Abstract
Through-tenon joints are widely used in ancient timber buildings. To study the influence of the gaps between mortise and tenon shoulder on the seismic performance of through-tenon joints, a 1:3.52 scaled model was constructed and used for low cyclic loading test. Finite element analysis was conducted to study the mechanical behavior of the through-tenon joint. The seismic performance parameters of the model such as moment-rotation hysteresis curves, envelope curves, degradation of rigidity, and energy dissipation capacity were compared. The analyses showed similar changing characteristics, which indicated that the finite element analysis results were reliable. Based on the results, 7 through-tenon joint finite element analysis models with different gaps between mortise and tenon shoulder were established. The seismic performance of each of the through-tenon joints with different gaps between mortise and tenon shoulder were studied. The moment-rotation hysteresis curve of the through-tenon joint had an obvious pinching effect, and the through-tenon joint had good rotational loading capacity and good deformation ability. The peak rotational loading capacity, initial stiffness, and energy dissipation capacity of the joint decreased, while the gap between mortise and tenon shoulder increased.
Download PDF
Full Article
Study on the Seismic Performance of Through-tenon Joints with Pullout Tenon Gaps between Mortise and Tenon Shoulder
Junhong Huan,a,b,c,d Zemeng Sun,a,b Xiaodong Guo,c,d,* Tianyang Chu,a,b Xiaoyi Zhou,a,b Wei Wang,c,d and Yating Yang c,d,e
Through-tenon joints are widely used in ancient timber buildings. To study the influence of the gaps between mortise and tenon shoulder on the seismic performance of through-tenon joints, a 1:3.52 scaled model was constructed and used for low cyclic loading test. Finite element analysis was conducted to study the mechanical behavior of the through-tenon joint. The seismic performance parameters of the model such as moment-rotation hysteresis curves, envelope curves, degradation of rigidity, and energy dissipation capacity were compared. The analyses showed similar changing characteristics, which indicated that the finite element analysis results were reliable. Based on the results, 7 through-tenon joint finite element analysis models with different gaps between mortise and tenon shoulder were established. The seismic performance of each of the through-tenon joints with different gaps between mortise and tenon shoulder were studied. The moment-rotation hysteresis curve of the through-tenon joint had an obvious pinching effect, and the through-tenon joint had good rotational loading capacity and good deformation ability. The peak rotational loading capacity, initial stiffness, and energy dissipation capacity of the joint decreased, while the gap between mortise and tenon shoulder increased.
DOI: 10.15376/biores.19.1.322-344
Keywords: Finite element analysis; Through-tenon joint; Seismic performance; Gap between mortise and tenon shoulder
Contact information: a: School of Civil Engineering, Shijiazhuang Tiedao University, Shijiazhuang 050043, China; b: Key Laboratory of Roads and Railway Engineering Safety Control (Shijiazhuang Tiedao University), Ministry of Education, Shijiazhuang 050043, China; c: Faculty of Architecture, Civil and Transportation Engineering, Beijing University of Technology, Beijing 100124, China; d: Key Scientific Research Base of Safety Assessment and Disaster Mitigation for Traditional Timber Structure (Beijing University of Technology), State Administration for Cultural Heritage, Beijing 100124, China; e: The College of Urban Construction, Hebei Normal University Of Science & Technology, Qinhuangdao 066000, China; *Corresponding author: guoxd7797@163.com
GRAPHICAL ABSTRACT
INTRODUCTION
As an important part of Chinese civilization, Chinese ancient building has its own unique architectural system. Usually, timber structure is the main load bearing system of Chinese ancient timber buildings, which is quite different from reinforced concrete structure. However, many ancient timber buildings were damaged and destroyed in earthquakes. Therefore, some researchers began to study seismic performance of ancient timber buildings. Xiong et al. (2013) investigated the damage situation of ancient timber buildings in the Lushan 7.0 earthquake, Sichuan. Xie et al. (2010) analyzed the earthquake damage situation of Chinese ancient buildings in the Wenchuan earthquake. They found that timber ancient buildings have good seismic performance. Timber structure has good isolation and energy dissipation ability. The mortise-tenon joint is a unique energy dissipation structure in timber structure (Ma et al. 2003). Research (Fang et al. 2001) has shown that mortise-tenon joints have semi-rigid character. On this basis, many scholars have conducted in-depth studies on the mechanical properties of mortise-tenon joints. Hu et al. (2020) established three types of mortise-tenon joint finite element models, which included a whole rigid model, a tie rigid model, and a semi-rigid model. The results showed that the semi-rigid model performed was the best. All these studies show that mortise and tenon joints are a key point of ancient timber buildings.
