Experimental Investigation of Full Hole Embedment Behavior of Bamboo Scrimber with Dowel-Type Fasteners

Abstract

A comprehensive understanding of the embedment behavior is of great importance in the design of contemporary bamboo constructions with connections utilizing dowel-type fasteners. The objective of this research was to assess the embedment behavior of bamboo scrimber using full-hole embedment tests. To investigate the effect of the loading angle and bolt diameter, a series of tests were performed using bolts of varying diameters (16 mm, 18 mm, and 20 mm) and loading angles (0° to 90°, with an increment of 15°). The experimental results demonstrated that the loading angle has a considerable influence on the embedment behavior. As the loading angle was increased, the failure mode underwent a change from a brittle failure mode, which was dominated by shear mechanisms, to a ductile failure mode, which was dominated by fiber crushing. The yield and ultimate embedment strengths showed an M-shaped response to changes in the loading angle, with the lowest values being 0°, 45°, and 90°. The bolt diameter was found to have no impact on the failure mode of the specimen. However, an increase in bolt diameter resulted in a reduction in the embedment strength when the specimen was loaded at 90°.

1. Introduction

The growing emphasis on energy conservation and environmental sustainability has prompted the utilization of bamboo composite materials in the field of civil engineering [1,2]. Bamboo scrimber, a low-carbon, renewable bamboo-based fiber composite, is manufactured from round bamboo through various processes. These include, but are not limited to, splitting, carbonization, dipping, drying, and hot pressing [3,4,5]. This innovative material overcomes the constraints associated with conventional round bamboo, including limitations regarding its cross-sectional dimensions, shape, and structural connection, and is therefore suitable for use in the construction of low- and multi-story buildings [6,7]. Over the past two decades, there has been extensive research conducted on the mechanical properties of bamboo scrimber [5,8,9,10], as well as its potential applications in structural components [11,12,13,14].

In contemporary bamboo structures, the most prevalent connection type is that which employs dowel-type fasteners. In contrast to the traditional mortise and tenon method, this approach utilizes standardized metal fasteners to connect engineered bamboo components, exhibiting a straightforward structural design and unambiguous force transmission. The load-bearing capacity of structural connections is of the utmost importance, as it directly affects the overall stability and seismic resistance of buildings. Johansen’s yield theory [15] provided a fundamental framework for estimating the bearing capacity of dowel-type connections, which suggests that the bearing capacity of dowel-type connections is dependent on two key factors: the yield moment of the fastener and the embedment strength of wood/bamboo members. At present, it has been incorporated into the design specifications for timber structures in numerous countries, including the European standard EN 1995-1 [16], the Canadian standard CSA O86 [17], and the Chinese standard GB 50005 [18].

In the assessment of embedment strength, two test set-ups are commonly utilized, namely half-hole and full-hole tests, as illustrated in Figure 1. The half-hole test involves the application of a uniform compressive load along the length of the fastener, which results in an even distribution of stress beneath the fastener’s surface, across the thickness of the specimen. In the full-hole test, the compressive load is applied to the protruding ends of the fastener, resulting in an uneven stress distribution along the specimen thickness. The half-hole test setup is based on the experimental research conducted by Wilkinson [19]. He conducted half-hole compression tests on a variety of wood species, including Douglas fir, southern pine, and other wood products, considering different fastener diameters and the specimen thickness. The full-hole test is based on a series of studies conducted by Whale and Smith [20,21]. They assessed the embedment strength of five European softwoods and two hardwoods through tensile or compression full-hole tests, with an investigation of the impact of bolts with different diameters. The American standard ASTM D5764 [22] employs the half-hole test method for the assessment of the embedment strength of wood and wood-based products. In contrast, the European Standard EN 383 [23] employs the full-hole test for evaluation purposes. The two distinct test set-ups have different requirements with respect to the test specimen. Figure 1 illustrates the specific requirements for the test specimen as set forth in ASTM D5764 [22] and EN 383 [23].

