investigated the effect of member moisture content on the mechanical properties o f "koshikake-ari", a type o f dovetail joint. Japanese Cedar (Cryptomeria Japonica) specimens were processed using a pre-cut processing machine. Four-point bending tests were carried out and it was found that there were no significant differences between the green, kiln-dried ( M C ~ 15%), and over-kiln-dried ( M C ~ 5%) specimens. There was no clear trend detected between moisture content and bending moment, probably due to the insufficient number of specimens used in the study. 2.3 Preliminary Investigations of Rounded Dovetail Connections Preliminary studies at the University of Brit ish Columbia have been carried out on rounded dovetail connections to gain insight into their structural behaviour and load-bearing mechanisms. Tests on the double rounded dovetail connections indicated that a less brittle failure mode could be achieved since the load is transferred over two areas, further dispersing stress concentrations. Work at the University of Brit ish Columbia demonstrated that rounded dovetail joint under shear loading show a linear elastic behaviour until reaching a load that leads to crack development, with the double dovetail connection being on average 19% stronger. The double dovetail joint provides higher shear capacity while not showing significantly different structural performance under other loading conditions, demonstrating considerable resistance in tension, vertical bending and horizontal bending of this innovative joint. Double dovetail joints were less brittle and presented higher capacities due to better load distribution and therefore 17 reduced stress concentrations at the mortise base. The presented work addresses geometric parameters that affect the structural performance of the dovetail joint in order to comprehend its complex load carrying mechanism (Steiniger 2004). Shear tests were conducted on CNC-processed dovetail joints to provide preliminary information on their strength performance. Different joint geometries were examined such as the single and double dovetail configurations. Tests were carried out using LSL-Timberstrand 1.9E and kiln-dried Hemlock to compare results of both solid wood and composite material. Results on the maximum strength of the connection, L S L specimens were found to have, on average, 50% higher strength than Hemlock specimens (Tannert and Lam 2006a) Dietsch (2005) developed a finite element model to evaluate the distribution of different stresses in the connection at failure. A linear elastic material model was used since experimental tests have indicated that brittle failure governed the load bearing capacity of the connection and that crack formation and propagation occurred within the elastic range. R D C , even in their basic form, have a complex load transfer mechanism, governed by distinct geometric features (Figure 2- 3). To aid in possible improvements in connection efficiency, a sensitivity study of the three most important geometric parameters was conducted at the University o f Brit ish Columbia, (Dietsch 2005). These parameters were the angle p between the dovetail flanges, the dovetail angle a, and the dovetail height h i . The study revealed that the flange angle p has a substantial impact on the load carrying capacity o f the dovetail connection with optimal stress conditions for 18 angles below 15°. The dovetail angle a was found to have little influence on the load carrying capacity of the dovetail connection. A n optimum height hi of the dovetail was found to be 120 mm when the height of both beams was 200 mm. A project was initiated at the University of Brit ish Columbia to study the load bearing capacity of rounded dovetail connections under different static loading conditions and to examine the connections stiffness and failure modes. The major objective of the work was to identify the structural performance of single and dovetail joints in shear and tension tests and in vertical and horizontal bending tests as well. The tests have demonstrated that although the rounded dovetail joint is designed and used as a shear connection, it withstands some tension load and has substantial moment resistance in vertical and horizontal bending. Significant differences in the performance of single and double joints were shown only under shear tests. Therefore, it was recommended to pursue further studies of this connection to establish the effect of various factors, such as geometric parameters and climatic conditioning (Tannert and Lam 2006a). 19 3/2 Figure 2- 3 Geometric parameters of rounded dovetail connection The behaviour of a timber frame is dependent upon the structural performance of the joints. Joint geometry should be examined to understand their manufacturing and construction potentials (Erman 1997). Tannert and Lam (2006b) studied the effect of joint geometry on the structural performance of rounded dovetail connections by considering the dovetail height (hi) and dovetail angle (a). Based on the results, it was recommended to produce rounded dovetail connections with dovetail angles between 10° and 15° and dovetail heights of approximately two thirds of the member height. 20 CHAPTER 3. EFFECT OF CLIMATIC CONDITIONS 3.1 Effect of Moisture on Timber Behaviour For many timber joints, connection tolerances can have a significant impact on the load bearing performance. One of the key problems of timber joints results from the shrinkage characteristic of the material itself because wood shrinks and swells i f exposed to moisture fluctuations below the fibre saturation point - very little in the axial, but substantially in the radial and tangential direction. This characteristic becomes important when the moisture content (MC) of the timber at the time of manufacturing differs significantly from the M C under service conditions. Tight joinery may become loose or tighten up over time, thus possibly resulting in a different bearing performance. In preliminary studies it was proven that excellent tight fitting connections can be produced, however, the issue of lack of fit of the connections resulting from drying out of the solid timber after manufacturing was also encountered. N o research on the influence of seasoning on the structural performance of dovetail joints under load has been done. Often timber frames structures are constructed with green timber that dries while in service conditions. Therein lies the motivation for this research. Lumber equilibrates to a wide range of moisture content ( M C ) levels in use. For example, lumber installed green in timber bridges may remain at or near the fibre saturation point for several years after bridge installation. In contrast, lumber used in attic may encounter a drier condition. 21 Timber is very sensitive to moisture fluctuations, in which it swells when takes up moisture and it shrinks when dries. The shrinkage and swelling usually take place at moisture contents below the fibre saturation point (-25%). However, these dimensional changes are not the same in al l directions. Small dimensional changes occur in the longitudinal direction. Shrinkage in the tangential direction (parallel to the growth rings) is always a little larger than radial shrinkage because the latter is restrained by rays. Environmental changes have a significant influence on the structural performance of wood and wood-based materials in many applications. Timber structures are actually used in various conditions and the performance of the joints of these structures should be properly evaluated according to the conditions in which they are used (Nakajima 2000). The mechanical properties of timber are dependent on moisture content and often dimensional changes induced by moisture variation lead to greater displacements compared to ones caused by mechanical loading. In addition, the interaction of mechanical loading and moisture changes can cause excessive creep in timber structures (Martensson and Thelandersson 1990). The study examines four different climatic conditioning strategies to study their influence on the load carrying capacity, stiffness, and failure modes of single and double dovetail (DD) joints under static shear loading. D D specimens are manufactured and tested in the dry condition. Such a case simulates a condition where the lumber is manufactured and erected into a structure while it is dry. D W D specimens are manufactured dry, subjected to high humidity, dried and tested at low M C . This type of conditioning simulates a case where dry lumber is cut, 22 stored in a wet condition, for example, outside on rainy days, and the lumber dries before erecting. The W D specimens are manufactured wet; dried, and tested at low M C . The W W specimens are manufactured and tested in the wet condition. This situation occurs when green lumber is processed, and erected into the structure while it is still green. 