Seismic Response of Two-Storey Light Timber Frame Structure with OSB Sheathing and GFB Cladding fastened with Staples: Dataset of Shaking Table Tests

This dataset contains data from experimental shake table tests conducted on a two-storey timber structure with light timber-frame wall (LTFW) diaphragms, involving linear sweep, white noise, impulse, harmonic single-frequency, and seismic excitations (see 'Load_Protocols_v1.0.0.pdf'). The experiments were carried out by the Institute of Steel Construction (STB) and the Chair of Structural Analysis and Dynamics (LBB). The institutes cooperate within the Center for Wind and Earthquake Engineering (CWE) at RWTH Aachen University. For the tests, the biaxial shaking table at the RWTHDynLab was used. The experimental campaign was developed to gain a better understanding of the structural behaviour of LTFWs under seismic loads. The lateral bracing system was tested in two construction stages: (i) using LTFWs with OSB sheathing, and (ii) using LTFWs with OSB sheathing and an additional cladding layer of gypsum fibre board (GFB). The test structure was subjected to impulse, noise, sweep, harmonic single-frequency, and earthquake excitations with different spectral properties. The main objectives of the test campaign were: Identification of the modal properties of the test structure with and without additional cladding, Measurement of the floor response in terms of acceleration and displacement, Determination of the maximum capacity and failure mode. Test structure: The test structure consisted of a two-storey timber structure. For the bracing system LTFWs with single-sided sheathing were used. Two stages of the specimen were tested: (i) the LTFWs had only the OSB sheathing layer, and (ii) an additional GFB cladding layer was applied on top of the sheathing layer to the LTFWs. For the panel material OSB/3 sheathing panels with a thickness of 15 mm from Egger Holzwerkstoffe Wismar GmbH & Co. KG and GFB cladding panels with a thickness of 18 mm from James Hardie Europe GmbH (fermacell®) were used. The butt joints of the cladding panels were glued together using Greenline adhesive from fermacell®. Both sheathing and cladding panels were fastened with staples into the framing – and so contributing to the sheathing-to-framing connection. The fasteners used were resin-coated staples from ITW Befestigungssysteme GmbH (Haubold). The staple type was KG 764 with a diameter of 1.53 mm, crown width of 11,3 mm and a length of 64 mm. On the 2nd storey of the structure, the sheathing and cladding were each fastened with a staple spacing of 240 mm. On the 1st storey, the staple spacing was reduced to 150 mm for each sheathing and cladding, to account for higher horizontal forces. The staples were all arranged at an angle of 30°. The floors consisted of CLT with three layers (40-40-40 mm) of strength class C24 produced by Pfeifer Group. The tensile forces between the walls of the 1st and 2nd storey were transferred using tension straps (four in total) that connect the perimeter studs of the vertically aligned walls in excitation direction to each other. To attach the tension straps, ZYKT69 connectors from Simpson Strong-Tie with ø8/300 mm screws were used. For this purpose, seven fasteners were attached per stud. The tension straps measure 780/180/3 mm. The test specimen was anchored to the shaking table using specially developed anchors that can transfer both horizontal shear forces and vertical tensile forces. The anchors were fastened by means of diagonal screws into the perimeter studs of the walls in loading direction of the 1st storey – 17 ZYKT69 connectors from Simpson Strong-Tie with ø8/300 mm fully threaded screws were used for this purpose. In addition, CCS 50/110 multifunction connectors from Simpson Strong-Tie were used, each of which is screwed into the perimeter studs with eight ø6/110 mm screws. The multifunction connectors have an M20 thread, which was used to connect the studs firmly to the anchors with a threaded rod in order to transfer any offset moments. The shaking table