Analysis of the influence of sheathing board orientation on the horizontal load-carrying capacity and stiffness of wall panels in timber buildings

This paper deals with the determination and comparison of the horizontal load-carrying capacity and stiffness of wall panels in the system of a lightweight frame structure with OSB (oriented strand board) sheathing. Wall panels with different types of load-carrying sheathing were loaded with a horizontal force to monitor their deformation response to the load. The results of experimental tests were subsequently compared with analytical standard calculations and numerical models. Experimental tests demonstrated the suitability of using some sheathing methods in practice. Numerical models showed good agreement with measured data.

The walls of timber buildings made in the system of a lightweight frame structure with load-carrying sheathing (large-sized boards made of wood-based materials) serve an important role not only in ensuring the stability of vertical studs and transferring vertical loads but also in transferring horizontal loads and providing horizontal stiffness of the entire timber building. Currently, the diffusion-open system of the outer wall composition represents the prevailing solution of walls in timber buildings, where the lightweight frame structure is sheathed by one-sided OSB (oriented strand board) boards, which are attached with metal staples. The valid European standards for the design of timber structures [1] require a vertical orientation of these boards. Two simplified methods are given in the standard [1] to analyse the wall diaphragms and determine their horizontal load-carrying capacity (whether sheathed on one side or on both sides). However, horizontally laid OSB boards with the tongue-and-groove joints can be also found in the practical application. OSB boards are attached to the supporting frame with metal staples, or possibly are also provided with glued tongue-and-groove joints between boards. This solution is not supported by the European standard for the design of timber structures [1]. The aim of this paper is a comparison of the horizontal load-carrying capacity (racking resistance) and stiffness of different wall panel systems with one-sided OSB sheathing, using experimental tests on several large-scale specimens. The test results are then compared with the results from numerical models and analytical calculation based on the standard [1].

In practical applications of timber buildings, especially in Central Europe, construction companies often adopt simplified sheathing techniques that are not fully supported by current design standards. One such commonly used method involves the horizontal installation of OSB boards without glued tongue-and-groove joints. This approach is frequently applied due to its simplicity and speed of construction, despite the fact that it lacks normative backing in the European standard for timber structures [1]. During field visits and consultations with contractors, the authors encountered several buildings utilizing this non-standard sheathing configuration. This observation served as the primary motivation for the presented research. The aim was to experimentally verify whether such construction practice can be considered structurally safe and whether it provides adequate horizontal load-carrying capacity and stiffness comparable to standard-compliant methods.

Over the past decades, many researchers studied the mechanical behaviour of sheathed wall panels in timber buildings. Early insights into the design principles and mechanical behaviour of wood-based wall panels are summarized in a publication dating knowledge from 1927 to 1982 [2]. Since then, increasing attention has been paid to the ability of shear walls to transfer horizontal loads. The approach for determining the horizontal load-carrying capacity, which assumes linear-elastic compliance of connections between the supporting frame and the sheathing up to failure, with hinged connections between individual frame elements, is based on research works [3, 4]. Two similar design methods (Method A and Method B) can be used to determine the horizontal load-carrying capacity (racking resistance) according to the European standard [1]. Both methods are based on the assumption that the wall is sheathed with boards that have sufficient in-plane strength and stiffness to transfer the horizontal shear stress. The boards should be attached to the timber frame by fasteners with plastic properties, which allow uniform shear distribution around the perimeter of the wall panel. The simplified Method A is based on the plastic method presented in [5]. This method assumes that the shear resistance of the wall panel is given by the sum of the partial transverse resistances of flexible fasteners with constant spacing around the perimeter of the panel, where shear stress is transferred. It is also assumed that the frame and sheathing elements are rigid and are connected by hinges. The simplified Method B is limited only to wall panels sheathed with boards made of wood-based materials, which are attached to the supporting frame using screws or nails with constant spacing around the perimeter of the panel. A sufficiently load-carrying anchorage of wall panels to the foundations [6, 7] or the lower floor in multi-storey buildings [8, 9] is also an important factor to ensure their high horizontal load-carrying capacity and stiffness. However, it should be noted that the horizontal stiffness of wall panels is also influenced by some other factors that are not usually taken into account in simplified standard calculations and that can only be captured by a more complex rheological model expressed as a system of fictitious springs [10]. Several advanced analytical models have been developed to account for this, including models that incorporate the nonlinear behaviour of compliant connections [11, 12]. The four-part handbook deals with the design of horizontally loaded wall panels using the plastic design method. The comprehensive series of publications presents design principles for horizontal stabilization of wall panels [13], the possibilities of joints and anchoring devices [14], specifies the theoretical assumptions and foundations of the plastic design method [15] and deals with design in the ultimate limit state [16]. Since the horizontal stiffness and load-carrying capacity of wall panels are mainly influenced by the method of construction, and mechanical properties of the connections between the sheathing and the supporting frame, some research works also focused specifically on these connections [17, 18].