3.2 Statistical Experimental Design Three different climatic conditions were compared to the control geometry for both, the single and the double dovetail configuration. The experimental layout for the climatic conditions study is a two-way analysis of variance. One factor is the climatic conditioning with four levels (control and three variations) and the second factor is the joint configuration with two levels (single and double dovetail). To enable comparisons with previous studies, the dovetail geometry of the control specimen was chosen as: width bi = 50 mm, depth t = 28 mm, dovetail angle a = 15°, and flange angle p = 15°. The connection back-cut B C , which is the difference between member height h and dovetail height hi was chosen as 65 mm. Therefore the dovetail height was hi = 123 mm. For the double dovetail joints, the back-cut of the second dovetail was chosen as: BC2 =105 mm, resulting in a second dovetail height b.2 = 83 mm, the width of the second dovetail was chosen as b2 = 57.5 mm (Figure 3- 1). 23 Figure 3-1 Joint Configuration of single (left) and double (right) dovetail connection A total of 50 specimens (a specimen consisting of a main beam and a joist) were used in the climatic conditioning tests, 25 of which had a single joint configuration and the other 25 with a double joint configuration. One additional dummy specimen was used to establish a rough estimate of the connection ultimate capacity and to calibrate the test apparatus. 3.3 Materials and Joint Configuration Kiln-dried western hemlock (Tsuga heterophylla) was used in the study. Naturally grown 90-year-old Western Hemlock stands represent much of the emerging timber supply in the B C coastal forest. Studies on dovetail connections in Europe (Kreuzinger and Spengler 1999; Hochstrate and Kessel 2000; Bobacz 2002) have used Norway spruce {Picea abies), a species that presents similar properties to western hemlock, making it 24 possible to compare their findings with results from the test program at the University of Brit ish Columbia. The test results from the control specimens are also used to investigate the influence of geometry parameters on the structural performance of rounded dovetail connections and to compare the performance of structural composite lumber to that of western hemlock. The R D C was applied to connect a joist to a main beam. The member dimensions b and h were chosen as 89 mm and 188 mm, respectively. The length of the joist and the main beam was chosen as 800 mm and 600 mm, respectively. The material was conditioned before testing and the moisture content ( M C ) and the apparent density (based on specimen as tested weight and volume) of each specimen were determined. The average M C was 14.1%, with a standard deviation of 4.4% and extreme values of 10.5% and 28.5%, respectively. The average apparent density was 484 kg/m 3 with a standard deviation of 52.9 kg/m 3 and extreme values of 397.7 kg/m 3 and 638.2 kg/m 3 , respectively. Table 3-1 and Table 3- 2 show a summary of the material properties of main beam and joists used in the climatic conditions tests. Short-term tests to failure are conducted on specimens that are subjected to different climatic conditions with practical relevance. These conditions are: D D condition: specimen dry at manufacturing and tested with low M C - D W D condition: specimen dry at manufacturing, subjected to high humidity, then dried and tested with low M C - W D condition: specimen subjected to high humidity, wet at manufacturing, then dried and tested with low M C 25 W W condition: specimen wet at manufacturing and tested with high M C Table 3- 1 Summary of material properties of main beams used in the climatic condition tests Moisture Content (%) Apparent Density (kq/m3) Mean Max Min St. Dev. Mean Max Min St. Dev. Control 1DT 12.9 14.3 10.6 1.2 474 513 425 32 Control 2DT 13.7 16.3 11.0 1.5 512 578 449 44 DWD 1DT 11.0 11.4 10.5 0.3 421 449 398 26 DWD 2DT 11.3 12.2 10.9 0.5 479 495 458 14 WD 1DT 10.8 11.0 10.6 0.2 450 486 407 30 WD 2DT 11.2 11.5 11.0 0.2 445 453 435 9 W W 1DT 23.4 28.5 20.2 3.3 539 638 499 57 W W 2 D T 22.7 24.8 19.8 2.1 572 620 490 48 Table 3 -2 Summary of material properties of joists used in the climatic condition tests Moisture Content (%) Apparent Density (kg/m3) Mean Max Min St. Dev. Mean Max Min St. Dev. Control 1 DT 12.3 15.0 11.0 1.5 498 560 456 39 Control 2DT 11.7 12.0 10.9 0.4 441 470 400 27 DWD 1DT 12.2 13.6 11.1 1.2 482 533 413 48 DWD 2DT 12.2 12.8 11.3 0.6 526 576 426 58 WD 1DT 11.6 11.9 11.3 0.2 463 510 430 38 W D 2 D T 11.5 11.8 11.2 0.2 444 453 426 10 W W 1DT 21.8 23.3 20.5 1.2 497 593 446 58 W W 2 D T 21.9 23.5 20.3 1.5 504 553 443 47 3.4 Joint Test Procedure Tests were performed similar to other research conducted on R D C connections to be able to compare results. The test specimens were mounted on a test apparatus by supporting the main beam on two steel plates of 100 mm x 100 mm x 10 mm. The main 26 beam was not clamped thus allowing rotation about its long axis, a state that closely simulates the supporting condition in a real structure. Two small steel plates of 50 mm x 50 mm x 5 mm prevented the specimen from moving backward under the applied load. The free end of the joist was simply supported on a steel plate of 100 mm x 100 mm x 10 mm, which rested on a damper with a spring stiffness of approximately 1 kN/mm in order to simulate a longer beam. The load was applied at a distance of 350 mm from the joint and distributed onto the joist with a steel plate with a diameter of 100 mm. A n M T S hydraulic actuator M T S type was used. The applied load from the actuator and the load at the support (with a load cell) of the free end of the joist were recorded. The vertical movement of the joint was recorded on both sides of the joint (as the relative displacements between main beam and joist), the horizontal movement of the joint, the deflection of the main beam, as wel l as the deflection of the joist at the load application point were recorded with calibrated electronic transducers. The complete test set is shown in Figure 3- 2. 27 Figure 3-2 Dovetail joint set-up test One dummy specimen (1-DD-O) was used to estimate the capacities for the western hemlock specimens. Using the data, the load regime was load controlled following the testing standard E N 26891 (1991) that has been used previously to test dovetail connections, (Hochstrate and Kessel 2000). E N 26891 (1991) applies to joints made with mechanical fasteners used in statically loaded timber structures; it covers the conditioning of test specimens and explains the apparatus to be used for measuring the geometry of the test specimens, and details the loading procedures. A n initial load was applied until a reading was shown, then all the transducer readings were set to zero. In the first load loop, 40% of the estimated connection capacity was applied in 120 sec, the load was maintained for 30 sec, then was reduced in 90 sec to 10% of the estimated capacity and then increased until failure of the joint was achieved with the same rate of loading. Figure 3- 3 shows the load application regime according to E N 26891 standards. 