Following the Bad Reichenhall ice-arena collapse, numerous expertises on the structural safety of wide-span timber structures were carried out at the Chair of Timber Structures and Building Construction. It became evident that inadequate structural design and detailing as well as inadequate manufacturing principles were the main reasons for observed failures. The design and manufacture of connections in wide-span timber structures are still amongst the most challenging tasks for both the structural engineer as well as the executing company. This paper will, on the basis of two exemplary expertises, discuss specific issues in the structural reliability of connections in wide-span timber trusses and give recommendations towards a state-of-the art design of such connections.
Most of the timber used in the Australian built environment is presently for low-rise residential construction. This market share is under constant erosion from competitive systems; therefore, entry into non-traditional sectors would benefit the industry through a wider market portfolio of building type applications, and a higher value product system development.
The project analysed building designs in order to estimate the size and value of the market sector in commercial and high-rise residential buildings; established the major building systems used in these sectors, and why these systems are popular (major attractiveness of current systems) and scoped two current timber systems (Cassette Flooring and Access floors) that have the opportunity to increase timber volumes in these markets.
This thesis deals with experimental tests and methods for strength analysis of glulam beams with holes. Test results and methods for strength analysis available in literature are compiled and discussed. The methods considered comprise both code strength design methods and more general methods for strength analysis.
New strength tests of beams with quadratic holes with rounded corners are presented. The test programme included investigations of four important design para\-meters: material strength class, bending moment to shear force ratio, beam size and hole placement with respect to beam height. One important finding from these tests is the strong beam size influence on the strength. This finding is in line with previous test results found in literature but the beam size effect is however not accounted for in all European timber engineering codes.
A probabilistic fracture mechanics method for strength analysis is presented. The method is based on a combination of Weibull weakest link theory and the mean stress method which is a generalization of linear elastic fracture mechanics. Combining these two methods means that the fracture energy and the stochastic nature of the material properties are taken into account. The probabilistic fracture mechanics method is consistent with Weibull weakest link theory in the sense that the same strength predictions are given by these two methods for an ideally brittle material. The probabilistic fracture mechanics method is also consistent with the mean stress method in the sense that the same strength predictions are given by these two methods for a material with deterministic material properties.
A parameter study of the influence of bending moment to shear force ratio, beam size, hole placement with respect to beam height and relative hole size with respect to beam height is presented for the probabilistic fracture mechanics method.
Strength predictions according to the probabilistic fracture mechanics method is also compared to the present and previous test results found in literature and also to other methods for strength analysis including code design methods. The probabilistic fracture mechanics method shows a good ability to predict strength, with the exception of very small beams.
The main sources of lateral loads on buildings are either strong winds or earthquakes. These lateral forces are resisted by the buildings’ Lateral Load Resisting Systems (LLRSs). Adequate design of these systems is of paramount importance for the structural behaviour in general. Basic procedures for design of buildings subjected to lateral loads are provided in national and international model building codes. Additional lateral load design provisions can be found in national and international material design standards. The seismic and wind design provisions for engineered wood structures in Canada need to be enhanced to be compatible with those available for other materials such as steel and concrete. Such design provisions are of vital importance for ensuring a competitive position of timber structures relative to reinforced concrete and steel structures.
In this project a new design Section on Lateral Load Resisting Systems was drafted and prepared for future implementation in CSA O86, the Canadian Standard for Engineering Design in Wood. The new Section was prepared based on gathering existing research information on the behaviour of various structural systems used in engineered wood construction around the world as well as developing in-house research information by conducting experimental tests and analytical studies on structural systems subjected to lateral loads. This section for the first time tried to link the system behaviour to that of the connections in the system. Although the developed Section could not have been implemented in CSA O86 in its entirety during the latest code cycle that ended in 2008, the information it contains will form the foundation for future development of technical polls for implementation in the upcoming editions of CSA O86.
Some parts of the developed Section were implemented in the 2009 edition of CSA O86 as five separate technical polls. The most important technical poll was the one on Special Seismic Design Considerations for Shearwalls and Diaphragms. This technical poll for the first time in North America includes partial capacity design procedures for wood buildings, and represents a significant step forward towards implementing full capacity-based seismic design procedures for wood structures. Implementation of these design procedures also eliminated most of the confusion and hurdles related to the design of wood-based diaphragms according to 2005 National Building Code of Canada. In other polls, the limit for use of unblocked shearwalls in CSA O86 was raised to 4.8 m, and based on the test results conducted during the project, the NLGA SPS3 fingerjoined studs were allowed to be used as substitutes for regular dimension lumber studs in shearwall applications in engineered buildings in Canada.