In addition to the traditional one-sided sheathing with OSB boards, experimental tests were carried out on wall panels with sheathing made of other materials [19,20,21], or wall panels with two-sided sheathing [22, 23], or wall panels with sheathing achieving higher fire resistance [24]. Experimental studies on entire modules of modular timber buildings also exist [9, 25]. Due to the generally high ductility of timber structures, lightweight frame wall panels can be conditionally applicable in areas with high seismicity. Experimental measurements on cyclically and dynamically loaded wall panels are carried out in addition to monotonic quasi-static tests [20, 23], often using reinforced frame structures [26] or with special use of composite materials [27]. The construction of shear wall panels is not necessarily limited to the use of a lightweight frame structure made of KVH (Konstruktionsvollholz) beams. Recently, shear walls made of CLT (cross laminated timber) panels [28, 29] or frames using glued I-beams [30] have been under investigation. Experimental data are used for the calibration and validation of advanced numerical models [6, 21, 31, 32], or simplified practical approaches such as the use of a fictitious diagonal with a spring at the end, which simulates the horizontal stiffness of the entire wall panel [33, 34].

Description of testing specimens of wall panels

Six specimens of timber wall panels with a length of 5 m and a height of 3 m were manufactured for the experimental testing. Three different types of wall panels were tested, each differing in the method of OSB sheathing. Each panel consisted of nine vertical studs with an axial spacing of 625 mm (respectively, 595 mm for the outer studs) regarding to the dimensions of the OSB boards. The panels were provided with continuous horizontal top and bottom plates, to which the studs were attached using screws. Spacer elements, placed 500 mm below the top plate between continuous studs, were included in the case of two specimens. These spacer elements served to attach the sheathing and keep a continuous shear flow in the panel, in accordance with the German standard for the design of timber structures, DIN 1052 [35]. All studs, top/bottom plates and spacers were made of KVH profiles with a cross section of 60/120 mm, strength grade C24 (S10). The one-sided sheathing consisted of EGGER OSB 3 boards with a thickness of 15 mm. Two wall panels were made with vertically oriented OSB boards of 1250/2500 mm, with one horizontal joint at the location of the spacer element—designation B01 (see Fig. 1). The other four wall panels were constructed with horizontally oriented OSB boards of 675/2500 mm, using tongue-and-groove horizontal joints (see Fig. 2). The horizontal tongue-and-groove joints were glued with D4 Profi waterproof polyurethane adhesive in two of these specimens—designation A02. No glue was used for the two remaining specimens—designation A01. The OSB boards were attached to the supporting frame made of KVH profiles using metal staples with a leg length of 50 mm and a rectangular cross section of 1.44/1.57 mm. The staple spacing was approximately 50 mm around the perimeter of the panel and 100 mm for the intermediate studs. The legs of the staples, with a minimum tensile strength of 900 MPa, are characterized by their high withdrawal capacity which is enhanced by the adhesive coating material. This coating is activated during application with a pneumatic stapler.

All materials were conditioned in a standard environment at (20 ± 2) °C and (65 ± 5)% relative humidity. The mean density of C24 structural timber was 435 kg/m³, the mean density of OSB sheathing was 620 kg/m³. The moisture content of the used timber materials was around 12%. The geometry of the wall panels, their profiles and materials, and the assembly methods were chosen based on requirements and habits of construction practice.