28 Figure 3- 3 Load application regime according to E N 26891 The test for all specimens followed the same procedure: Measure M C and apparent density of test specimen, - Photograph unbroken specimen, - Place the main beam on the supporting plates, - Place the joist into the main beam and on the supporting plate, - Ensure tight fit of the joint between the main beam and the joist, - Attach all four transducers onto the test specimen, - Apply load of 0.2 k N to ensure alignment, Set all transducer readings to zero, - App ly load regime according to Figure 3-3 and record all readings, - Manually record load when first cracks appeared in specimen, - Load specimen until ultimate failure, Photograph broken specimen and measure location of cracks The force transmitted by the dovetail connection was calculated as the difference between the applied load and the load recorded at the free end of the joist. The vertical deflection of the joint was calculated as the average between the vertical movements of the two sides of the joint. The applied load regime allows for further data evaluation, but emphasis was focused on the analysis of the connection capacity. Continuous load-deformation curves were obtained including the unloading/reloading loop. For the purpose of studying the geometric parameters, these unloading/reloading loops were not required and are not presented in this work. The maximum force applied represents the ultimate capacity; further the applied forces at deformations of 3 mm and 5 mm were determined. The load at these two deformation values represent critical values for practical considerations, since larger deformations might lead to unacceptable damages of the surrounding structure. The joint stiffness C was calculated for the deformation range of 0 to 3 mm and the adjusted joint stiffness Cadj was calculated for the deformation range of 1 to 3 mm. The latter does not take the initial alignment slip into consideration. The recommendations from E N 26891 (1991) are to calculate the joint stiffness and the adjusted joint stiffness for deformations at 40% of the ultimate capacity and the linear part of deformation the between 10% and 40% of the ultimate capacity respectively. 30 3.5 Results and Analysis Brittle failure mainly governs the load carrying capacity of rounded dovetail connections. The load was transferred from the joist onto the main beam through the mortise and tenon base. Two failure modes were observed: tension perpendicular to grain failure at the mortise base or/and failure of the tenon with splitting of the joist member (see Figure 3- 4). A lso the mortise base experienced crushing under large deformation. The connection behaviour is dependent upon the material properties of the timber and the geometry of the connection itself. It was found that the single and double dovetail connection had equal performance at ultimate capacity; however at a deformation of 3 mm, the single dovetail joint was on average 70% stronger than the double dovetail joint. The low strength in the double dovetail joint can be attributed to the weaker wedging effect on the smaller dovetail. Figure 3-4 Crack formation in test specimen 31 Single rounded dovetail connections produced from Western Hemlock do not show any initial alignment behaviour. The load increases right at the beginning and the steep linear behaviour continues until reaching a deformation of approximately 3 mm and until reaching a load that leads to crack development. At that point the load bearing capacity drops followed by a load redistribution process where staple crack development is observed and the further increase in load is associated with larger increase in deformation until brittle failure occurs at ultimate capacity. The single dovetail specimens show a large variability in their ultimate capacity with values ranging between 14 kN and 30 kN. The specimens failed after reaching a deformation between 3 mm and 9 mm (see Figure 3-6). Figure 3-7 shows that the double rounded dovetail connections do not show any initial alignment behaviour either. The load-deformation curve is almost tri-linear. The load increases right at the beginning and a very steep load deformation curve occurs until reaching a deformation of approximately 1 mm. Then the curve levels until reaching a deformation of approximately 4 mm then it gets steeper again until reaching the ultimate capacity. The double dovetail specimens also showed a large variability in their ultimate capacity with values varying between 13 kN and 29 kN. The specimens failed at an average deformation of 6.6 mm. 32 35.0 r 30.0 -25.0 -(kN) 20.0 -•o o 15.0 -_ i 10.0 -5.0 -0.0 * 0.0 2.0 4.0 6.0 Deformation (mm) 8.0 10.0 Figure 3-5 Load-deformation curves for 10 single DD-dovetail specimens including the unloading load loop 35.0 r 10.0 Deformation (mm) Figure 3-6 Load-deformation curves for 10 single DD-dovetail specimens 33 35.0 r 30.0 -TJ 0.0 2.0 4.0 6.0 8.0 10.0 Deformatioon (mm) Figure 3-7 Load-deformation curves for 10 double DD-dovetail specimens The load-deformation curve for the single DWD dovetail specimens show that the load increases right at the beginning and the steep behaviour continues until reaching a load that leads to crack development. At that point the load bearing capacity drops followed by a load redistribution process where staple crack development is observed and the further increase in load is associated with larger increase in deformation until brittle failure occurs at ultimate capacity. The single dovetail specimens failed at an average ultimate capacity of 27 kN with an average maximum deformation of 8.7 mm (see Figure 3- 8). Double rounded dovetail connections show initial alignment behaviour due to imperfect seating of the tenon into the mortise and specimen surface roughness. The connections behave linearly between deformations of 1 and 5 mm. The load increases at an average rate of approximately 1.0 kN/mm until reaching a deformation between 5 and 10 mm, then the load increases at a much higher rate. Testing was halted for all the specimens before reaching failure because the joint achieved large deformations, and the 34 specimens were resting on the testing apparatus causing crushing. The D R D C specimens had an average ultimate capacity of 20.2 k N reaching an average deformation of 9.7 mm (see Figure 3-9). 35.0 r o 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Deformation (mm) Figure 3-8 Load-deformation curves for 5 single DWD-dovetai l specimens 35 Figure 3-10 shows that the load-deformation curve for the single W D specimens is almost linear where the load increases at a rate of 1.25 kN/mm. The specimens exhibit large variability in their ultimate capacity at an average value of 18.4 k N with a standard deviation of 7.1 k N . A s for the double dovetail specimens produced in the dry condition and tested in the wet condition, they exhibit less variability in their ultimate capacity with an average value of 22.8 k N and a standard deviation of 3.2 k N . The load-deformation curve is almost tri-linear, where the load increases at a low rate until reaching a deformation of 4mm for some specimens and 7 mm for others, and then increases at a higher rate until brittle failure occurs (see Figure 3- 11). The double dovetail specimens show more consistency in their behaviour as opposed to the single dovetail specimens. The single and double dovetail specimens failed at an average deformation of 6.2 mm and 9.0 mm, respectively. 35.0 30.0 25.0 i 20.0 8 15.0 10.0 5.0 0.0 0.0 2.0 4.0 6.0 8.0 10.0 12.0 Deformation (mm) Figure 3-10 Load-deformation curves for 5 single WD-dovetai l specimens 36 30.0 r 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Deformation (mm) Figure 3-11 Load-deformation curves for 5 double WD-dovetail specimens Single rounded dovetail connections produced and tested in the wet conditions show initial alignment behaviour for some specimens. The load-deformation curve is linear. The load increases at an average rate of 2 kN/mm, reaching a deformation of 5mm leading to crack development. The load then increases at the same rate to ultimate capacity where brittle failure occurs. The single dovetail specimens had an average ultimate capacity of 20.5 k N with a standard deviation of 7.4 k N . The specimens had an average maximum deformation of 9.0 k N (see Figure 3- 12). As for the double dovetail specimens, they do not show any initial alignment behaviour. Figure 3-13 shows that the load-deformation curve is almost linear. Cracks in the specimens are developed at a deformation of 5 mm and the load increases until brittle failure occurs. The specimens exhibit consistent behaviour with an average ultimate capacity of 21.9 k N with a standard deviation of 5.2 k N , and an average maximum deformation of 9.2 k N . The single and double dovetail specimens produced and tested in the wet condition exhibit similar behaviour. 37 35.0 r 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Deformation (mm) Figure 3-12 Load-deformation curves for 5 single WW-dovetai l specimens 35.0 r T3 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 Deformation (mm) Figure 3-13 Load-deformation curves for 5 double WW-dovetai l specimens Analysis of variance was conducted to test the hypothesis that climatic conditioning has an effect on the structural performance of rounded dovetail connections. 