To date, there has been a considerable amount of research on the embedment behavior of sawn timber and engineered wood products [24,25,26,27,28,29,30,31,32,33]. Schweigler [24] conducted 85 full-hole embedment tests on laminated veneer lumber (LVL) with two different fastener diameters. The loading angle was varied from 0° to 90° to account for the anisotropic behavior of LVL. A pronounced displacement-hardening effect was observed for loading angles of 60° or higher, resulting in an elevated embedment stress at large displacements. Additionally, the authors reported that the embedment displacements were associated with the anisotropic properties of wood and the densification effect in the vicinity of the dowel. The research on the embedment behavior of birch plywood conducted by Wang et al. [25] indicated that, as the loading angle and the fastener diameter increased, the embedment strength exhibited a slight decrease. The research findings were in alignment with the results reported by Schweigler [24]. Furthermore, Wang et al. compared the impact of test set-ups between half-hole and full-hole tests. The study revealed that there were no notable differences in the observed failure modes between specimens subjected to different testing configurations. Nevertheless, the full-hole test typically gives a higher embedded stiffness and strength than the half-hole test. Similar observations were reported by Aquino et al. [26], wherein the embedment strength and stiffness of two European softwoods were investigated in accordance with the ASTM D5764 [22] and EN 383 [23] standards. A thorough examination of the methodologies employed in embedment tests was presented by Ottenhaus et al. [27]. They conducted a total of 224 full-hole and half-hole tests on Australian sawn softwoods with two different dowel diameters. It was recommended that full-hole tests be conducted for the purpose of establishing the embedment strength, as this method most closely approximates the stress state that would be encountered in actual connections. In cases where the material properties result in an embedded strength exceeding the yield moment of the fastener, the half-hole test is deemed an appropriate method. Furthermore, the impact of additional parameters on the embedment strength, including the material density [28,29], moisture content [30,31], and surface conditions of the fastener [32,33], has been thoroughly investigated.

In contrast to the extensive literature on the embedment behavior of wood products, there is relatively limited research available on bamboo-based composites. Chen et al. [34] evaluated the dowel-bearing strength of laminated flattened bamboo lumber by conducting half-hole tests on specimens oriented both parallel and perpendicular to the grain. The impact of the flattened-bamboo configuration methodology and fastener diameter was also taken into account. The results demonstrated that the bearing strength perpendicular to the grain was significantly greater than that parallel to the grain. Furthermore, it was observed that the bearing strength decreased with an increase in bolt diameter. In the work performed by Li et al. [35], the effect of the moisture content and loading angle on the dowel-bearing behavior of bamboo scrimber was investigated. The loading angle has a considerable influence on the bearing strength and stiffness of bamboo scrimber. Moreover, as the loading shifts from the parallel direction to the perpendicular direction, the failure mode also undergoes a transition from brittle to ductile. Given the observations to date, research on the dowel-type bearing behavior of bamboo-based composites has been primarily based on half-hole embedment tests, with a paucity of reports on full-hole embedment tests.

This study was therefore conducted with the objective of examining the embedment behavior of bamboo scrimber using the full-hole embedment test. The experiment was conducted in accordance with ES 383 [23]. Three different dowel diameters and varying load-to-grain angles, spanning a range of 0° to 90° with an increment of 15°, were considered. The influence of the dowel diameter and load-to-grain angle on the embedment behavior of bamboo scrimber was discussed with regard to the failure mode, the load–displacement curve, and the embedment strength.