Experimental determination of behaviour of staple connections

Several additional tests were conducted to verify and compare the translational stiffness of the stapled connection between the OSB sheathing and the structural beam, as well as to provide input data for the numerical models [36, 37]. The specimens for the tensile tests were made of two KVH profiles with a cross section of 60/120 mm, which were connected on both sides with 15 mm thick OSB boards using staples. Various modifications of the stapled connection were tested, differing in the number of staples between the OSB boards and the beam. Modifications with twice one, twice two, twice four and twice seven staples between the OSB board and the beam on each side of the connection were manufactured. There were three test specimens for each modification. The protruding ends of both beams were attached using four M10 8.8 bolts to a hinge mechanism made of steel plates and pins. The plates of the hinge mechanism were then clamped into the jaws of the testing machine. The measured translational stiffness of the stapled connections was, therefore, influenced by the stiffness of the individual components (springs in series) of the entire loaded mechanism. This effect was later taken into account when evaluating the slip moduli (translational stiffnesses). The most cardinal impact of this effect was observed for the stiffest connection with twice seven staples on each side (with a difference of around 12%). On the contrary, the marginal impact of this effect was for the connection with twice one staple on each side (with a difference of up to 1.5%). The step-like behaviour observed in some specimens can be attributed to the progressive overcoming of adhesive friction as the coated staples were incrementally withdrawn from the timber during loading.

A summary of the test results for the stapled connections is given in Tables 1 and 2. Table 1 contains the maximum achieved forces of the connection F_max, their conversion to one staple F_max,1 (i.e., the maximum force of the connection F_max divided by the number of staples on one side of the connection n_s) for individual specimens and the average values ​​for each series F_max,AVG and F_max,1,AVG. Table 2 shows the individual measured slip moduli of the connection k_s, their conversion to one staple k_s,1 (i.e., the slip modulus of the connection just on one side, calculated as 2k_s, divided by the number of staples on one side of the connection n_s) for individual specimens and the average values ​​for each series k_s,AVG and k_s,1,AVG. The table also includes the average slip moduli of the connections after correction (i.e., taking into account the stiffness ratios of the entire loaded mechanism, including the steel hinges and timber beams) k_s,AVG,cor and k_s,1,AVG,cor. This correction was made by treating the system as a series of springs, where the total measured deformation was separated into contributions from the test fixture and from the stapled joint itself. The corrected slip modulus was then calculated by subtracting the deformation attributed to the fixture, based on the principle of springs connected in series (see Fig. 3).

The evaluated actual slip moduli of the specimens were obtained from the load–deformation curves (a set of measured points) using linear regression analysis. The load–deformation curves of stapled connections with twice seven staples is shown in Fig. 4. The part of the graph corresponding to a load between 10 and 40% of the estimated maximum load-carrying capacity was used for this linear approximation of the translational stiffness. The independent explanatory variable was the load and the dependent response variable was the deformation, with the minimum required correlation coefficient of 0.99. The slip modulus of the connection in the serviceability limit state k_ser,EC5 according to the EN 1995年1月1日 standard equation (1) is also given for the comparison:

where \(\rho\)_m is average mean density of timber beam and OSB board [kg/m³]; dis average cross-sectional dimension of staples [mm]; n_s is number of staples in the connection [mm].

The evaluation of the test results after correction shows that the calculated slip modulus in the serviceability limit state k_ser,EC5 according to the standard is comparable to the measured slip moduli for specimens with more fasteners (twice four and twice seven staples on each side of the connection). On the contrary, the measured slip moduli are lower than the standard estimate for connections with fewer fasteners (twice one and twice two staples on each side of the connection). This difference in stiffness can be attributed to the fact that any manufacturing imperfections are amplified as the number of fasteners (staples) in the connection decreases. However, it is important to emphasize that there is always a larger number of compliant fasteners with a certain spacing in real wall panels, where a stress redistribution is allowed, and the wall panel acts as one whole. While we acknowledge the significant variation observed in the measured slip moduli, especially for low-staple configurations, average values are still used here to enable meaningful comparison and to provide practical input for numerical modelling. Moreover, the values obtained for specimens with a higher number of staples showed good agreement with the theoretical slip modulus, which confirms the suitability of the standard value for design and simulation purposes.

