Experimental and numerical analyses of rounded dovetail timber connections (RDC) under fire conditions
Graphical abstract
Introduction
The fire performance of a timber structure is largely influenced by the behaviour of its connections [1]. The size of the cross section of a timber element decreases gradually under fire conditions. Eventually, this loss leads to the collapse of the element. The time between the ignition and the collapse of the element supporting an external load is defined as the load-bearing capacity (R) of the element [2] (see Table1).
The European timber construction code (Eurocode 5, part 1.2 [3]) provides methodologies to design timber connections able to withstand fire conditions. However, timber-to-timber carpentry connections are not included in this code. These connections join timber elements by cutting and fitting them, without nails, screws or bolts. The forces are transmitted from one piece to another through mortises and tenons, or notches and pins. The axial forces are transmitted through compressions and tangential forces [4].
This kind of connections were very time consuming and expensive, and consequently rarely used in building industry. In recent years, the use of carpentry connections has enjoyed a comeback due to advanced software packages and techniques to design and manufacture them. These include Computer Aided Design or Computer Aided Manufacturing (CAD/CAM) and Computer Numeric Control (CNC) [5].
The rounded dovetail connection (RDC) is a carpentry connection, particularly used in roof and floor frames. It transmits loads from secondary structural elements, such as joists, to primary structural elements such as beams. The mechanical behaviour of RDCs at ambient temperature is well known, and the critical parameters of its design have been studied [6], [7], [8]. The results show that these connections not only transfer vertical shear forces but also carry load in tension and bending. These works also show that the angle and height of the dovetail flange affect the structural behaviour of RDCs significantly. Furthermore, a probabilistic method was proposed for the improvement of RDC design. Tannert [9] conducted a series of experiments on RDC to study different methods to increase the stiffness of these structural elements. The research shows that stiffness is improved by increasing the size of the tenon or by reinforcing the joints.
Connections with dowel-type fasteners are commonly used in timber structures, and their mechanical behaviour under a fire event have been studied experimentally since the late 1970s and early 1980s [10]. In the late 1990s, fire tests on three-member timber-to-timber and steel-to-timber (with an internal steel plate) connections, and exposed to fire on all sides were performed [11], [12], [13]. These results were included in EN 1995-1-2:2004. Fire tests on other connections typologies were carried out in the 2000s by [14] and in the 2010s by [15], [16], [17]. However, the research into the performance of carpentry-type timber connections under fire conditions is still limited. There is little work on this topic, and the one that has been made, is all about dovetail connections [18], [19], [20]. No research about any other kind of carpentry-type timber connection has been found in scientific literature. Racher et al. [18] studied dovetail connections under fire conditions. All but one of the tests exceed 15 min of fire exposure. Zhang et al. [19] conducted several experimental tests on straight-line dovetail joints under fire conditions. The experimental tests include protected and unprotected dovetail joints. In both cases, a gap between the tenon and the mortise was identified. The results show that the existing gap had a negligible effect on heat transfer. Furthermore, the fire-retardant coating improves the performance and significantly increases the fire protection.
These previous works do not specify the compliance with the Eurocode 5 [3] requirements for timber elements under fire conditions (R).
Experimental tests under fire conditions are extremely expensive and complicated. Numerical analysis can be used to study different geometrical joints and thermo-mechanical parameters. In this context, Regueira et al. [20] developed a finite element (FE) model to predict the mechanical performance of RDC under fire conditions. This numerical analysis, which is based on the software ANSYS, was validated using experimental results, leading to two conclusions. Firstly, the numerical results show that the temperature on the sides of the joint is influenced by its geometry. Secondly, the performance of the RDC does not meet the R30 performance criteria for fire resistance. Zhang et al. [21] developed a 3D FE model to simulate the thermo-mechanical behaviour of dovetail joints under fire conditions. The approach was validated using experimental data. The gap between the mortise and tenon is not considered in the numerical model and there is a discrepancy regarding numerical and experimental results at the beginning of tests. Despite this, there is good agreement with the experimental results.