With the US being the largest export market for the Canadian forest products industry, participation at code development committees in the field of structural and wood engineering in the US is of paramount importance. As a result of extensive activities during this project, for the first time one of the AF&PA Special Design Provisions for Wind and Seismic includes design values for unblocked shearwalls that were implemented based on FPInnovations’ research results. In addition, the project leader was involved in various aspects related to the NEESWood project in the US, in part of which a full scale six-storey wood-frame building will be tested at the E-Defense shake table in Miki, Japan in July 2009. Apart from being built from lumber and glued-laminated timber provided from Canada, the building will also feature the innovative Midply wood wall system that was also invented in Canada. The tests are expected to provide further technical evidence for increasing the height limits for platform frame construction in North America.
Building construction - Design
Earthquakes, Effect on building construction
Glued joints - Finger
Grading - Lumber
spIn this report, the seismic performance of 6-storey wood frame residential buildings is studied. Two building configurations, a typical wood-frame residential building and a building to be tested under the NEESWood project, were studied. For each building configuration, a four-storey building and a six-storey building were designed to the current (pre-April 6, 2009) 2006 BC Building Code (BCBC) and to the anticipated new requirements in the 2010 National Building Code of Canada (NBCC), resulting in four buildings with different designs. The four-storey building designed to the current 2006 BC Building Code served as the benchmark building representing the performance of current permissible structures with common architectural layouts.
In the design of both four-storey and six-storey buildings, it was assumed that the buildings are located in Vancouver on a site with soil class C. Instead of using the code formula, the fundamental natural period of the buildings was determined based on the actual mass and stiffness of wood-based shearwalls. The base shear and inter-storey drift are determined in accordance with Clauses 220.127.116.11.(3)(d)(iii) and 18.104.22.168.(3)(d)(iv) of BCBC, respectively.
Computer programs DRAIN 3-D and SAPWood were used to evaluate the seismic performance of the buildings. A series of 20 different earthquake records, 14 of the crustal type and 6 of the subcrustal type, were provided by the Earthquake Engineering Research Facility of the University of British Columbia and used in the evaluation. The records were chosen to fit the 2005 NBCC mean PSA and PSV spectra for the city of Vancouver.
For representative buildings designed in accordance with 2006 BCBC, seismic performance with and without gypsum wall board (GWB) is studied. For representative buildings designed in accordance with the 2010 NBCC, the seismic performance with GWB is studied. For the NEESWood building redesigned in accordance with 2010 NBCC, seismic performance without GWB is studied. Ignoring the contribution of GWB would result in a conservative estimate of the seismic performance of the building.
In the 2006 BCBC and 2010 NBCC, the inter-storey drift limit is set at 2.5 % of the storey height for the very rare earthquake event (1 in 2475 year return period). Limiting inter-storey drift is a key parameter for meeting the objective of life safety under a seismic event.
For 4-storey and 6-storey representative wood-frame buildings where only wood-based shearwalls are considered, results from both DRAIN-3D and SAPWood show that none of the maximum inter-storey drifts at any storey under any individual earthquake exceed the 2.5% inter-storey drift limit given in the building code. With DRAIN-3D, the average maximum inter-storey drifts are approximately 1.2% and 1.5% for 4-storey and 6-storey buildings designed with 2006 BCBC, respectively.
For the NEESWood wood-frame building, none of the maximum inter-storey drifts at any storey under any individual earthquake exceed the 2.5% inter-storey drift limit for 4-storey building obtained from SAPWood and 6-storey building obtained from DRAIN-3D and SAPWood. For any 4-storey building analysed with DRAIN-3D, approximately half of the earthquakes resulted in the maximum inter-storey drifts greater than 2.5% inter-storey limit. This is partly due to the assumptions used in Drain-3D model in which the lumped mass at each storey is equally distributed to all the nodes of the floor. As a result, the total weight to counteract the uplift force at the ends of a wall would be much smaller than that anticipated in the design, thus causing hold-downs to yield and large uplift deformations to occur.
Based on the analyses of a representative building and a redesigned NEESWood building situated in the city of Vancouver that subjected the structures to 20 earthquake records, 6-storey wood-frame building is expected to show similar or smaller inter-storey drift than a 4-storey wood-frame building, which is currently deemed acceptable under the current building code.