2. Materials and Methods

2.1. Bamboo Scrimber and Steel Bolts

The ISO 23478 standard [4] defines bamboo scrimber as a panel or lumber product made from compressed bamboo fiber bundles strips or sheets. Bamboo scrimber displays a number of favorable characteristics, including a high specific strength, high specific stiffness, ease of assembly, and so forth [10,36]. These characteristics have contributed to the accelerated development of bamboo scrimber as a construction material in China [37]. Bamboo scrimber is typically derived from 5-year-old moso bamboo (Phyllostachys edulis) as a raw material, which then undergoes a series of manufacturing processes, including cutting and splitting, the removal of the inner and outer parts, drying, dipping, assembly, and hot pressing. The oven-dried density of the bamboo scrimber was found to be 1.22 g/cm³. Before the test, the specimens were conditioned in an environmental chamber with a humidity of 65 ± 5% and a temperature of 20 ± 2 °C until equilibrium was reached. The equilibrium moisture content of the specimens was determined to be 8.01%, with a compressive strength of 88.25 MPa parallel to the grain and 23.14 MPa perpendicular to the grain. Additionally, it displayed a shear strength parallel to the grain of 8.21 MPa and a tensile strength perpendicular to the grain of 4.43 MPa.

To account for the influence of the bolt diameter, 8.8 grade high-strength bolts with diameters of 16 mm, 18 mm and 20 mm were selected. Embedment tests were performed on bamboo scrimber using three bolt diameters and two orientations: parallel and perpendicular to the grain. Furthermore, in order to assess the influence of the loading angle, embedment tests were carried out at load-to-grain angles of 15°, 30°, 45°, 60° and 75°, with a bolt diameter of 18 mm.

2.2. Full-Hole Embedment Test

The test schematic is presented in Figure 2, and the dimensions of the specimen used in the embedment test are presented in

Table 1. Dimensions of the tested specimens.

Group	L [mm]	h1 [mm]	h2 [mm]	T [mm]	d [mm]	$\mathit{\theta }$ [°]
A	224	112	112	30	16	0
						90
B	252	126	126	30	18	0
						15
						30
						45
						60
						75
						90
C	280	140	140	30	20	0
						90

. The test specimens were categorized into three distinct groups based on the bolt diameter. Group A, Group B, and Group C correspond to bolt diameters of 16 mm, 18 mm, and 20 mm, respectively. The inter-group comparison primarily focuses on analyzing the influence of the bolt diameter, while the intra-group comparison within Group B is concerned with the influence of the load-to-grain angle. L and H represent the length and height of the specimen, respectively. h₁ and h₂ are the heights of the upper and lower halves of the specimen, respectively. t signifies the thickness of the specimen, whereas d denotes the diameter of the steel bolt. The symbol $\theta$ represents the load-to-grain angle. $\theta =0$ indicates parallel to the grain direction, while $\theta =90$ indicates perpendicular to the grain direction. Each loading condition was subjected to five repetitions.

The experimental work was conducted on a 10-ton universal testing machine with 0.5% precision, which was located within the laboratory of the National Engineering Research Center of Nanjing Forestry University. The full-hole embedment tests were performed in accordance with the specification EN 383 [23]. A downward load was applied at a crosshead rate of 1 mm/min until either the specimen failed or a displacement of 10 mm was reached. The displacement of the crosshead was recorded throughout the testing process in order to facilitate the subsequent construction of load–displacement curves and the calculation of the embedment strength. Furthermore, it is essential to observe and record any notable experimental occurrences during the experiment. Details such as the failure modes’ sequence of crack formation should be included, as they will inform the discussion of the test results.

2.3. Embedment Strength

The embedment strength is a crucial factor in the design of contemporary bamboo structures with connections using dowel-type fasteners. It is defined as the nominal stress related to the projected area of the dowel, which is the product of the dowel diameter and the specimen thickness. In this study, the embedded strength was evaluated using two nominal stresses: the yield embedment strength $f_y$, determined through the 5% off-set method, and the ultimate embedment strength $f_u$ within the initial 5 mm displacement. The equations below are used to calculate the yield and ultimate embedment strength.

$$f_y=F_y/\left(d\times t\right)$$

(1)

$$f_u=F_u/\left(d\times t\right)$$

(2)

where $F_y$ represents the load at the intersection of the load–displacement curve and the 5% dowel diameter off-set line; $F_u$ is the maximum load within the initial 5 mm displacement. The aforementioned characterizing values are schematically illustrated in Figure 3. It is noteworthy that in some cases, a descending segment may not be observed within the initial 5 mm displacement in the load–displacement curve, particularly when the load-to-grain angle is large.