Ancient timber buildings have stood for hundreds of years. Many of them have survived many earthquakes. In the course of such events, different degrees of structural damage have occurred in the buildings, especially in the mortise and tenon joints. Therefore, many scholars have studied the reinforcement methods to improve the seismic performance of mortise-tenon joints. For example, Hu et al. (2023) proposed a novel method of enhancing the mechanical strength of the dovetail joint using a dowel reinforcement. Wang et al. (2021) investigated the in-plane rotational behavior and wood damage evolution of wood pegged semi mortise and tenon connections. Huan (2019) found that mortise and tenon joints retained semi-rigid characteristics after flat steel reinforcement. Seismic performance of the reinforced joints can be effectively improved. Xie et al. (2018) found that the residual deformation of the mortise-tenon joints strengthened with shape memory alloy is approximately 20% less than that strengthened with the conventional strengthening methods，such as cramp, carbon fiber, steel pins, U-shaped hoop, angle steel, curved plate, etc. Shi et al. (2022) studied the influence of six reinforcement methods such as nails on the seismic performance of mortise-tenon joints. Their studies provided a reference for the repair and seismic pre-reinforcement of similarly damaged timber structure buildings. However, these studies were focused on the reinforcement effect on healthy joints without considering the negative effect of structural damage on the joints.
There are many factors affecting seismic performance of mortise and tenon joints, such as the type of mortise and tenon joint, their geometry, and the type of reinforcement. Through-tenon joints, dovetail tenon joints, and half tenon joints are widely used in ancient buildings of large palaces, whereas continuous tenon joints are mostly used in traditional residential buildings. Chen and Qiu (2016), Chun et al. (2016), and Xu et al. (2021) respectively studied the mechanical properties of through-tenon joints and obtained its main failure modes, flexural behavior, bearing capacity, and other mechanical properties. Chen et al. (2016) obtained the moment-rotation curves and failure modes of through-tenon joints through tests and numerical simulation. Xie et al. (2016) investigated the force mechanism of a dovetail mortise-tenon joint and deduced the moment-rotation relationship according to the mechanical equilibrium and deformation coordination. In order to rationalize the dimensional design of mortise and tenon joints, some scholars investigated the effect of tenon geometric dimensions (length, width, and thickness) on withdrawal and bending load capacities of mortise-and-tenon (M-T) joints (Hu et al. 2021; Zhang and Hu 2021). The results of Xue’s (2019) study show that pinching effect of the moment-rotation hysteresis curves of continuous tenon joint becomes more obvious as the section height of Fang increases. Apart from these factors, damage has been found to affect the seismic performance of mortise-tenon joints (Sha et al. 2019). Inner gaps between mortise and tenon is a common form of structural damage. He (2021) investigated the mechanical performance of through-tenon joint involving gaps, and a theoretical model of the bending moment for a loose mortise-tenon joint was proposed. Yu (2022) studied the behavior of a wooden portal frame with through-tenon joints and column foot connections under transverse load. An analytical model for a planar loose through-tenon joint was proposed based on the load-displacement relationship. Results indicate that the lateral deflection mechanism of a wooden frame is strongly affected by the loose through-tenon joints and the P-Δ effect of the supporting columns. Hu (2022) analyzed the stress mechanism of a straight mortise-tenon joint with wooden pegs in traditional residential wooden structures. A theoretical moment-rotation model of the joint was derived, and the parameters of the theoretical model were analyzed. Yang et al. (2020) studied the load resisting mechanism of the mortise-tenon joint with gaps under in-plane forces and moments. Chang et al. (2022) investigated the seismic behavior of Gutou mortise-tenon joint under loose damage. Ogawa et al. (2016) derived a method of theoretical estimation with a gap as parameter for the mechanical performance of mortise-tenon joint. In addition, it experimentally validated the method of estimation and numerically analyzed the influence of a size of such gap on mechanical properties. The numerical analysis clarified the large influence of the gap at joints on the mechanical properties. Xue (2018) analyzed the force mechanism of the through-tenon joints, and the theoretical formulas of the moment-rotation angles of the joints with different loosening degrees were deduced. Research results show that the bearing capacity of the joint is reduced gradually with the increase of the looseness.