Building construction - Design
Building construction - Specfications
Earthquakes, Effect on building construction
January 26-28, 2009, San Francisco, California, USA
Fire-resistive wood construction is achieved either by having the structural elements be part of fire-rated assemblies or by using elements of sufficient size that the elements themselves have the required fire-resistance ratings. For exposed structural wood elements, the ratings in the United States are calculated using either the T.T. Lie method or the National Design Specifications (NDS) Method. There is no widely accepted methodology in the United States to determine the fire-resistance rating of an individual structural wood element with the protective membrane directly applied to the exposed surfaces of the element. In these tests, we directly applied one or two layers of 16-mm thick fire-rated gypsum board or 13-mm thick southern pine plywood for the protective membrane to the wood element. The wood elements were Douglas-fir laminated veneer lumber (LVL) specimens and Douglas-fir gluedlaminated specimens that had previously been tested without any protective membrane. The methodology for the tension testing in the horizontal furnace was the same used in the earlier tests. The fire exposure was ASTM E 119. For the seven single-layer gypsum board specimens, the improvements ranged from 25 to 40 min. with an average value of 33 min. For the three double-layer specimens, the improvement in times ranged from 64 to 79 min. with an average value of 72 min. We concluded that times of 30 min. for a single layer of 16-mm Type X gypsum board and at least 60 min. for a double layer of 16-mm Type X gypsum board can be added to the fire rating of an unprotected structural wood element to obtain the rating of the protected element.
There is a discrepancy between the estimated modulus of elasticity (MOE) of glulam based on the dynamic MOE of laminates and measured MOE. The discrepancy is greater for glulam manufactured with mixed species. This study was undertaken to reduce the discrepancy between those MOE values. The error rate of predicting MOE of glulam by the transformed section method, without considering tension and compression modulus differences, was about 30%. To estimate the MOE of glulam more accurately, the differences between compression and tension modulus should be taken into account in the transformed section method. The measured tensile and compressive strain at the center of glulam under a bending load showed the movement of neutral axis toward the tension side of glulam. Therefore, the compression and tension modulus differences for each species should be identified before estimating the MOE of glulam. The prediction of glulam MOE was improved significantly by reflecting the ratio of compression and tension modulus vs dynamic MOE of laminates. The outermost of laminates in the compression side under bending load experienced plastic behavior and failure. This caused the neutral axis to move to the tension side and increased tension stress to cause the glulam to fail abruptly in tension. To improve the bending performance of glulam, reinforcing compression laminates need to be considered.
In October 2007 a series of seismic tests were carried out on a 7-storey building made of cross laminated (XLam) wooden panels in natural scale on a shaking table E-Defence in Japan within the SOFIE project. The paper presents calculation procedure, prediction of dynamic behaviour of the tested structure excited by the earthquake record "Kobe JMA 1995" and comparison between predicted, that means calculated and measured response. Due to blind prediction approach some construction details were not known before dynamic time history response calculation. Therefore some assumptions, engineering judgment and rough static analyses were needed to define all construction parts which were in modelling approach assumed as important and could have had influence on dynamic response of the analyzed structure. The most important assumptions related to the definition of the stiffness and load bearing capacity of mechanical connections, types of anchors and their positions in each floor level, were determined on the basis of static analysis where the structure was loaded with equivalent horizontal seismic forces and then were used in dynamic analysis. A mathematical model was developed in program SAP2000 where modal and time history analyses were carried out. Comparison of calculated and measured results is described and evaluated on the basis of the model assumptions and its simplification.
This project has developed technologies for prefabricated structural systems constructed from engineered wood products for floors and building frames, suitable for buildings up to eight stories in height. The project included the design of a virtual multi-storey timber building, a review of commercial flooring systems, and the development of interim design procedures for timber concrete composite (TCC) floors. Compared with either solid concrete or timber floors, TCC floors provide an excellent balance between increased stiffness, reduced weight, better acoustic separation and good thermal mass.
Outcomes from the project have confirmed TCC floors as a viable alternative to conventional flooring systems. The life cycle analysis of the virtual timber building has highlighted the potential advantages of timber-based building systems for commercial applications. The project also resulted in the formation of the Structural Timber Innovation Company, a research company that will continue to develop timber building systems in non-residential buildings in Australia and New Zealand.
In this paper, some of the results are presented from a series of quasi-static tests on CLT wall panels conducted at FPInnovation-Forintek in Vancouver, BC. CLT wall panels with various configurations and connection details were tested. Wall configurations included single panel walls with three different aspect ratios, multi-panel walls with step joints and different types of screws to connect them, as well as two-storey wall assemblies. Connections for securing the walls to the foundation included: off-the-shelf steel brackets with annular ring nails, spiral nails, and screws; combination of steel brackets and hold-downs; diagonally placed long screws; and custom made brackets with timber rivets. Results showed that CLT walls can have adequate seismic performance when nails or screws are used with the steel brackets. Use of hold-downs with nails on each end of the wall improves its seismic performance. Use of diagonally placed long screws to connect the CLT walls to the floor below is not recommended in high seismic zones due to less ductile wall behaviour. Use of step joints in longer walls can be an effective solution not only to reduce the wall stiffness and thus reduce the seismic input load, but also to improve the wall deformation capabilities. Timber rivets in smaller groups with custom made brackets were found to be effective connectors for CLT wall panels. Further research in this field is needed to further clarify the use of timber rivets in CLT.