The test arrangement and specimens after failure are shown in Figs. 5 and 6.

Experimental determination of behaviour of glued connections

Several additional tests were conducted to verify the shear load-carrying capacity and translational stiffness of the glued tongue-and-groove joint between two OSB boards, as well as to provide input data for the numerical models (see Fig. 7). The specimens were adapted to the testing machine and the loading procedure. The outer parts consisted of two OSB boards with a thickness of 15 mm, a height of 360 mm and a width of 140 mm, between which a spacer element made of a KVH profile with a cross section of 60/120 mm was attached. Both sides of the sheathing had protruding edges with a tongue on one edge and a groove on the other edge for connecting the intermediate middle part. The middle part was made in a similar way with double-sided sheathing using OSB boards with a thickness of 15 mm, a width of 675 mm, and a height of 360 mm high, where the horizontal spacer profiles were situated near the upper and lower edges. The middle part was inserted between the outer parts through the tongue-and-groove joint and it protruded 60 mm above the outer parts. The mutual connection was ensured by gluing with D4 Profi waterproof polyurethane adhesive. The total length of the glued shear connection was twice 300 mm on each side of the connection. The entire specimen was placed on the lower base of the testing machine. A spreader beam was placed on the upper edge of the protruding middle part, and it was loaded by compression.

A summary of the test results for the glued connections is given in Table 3. The table contains the maximum achieved forces of the connection F_max, their conversion to one meter of glued length for individual specimens and the average values ​​for each series F_max and F_max,AVG. The measured translational stiffnesses of the glued connection k_s, and their conversion to one meter of glued length are also shown in the table. The evaluated actual translational stiffnesses of the specimens were obtained from the load–deformation curves (a set of measured points) using linear regression analysis. The part of the graph corresponding to a load between 10 and 40% of the estimated maximum load-carrying capacity was used for this linear approximation of the translational stiffness. The independent explanatory variable was the load and the dependent response variable was the deformation, with the minimum required correlation coefficient of 0.99.

Experimental determination of horizontal load-carrying capacity and stiffness of wall panels

An auxiliary external steel structure was designed, manufactured and assembled for the experimental testing of the wall panels (see Fig. 8). The steel structure consisted of two parallel, interconnected frames, formed by a system of vertical and horizontal HEA profiles with diagonal rods made of L profiles. The structure was also equipped with a stiffened steel cross element just below the upper corner, providing an area for the placement and anchoring of the test device. The connections between the steel elements were designed as bolted. All steel structures were made of structural steel S235J0. The steel structure was anchored to an eight-ton reinforced concrete beam. This anchoring was implemented at six different locations (in the area of ​​the bottom horizontal plates near the vertical studs) using a steel connector and three M12 8.8 threaded rods with the HILTI HVU2 chemical anchor.

The tested wall panels were inserted between the two parallel steel frames on a laminated veneer lumber (LVL) plank with a width of 300 mm and thickness of 40 mm, which was anchored to the reinforced concrete beam with SFS Intec fasteners for timber-concrete connections. The densest concentration of SFS fasteners was placed in the area of ​​the expected highest tensile stress. The bottom horizontal plate was attached to the LVL plank using 48 (six screws in each space between the studs) VGZ Rothoblaas fully threaded screws for timber structures. The first pulled stud of the panel was additionally provided with a tie-down anchor (L-shaped steel connector with stiffeners), which was anchored to the reinforced concrete beam using one M20 8.8 threaded rod and HILTI HVU2 chemical anchor. The tie-down anchor was connected to the first stud using fourteen KOP screws with hexagon head (M12 8.8).

The tested panels were equipped with sixteen linear variable differential transformers (LVDT) for measuring displacements in selected areas of the wall panels and auxiliary structures (see Fig. 9). The individual LVDTs were typically arranged in pairs (measurement of horizontal and vertical deformations) at three locations on the bottom plate and three locations on the top plate (at the beginning, at the end and in the middle of the panel length). Three other LVDTs measured the vertical deformation of the bottom plate, the LVL plank and the reinforced concrete beam near the tie-down anchor. The last LVDT measured the horizontal displacement of the wall panel in the area of ​​load application.

