Abstract
This study examined the flexural behavior of the poplar laminated veneer lumber (LVL) box floor with orthogonal rib beams. Four 3.6 m × 4.8 m box floor samples made of poplar LVL orthogonal rib beams and oriented strand board (OSB) plates were tested under vertical uniform loading, from which the bearing capacity, stiffness, and failure characteristics were analyzed. There was no damage in all box floor samples at the normal service load of 2.5 kN/m2, and the maximum deflection was far less than the allowable value. When the maximum load was applied, the load-displacement curve of each floor sample exhibited a linear relationship without obvious failure. However, localized failure was manifested as the dislocation slip of the rib beams relative to the upper and lower floor slabs at the corner nodes and the joint expansion and staggered floors at the bottom plate, with obvious failure signs. The rib beam height had the most significant impact on the floor stiffness, followed by the spacing of short-side rib beams, whereas the OSB plate thickness had lest impact. The mid-span deflections of poplar LVL orthogonal ribbed box floor samples, which were calculated using the analog slab method, were in good agreement with the experimental results with an error being less than 10%.
Download PDF
Full Article
Experimental Study on Flexural Behavior of Box Floors with Orthogonal Rib Beams Made of Poplar Laminated Veneer Lumber
Xiangyu Su,a,b Jing Chen,b Yan Liu,b,* Xufeng Sun,b Gong Meng,c and Haida Wang b
This study examined the flexural behavior of the poplar laminated veneer lumber (LVL) box floor with orthogonal rib beams. Four 3.6 m × 4.8 m box floor samples made of poplar LVL orthogonal rib beams and oriented strand board (OSB) plates were tested under vertical uniform loading, from which the bearing capacity, stiffness, and failure characteristics were analyzed. There was no damage in all box floor samples at the normal service load of 2.5 kN/m2, and the maximum deflection was far less than the allowable value. When the maximum load was applied, the load-displacement curve of each floor sample exhibited a linear relationship without obvious failure. However, localized failure was manifested as the dislocation slip of the rib beams relative to the upper and lower floor slabs at the corner nodes and the joint expansion and staggered floors at the bottom plate, with obvious failure signs. The rib beam height had the most significant impact on the floor stiffness, followed by the spacing of short-side rib beams, whereas the OSB plate thickness had lest impact. The mid-span deflections of poplar LVL orthogonal ribbed box floor samples, which were calculated using the analog slab method, were in good agreement with the experimental results with an error being less than 10%.
DOI: 10.15376/biores.17.4.6019-6035
Keywords: Poplar laminated veneer lumber (LVL); Box floor; Orthogonal rib beam; Static test; Flexural behavior
Contact information: a: Management Committee of Hai’an Shanghu Innovation Zone (Sino-Italian Hai’an Ecological Park), Nantong, Jiangsu 226600, China; b: College of Architectural Science and Engineering, Yangzhou University, Yangzhou, Jiangsu 225127, China; c: Wood Science and Technology Center, University of New Brunswick, Fredericton, New Brunswick, E3C 2G6, Canada;
* Corresponding Author: liuyan@yzu.edu.cn
INTRODUCTION
In the 1960s and 1970s, China successfully introduced poplar (Populus euramevicana cv. I-214) from Italy, and it has become popular in China. After decades of development, the domestic poplar resources are quite rich, and a modern industrial development pattern has been formed, from seed selection and cultivation to expanded reproduction. Poplar laminated veneer lumber (LVL) is widely used in packing materials and furniture components in China. However, it is seldom used in building construction. Therefore, research on the application of poplar LVL in construction will promote the sustainable development of modern timber structures in China. Therefore, the research team has completed a series of studies on the mechanical performance of poplar LVL members, frames, and trusses. This paper mainly introduces the study on the mechanical performance of poplar LVL box floor.