3. Results and Discussion

3.1. Failure Modes

Regarding the bolt diameters examined in this article, the bolt diameter does not affect the failure mode of the specimen in the full-hole embedment test. The specimen displays brittle splitting damage when subjected to compressive loads parallel to the grain at all bolt diameters. When subjected to compressive loads perpendicular to the grain, the specimen exhibits a failure mode characterized by a combination of fiber fracture and fiber crushing. Therefore, the discussion primarily focused on failure modes associated with different load orientations.

Figure 4 presents the failure modes of all specimens with a bolt diameter of 18 mm under different loading orientations. The 0–30° specimens exhibited splitting cracks that extended throughout the specimen. Consequently, Figure 4a–c show images taken of the entire field of the specimen. In contrast, the damage observed in the 45–90° specimens was clearly concentrated in the vicinity of the bolt hole. Therefore, Figure 4d–g show the damage in the area around the bolt hole. In specimens loaded at 0°, the failure mode induced by embedded loads is characterized by fiber splitting. As illustrated in Figure 4a, two cracks are evident in the region beneath the fastener, while a single crack is observed in the upper half of the specimen. It is proposed that fiber cracking is attributable to tension, shear force or a combination of both acting perpendicular to the grain. Additionally, no evidence of matrix crushing was observed in the 0° bamboo scrimber specimens, which differs from the findings of Wang et al. [25] regarding birch plywood. The 15° and 30° specimens exhibited analogous failure modes, both of which demonstrated brittle failure. Two misaligned cracks emerged in the upper and lower regions of the specimen, subsequently extending throughout the entirety of the specimen. As illustrated in Figure 4d–g, the failure mode observed in the specimens subjected to loading in the range of 45–90° is identical. The damage is restricted to a limited area in close proximity to the bolt hole, characterized by fiber crushing and drumming under the fastener and fiber cracking on the sides of the bolt hole. In the case of 90°, fiber cracking on the side is classified as a tensile failure perpendicular to the grain, whereas in the cases of 45–75°, it is classified as the combination of shear and tensile failure. Due to the movement of the fasteners, the bolt holes on the 45–90° specimens were pulled into elongated circles, which is indicative of significant ductile embedment behavior.

3.2. Effect of Bolt Diameter on Embedment Behavior

The present study comprises three embedment test series, each utilizing bolts with diameters of 16 mm, 18 mm, and 20 mm, respectively. Figure 5 illustrates the embedment load–displacement curves obtained from the three test series. The shaded area in the figure represents the interval within which the results of five repeated experiments are located, thereby indicating the variability of the experimental data. The black solid line represents the median curve. The yield and ultimate embedment strengths of bamboo scrimber for bolt diameters of 16 mm, 18 mm, and 20 mm were calculated using Equations (1) and (2). The statistical values for the embedment strength, including the mean value and standard deviation, are given in

Table 2. Embedment strength for bolt diameters of 16 mm, 18 mm, and 20 mm.

d [mm]	$\mathit{\theta }$ [°]	${\mathit{f}}_{\mathit{y}}$ [MPa]		${\mathit{f}}_{\mathit{u}}$ [MPa]
		Mean	Std *	Mean	Std *
16	0	149.72	4.54 (3.03%)	162.56	3.88 (2.39%)
18		142.19	14.92 (10.50%)	148.44	14.58 (9.82%)
20		145.29	5.33 (3.67%)	155.13	3.57 (2.30%)
16	90	162.28	12.27 (7.56%)	179.98	12.45 (6.92%)
18		138.55	7.73 (5.58%)	159.69	9.61 (6.02%)
20		133.06	3.00 (2.66%)	142.95	1.69 (1.18%)

* “std” refers to the standard deviation, which indicates the degree of variation or dispersion of the values in a dataset from their mean. The values in parentheses represent the coefficients of variation.

.