The gap between mortise and tenon shoulder is a common phenomenon of damage to mortise-tenon joints in post-earthquake ancient timber architecture (Li et al. 2022). Therefore, some scholars have studied the influence of the gap between mortise and tenon shoulder on the seismic performance of mortise-tenon joints. It can be found that most previous studies mainly focused on the seismic performance of mortise and tenon joints under non-destructive conditions. A minority of studies focused on the influence of gaps between surfaces of tenon and mortise on the seismic performance of joints. However, little research has been done to study the pullout tenon gaps between mortise and tenon shoulder. According to historical earthquake damage data by Xiong et al. (2013), the pullout tenon gaps between mortise and tenon shoulder is one of the main earthquake damage forms of ancient timber buildings. Xie et al. (2019) studied the cyclic behavior of straight mortise-tenon joints with pullout tenons. The results showed that the strength and deformation of a joint with a pulled-out tenon decreases with the increase of the pulled-out length. Yang et al. (2015) conducted a static test on the dovetail joint with the gap between mortise and tenon shoulder and analyzed the mechanical properties of the wooden frame in the damaged state. The main failure modes of the joints, the influence of the gap between mortise and tenon shoulder on the failure of the frame, the ultimate bearing capacity, and the value method of the ultimate load were obtained. Xu et al. (2014) studied the mechanical properties of through-tenon joints under the state of pull tenon, and the influence of the gap between mortise and tenon shoulder on the overall performance of the structure was obtained.
In this paper, the seismic performance of through-tenon joint was investigated by low-cycle loading tests. Finite element analysis was used to study the seismic performance of through-tenon joints. The finite element analysis results were compared with the test results to verify the reliability of the finite element analysis model. Based on the model, six finite element analysis models of through-tenon joint with different gaps were built. The seismic performance of these models was compared and analyzed.
EXPERIMENTAL
According to the building code published during the Song dynasty, Yingzao Fashi (Li 1950), a 1:3.52 scaled through-tenon joint model was fabricated. The tested timber structure consists of two main parts, column and beam. The dimensions of specimen are shown in Fig. 1. As Pinus sylvestris timber is a widely used material in construction and restoration of ancient timber structures, the model used in this test was made of Pinus sylvestris timber. The wood for test is sourced from northeast of China. Moisture content and air-dry density of the wood were 11% and 0.47 g/m3, respectively. Table 1 shows the mechanical parameters of wood material, and these parameters were also used in the finite element model, which were obtained from the previous studies (Meng et al. 2018).
Fig. 1. Dimensions and structure of through-tenon joint model. (a) Plan view of mortise and tenon joint (mm) (b) 3D view of mortise and tenon joint (L, R, and T are the longitudinal, radial, and tangential directions of timber)
All the specimens were handmade by carpenters. After the specimens were manufactured, they are transported to the test set. Some handmade devices are shown in Fig. 2.
Fig. 2. Handmade devices
Table 1. Mechanical Parameters Used in Finite Element Model
Loading Scheme
Tests were done according to the loading procedure in Building Seismic Test Regulations (JGJ/ T101-2015) (Wu 1997). A universal testing machine was used for loading, using the device shown in Fig. 3. Vertical displacement load was applied to the end of beam to make the tenon rotate until displacement reached the maximum range of the loading machine. Vertical upward loading was set as the negative loading direction, and the opposite was set as the positive loading direction. Control displacements were selected as ± 1, ± 3, ± 5, ± 10, ± 20, ± 30, ± 40, and ± 50 mm (1/700, 3/700, 1/140, 1/70, 1/35, 3/70, 2/35, and 1/14 rad, respectively). When the control displacement was 1 to 5 mm, the load speed was 1 mm/min, whereas when the control displacement was 10 to 50 mm, the load speed was 10 mm/min. Each displacement load was cycled three times. The loading procedure for the cyclic test is shown in Fig. 4. The loading position was 700 mm away from the shoulder of the tenon. According to the Ying Zao Fa Shi standard and scale factor, the computed roof weights were 4 kN. The loads were applied on the top of column by a hydraulic jack.