In this research work, the load-bearing capacity of a timber beam connected to a joist using an RDC is experimentally studied under fire conditions. Then, numerical simulations using the finite element method (FEM) are developed. Firstly, two thermal models are developed using different thermal boundary conditions to study the self-protection performance of the connection. Then coupled thermal-structural models are studied and compared with the experimental results
Section snippets
Experimental setup
This research work studies the structural behaviour of an RDC made of common spruce timber (Picea abies) under fire conditions. The hygrothermal conditions of the specimens were controlled before the tests, and they had a moisture content of 13.2% and its density had a value of 481 kg/m3. The density was measured weighing the specimens using a scale. The beams and joists volumes were obtained measuring their dimensions using a measuring tape. The tenon and mortise volumes which were determined
Finite element model
A finite element analysis (FEA) was carried out using the software ANSYS [28], which has been proved to be a valid commercial finite element software package to model timber connections under fire conditions [29]. The numerical analysis coupled a transient thermal analysis with a static structural analysis. This methodology was successfully used to simulate the behaviour of a timber connection under fire conditions [30]. In this work, the thermal performance of the timber connection is studied
Near the mortise-tenon zone
Fig. 17 compares the evolution of temperature in models FEM A and FEM B with the experimental results. The temperature in model A for virtual probes 1, 2 and 3 is lower than in the experimental tests. However, the evolution of temperature in model B has a similar thermal behaviour to the experimental results. Model FEM B is closer to the experimental results than FEM A.
The centre of the joist
The comparison between models FEM A and FEM B shows that the boundary conditions in the connection have no influence on the
Conclusions
This research work presents an experimental and numerical study of the thermal-structural behaviour of timber RDC under fire conditions. Experimental results show the mechanical failure before fifteen minutes of fire exposure.
The main experimental conclusions are summarized:
- •
The timber RDC does not meet R30 criterion. The bearing capacity of the RDC was maintained for 720 s (12 min).
- •
The charring causes loss of material in the connection, which in turn causes the sliding of the RDC of the joist
CRediT authorship contribution statement
Rubén Regueira: Methodology, Conceptualization, Formal analysis. Juan Enrique Martínez-Martínez: Writing - review & editing, Visualization. Mar Alonso-Martínez: Data curation, Writing - original draft. Felipe Pedro Álvarez Rabanal: Validation, Investigation, Resources. Manuel Guaita: Supervision. Juan José del Coz Díaz: Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors of the research presented in this paper acknowledge the financial support provided by the spanish Ministry of Science, Innovation and Universities through the National Project PGC2018-098459-B-I00 and by the Gobierno del Principado de Asturias and FICYT under Research Project GRUPIN-IDI/2018/000221, both co-financed with FEDER funds. Furthermore, authors also thank the manufacturer Maderas Besteiro S.L. for the timber elements provided and finally, to Swanson Analysis Inc. for the
References (36)
Improved performance of reinforced rounded dovetail joints
Constr Build Mater
(2016)- et al.
Numerical simulation of the fire behaviour of timber dovetail connections
Fire Saf J
(2018) - et al.
Thermo-mechanical behaviour of dovetail timber joints under fire exposure
Fire Saf J
(2019) - et al.
Comparative study of LightWeight and Normal Concrete composite slabs behaviour under fire conditions
Eng Struct
(2020) - et al.
Fire design of steel-to-timber dowelled connections
Eng Struct
(2010) - Frangi A, Palma P, Hugi E, Cachim P, Cruz H. Fire resistance tests on beam-to-column shear connections; 2014....
- Science Technical Research Institute of SwedenScience Technical Research Institute of Sweden, Östman B, Winter S,...
- AENOR (Asociación Española de Normalización y Certificación). Eurocode 5: Design of timber structures - Part 1-2:...
- et al.
Estructuras de madera
Uniones.
(2015) - Regueira R. Aplicación de métodos numéricos al análisis de estructuras de madera en situación de incendio. Santiago de...
Structural performance of rounded dovetail connections
Structural performance of rounded dovetail connections: experimental and numerical investigations
Eur J Wood Wood Prod
Probabilistic models for structural performance of rounded dovetail joints
J Struct Eng (United States)
Load-bearing capacity of nailed joints exposed to fire
Fire Mater
Cited by (2)
Experimental study and finite element analysis on residual carrying capacity of CLT wall-floor angle bracket connections after fire
2022, Construction and Building MaterialsCitation Excerpt :Experimental and numerical studies of the fire-resistance limit of timber joints subject to static loads conducted by Laplanche [22] showed that a smaller load ratio from 0.3 to 0.1 improved the fire resistance limit from 20 min to 25 min. Regueira et al. [23] conducted tests and finite element analyses of the fire resistance of rounded dovetail connections, showing that this kind of connection joint lacks self-protection under fire conditions and does not achieve R30 fire resistance rating (Eurocode 4). In practice, it is necessary to fill the joint gap with fireproof materials to ensure the safety of the joint.
CT imaging of Lindner tested, burnt timber
2024, Journal of Thermal Analysis and Calorimetry