The embedment load–displacement curve obtained from the full-hole embedment test demonstrated a consistent trend regardless of the variation in bolt diameter. These findings are in accordance with the observations reported in numerous previous studies [24,25,27,34]. This section does not address the comparison of the embedment load–displacement curve parallel and perpendicular to the grain. However, this comparison will be presented in Section 3.3.

Several empirical formulas for predicting the embedment strength of dowel-type fasteners are available for consultation in the references [38,39,40]. Sawata et al. [38] and Ottenhaus et al. [27] have highlighted that the yield embedded strength is independent of the fastener diameter. The empirical prediction formulas are almost based on the ultimate embedment strength, with the wood density (wood species) and fastener diameter serving as parameters. Eurocode 5 [16] states that the ultimate embedment strength of bolts with a diameter of up to 30 mm is $f_h=0.11\left(1-0.01d\right)\rho$ in plywood A similar prediction formula was proposed by Franke and Quenneville [39], though the coefficients were different. In addition to the linear formulas, Leijten et al. [40] proposed an exponential prediction formula. The aforementioned prediction formulas all demonstrate a negative correlation between the bolt diameter and ultimate embedded strength. As shown in Table 2, both the yield embedment strength and the ultimate embedment strength demonstrate a decline with an increase in bolt diameter when subjected to compressive loading perpendicular to the grain. However, no discernible trend was evident when the specimens were subjected to compressive loading parallel to the grain direction. As shown in Figure 5c and Table 2, the coefficient of variation for the test results is approximately 10% for d = 18 mm and the loading angle is 0°. This value is markedly higher than that observed in other cases. It is possible that the relatively low number of repetitions of the experiment resulted in a large coefficient of variation, which in turn impeded the observation of the effect of the bolt diameter on embedded strength. It is therefore recommended that further research be conducted in future work by increasing the number of experimental repetitions in order to obtain more accurate results.

3.3. Effect of Load-to-Grain Angle on Embedment Behavior

A total of seven series of embedment tests were conducted, with each series representing a specific load-to-grain angle ranging from 0° to 90°, with a 15° increment. The representative embedment load–displacement curves for each loading angle are presented in Figure 6. The curves of the 15° and 30° specimens exhibit a sudden decline at a displacement of approximately 5 mm, accompanied by the presence of cracks throughout the specimen (Figure 4b,c). This observation aligns with the characteristics of brittle failure. Cracks were also observed on the 0° specimen, which subsequently failed when the displacement reached 9 mm. The embedment load–displacement curves at other loading angles demonstrate notable ductility, which is in accordance with the failure mode illustrated in Figure 4.

Based on the load–displacement curves, the yield and ultimate embedment strengths were calculated as the 5% d off-set stress and the maximum stress prior to 5 mm displacement, respectively. Figure 7 illustrates the variation in the embedment strength with loading angle. The yield embedment strength obtained for the bamboo scrimber is between 138.6 MPa and 163.8 MPa, significantly higher than the value obtained from the half-hole embedment test, ranging from 63.3 MPa to 85.0 MPa [25]. This confirms the contribution of the upper half of the specimen to the embedment strength. As illustrated in Figure 7, the embedment strength at varying loading angles displays an “M”-shaped profile. The embedment strength exhibits the lowest values at load-to-grain angles of 0°, 45°, and 90°. It was observed that the strength increased when the loading angles were 15°, 30°, 60°, and 75°.

Analysis of variance (ANOVA) is a statistical method used to test for significant differences between the means of three or more groups. One-way ANOVA is the simplest form of ANOVA, involving a single independent variable that has three or more levels. In this study, one-way ANOVA was performed to investigate whether the loading-to-grain angle significantly affected the embedment strength. The results of the ANOVA are presented in

Table 3. Results of One-Way ANOVA.

	Index	SSA	SSE	MSA	MSE	p-Value
—
yield strength		2963.4	2370.3	493.9	84.7	5.0 × 10−4
ultimate strength		2477.3	2437.9	412.9	87.1	1.9 × 10−3

Note: SS represents the sum of squares for error, and MS denotes the mean square error. The subscript ‘A’ signifies the values obtained from differences between different angular configurations, whereas the subscript ‘E’ signifies the values obtained from differences within all angular configurations.