has a hole pattern of 400 x 400 mm with M24 internal threads. Since the existing hole pattern and threads of the shaking table were not compatible with the existing anchors, steel anchor plates measuring 625 x 600 x 50 mm were manufactured. These anchor plates are equipped with five M20 internal threads, which were used to fasten the anchors. There are four holes without threads in the corners of each anchor plate, which were used to fix the plate to the shaking table using M24 screws. Two BTALU505 beams from Simpson Strong-Tie were used to transfer shear forces between the 1st and 2nd storey. These connected the 1st CLT floor slab and the two shear walls in excitation direction (W1 and W3) of the 2nd storey. The connectors were placed in the middle of the bottom rail to avoid tensile forces. The BTALU505 was connected through its flange into the 1st CLT floor slab using 100 CSA (ø5/80 mm) screws. Ten drift bolts ø12/180 mm were used to connect the bottom rail and the BTALU505. The walls perpendicular to the direction of excitation (W2 and W4) were structurally connected with four pairs of fully threaded screws (ø8/160 mm) between the bottom rail and the 1st CLT floor slab. These were screwed diagonally from both sides of the bottom rail into the floor slab. The floor slabs were connected to the walls below using a circumferential diagonal screw connection with two fully threaded screws, which were offset by 15 mm. For the 1st floor, ø10 mm screws with a length of 320 mm and a maximum spacing of 280 mm and on the 2nd floor, ø10 mm screws with a length of 300 mm and a maximum spacing of 330 mm were used. The wall joints (between W1/W3 and W2/W4) were structurally connected by five fully threaded screws (ø12/300 mm) distributed over the height. On all test days, the timber moisture content was measured in all perimeter studs of walls W1 and W3 of the 1st and 2nd floor. Each stud was measured at three positions (top, centre, and bottom). The mean moisture content of these measurements for each test day were the following values: Date Mean moisture content of timber members 08.10.2024 16.5 % 11.10.2024 15.9 % 15.10.2024 15.7 % 22.10.2024 15.8 % 29.10.2024 15.3 % 04.11.2024 15.1 % 05.11.2024 15.1 % The dimensions of the test specimen were: 2.57 m in length, 2.57 m in width and 3.84 m in total height (1st storey: 1.92 m; 2nd storey: 1.92 m). Four large steel I-sections, each with a dead weight of 1700 kg, were attached to the specimen as masses and secured by U-Profiles – two on the 1st and two on the 2nd floor. An illustration of the test specimen is depicted in 'Test_Specimen_Sketch_v1.0.0.pdf'. Test setup: The biaxial shaking table specifications are: Table size: 3.0x3.0 m Max. specimen mass: 10 t Max. overturning moment: 30 m t Max. actuator stroke: +/- 250 mm Max. table velocity: +/- 1 m/s at rated load Max. table acceleration: +/- 1g at rated load Test frequency: 0 to 50 Hz The instrumentation scheme of the test setup consisted of accelerometers and displacement transducers, measuring the excitation provided by the shaking table and the response of the structure. Regarding the global response of the test structure, the recordings of the accelerometers and displacement transducers indicated in the uploaded file 'Instrumentation_Scheme_v1.0.0.pdf' are provided. Two data acquisition systems (DAQ) were used simultaneously and later synchronised by cross-correlation of sensors a10s and a30. The properties of the 3g accelerometers (a10s, a1 to a4) are: Type: M3701-series Manufacturer: PCB Piezotronics, Inc. Measurement range: +/- 3g Frequency range (+/- 5%): 0-100 Hz DAQ system 1 The properties of the 20g accelerometers (a10, a20, a30) are: Type: M3701-series Manufacturer: PCB Piezotronics, Inc. Measurement range: +/- 20g Frequency range ( +/- 5%): 0-100 Hz DAQ system 2 The properties of the displacement tranducers are: Type: LZW-M-500 Manufacturer: WayCon Positionsmesstechnik GmbH Measurement range: +/- 250 mm Linearity: +/- 0.05% Repeatability: 0.01 mm Displacement force: ≤15 N Displacement speed: ≤5 m/s DAQ system 2 