38 A n A N O V A was carried out for the ultimate capacity of the connections, capacity at both 3 -mm and 5 -mm deformations, shear stiffness, adjusted shear stiffness and maximum deformation. Test results show that different climatic conditions have no significant effect on the maximum capacity of the connection. The single and double dovetail specimens that were manufactured and tested in the dry conditions showed equal performance. Figure 3 -14 shows that the double dovetail performed better and showed less variability in behaviour at ultimate capacity than the single dovetail in all climatic conditioning except for the D W D condition. DD 1DT DD 2DT DWD 1DT DWD 2DT WD 1DT WD 2DT WW 1DT WW 2DT Figure 3 - 1 4 Comparison for ultimate capacities for different climatic conditions 3 9 Table 3-3 A N O V A results for climatic conditions effect on the connection ultimate capacity Source DF Sum of Squares Mean Square F value Pr>F Model 7 284.69 40.67 1.33 0.2613 Conditioning 3 103.30 34.43 1.12 0.3501 Dovetail 1 1.63 1.63 0.05 0.8189 Interaction 3 180.37 60.12 1.96 0.1342 Error 42 1286.24 30.62 Total 49 1856.22 For the capacity at a deformation of 5mm, the specimens subjected to different climatic conditions exhibited a lower capacity than those manufactured and tested in the dry condition. In addition, the single dovetail specimens displayed a higher capacity than the double dovetail joint for all the climatic conditions except for the wet-wet condition. Furthermore, the specimens subjected to different climatic conditions show larger deformations. Since the interaction of climatic conditions and dovetail types has a significant effect on the performance of the connection, the two factors cannot be interpreted separately (see Figure 3-15). 40 0 - J -DD DD DWD DWD WD WD WW WW 1DT 2DT 1DT 2DT 1DT 2DT 1DT 2DT Figure 3-15 Capacities at 5 mm deformation for different climatic conditions Table 3-4 A N O V A results for climatic conditions effect on the connection capacity at 5mm deformation Source DF Sum of Squares Mean Square F value Pr>F Model 7 655.02 93.57 7.16 <0001 Conditioning 3 170.35 56.78 4.34 0.0094 Dovetail 1 245.18 245.18 18.75 <0001 Interaction 3 241.22 80.41 6.15 0.0015 Error 42 549.28 13.08 Total 49 1204.30 The connections performance at 3 mm deformation is similar to that at 5 mm deformation. Since the interaction between climatic conditions and dovetail types is significant, the two factors cannot be interpreted independently. However, Figure 3-16 clearly shows that the connection manufactured and tested in the dry condition displayed a higher capacity than those exposed to varying climatic conditions. In addition, similar observations are made regarding the performance of the single dovetail joint compared to 41 that of the double joint at 5 mm deformation; at 3 mm deformation the single dovetail specimens displayed a higher capacity than the double dovetail joint for all the climatic conditions except for the wet-wet condition. 20 -0 -DD DD DWD DWD WD WD WW WW 1DT 2DT 1DT 2DT 1DT 2DT 1DT 2DT Figure 3-16 Capacities at 3 mm deformation for different climatic conditions Table 3- 5 A N O V A results for climatic conditions effect on the connection capacity at 3 mm deformation Source DF Sum of Squares Mean Square F value Pr>F Model 7 569.26 81.32 13.21 <0001 Conditioning 3 156.26 52.09 8.46 0.0002 Dovetail 1 258.64 258.64 42.01 <0001 Interaction 3 93.16 31.05 5.04 0.0045 Error 42 258.56 6.16 Total 49 1335.89 42 A N O V A results indicate that the interaction between climatic conditions and dovetail types has a significant effect on the shear stiffness of the connections. However, the effect of both conditioning and dovetail types cannot be interpreted separately. Figure 3-17 shows that the single dovetail joint exhibited higher shear stiffness than that of double dovetail joint. The calculated shear stiffness for the single and double dovetail joint is 4.1 kN/mm and 2.6 kN/mm, respectively. C (kN/mm) 15 12 9 6 3 0 IT — DD DD DWD DWD WD WD WW WW 1DT 2DT 1DT 2DT 1DT 2DT 1DT 2DT Figure 3-17 Shear stiffness for different climatic conditions Table 3 -6 A N O V A results for climatic conditions effect on the connection shear stiffness Source DF Sum of Squares Mean Square F value Pr>F Model 7 64.63 9.23 8.48 <0001 Conditioning 3 28.54 9.51 8.73 0.0001 Dovetail 1 24.10 24.10 22.12 <0001 Interaction 3 9.15 3.05 2.80 0.0516 Error 42 45.75 1.09 Total 49 110.38 43 The A N O V A indicates that the interaction between climatic conditioning and dovetail types has a significant effect on the adjusted shear stiffness of the connections. Figure 3-18 shows that the single dovetail specimens in all conditions exhibited higher adjusted shear stiffness than those of the double dovetail specimens. 9 T 0 J -DD DD DWD DWD WD WD WW WW 1DT 2DT 1DT 2DT 1DT 2DT 1DT 2DT Figure 3-18 Adjusted shear stiffness for different climatic conditions Table 3-7 A N O V A results for climatic conditions effect on the connection adjusted shear stiffness Source DF Sum of Squares Mean Square F value Pr>F Model 7 112.85 16.12 15.90 <0001 Conditioning 3 14.69 4.90 4.83 0.0056 Dovetail 1 62.29 62.29 61.45 <0001 Interaction 3 17.82 5.94 5.86 0.0019 Error 42 42.58 1.01 Total 49 155.43 44 A N O V A results show that the interaction of conditioning and dovetail types has a significant effect on the maximum deformation of the connection. Although the dovetail type has no effect on the maximum deformation, Figure 3- 19 shows that the double dovetail joints failed at larger deformation compared to that of single dovetail joints. The maximum deformation for the single and double dovetail joint is 6.6 mm and 8.4 mm respectively. In addition, the connection manufactured and tested in the dry condition exhibit a lower deformation at failure compared to those exposed to the variable climatic conditions. The large deformation that the wet specimens exhibited is largely attributed to the stresses induced by moisture. Table 3- 9 shows a summary of the rest results of the climatic conditioning tests. 15 T 0 J -DD DD DWD DWD WD WD WW WW 1DT 2DT 1DT 2DT 1DT 2DT 1DT 2DT Figure 3-19 Max imum deformation for different climatic conditions 45 Table 3 - 8 A N O V A results for climatic conditions effect on the connection maximum deformation Source DF Sum of Squares Mean Square F value Pr>F Model 7 187.92 26.85 5.62 0.0001 Conditioning 3 129.34 43.11 9.02 <0001 Dovetail 1 34.53 34.53 7.22 0.0103 Interaction 3 18.45 6.15 1.29 0.2915 Error 42 200.75 4.78 Total 49 570.99 46 Table 3- 9 Summary of test results of the climatic conditioning tests Control Control DWD DWD WD WD WW IDT 2DT IDT 2DT IDT 2DT IDT Mean 19.89 19.97 27.42 20.20 18.44 22.77 20.54 Fu„ (kN) M a x 29.76 28.75 33.31 25.83 30.43 26.06 31.58 M i n 13.63 12.71 21.81 17.04 12.18 17.79 12.46 StDev 5.61 5.50 5.14 3.80 7.06 3.19 7.40 Mean 18.21 14.68 18.42 6.41 14.58 9.90 13.21 F@5mm M a x 22.57 19.61 24.12 10.99 17.14 16.34 21.10 (kN) M i n 13.63 10.72 14.46 4.55 12.18 5.32 10.40 StDev 3.22 3.01 4.28 2.67 2.29 5.63 4.47 Mean 14.87 8.61 10.77 4.21 11.88 5.28 8.40 F@3mm M a x 18.00 11.29 16.52 7.63 15.01 7.41 13.90 (kN) M i n 9.63 5.25 5.10 2.91 10.11 3.62 5.34 StDev 2.73 1.68 4.44 1.94 1.87 1.37 3.31 Mean 5.00 3.46 3.56 1.36 3.94 1.73 2.78 M a x 5.98 6.25 5.48 2.50 5.00 2.45 4.62 C (kN/mm) M i n 3.21 1.73 1.68 0.95 3.36 1.18 1.75 StDev 0.92 1.48 1.48 0.64 0.63 0.46 1.11 Mean 5.39 2.05 4.50 1.03 3.46 1.27 3.05 a^dj (kN/mm) M a x 7.93 3.08 5.86 1.73 5.61 2.28 4.87 M i n 4.05 1.16 2.35 0.74 2.30 0.75 2.17 StDev 1.32 0.61 1.43 0.41 1.31 0.60 1.10 Mean 4.62 6.62 8.71 9.74 6.17 9.95 9.04 dmax (mm) M a x 8.85 8.72 13.60 11.10 9.70 11.95 11.62 M i n 2.82 4.37 6.15 8.07 3.34 8.54 6.12 StDev 2.27 1.48 3.14 1.26 2.71 1.32 2.36 47 CHAPTER 4. EFFECT OF MANUFACTURING TOLERANCES 4.1 Introduction Varying connection tolerances have a significant impact on the load bearing performance of many wood-to-wood connections. Therefore, reducing variations in the joint geometry and improving the quality of workmanship directly affects the strength and performance of these joints. Eliminating the variability of manual labour in the manufacture of traditional joinery, CNC-processing technology provides repetitive, identical, very accurate and precise joint geometry. However, the individual set-up and the service condition of the processing machinery (machine and tooling) as well as the skills of the operator may affect the cut-quality and geometry, hence, the overall performance of the joint. The processability of timber members and the quality of the fabricated joinery directly depend on the material. Timbers with undesired wood characteristics such as large knots and major cracks, warp and twist are not suitable for CNC-processing. High-grade wood products such as kiln dried timber, Glulam and Engineered Structural Composite Lumber (SCL) provide desired characteristics, a low moisture content (10-16%MC) and i f not exposed to moisture fluctuations, dimensional stability. Taking into account these considerations, the following questions are presented: 48 4.2 Statistical Experimental Design Four different tolerance conditions were compared to the control geometry for the single dovetail configuration. The experimental layout for the manufacturing tolerances study is a one-way analysis of variance. For the analysis the only factor is the tolerance conditions with five levels (control and four variations). The dovetail geometry of the control specimen was chosen as: width b l = 50 mm, depth t = 28 mm, dovetail angle a = 15, and flange angle p = 15. The connection back-cut B C , which is the difference between member height h and dovetail height h i was chosen as 65 mm. Therefore the dovetail height was h i = 123 mm. The specimens with a 1 mm gap and a 2 mm gap were produced with a tenon smaller than the mortise by 1 mm and 2 mm, respectively, on each side of the dovetail flange. A total of 30 specimens (a specimen consisting of a main beam and a joist) were used in the manufacturing tolerances tests. Ten control specimens (DD series) were manufactured at low machining speed and without a gap. Five specimens each were manufactured with a 1 -mm gap and a 2-mm gap, and at medium and high speed. 4.3 Material and Joint Configuration Kiln-dried western hemlock (Tsuga heterophylla) was used in the study. Naturally grown 90-year-old Western Hemlock stands represent much of the emerging timber supply in the B C coastal forest. 49 The member dimensions b and h were chosen as 89 mm and 188 mm, respectively. The length of the main beam was chosen as 600 mm, the length of the joist as 800 mm. The material was conditioned before testing and the moisture content ( M C ) and the apparent density (based on specimen as tested weight and volume) of each specimen were determined. The average M C of the tested specimens was 14.0%, with a standard deviation of 1.9% and extreme values of 10.6% and 17.9%, respectively. The average apparent density of the specimens was 473.9 kg/m 3 with a standard deviation of 37.3 kg/m 3 and extreme values of 409.7 kg/m 3 and 559.6 kg/m 3 , respectively. Table 4- 1 Summary of material properties of main beams used in the tolerances tests Moisture Content (%) Apparent Density (kg/m3) Mean Max Min StDev Mean Max Min StDev Control 1 DT 12.9 14.3 10.6 1.2 474 513 425 32 Control 2DT 13.7 16.3 11 1.5 512 578 449 44 S5 14.9 17 13.2 1.5 465 486 449 17 S9 15.3 17.5 14.1 1.5 481 523 449 30 G1 15.9 17.9 13.2 1.8 491 527 444 41 G2 14.3 16.7 12.1 1.9 521 560 495 27 Table 4- 2 Summary of material properties of joists used in the tolerances tests Moisture Content (%) Apparent Density (kg/m3) Mean Max Min StDev Mean Max Min StDev Control 1 DT 12.3 15 11 1.5 498 560 456 39 Control 2DT 11.7 12 10.9 0.4 441 470 400 27 S5 14.8 17.7 12 2.2 466 483 440 20 S9 14.5 17.1 13.1 1.6 456 480 430 19 G1 14.4 17.4 11.6 2.3 419 433 410 9 G2 13.5 14.1 12.8 0.5 444 463 420 18 50 4.4 Joint Test Procedure The tests conducted for the manufacturing tolerances study were carried out in the same manner as the tests for the climatic conditioning study. The test specimens were mounted on a test apparatus by supporting the main beam on two steel plates o f 100 mm x 100 mm. The main beam was free to rotate, a state that closely simulates the supporting condition in a real structure. Two small steel plates of 50 mm x 50 mm prevented the specimen from moving backward under the applied load. The free end of the joist was simply supported on a steel plate of 100 mm x 100 mm. That steel plate rested on a damper with a spring stiffness o f approximately 1 kN/mm in order to simulate a longer beam. The load was applied at a distance of 350 mm from the joint and distributed onto the joist with a steel plate with a diameter of 100 mm. A hydraulic actuator M T S type was used. The applied load from the actuator and the load at the support (with a load cell) of the free end of the joist were recorded. The vertical movement of the joint was recorded on both sides of the joint, the horizontal movement of the joint, the deflection of the main beam, as wel l as the deflection of the joist at the load application point were recorded with calibrated electronic transducers. 4.5 Results and Analysis Single rounded dovetail connections produced at medium speed do not show any initial alignment behaviour. The specimens exhibit high variability in their behaviour at an average ultimate capacity of 18.1 k N with a standard deviation of 2.5 k N , and a 51 maximum deformation 5.2 k N (see Figure 4- 2). Single rounded dovetail connections produced at high speed show initial alignment behaviour. The load increases at a slow rate where large deformation is accompanied until reaching a deformation between 1 mm and 4 mm, where the load rises at higher rates until brittle failure occurs at ultimate capacity. The specimens had an average ultimate capacity of 18.8 k N with a standard deviation of 5.0 k N . In addition, the specimens failed after reaching a deformation between 3.5 mm and 9 mm (see Figure 4- 3). Figure 4- 1 Load-deformation curves for 10 single dovetail specimens produced at low speed and without a gap 52 25.0 20.0 "O CO o 0.0 1.0 2.0 3.0 4.0 Deformation (mm) 5.0 Figure 4- 2 Load-deformation curves for 5 single dovetail specimens produced at medium speed 30.00 -I 25.00 -20.00 -T3 15.00 -CO O 10.00 -5.00 -0.00 -0.00 2.00 4.00 6.00 Deformation (mm) 8.00 10.00 Figure 4- 3 Load-deformation curves for 5 single dovetail specimens produced at high speed Figure 4- 4 illustrates that the single rounded dovetail connections produced from Western Hemlock with a 1 mm gap do not show any initial alignment behaviour. The 53 load increases right at the beginning and the steep linear behaviour continues until reaching a deformation of approximately 2 mm and until reaching a load that leads to crack development. A t that point the load bearing capacity drops, stable crack development is observed and the load further increases associated with bigger increases in deformation until brittle failure occurs at ultimate capacity. The R D C specimens show a small variability in their ultimate capacity with an average value of 16.2 k N , except for one specimen which had an ultimate capacity of 36.6 k N . The specimens failed after reaching a deformation between 3 mm and 8 mm, with one specimen failing at a deformation of 16.7 mm. Single rounded dovetail connections produced with a 2 mm gap do not show any initial alignment behaviour either. The load picks up right at the beginning and then flattens at a deformation of 1 mm where the load increases at a slow rate, associated with large deformation until brittle failure takes place at ultimate capacity. The R D C specimens showed small variability in their ultimate capacity with an average value of 15.6 k N and a standard deviation of 4.7 k N . The specimens failed at an average deformation of 8.1 mm with a standard deviation of 4.5 mm (see Figure 4- 5). 54 0.0 5.0 10.0 Deformation (mm) 15.0 20.0 Figure 4- 4 Load-deformation curves for 5 single dovetail specimens produced with a 1-mm gap 25.0 r Deformation (mm) Figure 4- 5 Load-deformation curves for 5 single dovetail specimens produced with a 2-mm gap Analysis of variance was conducted to test the hypothesis that the production parameters have an effect on the structural performance of rounded dovetail connections. 55 A N O V A was carried out for the ultimate capacity of the connections, capacity at both 3-mm and 5-mm deformations, shear stiffness, adjusted shear stiffness and maximum deformation. The control specimens were manufactured without an existing gap between the mortise and tenon. The other specimens were produced with either a 1mm gap or a 2mm gap. The aim was to determine whether a tight or a loose joint does affect the overall performance of the connection. Although A N O V A results show that production parameters do not affect the ultimate capacity of the connections (Table 4- 3), Figure 4- 6 indicates that the control specimens exhibited, on average, a 25% higher capacity than those produced at a higher speed and with a gap. o ro Q. co O 40 35 30 25 20 15 10 5 0 Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4- 6 Comparison for ultimate capacities for different production parameters 56 Table 4- 3 A N O V A results for production parameters effect on the connection ultimate capacity Source DF Sum of Squares Mean Square F value Pr>F Model 4 68.87771 17.2194275 0.5 0.7356 Error 25 860.20572 34.4082288 Total 29 929.08343 A N O V A results shows that the effect of manufacturing parameters on the capacity of the connections at 5 mm deformation is statistically significant (a = 0.05), see Table 4- 4. Regarding production speed, the control specimens had a capacity of 18.2 k N , while the specimens produced at medium and high speed had a capacity of 16.0 k N and 13.8 k N , respectively. A s for the comparison of specimens produced with and without a gap, the specimens produced with a 1mm and a 2mm gap had a capacity of 14.0 k N and 12.3 k N , respectively (see Figure 4- 7). 