. The p-value is less than the chosen significance level (0.05 for this analysis), indicating that the variability between groups is significantly greater than that within the groups. This suggests that the load-to-grain angle has a significant effect on the embedment strength of bamboo scrimber. Furthermore, based on the ANOVA results, multiple comparisons were conducted to determine if there were any significant differences between any two sets of angular configurations. The multiple comparisons revealed significant differences between the angular configurations of 0°, 45°, and 90°, and those of 60° and 75°. However, no significant differences were observed between the angular configurations of 0°, 45°, and 90°, and those of 15° and 30°.

As an anisotropic material, bamboo scrimber demonstrates a compressive strength of 23.14 MPa and a tensile strength of 4.43 MPa perpendicular to the grain direction. Its shear strength parallel to the grain is 8.21 MPa. Figure 8 provides a schematic diagram of the stress occurring in bamboo scrimber specimens at varying loading angles, which facilitates an understanding of the underlying causes of the variation in embedment strength. The specimen fails as a result of shear action at 0°. The embedded strength can be attributed to the contribution of shear strength parallel to the grain. When the loading angle is below 45°, the specimens still undergo shear failure. Nevertheless, the shear force acting on the failure surface is the parallel-to-grain component of the embedment load. Consequently, greater embedment loads are required to induce shear failure. As illustrated in Figure 8, when the load-to-grain angle is 90°, the fiber situated directly beneath the bolt is compressed, and the materials on both sides of the bolt are subjected to tensile stress perpendicular to the grain direction, resulting in a pronounced tearing phenomenon (Figure 4). The embedment strength of the 90° specimen is dependent on both the tensile strength and the compressive strength of bamboo scrimber perpendicular to the grain direction. When the load-to-grain angle is between 45° and 90°, the fiber crushing and tearing phenomena are less severe than at 90° due to the action of the perpendicular-to-grain component. This consequently yields a greater embedment strength than that obtained at 90°. When the load is applied at 45°, the two orthogonal components work together to form a combined shear and tension area in the vicinity of the bolt hole. As previously stated, the tensile and shear strengths of the bamboo scrimber perpendicular to the grain direction are relatively low, resulting in a reduction in the embedment strength at 45°.

4. Conclusions

This study was conducted with the objective of investigating the embedment behavior of bamboo scrimber using dowel-type fasteners. A total of 55 specimens were subjected to full-hole embedment tests, with three different bolt diameters and seven load-to-grain angles. The effect of various diameters and loading angles was thoroughly examined, resulting in the following conclusions:

(1)

In cases where the loading angle was less than 45°, the bamboo scrimber specimen was observed to fail in a brittle manner, with cracks propagating throughout the specimen and exhibiting a shear failure. Upon increasing the loading angle to 45° or above, the damage was confined to the region in close proximity to the bolt hole, including fiber crushing beneath the fastener and fiber rupture on both sides. This damage displays ductile characteristics.

(2)

When the compressive load was applied at 90°, as the bolt diameter increased from 16 mm to 20 mm, there was a corresponding decrease in both the yield embedment strength (from 162.28 MPa to 133.06 MPa) and ultimate embedment strength (from 179.98 MPa to 142.95 MPa). When the load was applied at 0°, no evident change in the embedment strength was observed.

(3)

The loading angle has a considerable effect on the embedding strength of bamboo scrimber, which can be attributed to the anisotropic property of the material. The strength exhibited an M-shaped variation with an increase in the loading angle, reaching a minimal value at load-to-grain angles of 0°, 45°, and 90°. At other loading angles, an increase was observed.

In the experimental data, higher variability was observed under certain conditions. This higher variability may be attributed not only to the variations in the manufacturing process of the specimens and the inherent heterogeneity of the material, but also to the complex stress distribution and failure mechanisms at specific conditions. It is recommended that future studies consider a larger sample size and more rigorously control the testing conditions to reduce variability and improve the generalizability of the findings.