The horizontal load was applied at the top corner of the wall panel, at the level of the top horizontal plate (see Fig. 10). The load was generated by the piston stroke of an Enerpac hydraulic cylinder with a capacity of 300 kN, operated manually by means of a hand pump. This method does not allow displacement-controlled loading, and the pressure in the hydraulic system is gradually increased in steps by the operator. During the initial phase of loading, the specimen typically undergoes settling, including the closing of gaps and local compliance at contact points, which results in temporary drops in the measured force. This effect may affect the shape of the load–deformation diagram. However, this is a known phenomenon inherent to force-controlled loading and does not compromise the evaluation of the main parameters such as stiffness, maximum load, or the overall trend of the structural response. A spreader steel plate was inserted between the piston and the contact surface of the loaded wall panel.

The loading procedure followed the standard for determining the racking strength and stiffness of timber frame wall panels EN 594 [38], with minor modifications depending on the specific geometry and test arrangement of the tested wall panels. The tested panels were larger (3.0 ×ばつ 5.0 m) than the standard 2.4 ×ばつ 2.4 m specimens, which is permitted according to Annex A1. The panels were tested without the application of a vertical load. All other aspects of the testing procedure followed the standard methodology. The monotonic loading rate was chosen with a constant increase of force-controlled loading (piston pressure), so that 90% of the maximum load was reached within (300 ± 120) s.

The horizontal racking stiffness of the panel R was evaluated according to Eq. (2) and also the maximum horizontal racking capacity (load) F_max was determined from the measured data [38]:

where F₀₄ is racking load of 0,4 F_max [N]; F₀₂ is racking load of 0,2 F_max [N]; u₀₄; u₀₂ are corresponding deformations [mm].

Calculation of the racking resistance according to Eurocode 5

The standard for the design of timber structures [1] provides two simplified methods for analysing wall diaphragms to calculate the horizontal load-carrying capacity (racking resistance) (Method A, Method B). Both simplified methods assume sufficient anchoring to the foundation to effectively prevent overturning or sliding (tie-down anchor), as well as in-plane stiffening to transfer horizontal shear stress (e.g., board materials, diagonal bracing, moment connections).

The racking capacity of the wall panel with vertically oriented OSB boards was calculated using the simplified method A (according to [1]) based on the panel geometry, the load-carrying capacity of metal staples and their spacing. The conditions for using the procedure from the standard were: keeping the continuity of shear flow around the perimeter of the wall panel and the presence of a tie-down anchor at its end. The characteristic value of the racking capacity is F_v,Rk = 69.99 kN. The design racking capacity using the modification factor k_mod = 1.0 and the partial factor for connections ɣ_M = 1.3 is then F_v,Rd = 53.54 kN.

Experimental testing of horizontal load-carrying capacity and stiffness

This section summarizes the results of experimental testing on large-scale wall panels with different OSB sheathing configurations: A01 (horizontally oriented sheathing without any glued joints), A02 (horizontally oriented sheathing with glued tongue-and-groove joints), and B01 (vertically oriented sheathing). Load–deformation diagrams of the wall panels (horizontal force vs. corrected horizontal deformation based on LVDT no. 15, adjusted using LVDT no. 11 and no. 10 to remove global slip and uplift effects) were recorded during testing to evaluate the horizontal load-carrying capacity (racking resistance) and horizontal stiffness. The graphs below show the load–deformation curves for tested wall panels in three different sheathing configurations (see Figs. 11, 12, 13). The maximum load and the corresponding deformation at failure for each test specimen can also be read from the graphs. The limit horizontal deformation of 20 mm (1/150 of the panel height) is highlighted with a black dashed line in the graphs. The characteristic load-carrying capacity based on the standard calculation is highlighted with a yellow dashed line in the graphs of configurations supported by the standard. The graphs are accompanied by a table summarizing the experimentally determined load-carrying capacities and stiffnesses—average values for each sheathing configuration (see Table 4). The first column specifies the sheathing of the wall panel. The second column shows the average maximum force achieved during testing F_max. The third column contains the average horizontal stiffness R, which was evaluated on the basis of the measured data according to Eq. (2).