25 T 0 Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4- 7 Comparison for capacities at 5 mm deformation for different production parameters 57 Table 4- 4 A N O V A results for production parameters effect on the connection capacity at 5mm deformation Source DF Sum of Squares Mean Square F value Pr>F Model 4 151.66 37.91 5.24 0.0033 Error 25 180.75 7.23 Total 29 332.41 A N O V A results (Table 4- 5) show that the effect of production parameters on the capacity of the connections at 3mm deformation is statistically significant (a = 0.05). Figure 4- 8 shows that the specimens produced at low speed had a capacity of 14.9 k N , while the specimens manufactured at medium and high speed had a capacity of 11.6 k N and 7.9 k N , respectively. In addition, regarding the effect of gap on the capacity of the connection, the specimens produced with a 1 mm and 2 mm gap had a capacity of 11.7 k N and 10.1 k N , respectively, compared to the control specimens which had a capacity of 14.9 k N . The control specimens exhibited, on average, 45% higher capacity than the other specimens. 58 CO O 5 -0 -Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4- 8 Comparison for capacities at 3 mm deformation for different production parameters Table 4- 5 A N O V A results for production parameters effect on the connection capacity at 3 mm deformation Source DF Sum of Squares Mean Square F value Pr>F Model 4 185.75 46.44 4.66 0.006 Error 25 249.34 9.97 Total 29 435.09 Figure 4- 9 shows that the control specimens (produced at low speed and without a gap) exhibited a higher shear stiffness compared to those specimens manufactured at higher speed and with gaps. The specimens produced at low had a shear stiffness of 5.0 kN/mm, while those produced at medium and high speed had a shear stiffness of 3.9 kN/mm and 2.6 kN/mm, respectively. In addition, the specimens manufactured with a 1mm and 2mm gap had a shear stiffness of 3.9 kN/mm and 3.3 kN/mm, respectively. 59 A N O V A results show that the effect of production parameters on the shear stiffness of the connections is statistically significant (a=0.05), see Table 4- 6. Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4- 9 Shear stiffness for different production parameters Table 4- 6 A N O V A results for production parameter effect on the connection stiffness Source DF Sum of Squares Mean Square F value Pr>F Model 4 21.92 5.48 4.96 0.0044 Error 25 27.64 1.11 Total 29 49.57 • A N O V A results (Table 4- 7) show that the effect of production parameters on the adjusted shear stiffness of the connections is statistically significant (a=0.05). Figure 4-10 illustrates that the control specimens exhibited higher adjusted shear stiffness than those specimens manufactured at higher speeds and with gaps. The specimen produced at low speed had an adjusted shear stiffness of 5.4 kN/mm, while the specimens produced at 60 medium and high speed had a stiffness of 4.1 kN/mm and 2.7 kN/mm, respectively. In addition, the specimens with a 1mm and 2mm gap had a stiffness of 2.9 kN /mm and 2.4 kN/mm, respectively. 9 T 0 Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4- 10 Adjusted shear stiffness for different production parameters Table 4- 7 A N O V A results for production parameter effect on the connection adjusted stiffness Source DF Sum of Squares Mean Square F value Pr>F Model 4 46.50 11.62 8.96 0.0001 Error 25 32.44 1.30 Total 29 78.93 A N O V A results (Table 4- 8) show that the effect of production parameters on the maximum deformation of the connections is not statistically significant. Figure 4- 11 shows that the specimens manufactured at low speed and without a gap failed at a lower deformation than those specimens produced at higher speeds and with gaps. The control 61 specimens exhibited a maximum deformation of 4.6 mm, while the specimens produced at medium and high speed failed at a deformation of 5.2 mm and 6.1 mm, respectively. Furthermore, the specimens produced with a 1mm and 2mm gap failed at an average deformation of 8.2 mm and 8.1 mm, respectively. 20 j 16 -0 Speed 1 Speed 5 Speed 9 Gap 1mm Gap 2mm Figure 4-11 Max imum deformation for different production parameters Table 4- 8 A N O V A results for production parameters effect on the connection maximum deformation Source DF Sum of Squares Mean Square F value Pr>F Model 4 69.51 17.38 1.76 0.1679 Error 25 246.37 9.85 Total 29 315.88 62 Table 4- 9 Summary of test results of the production parameters tests Control IDT Gl G2 S5 S9 Mean 19.89 20.27 15.85 18.08 18.79 Fult(KN) M a x 29.76 36.62 23.00 20.69 26.80 M i n 13.63 12.96 11.33 15.44 13.09 StDev 5.61 9.51 4.71 2.47 5.04 Mean 18.21 13.96 12.30 16.03 13.84 F 5 m m (KN) Max M i n 22.57 13.63 14.85 12.96 13.39 11.33 19.56 11.55 18.97 9.78 StDev 3.22 0.82 0.75 3.12 3.29 Mean 14.87 11.73 10.08 11.64 7.89 F 3 m m (KN) M a x 18.00 11.97 11.21 15.59 17.04 M i n 9.63 11.46 9.22 7.80 1.67 StDev 2.73 0.18 0.76 3.27 5.86 Mean 5.00 3.90 3.33 3.86 2.61 C Max 5.98 3.98 3.71 5.16 5.66 (KN/mm) M i n 3.21 3.81 3.06 2.60 0.54 StDev 0.92 0.06 0.25 1.08 1.95 Mean 5.39 2.85 2.39 4.11 2.67 Cadj M a x 7.93 3.61 3.11 5.25 4.37 (KN/mm) M i n 4.05 2.04 1.75 2.81 0.54 StDev 1.32 0.59 0.64 1.14 1.47 Mean 4.62 8.22 8.14 5.19 6.07 dmax (mm) Max M i n 8.85 2.82 16.74 3.67 13.27 4.29 5.95 3.61 8.87 3.66 StDev 2.27 5.02 4.46 0.97 1.99 CHAPTER 5. CONCLUSIONS AND RECOMMENDATIONS 5.1 Conclusions Single dovetail specimens show inconsistent and highly variable performance throughout the load deformation curve until a deformation is accompanied with crack development. Some specimens exhibit large stiffness initially followed with a levelling off of the load-deformation curve until brittle failure occurs. As for the double dovetail specimens, they show more consistent behaviour with less variability in ultimate capacity and maximum deformation. Contrary to results of previous research at U B C , few double dovetail connections developed two cracks along the two planes with high tension perpendicular to the grain stresses. The main beam was clamped at the ends in the previous study, whereas is this study, it was simply supported. Both single and double dovetail specimens exhibited brittle failure. Both climatic conditioning and manufacturing tolerances do not have a statistically significant effect on the ultimate capacity of either single and double dovetail connections. This can be attributed to the fact that most specimen failure was accompanied with large deformation, causing severe damage in the connections. However, regarding the connection capacity at a deformation of 3mm and 5mm, results concluded that there is statistically significant difference between specimens subjected to varying climatic conditions. It was found that the specimens manufactured and tested in the dry condition outperformed those specimens under different climatic conditions. In addition, the single dovetail joints have higher design capacity, higher stiffness, and smaller deformation at failure than the double dovetail joints, except in W W conditions. 64 In the study of different manufacturing tolerances, it was determined that the specimens produced at low speed and without a gap had a higher capacity than those produced at higher speed and with a gap. L o w cutting speed produces a clean cut and good fit joint. Whereas, high cutting speed causes vibration of the machine resulting in poor quality and misfit of the tenon into the mortise of rounded dovetail joints. Results have shown that manufacturing tolerances had a statistically significant effect on the capacity of the specimens at a deformation of 3mm and 5mm, and on the shear stiffness of the joints. However, it was concluded that there was no statistically significant difference in terms of maximum deformation between specimens produced at low speed and without a gap and the specimens produced at higher speed and with a gap. Based on the results presented in this study, it is recommended that rounded double dovetail joints be manufactured using dry lumber and ensure no changes in moisture content take place. This ensures a good performance of the joint. It is also proposed that the joints be produced at low speed on the cutting machine and without a gap between the tenon and the mortise, to ensure a good quality and good fit joint, resulting in a higher structural performance. 