Floor panels are an important component of prefabricated wooden buildings. However, traditional wooden floors composed of girders, grids, cross braces, and floor slabs have many disadvantages, such as low stiffness, large vibration response, and low comfort. As early as the 1950s, Countryman (1952) studied the effects of floor slab layout, floor slab thickness, and grid connection on the stiffness of wooden floor slabs through the static test. The results showed that the strength of the nail connection between the grid and the floor slab exhibited the most significant effect on the stiffness of wooden floors. Countryman and Colbenson (1954) investigated the effects of floor slab-grid top spacing and depth-span ratio on the lateral stiffness of wooden floors, discovering that nail spacing, nail size, and floor slab thickness were the most influential factors. Zagajeski et al. (1984) conducted an experimental study on the in-plane shear response of plywood floors under monotonic and cyclic loads. They analyzed the effects of cross brace, opening, plywood thickness, and nail size on the in-plane shear deformation, noting that repeated loading could reduce the floor stiffness. Johnson (1994) tested the field vibration characteristics of 86 wooden floor samples, analyzed the data, and proposed the comfort standard that the basic frequency should be greater than 15 Hz. Dolan et al. (1999) further clarified this standard by tracking the performance of these floor samples and proposed a standard of greater than 14 Hz for occupied wooden floors. Suzuki et al. (1995) conducted an excitation test with the vibration exciter and shaking table for an 11.18 by 6.38 m full-scale wooden floor, showing that the shear vibration theory could more accurately calculate the in-plane natural frequency and mode of the wooden floor. Filiatrault et al. (2002) also conducted a low-cycle test on the wooden floor of a full-scale two-story wooden frame house to study the influence of floor parameters on in-plane internal stiffness. The results showed that the edge cross brace of the floor slab and the upper and lower walls played the most significant role in improving the in-plane stiffness.
As for the research on structural types of wooden floors, Xiong et al. (2012) studied the flexural behavior of composite beams of wood-based structural plates and rectangular-section wood joists through the two-point loading tests of 42 wooden beam samples. They found that the stiffness and flexural capacity were significantly improved when the structure was changed from a rectangular-section beam to a T-shaped composite beam and then to an I-shaped composite beam. Awaludin et al. (2017) also conducted full-scale sample bending experiments on LVL wooden floors with various rib beam forms to find the optimum structure. The results showed that the box rib beam floor exhibited better bending stiffness and bearing capacity compared with the I-shaped rib beam floor. Wang (2012) designed a two-way wood truss floor system by substituting the truss for the grid and cross brace and provided improvement measures based on the test analysis. The results revealed that the main factor affecting the floor stiffness and bearing capacity was the cooperation of the trusses in two directions. Xue et al. (2019) used herringbone and monoclinic parallel chord trusses as the grid of light wood floors and applied uniform loads to test the bearing performance and deflection. The results showed that the deformations of the two parallel chord trusses were similar at midspan and the ends, which were far lower than the specified deflection limit. Li et al. (2016) used C-shaped thin-walled steel members to form a joist and connected the wood plate to the steel joist with screws to form a new steel wood hybrid floor. The in-plane cyclic loading test suggested that the shear deformation of this floor type accounted for 90% of the in-plane deformation. The bearing performance exhibited obvious nonlinear characteristics, and the ductility of the floor was closely related to the failure mode of the threaded connection between the steel joist and wood plate. Loss and Frangi (2017) designed a prefabricated modular steel wood composite floor unit, in which cross-laminated timber (CLT) was used for the wooden floor. Through the in-plane monotonic and cyclic loading tests of the full-scale sample, it was revealed that the deformation of this floor type mainly originated from the connecting steel joints between beams, which avoided the degradation of material properties in the CLT floor under reciprocating loads.
This study proposed a new type of wooden floor structure with an aim at enhancing the floor stiffness based on the previous studies (Loss and Frangi 2017; Xing 2020). The floor structure proposed adopted two-way orthogonal Poplar LVL rib beams connected by metal components, and the oriented strand board (OSB) as the top and bottom plates to form the box floor. The hollow box structure facilitated heat and sound insulation by filling it with air and reducing the thickness of the floor structure layer. Through the flexural performance tests under local and overall loading, the effects of rib beam depth-span ratio, rib beam spacing, and OSB plate thickness on the mechanical performance of floor slabs were analyzed, providing support for the engineering applications of orthogonal rib beam box floors made of poplar LVL.
EXPERIMENTAL
Test Sample Design
Figure 1 illustrates the floor structure designed in this study. Four types of 3600 mm× 4800 mm planar poplar LVL box floor samples were designed and manufactured by the research group according to the Standard for Design of Timber Structures (GB 50005 2017). Each type had one sample that was built and tested. The distance between the long-side rib beams was 600 mm in all samples. The other structural parameters and component layout are illustrated in Table 1 and Fig. 2.