Numerical modelling of horizontal load-carrying capacity and stiffness

Numerical models were created to take into account the boundary conditions and the real geometry of the tested wall panels and the auxiliary external steel structure (see Fig. 14). The numerical model also incorporates the anchorage of the wall panel to the lower reinforced concrete beam using the LVL plank, steel connectors with screws, and anchors. The connections between elements were modelled with eccentricities and slips.

The numerical model was created by combining 1D beam and 2D shell elements. Material and geometric nonlinearities with imperfections were included into the model. The numerical model also takes into account the contact nonlinearities and slip moduli of the connections (staples and other mechanical fasteners). The fineness of the finite element mesh was determined based on experience with similar numerical models.

The material model for the OSB boards was considered orthotropic, taking into account the strand orientation. The orthotropic behavior was implemented based on the general relations provided on the software’s official website [39]. Input parameters were adopted from the relevant standard, with values as follows: E₁ = 3.8 GPa (Young’s modulus in the strand direction); E₂ = 2.6 GPa (Young’s modulus perpendicular to the strand direction); and μ = 0.34 (Poisson’s ratio for transverse strain). The material model for the 1D beam elements was considered as isotropic linear elastic material (steel and timber). Some steel elements were modelled as 2D shell elements to reach more accurate analysis of specific areas. They were considered with a bilinear stress–strain curve with isotropic hardening.

The OSB sheathing was connected to the timber frame using rigid beam elements that replicated the actual spacing of the staples. At the end of each rigid beam, a nonlinear spring was introduced to represent the translational stiffness in two in-plane directions, based on additional experimental tests on stapled connections. This configuration of rigid links between the OSB panels and the timber frame was applied for all configurations. In modification A02, the translational stiffness of the glued tongue-and-groove joints between individual OSB boards was also simulated. OSB boards were interconnected in the model by rigid beam elements with nonlinear springs at their ends. The translational stiffness of these springs was derived from additional experimental tests on glued connections. In modification A01, the frictional interaction between adjacent OSB boards in the horizontal tongue-and-groove joints was represented by rigid beam elements with nonlinear springs at their ends. The tongue-and-groove contact was configured to transmit only compressive (contact) forces, while it did not carry any tensile load. The curves and data of nonlinear springs used in numerical models are shown in Fig. 15.

The steel structure and the wall panel were subjected to deformation-controlled loading via temperature, where the intermediate element stretches and thus the steel structure and wall panel interact with each other. The external and internal boundary conditions were modelled to be as close as possible to real connections.

The following figures illustrate the horizontal deformation of individual types of wall panels based on numerical models. The deformations correspond to a load of approximately 29 kN for the panel with horizontally oriented OSB boards and unglued tongue-and-groove joints A01 (see Fig. 16), and approximately 70 kN for both the panel with horizontally oriented OSB boards and glued tongue-and-groove joints A02 (see Fig. 17) and the panel with vertically oriented OSB boards B01 (see Fig. 18).

The value of 29 kN represents the maximum load actually reached during the experimental test of the A01 configuration, while the value of 70 kN corresponds to the characteristic racking resistance calculated according the Eurocode 5 standard. These values were selected to enable direct comparison between the measured performance of the tested panels and the expected normative requirements.

Figures 19, 20, 21 show comparisons of the load–deformation diagrams obtained experimentally and from numerical analysis for all tested wall panel configurations. Figure 19 presents the results for the panel with horizontally oriented OSB boards and unglued tongue-and-groove joints (A01), Fig. 20 for the panel with glued tongue-and-groove joints (A02), and Fig. 21 for the panel with vertically oriented OSB boards (B01).

In all cases, the numerical models demonstrate good agreement with the experimental measurements, particularly in the elastic range and around the serviceability limit state. Minor differences at higher load levels are attributable to localized effects and material nonlinearities not fully captured in the simplified material models. These results support the validity and applicability of the numerical modelling approach used in this study.

The experimental data presented in Table 4 show that wall panels with horizontally oriented OSB boards without glued tongue-and-groove joints (A01) exhibit the lowest horizontal load-carrying capacity and stiffness. Compared to panels with glued joints (A02), the stiffness is approximately 3.9 times lower, and compared to panels with vertically oriented OSB boards (B01), it is about 2.8 times lower. The highest horizontal stiffness is achieved by the A02 configuration, where OSB boards are glued along the tongue-and-groove joints, creating a structurally continuous and well-integrated panel. This construction method results in stiffness values even exceeding those of vertically oriented panels (the stiffness ratio of A02 to B01 is approximately 1.4:1).