5.2 Recommendations for Future Work Timber used as a structural member exhibits mechano-sorptive behaviour, where it undergoes mechano-sorptive creep as a result of combined long-term loading and moisture content changes. Moisture induced stresses can cause greater deformation in wood than long term loading stresses. Therefore, for future studies, it is important to study the progress of creep in rounded dovetail connections during concurrent moisture 65 changes, since design codes do not specify requirements for long-time effects (creep) on the overall performance of structures built with traditional timber connections (Madsen 1998). 66 BIBLIOGRAPHY Bamford, S. (2003). "The effects of Advanced Wood Processing Technologies on Contemporary Timber Design". B.Sc. Thesis, University of Brit ish Columbia, Vancouver, B C . Benson, T. and Gruber, J . (1980). "Bui ld ing the Timber Frame House, The Revival of a Forgotten Craft". New York: Charles Schribner's Sons. Bobacz, D. (2002). "In CNC-Technik Gefertigte Zimmermannsmaafiige Verbindungsmittel- Untersuchung des Schwalbenschwanzzapfens". Universitat Fi ir Bodenkultur, Wien, Diplomarbeit. Brungraber, R. L. (1985). "Traditional Timber Joinery: A Modern Analysis". Ph.D thesis. Stanford University, Palo Al to, California. Brungraber, R. L. (1992a). "Engineered Tension Joinery". Timber Framing, Journal of the Timber Framers Gui ld , 23:10-12. Brungraber, R. L. (1992b). "Assessing Capacities of Traditional Timber Connections". Wood Design Focus. 3(4), 17-21. Brungraber, R. L. (1999). "How I Relaxed and Learned to Love the Machine". Proceedings o f Timber Framers Gui ld o f North America 1999 Western Conference, Apr i l 9-11, Whistler, Brit ish Columbia. Bulleit, W. M . , Sandberg, B. L., Drewek, M . W. and O'Bryant, T. L. (1999). "Behaviour and Model l ing of Wood-Pegged Timber Frames". A S C E Journal of Structural Engineering, 125(1), 3-9. 67 Bulleit, W . M . , Sandberg, L. B., O'Bryant, T. T. Weaver, D. A . and Pattison, W . E. (1996). "Analysis of Frames with Traditional Timber Connections". Proceedings of International Wood Engineering Conference, Oct. 28-31, New Orleans, L A . Campbell , D., Forsyth, M . , Mi l ler , J . , Parker, R. and Weber, D. (2001). "Heavy Timber Dovetail Connections: Some Preliminary Design Approaches for Engineers". Unpublished Research Report, University of Brit ish Columbia, Vancouver, B C . Chang, W S . and Hsu, M F . (2005). "Mechanical Characteristics of Traditional Go-Dou and Stepped Dovetail Timber Connections in Taiwan (in Chinese)". Taiwan Forest Science Journal, 20:61-71. Chang, W S . , Hsu, M F . and Komatsu, K. (2006). "Rotational Performance of Traditional Nuk i Joints with Gap I: Theory and Verif ication". Wood Science Journal, 52: 58-62. Chappell, S. (1998). " A Timber Framer's Workshop". Fox Maple Press. Brownfield, Maine. Charles, F. W . B. (1984). "Conservation of Timber Buildings," Hutchinson & Co. (Publishers) Ltd., 10-12. Dietsch, P. (2005). "Development of a Finite - Element Model for Parameter Studies of Rounded Dovetail Connection". Engineering Thesis, University of Brit ish Columbia, Vancouver, B C . Drewek, M . W. (1997). "Model l ing the Behaviour of Traditional Timber Frames". M.S . Thesis, Michigan Technological University, Houghton, Michigan. Ell iot, S. and Wallas, E. (1977). "The Timber Framing Book," Housesmiths Press, 1-4 Erikson, R. G . (2003). "Behaviour of Traditional Timber Frame Structures Subjected to Lateral Load" . Ph.D thesis, University of Wyoming, Laramie, Wyoming. 68 Erkison, R. G . and Schmidt, R. J . (2001). "Laterally Loaded Timber Frames: I. One-Story Frame Behaviour". Timber Framing, Journal of the Timber Framers Gui ld , 63:8-11 Erikson, R. G. and Schmidt, R. J . (2002a). "Laterally Loaded Timber Frames: II. Two-Story Frame Behaviour". Timber Framing, Journal of the Timber Framers Gui ld , 64:10-13 Erman, E. (2002). "Timber Joint Design: The Geometric Breakdown Method". Bui lding Research and Information, 30(6), 446-469. Erman, E. (1997). "Demountable Joints for Timber Structures". Ph.D thesis, Middle East Technical University, Ankara. Gerner, M . (1992). "Handwerkliche Holzverbindungen der Zimmerer". Deutsche Verlagsanstalt ( D V A ) , Stuttgart. Goetz, K., Hoor, D., Moehler, K. and Natterer. (1989). "Timber Design and Construction Sourcebook". McGraw H i l l Publishing Company. New York, New York Gorlacher, R. and Kromer, M . (1993). "Tragfahigkeit von Zugbeanspruchten Blattverbindungen in Historischen Holzkonstruktionen". Bauen M i t Ho lz 4/93 Harada, M . , Hayashi, Y . , Hayashi, T., Karube, M . and Ohgama, T. (2005). "Effect of Moisture Content of Members on Mechanical Properties of Timber Joints". Journal of Wood Science. 51,282-285 Heimeshoff, B. and Kohler, N . (1988). "Untersuchungen uber das Tragverhalten von ZimmermannsmaBigen Holzverbindungen". Forschungsbericht Lehrstuhl fur Baukonstruktionen und Holzbau, Technische Universitat Miinchen, 1988 Hewett, Ceci l A . (1980). "Engl ish Historic Carpentry". Phill imore and Co. , Glossary. 69 Hinkes, F. J . (1987). "Experimentelle und rechnerische Untersuchungen zur Ermittlung der Tragfahigkeit von Zapfenverbindungen". Thesis, Universitat Hanover. Hochstrate M . and Kessel, M . H . (2000). "Untersuchungen zum Tragverhalten von CNC.gefertigten Schwalbenschwanz-Zapfenverbindungen". Diplomarbeit, Fachhochschule Hildesheim. Holzner, H. and Spengler, R. (1999). "Entwicklung Eines Nachweisverfahrens zur Bemessung von Speziellen (Maschinell Gefertigten) Zapfenverbindungen. Diplomarbeit, Technische Universitat Munchen, Institut fur Tragwerksbau -Fachgebiet Holzbau Kessel, M . H. (1989). "Rechnergestutztes Konstruieren ( C A D ) im Holzbau - Stand der Technik". Bauen mit Holz , 19: 160-169 Kessel, M . H. and Augustin, R. (1995). "Load Behaviour of Connections with Oak Pegs". Peavy, M . D.; Schmidt, R. J . , trans. Timber Framing. Journal of the Timber Framers Gui ld , 38:6-9 Kessel, M . H. and Augustin, R. (1996). "Load Behaviour of Connections with Oak Pegs II". Peavy, M . D.; Schmidt, R. J . , trans. Timber Framing. Journal of the Timber Framers Gui ld , 39:8-11 K ing , W. S., Yen , R. J . Y . and Yen, A . Y . N . (1996). "Joint Characteristics of Traditional Chinese Wooden Frames". Engineering Structures, 18(8), 635-644 Kreuzinger, H. and Spengler, R. (1999). "Zum Tragverhalten von Maschinel l Abgebundenen Zapfenverbindungen aus Konstruktionsvollholz Zwischen Haupt- und Nebentrager". Report, Technical University Munich 70 Lam, F., Barrett, J . D. and Nakajima, S. (2001). "Engineering Properties of Hem-Fir Used in Japanese Post and Beam Housing". Forest Products Journal, 51(10), 79-87. Lewandoski, J . L. (1992). "Historic Timber-Frame Construction in North Amer ica" . Wood Design Foucs. 3(4), 8-11 Madsen B. (1998). "Reliable Timber Connections". Progress in Structural Engineering and Materials, 1(3), 245-252 Martensson, A . and Thelandersson, S. (1990). "Effect of Moisture and Mechanical Loading on Wooden Materials". Wood Science and Technology, 24: 247-261 Mi l ler , J . F. (2004). "Capacity of Pegged Mortise and Tenon Joints". M.S . Thesis, University of Wyoming, Laramie, Wyoming. Nakajima, S. (2000). "The Effect ofthe Moisture Contents on the Strength and Stiffness Properties of Nai led Joints and Shear Wal ls" . Proceedings of World Conference on Timber Engineering, Whistler, B C . August 2000 O'Connel l , T and Smith, P. (1999). "The North American Timber Framing Housing Industry". Forest Products Journal, 49(1), 36-42. Parisi, M . A . and Piazza. M . (2000). "Mechanics of Plain and Retrofitted Traditional Timber Connections". Journal of Structural Engineering, 126: 1395-1403. Price, C. (1996). "Timber Framing: Age Old Craft Enjoys Revival" . Country Journal, 23(4), 50-54. Sandberg, L. B., Bulleit, W. M . , O'Bryant, T. L., Postlewaite, J . J . and Schaffer, J . J . (1996). "Experimental Evaluation of Traditional Timber Connections". Proceedings of International Timber Engineering Conference, Oct. 28-31, New Orleans, L A . 