Post processing: The data were recorded by both DAQ systems with a sampling frequency of 500 Hz. The post-processing includes the following steps: Due to occasional hardware-related spikes at the beginning of the measurements, the first second (500 samples) is removed. Time-dependent trends in all acceleration signals are eliminated. An offset correction is applied to all measurements using the mean value of the first half second (250 samples). The acceleration signals are filtered with a 3rd-order Butterworth filter with a passband of 1 Hz to 50 Hz. The measurements from both DAQ systems are synchronised using the cross-correlation of the shake table acceleration signals measured with a10s and a30. The impulse and earthquake signals are then limited to the relevant response duration. After trimming, the signals are resampled to 500 Hz and synchronised to a common time array. For the earthquake signals, the shaking table iteratively increases the applied acceleration to match the target ground acceleration. To avoid redundant data, only the first iteration in which the target excitation is successfully reached is included in the dataset. The complete testing history, including iterations that were not incorporated into the dataset, is documented in the load protocol file 'Load_Protocols_v1.0.0.xlsx', where excluded measurements are indicated in grey. Files: Data_v1.0.0.zip Contains all data files according to the load protocols. The experimental data is provided as .csv files for each load protocol. Instrumentation_Scheme_v1.0.0.pdf .pdf file illustrating the sensor placements on the test structure. Load_Protocols_v1.0.0.pdf / .xlsx .pdf and .xlsx files listing all load protocols applied to the structure. Shake_Table.pdf Photo of the shaking table without any specimen. Test_Structure_v1.0.0.pdf Photo of the shaking table including the test structure. Test_Structure_Sketch_v1.0.0.pdf .pdf file illustrating the test structure. Time_Histories_v1.0.0.pdf .pdf file including plots of the measurement data. File format of the datasets: The data is stored in .csv files, where each file contains the following columns (see also 'Instrumentation_Scheme_v1.0.0.pdf'): Time: Time in seconds since the start of the test (time step equals 0.002 s). W1: Displacement signal measured in mm on the shaking table in x-direction (Axis C-C, centre right). W2: Displacement signal measured in mm on the first floor in x-direction (Axis B-B, centre right). W3: Displacement signal measured in mm on the second floor in x-direction (Axis A-A, centre right). D1: Strain signal measured in με in anchorage 1 (Axis C-C). D2: Strain signal measured in με in anchorage 2 (Axis C-C). D3: Strain signal measured in με in anchorage 3 (Axis C-C). D4: Strain signal measured in με in anchorage 4 (Axis C-C). D5: Strain signal measured in με in tension strap 1 (Axis B-B). D6: Strain signal measured in με in tension strap 2 (Axis B-B). D7: Strain signal measured in με in tension strap 3 (Axis B-B). D8: Strain signal measured in με in tension strap 4 (Axis B-B). a10: Acceleration signal measured in m/s2 on the shaking table in y-direction (Axis C-C, upper left). a20: Acceleration signal measured in m/s2 on the shaking table in y-direction (Axis C-C, upper right). a30: Acceleration signal measured in m/s2 on the shaking table in x-direction (Axis C-C, centre right). a10s: Acceleration signal measured in m/s2 on the shaking table in x-direction (Axis C-C, centre right). a1x: Acceleration signal measured in m/s2 on the first floor in x-direction (Axis B-B, upper left). a2x: Acceleration signal measured in m/s2 on the first floor in x-direction (Axis B-B, lower right). a3x: Acceleration signal measured in m/s2 on the second floor in x-direction (Axis B-B, upper left). a4x: Acceleration signal measured in m/s2 on the second floor in x-direction (Axis B-B, lower right). a1y: Acceleration signal measured in m/s2 on the first floor in y-direction (Axis B-B, upper left). a3y: Acceleration signal measured in m/s2 on the second floor in y-direction (Axis B-B, upper left). a4y: Acceleration signal measured in m/s2 on the second floor in y-direction (Axis B-B, lower right) These data files can easily be imported using the pandas library in Python. For example by: import pandas as pd df = pd.read_csv('001_TP1_20241008_SW_N_x_0p13mm.csv') Time = df["Time"] W1 = df["W1"] W2 = df["W2"]W3 = df["W3"] Name convention: To ensure a well-organized and accessible dataset, the final csv data are named in the following name convention. Description Example # Global number of measurement corresponding to the data matrix. 