Fig. 1. Orthogonal rib beam box floor made of poplar LVL
Table 1. Parameters of the Floor Sample
- Samples L1, L3, L4
(b) Sample L2
Fig. 2. Layout details of the floor components
The internal rib beam joints of each sample adopted the 3 mm thick Q235 steel L-shaped connector, which was connected by M4 × 20 mm cross-countersunk self-tapping screws. The size of connectors and the number of screws were increased at the end nodes of the rib beam, which were connected to the edge-sealing rib beam by M4 × 40 mm cross-countersunk self-tapping screws (Liu et al. 2022). The node structure is detailed in Fig. 3.
Fig. 3. Structural details of the node connector
OSB plates were used as the upper and lower floor slabs of the box floor, which were connected to the rib beam by 2.8 × 50 mm round nails. The spacing between the round nails was 150 mm and 300 mm at the edge-sealing rib beam and the internal rib beam, respectively. The installed floor sample is displayed in Fig. 4.
Fig. 4. Floor sample after installation
The poplar LVL used in this study was purchased from Jiangsu Jiuhe Timber Company (Jiangsu, China), and the OSB, nails, and cross countersunk head tapping screws were all purchased from a local market. The 3 mm thick Q235 steel L-joint was processed with the help of Yangzhou Huayun Electric Co., Ltd. (Jiangsu, China).
The physical and mechanical properties of poplar LVL based on the experimental results (Liu et al. 2017) were used for the rib beam (Table 2).
Table 2. Physical and Mechanical Parameters of Poplar LVL
Loading Scheme
A self-designed water tank was used for the loading test, which was composed of a steel frame and a thickened flexible plastic canvas water bag, as displayed in Fig. 5. The clear height of the steel frame was 1500 mm, and the clear height of the canvas water bag was 1600 mm. The reserved 100 mm height difference could allow fitting of the water bag and roof during the loading process so that the load could act evenly on the floor sample and allow the bag to deform with the floor (Xing 2012).
(a) Schematic of the loading device (b) Installed loading device
(c) Interior of the water tank when loading
Fig. 5. Loading device
A multi-stage loading process was adopted according to GB/T 50329 (2012) and ASTM E2322–03 (2015). First, 5% of the ultimate load (the ultimate load here was estimated by finite element analysis) was preloaded and held for 15 min. During this time, check to see if the instruments are working properly, and then empty the water tank for unloading. During formal loading, the load was set at 10% of the ultimate load for the first level and increased by 5% of the ultimate load for each new level. The holding time was fixed at 15 min, and the data were recorded when the reading of the measuring instrument was stable.
Layout of Measuring Points
The symmetry in load and structure of the box floor was considered in the test. All the displacement gauges were arranged below the rib beam node, and the layouts of displacement gauges for samples L1 through L4 are illustrated in Fig. 6.
Strain gauges were installed on the bottom plate, which were arranged at the central axis considering the sample symmetry, as displayed in Fig. 7.
Fig. 6. Layout of displacement measuring points
Fig. 7. Layout of strain measuring points
RESULTS AND DISCUSSION
Failure Morphology
No damage was present when the load was set to the standard GB 50009-2012 (2012) normal floor service load of 2.5 kN/m2. Considering the reliability and safety of the test device, the test was terminated when the load was 15 kN/m2 for samples L1, L3, and L4 and 13 kN/m2 for sample L2. The following paper only discussed the test phenomena when the samples were loaded to the levels mentioned above.
1) Sample L1: When the load reached 12 kN/m2, a few nail joints at the bottom of the midspan of the specimen began to fail, and a continuous sound came from the floor slab. With the increase of flexural deformation as the load was increased to 15 kN/m2, dislocation slip appeared between the beam ends at the four corner nodes and the upper and lower floor slabs, as illustrated in the red square boxes in Fig. 8a. All local damage parts are also labelled using the red boxes in the following figures. In the meantime, expansions with a width of 5 to 7 mm appeared at the joints of multiple bottom plates (Fig. 8b), and their positions are displayed in Fig. 8c. Many connectors were separated from the rib beam, and the screws at A-5 were pulled out (Fig. 8d) after unloading and removing the upper and lower floor slabs. The specific locations are displayed in Fig. 8e.
Fig. 8. Locations and failure in Sample L1
2) Sample L2: When the load was 13 kN/m2, expansions with a width of 4 to 7 mm occurred at several bottom plate joints, as displayed in Fig. 9a The camera placed inside the floor observed that the bottom plate was separated from the rib beam, resulting in staggered layers of about 3 mm at multiple joints (Fig. 9b). After removing the upper and lower floor slabs, many connectors were separated from the rib beam. Some gaps appeared at the rib beam joints, and the screws at A-5 were pulled out. The specific locations are displayed in Fig. 9c.