Notably, the A01-2 specimen showed multiple sudden drops in the load–deformation diagram, which were not observed in A01-1. These drops are attributed to frictional slips between horizontally placed OSB boards without glued joints. As the applied load increases and exceeds local frictional resistance, boards shift relative to each other, resulting in sudden loss of load transfer capacity. This phenomenon is analogous to the step-like behaviour seen in the joint tests with stapled connections, where adhesive-coated staples exhibited staged withdrawal after frictional resistance was overcome.

The load-carrying capacity (racking resistance) of the A01 configuration is significantly below the calculated value for vertically oriented panels based on Eurocode 5 [1]. As shown in Figs. 10 and 11, the horizontal force transferred at the serviceability deformation limit (H/150 = 20 mm) does not exceed 20 kN. Such a sheathing approach is, therefore, deemed structurally inadequate and is not recommended for use in lightweight timber frame construction. Despite its use in practice by some construction companies, our findings highlight the considerable structural risk associated with this method.

By contrast, the achieved load-carrying capacities are significantly higher for wall panels with horizontally oriented OSB boards with glued tongue-and-groove joints A02 and with vertically oriented OSB boards B01, because these structural configurations keep a continuous shear flow in the panel. It ensures the interaction of all panel components with regard to horizontal load-carrying capacity and stiffness. Total load-carrying capacities at failure are above 100 kN. As shown in Figs. 12 and 13, the transferred horizontal force at the limit horizontal deformation (H/150 = 20 mm) exceeds 60 kN for the wall panel A02 and around 50 kN for the wall panel B01. It should be emphasized that the horizontal load within the serviceability limits will not exceed these values in the case of a statically correctly designed structural system and structural elements of a timber building in the system of a lightweight frame structure. In general, Eurocode 5 [1] recommends the use of vertically oriented sheathing for the needs of construction practice. The results of experimental testing also demonstrated the possibility of using horizontally oriented sheathing, where the individual sheathing boards must necessarily be glued into one rigid panel (diaphragm). This also depends on the properties of the adhesive and the quality of the glued tongue-and-groove joints.

It is important to note that the calculated reference value used for comparison is based on the Eurocode 5 approach for vertically oriented OSB boards. This method assumes a continuous, rigid diaphragm behaviour with proper anchorage and shear transfer through the sheathing. Although this method is not explicitly intended for glued horizontal OSB configurations, the structural behaviour of glued joints in A02 panels justifies using it as a practical and conservative design basis. Therefore, glued horizontal joints can be recommended for construction practice, whereas unglued horizontal joints must be avoided.

The analytical model is able to correctly capture the load-carrying capacity of the panel, but it is more beneficial to create a comprehensive numerical model for a correct description of its horizontal stiffness. The advantage of the numerical model is (after validation by a physical test) its ability to determine the redistribution of internal forces in connections and structural elements and it also provides insight 364 into the stress state of OSB boards and their deformation.

The numerical model is able to determine the stability behaviour of the panel and provides a very valuable insight into the behaviour of the diaphragm under vertical load. A calibrated numerical model is able (with relatively minor changes in load and boundary conditions) to provide information on the mechanical behaviour of the structure without the need for demanding experimental measurements. The numerical models showed a good agreement with the experimental data. The agreement of these two approaches for determination of the horizontal stiffness of the wall panel depends primarily on the input slip moduli for the numerical model.

This paper presents the research findings on the horizontal load-carrying capacity and stiffness of on-sided sheathed wall panels in timber buildings using the lightweight frame structure. This structural solution is one of the most widely used systems in the field of residential and civil construction, thanks to its use of timber materials, which contribute to low carbon footprints. The stiffness, stability and load-carrying capacity of these walls in the vertical and especially in the horizontal plane are primarily ensured by the sheathing.