001_002_ TP1 Test program corresponding to the data matrix to distinguish different research goals while testing. 001_TP1_014_TP2_ 20241008 Date of measurement: year, month, day. 001_TP1_20241008_014_TP2_20241011_ SW / CW Structure setup: only sheathing (SW) or with cladding (CW). 001_TP1_20241008_SW_022_TP3_20241015_CW_ N / S / I / E / H Type of excitation: noise (N), frequency sweep (S), impulse (I), earthquake (E), harmonic with single frequency (H). 001_TP1_20241008_SW_N_003_TP1_20241008_SW_S_004_TP1_20241008_SW_I_015_TP2_20241011_SW_E_103_TP5_20241029_CW_H_ x / xy Direction of excitation: uniaxial (x) and biaxial (xy). 001_TP1_20241008_SW_N_x_007_TP1_20241008_SW_S_xy_ Description Noise RMS of the applied table displacement. 001_TP1_20241008_SW_N_x_0p2mm Sweep Amplitude and frequency range of table displacement. 003_TP1_20241008_SW_S_x_0p2mm_1to30Hz Impulse Amplitude of table displacement. 004_TP1_20241008_SW_I_x_4mm Earthquake Earthquake name, iteration number*, absolute maximum of the table acceleration. 015_TP2_20241011_SW_E_x_Lefkada_I6_0p14g Harmonic Amplitude and frequency of table displacement. 103_TP5_20241029_CW_H_x_0p2mm_5p8Hz * The shaking table iteratively increases the applied acceleration to match the target ground acceleration. Contact: Please send your enquiries regarding the shaking table to [e-mail]. Further information can be found on our website. Usage/License: The data is licensed under CC BY-SA 4.0. If you have used our data and are publishing your work, we ask you to please reference both this dataset by its DOI, as well as any associated publications. Associated publications: L. Rauber; J.-W. Hoffmann; G. Balaskas; F. Kolisch; N. Lenzen; B. Hoffmeister and S. Klinkel, Seismic performance of light timber frame structures with sheathing and cladding: Experimental shake table tests, Engineering Structures, Volume 365, 2026. DOI: https://doi.org/10.1016/j.engstruct.2026.123248 Further publications: L. Rauber and B. Hoffmeister, Experimental Investigations of the Load-Carrying Behaviour of Light Timber-Frame Structures with Cladding : Test Structure 1 - Pushover and Vibration Behaviour, Aachen: RWTH Publications, 2025. DOI: 10.18154/RWTH-2025-01547 L. Rauber and B. Hoffmeister, Experimental Investigations of the Load-Carrying Behaviour of Light Timber-Framed Shear Walls with Cladding : Series 1 - Single-Sided Sheathing, Aachen: RWTH Publications, 2025. DOI: 10.18154/RWTH-2025-01547 L. Rauber and B. Hoffmeister, Experimental Investigations of the Load-Carrying Behaviour of Light Timber-Framed Shear Walls with Cladding : Series 2 - Double-Sided Sheathing, Aachen: RWTH Publications, 2025. DOI: 10.18154/RWTH-2025-02576 L. Rauber and B. Hoffmeister, “Lateral Stiffness of Light Timber-Framed Shear Walls with Cladding - Proposal for an Analytical Model,” in INTER Proceedings, 58 - 15 - 1, Istanbul, Turkey, 2025 L. Rauber and B. Hoffmeister, “Proposal for an Analytical Model of Light Timber-Framed Shear Walls with additional Cladding,” in INTER Proceedings Meeting, 57 - 15 - 2, Padova, Italy, 2024 Fundings: Deutsche Forschungsgemeinschaft - Grant number: INST 222/1161-1 FUGG. Einaxialer Schwingtisch für dynamische Modell- und Bauteilversuche. Deutsche Forschungsgemeinschaft - Grant number: INST 222/1403-1 FUGG. Multiaxiales Testsystem für hybride Echtzeitsimulationen. Bundesministerium für Ernährung und Landwirtschaft and Fachagentur Nachwachsende Rohstoffe - Grant number: 2221HV075B. HELEPOLIS – Holztafelbauweise mit hybrider Beplankung für den mehrgeschossigen Holzbau.