Fig. 3. Universal testing machine
Fig. 4. Loading procedure for the cyclic test
Measuring Scheme
The horizontal loads and displacements of specimen were collected automatically by the program system. Two displacement measuring instruments were set to the top and bottom surfaces of the beam to measure the tenon pullout length. An inclinometer was used to measure the rotation angle of the beam. The test setup is shown in Fig. 5.
Fig. 5. Test setup Fig. 6. Cracks and plastic deformation on the tenon
General Observations from the Test
No obvious damage was observed during the course of the initial loading procedure (the rotation angles between ±1/700 rad to ±1/70 rad). When the rotation angle was beyond 1/70 rad, the mortise and tenon began to rub and squeeze each other harder and harder. When the rotation angle reached ±3/70 rad, slightly plastic deformation occurred on the tenon. The specimen began to creak and groan while loading rotation angle continued to grow. At first, there was no gap between mortise and tenon shoulder. However, the gap became wider and wider while the loading displacements increased. At the end of test, the tenon was pulled out of the mortise at a length of about 5 mm. Cracks on the tenon were observed. Slight plastic deformation had occurred at the end of the tenon. Details are shown in Fig. 6.
Finite Element Model Using ABAQUS
ABAQUS is software that uses the finite element method to generate highly approximate solutions. ABAQUS was used in this study to analyze the seismic performance of through-tenon joints. Established finite element model is shown in Fig. 7. Mechanical properties of timber are characterized by anisotropy, which can be simplified as orthotropic materials. The ideal elastic-plastic model was used to simulate parallel to grain compression property of wood, and the bilinear strengthening model was used to simulate perpendicular to grain compression property of wood. Details are shown in Fig. 8. Mechanical properties of wood in elastic stage were defined by Engineering Constants module of ABAQUS. Initial yield ratios in every direction of the plastic stage of wood were defined by a potential function. The model takes the parallel to grain compressive strength as the reference yield strength, and sets the initial yield stress ratio R11, R22, R33, R12, R13, and R23 of timber in every direction as 1, 0.124, 0.111, 0.326, 0.326, and 0.326, respectively. These ratios can be calculated by the method from Zhu (2015) and Xue et al. (2019). Eight-node hexahedral linear reduced solid units (C3D8R) were selected to build finite element model. Contact mode between cells was set as surface-to-surface contact. There were normal and tangential forces between contact surfaces, and hard contact was adopted in normal contact. The analysis model allowed the tenon to separate from the column during revolution of the beam. Friction-slip between tenon and mortise was expressed by a penalty friction formula in tangential behavior. Some studies show that the coefficient of friction between surfaces of wood is about 0.5 (Hu et al. 2020). Xie et al. (2019) suggest that the coefficient of friction is between 0.1 and 0.6. However, some studies show the coefficient of sliding friction is smaller than that of static friction, and the direction of wood texture will affect the friction coefficient (Wang 2014). The sliding friction coefficient between wood surfaces along to the grain and perpendicular to the grain ranges from 0.3 to 0.4. (Wang 2014). Considering the size error between test model and finite analysis model, the tenon is easier to slip during the test. Therefore, set the coefficient friction to 0.3 in finite element model. The surfaces of mortise were set as master surfaces, and surfaces of tenon were set as slave surfaces. In order to avoid the master surface penetrating the slave surface, the mesh size of tenon near mortise-tenon joint contact surface was 10 mm, and the mesh size of the part away from contact surface was 20 mm. The mesh size of mortise was 20 mm, and the mesh size of the parts far from mortise was 30 mm. Meshing of finite element model is shown in Fig. 7(a).