71 Schmidt, R. J . and Daniels, C. E. (1999). "Design Considerations for Mortise and Tenon Connections". Research Report, University of Wyoming, Laramie, Wyoming. Schmidt, R.J. and MacKay , R.B. (1997). "Timber Frame Tension Joinery". Research Report, University of Wyoming, Laramie, Wyoming. Schmidt, R.J. , MacKay , R.B. and Leu, B.L. (1996). "Design of Joints in Traditional Timber Frame Buildings". Proceedings of International Wood Engineering Conference. University of Wyoming, Laramie, W Y . U S A . Schmidt, R. J . and Scholl , G. F. (2000). "Load Duration and Seasoning Effects on Mortise and Tenon Joints". Research Report, University of Wyoming, Laramie, Wyoming. Shanks, J . and Walker P. (2003). "Tension Testing of Green Oak Connections". Report, University of Bath, England. (Frame 2003, St Fagans). Shanks, J . and Walker P. (2004). "Testing of Traditional Connections in Green Oak Carpentry". Proceedings of 8 t h World Conference on Timber Engineering, Lahti, Finland, June 2004, 183-186. Seo, J . M . , Cho i , I. K. and Lee, J . R. (1999). "Static and Cycl ic Behaviour of Wooden Frames with Tenon Joints under Lateral Load" . A S C E Journal of Structural Engineering, 125(3), 344-349. Sobon, J . A . (2002). "Historic American Timber Joinery - A Graphic Guide". Timber Framers Gui ld Sobon, J . and Schroeder, R. (1984). "Timber Frame Construction," Garden Way Publishing, 3-15. 72 Steiniger, M . (2004). "Preliminary Test Series Examining Strength and Failure properties of Modi f ied Rounded Dovetail Connections Using LSL-Timberstrand and Ki ln-dried Hemlock". Thesis Proposal, University of Brit ish Columbia, Vancouver, B C . Tannert, T. and Lam, F. (2006a). "Structural Performance of Value-Added Bui lding Components". Project No. : U B C 4 Final Report to the Value to Wood Program, funded by Natural Resources Canada. Department of Wood Science. Faculty of Forestry. University of B .C . Tannert, T. and Lam, F. (2006b). "Geometry parameters of Rounded Dovetail Connections". Proceedings of 9th Wor ld Conference on Timber Engineering Aug 6-10 2006 Portland Or. ( C D - R O M Proceeding). Technische Universitat Miinchen (1999). "Untersuchungsbericht: Zum Tagverhalten von Maschinel l Abgebundenen Zapfenverbindungen aus Konstruktionsvollholz Zwischen Haupt - Und Nebentrager". Institut fur Tragwerksbau - Fachgebiet Holzbau, T U Miinchen. 73 APPENDIX Table A 1 Material properties and test results for control (DD) specimens Specimen Nicest % MCn/iain % Djoist kg/m3 Dn/iain kg/m3 Fuit kN F@5mm kN F@3mm kN C kN/mm Cadj kN/mm dmax mm 1-DD-1 12.8 453 11.4 463 23.18 20.81 9.63 3.21 4.20 5.54 1-DD-2 13.5 444 11.4 466 14.26 14.26 14.26 4.75 6.68 2.98 1-DD-3 14.3 453 11.8 463 27.95 22.57 17.39 5.78 6.08 8.03 1-DD-4 13.2 425 12 496 21.30 19.84 11.65 3.88 4.35 5.48 1-DD-5 13.6 449 11 540 18.75 18.75 17.49 5.83 5.57 3.36 1-DD-6 13.9 499 15 496 16.91 16.91 16.91 5.83 4.21 2.90 1-DD-7 13.5 499 11 483 13.63 13.63 13.63 4.81 4.05 2.82 1-DD-8 12.1 509 13.9 560 18.27 18.27 18.00 5.98 7.93 3.13 1-DD-9 11.4 513 14.2 553 14.93 14.93 14.60 4.85 4.60 3.08 1-DD-10 10.6 495 11.7 456 29.76 22.17 15.13 5.04 6.20 8.85 2-DD-1 13.9 527 11.2 450 23.94 18.39 11.29 3.74 2.12 6.43 2-DD-2 16.3 564 11.8 470 27.85 19.61 9.50 3.16 3.08 6.75 2-DD-3 14.4 486 11.9 466 12.71 12.71 9.29 6.25 3.07 4.37 2-DD-4 11 578 12 453 28.75 17.59 8.95 5.89 1.96 8.72 2-DD-5 12.6 523 11.7 403 22.41 14.17 8.60 2.86 2.03 7.90 2-DD-6 15.4 532 11.9 400 18.27 10.94 6.42 2.12 1.57 7.83 2-DD-7 14.4 458 12 410 18.49 10.72 5.25 1.73 1.16 8.15 2-DD-8 13.8 449 11.8 440 15.63 14.76 8.68 2.88 1.79 5.18 2-DD-9 12.7 527 11.8 466 16.41 13.65 9.53 3.17 2.06 5.57 2-DD-10 12.7 472 10.9 456 15.23 14.21 8.55 2.84 1.67 5.34 74 Table A 2 Material properties for D W D series specimens before and after conditioning Before Conditioning After Conditioning Specimen MC J OiSt % MCn/iain % D Joist kg/m3 Diviain kg/m3 MC J o i s t % MCn/iain % Djoist kg/m3 DlVlain kg/m3 1-DWD-1 16.4 17.5 430 467 20.8 21.0 480 523 1-DWD-2 15.6 18.0 480 425 20.8 24.5 523 486 1-DWD-3 15.9 15.6 493 421 19.5 19.3 533 481 1-DWD-4 17.7 17.7 553 467 22.8 20.0 606 518 1-DWD-5 21.0 16.3 546 425 22.5 22.3 600 486 2-DWD-1 16.7 16.1 593 509 20.0 18.0 640 555 2-DWD-2 18.6 16.4 576 481 18.5 18.8 616 527 2-DWD-3 20.8 16.6 563 513 23.1 18.3 593 555 2-DWD-4 19.3 15.4 563 495 21.3 18.0 600 541 2-DWD-5 16.2 15.2 443 495 19.0 18.0 496 550 75 Table A 3 Material properties and test results for D W D series specimens before testing Specimen MCjoist M C M a i n Djoist D|«iain Fuit F@5mm F@3mm C Cadj dmax % % kg/m3 kg/m3 kN kN kN kN/mm kN/mm mm 1-DWD-1 11.1 11.3 413 449 22.20 14.46 8.50 2.81 3.93 6.78 1-DWD-2 11.3 11.4 466 407 29.57 24.12 16.52 5.48 5.67 6.15 1-DWD-3 11.5 11 473 398 21.81 15.85 5.10 1.68 2.35 6.90 1-DWD-4 13.6 11.1 533 449 33.31 21.84 13.60 4.50 5.86 13.60 1-DWD-5 13.3 10.5 523 402 30.21 15.81 10.11 3.35 4.66 10.11 2-DWD-1 12.8 11 576 486 18.75 10.99 7.63 2.50 1.73 8.91 2-DWD-2 12.5 11.3 553 458 17.09 6.55 3.88 1.18 1.05 9.87 2-DWD-3 12 12.2 530 495 25.83 4.93 3.42 1.10 0.81 11.1 2-DWD-4 12.3 11.4 543 481 22.29 5.05 3.22 1.05 0.80 8.07 2-DWD-5 11.3 10.9 426 476 17.04 4.55 2.91 0.95 0.74 10.73 76 Table A 4 Material properties for WD specimens after conditioning After Conditioning Specimen M C ^ ' M C ™ " D j o i s ! D m i " __ % % kg/m3 kg/m3 1-WD-1 23.5 21.3 490 476 1-WD-2 21.3 19.9 483 509 1-WD-3 21.5 19.5 553 509 1-WD-4 21.9 19.1 563 449 1- WD-5 21.8 21.4 483 536 2- WD-1 203 28TJ 476 476~ 2-WD-2 19.9 20.3 490 495 2-WD-3 20.5 20.4 483 504 2-WD-4 20.3 18.6 493 476 2-WD-5 19.8 23.0 493 504 Table A 5 Material properties and test results for WD series specimens before testing Specimen M C j o i s t % MC M ain % Djoist kg/m3 Diwain kg/m3 Fuit kN F@5mm kN F@3mm kN C kN/mm Cadj kN/mm dmax mm 1-WD-1 11.6 10.9 443 435 16.91 16.91 15.01 5.00 3.76 3.88 1-WD-2 11.6 10.6 433 462 30.43 17.14 11.86 3.95 5.61 9.70 1-WD-3 11.6 10.9 500 462 14.71 13.65 10.86 3.55 2.83 5.87 1-WD-4 11.9 10.7 510 407 17.98 13.04 10.11 3.36 2.30 8.06 1-WD-5 11.3 11.0 430 486 12.18 12.18 11.58 3.82 2.82 3.34 2-WD-1 11.8 11.2 426 435 26.06 15.73 7.41 2.45 2.28 9.91 2-WD-2 11.7 11.0 450 449 21.69 5.32 4.81 1.58 0.93 9.03 2-WD-3 11.2 11.4 443 453 24.23 16.34 5.40 1.76 1.11 8.54 2-WD-4 11.4 11.5 446 435 17.79 5.35 3.62 1.18 0.75 10.33 2-WD-5 11.5 11.0 453 453 24.06 6.78 5.15 1.69 1.25 11.95 78 Table A 6 Material properties and test results for WW series specimens Specimen M C j 0 i S t % MC|\/iain % Djoist kg/rrb Diviain kg/rrb Fuit kN Fsmm kN f"3mm kN C kN/mm Cadj kN/mm dmax mm 1-WW-1 20.5 24.3 493 499 13.73 8.90 4.36 1.45 1.97 8.28 1-WW-2 20.8 20.5 496 532 19.54 10.05 5.64 1.88 2.67 11.31 1-WW-3 22.8 28.5 456 638 14.72 8.49 5.68 1.89 2.02 11.62 1-WW-4 23.3 20.8 446 518 10.19 9.22 7.20 2.40 1.76 6.12 1-WW-5 21.5 22.8 593 509 25.84 18.51 11.37 3.79 3.98 7.89 2-WW-1 23.5 24.5 550 583 15.21 11.92 7.69 2.56 2.38 6.42 2-WW-2 23.0 22.8 553 583 24.83 16.88 8.76 2.92 3.05 9.28 2-WW-3 20.5 19.8 480 490 18.58 10.20 6.16 2.05 2.09 12.58 2-WW-4 22.0 24.8 443 583 16.76 10.32 6.59 2.2 1.74 11.6 2-WW-5 20.3 21.8 496 620 14.08 11.53 6.67 2.22 1.84 6.93 79 Table A 7 Material properties and test results for S5 and S9 series specimens Specimen MCjoist MCn/iain Djoist Dr/lain Fuit Fsmrn l"3mm C Cadj dmax % % kg/ni3 kg/m3 kN kN kN kN/mm kN/mm mm 1-S5-1 15.1 13.7 476 458 16.57 12.25 7.35 2.45 2.42 5.89 1-S5-2 13.4 13.2 450 453 16.93 15.98 12.73 4.24 4.28 5.54 1-S5-3 16.1 17.0 480 449 12.63 9.45 6.38 2.13 2.30 5.95 1-S5-4 17.7 14.8 483 486 15.00 15.00 11.40 3.80 3.94 4.95 1-S5-5 12.0 15.7 440 481 12.83 12.83 9.72 3.24 3.87 3.61 1-S9-1 13.5 16.0 480 481 13.58 11.20 5.77 1.92 2.42 3.85 1-S9-2 13.1 14.1 446 523 15.52 15.52 13.94 4.65 3.76 3.66 1-S9-3 13.8 14.3 460 495 10.71 10.71 7.57 2.52 2.85 4.94 1-S9-4 14.8 14.4 430 458 21.93 8.00 1.37 0.46 0.45 8.87 1-S9-5 17.1 17.5 466 449 15.13 11.14 3.60 1.20 1.65 7.01 8 0 Table A 8 Material properties and test results for Gl and G2 series specimens S p e c i m e n MCjoist % MCn/iain % Djoist kg/m3 DiVlain k g / m 3 Fuit kN Fsmm kN F3mm kN C k N / m m Cadj kN /mm dmax m m 1-G1-1 13.5 13.2 413 444 10.60 10.60 9.75 2.10 3.25 3.67 1-G1-2 16.0 16.0 433 449 13.24 11.64 9.56 1.66 3.19 7.03 1-G1-3 13.5 15.2 423 513 29.96 10.80 9.36 2.97 3.12 16.97 1-G1-4 17.4 17.2 410 523 12.53 11.87 9.58 2.55 3.19 5.76 1-G1-5 11.6 17.9 416 527 16.60 12.13 9.58 2.40 3.19 7.92 1-G2-1 13.1 13.6 463 527 11.86 10.91 8.38 2.39 2.79 5.95 1-G2-2 13.9 16.7 460 560 14.67 9.98 8.17 1.94 2.72 13.27 1-G2-3 14.1 15.7 420 527 18.82 9.89 7.49 2.53 2.50 12.71 1-G2-4 12.8 12.1 436 495 9.27 9.27 7.84 1.43 2.61 4.52 1-G2-5 13.8 13.3 443 495 10.23 10.23 9.15 1.42 3.05 4.29 81 

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.831.1-0074932/manifest

Comment

Related Items