The sheathing is typically designed from OSB boards, fastened with staples to a supporting frame of vertical studs and horizontal top/bottom plates. The OSB boards are usually laid vertically, and the horizontal load-capacity capacity of these walls can be determined relatively accurately by a simple calculation according to valid European standard [1] for this way of sheathing. However, the standard does not provide any procedure for determining the horizontal deformation and stiffness of the wall panels. It is quite common in the construction practice to encounter the horizontal orientation of OSB sheathing with glued or unglued tongue-and-groove joints. This method of laying OSB boards is not addressed in the standard; therefore, the authors of this paper decided to compare the horizontal load-carrying capacity and stiffness of wall panels using these different methods of laying OSB boards, based on experimental tests on large-scale specimens (see the description of the specimens in the previous text). The study demonstrated that wall panels with horizontally oriented OSB boards and glued tongue-and-groove joints exhibited significantly higher load-carrying capacity and stiffness compared to those without glued joints, as well as those with vertically oriented boards. The findings underline the importance of joint integrity, as glued joints contribute to maintaining continuous shear flow and enhancing the overall stiffness of the panel. Notably, the experimental results were compared with analytical calculations based on the European standard, revealing discrepancies, especially for wall panels with unglued tongue-and-groove joints.

The research incorporated computer simulations to create reliable numerical models for predicting the horizontal load-carrying capacity and stiffness of wall panels that would be further modified as needed. These simulations can be used, for example, to assess wall panels with different dimensions or with openings for doors and windows. The agreement between the experimental data and numerical models was promising and provided valuable insights for further development of design methods and more accurate predictions of the mechanical behaviour of wall panels in the system of the lightweight timber frame structure. In future research, the numerical models developed in this study will be further extended and applied to investigate additional aspects such as the effect of friction between adjacent OSB boards in panels without glued joints, the influence of staple withdrawal stiffness, and the progressive plastic behaviour of mechanical fasteners.

In conclusion, the results of this research highlight the importance of selecting the appropriate sheathing method to ensure optimal horizontal load-carrying capacity and stiffness in lightweight timber frame structures. The study also emphasizes the need for further validation of sheathing methods in design standards to improve the design and performance of these systems in practice.

The data generated or analysed during this research are available from the corresponding author upon reasonable request.

CLT:

Cross-laminated timber

KVH:

Konstruktionsvollholz

LVL:

Laminated veneer lumber

LVDT:

Linear variable differential transformer

OSB:

Oriented strand board

The authors thank to the Department of Structures and the Centre of Building Experiments and Diagnostics of the Faculty of Civil Engineering at the VSB—Technical University Ostrava for the technical and financial support.

This article has been produced with the financial support of the European Union under the REFRESH—Research Excellence For REgion Sustainability and High-tech Industries project number CZ.10.03.01/00/22_003/0000048 via the Operational Programme Just Transition.

Authors and Affiliations

1. Faculty of Civil Engineering, VSB – Technical University of Ostrava, Ludvíka Podéště 1875/17, 708 00, Ostrava, Czech Republic

   Pavel Dobeš, Antonín Lokaj, David Mikolášek, Marek Johanides & Petr Mynarčík

Contributions

Conceptualization, Pavel Dobeš and Antonín Lokaj; data curation, Pavel Dobeš, Marek Johanides, Petr Mynarčík; formal analysis, Pavel Dobeš and Antonín Lokaj; funding acquisition, Antonín Lokaj and Petr Mynarčík; investigation, Pavel Dobeš and Marek Johanides; methodology, Antonín Lokaj and Pavel Dobeš; project administration, Antonín Lokaj; resources, Antonín Lokaj and Pavel Dobeš; software, David Mikolášek; supervision, Antonín Lokaj; validation, Antonín Lokaj, Pavel Dobeš and Marek Johanides; visualization, Pavel Dobeš and Marek Johanides; writing—original draft, Pavel Dobeš; writing—review and editing, Antonín Lokaj.

Corresponding author

Correspondence to Pavel Dobeš.

Ethics approval and consent to participate

The authors confirm that this paper does not contain any studies with human participants or animals performed by any of the authors.

Competing interests

The authors confirm that there are no known conflicts of interest associated with this paper and there has been no significant financial support for this work that could have influenced its outcome.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions