A A A Experimental study on vibration response of wooden house façade to low-frequency outdoor sound Jinyu Liu 1 , Graduate School of Engineering, The University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan Naohisa Inoue 2 Maebashi Institute of Technology 460-1 Kamisadori, Maebashi-shi, Gunma 371-0816, Japan Tetsuya Sakuma 3 Graduate School of Engineering, The University of Tokyo 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan ABSTRACT Generally wooden house has low sound insulation performance for low-frequency outdoor noise due to the lightweight façade wall system. Moreover, the sound insulation performance is caused by the vibration responses of window and wall systems, that are strongly coupled with normal modes in a room. In order to clarify the complicated phenomena, field measurements of vibration characteristics are conducted for a test house with Japanese traditional wooden structure, under the cases with single/double window systems. As outdoor sound source, low-frequency band noise from 25 Hz to 160 Hz is emitted from a subwoofer, and vibration acceleration is measured on each element, such as window glazing, interior walls, floor, and ceiling. From the analysis of measured vibration transfer functions and indoor sound pressure distribution, the vibration characteristics of the whole coupled system are discussed, especially on the effect of different window systems. 1. INTRODUCTION Wooden house has a low sound insulation performance for low-frequency outdoor noise due to the lightweight façade wall system. Moreover, the sound insulation performance is caused by the vibration responses of window and wall systems, that are strongly coupled with normal modes in a room. In our previous work [1], airborne sound insulation measurements were performed in accordance with the low-frequency procedure in ISO 16283-3 [2], and the effect of soundproof measures, such as increasing the surface density of outer wall and installing double window system, was examined. The measured results suggested that soundproofing of façade wall is effective below 1 rika0626@g.ecc.u-tokyo.ac.jp 2 inoue@maebashi-it.ac.jp 3 sakuma@arch1.t.u-tokyo.ac.jp Sia inter noine 21-24 AUGUST SCOTTISH EVENT CAMPUS GLASGOW 40 Hz, the lowest eigenfrequency in the room, whereas the effect of double window system appears above the frequency. In order to clarify the whole mechanism of vibro-acoustic coupled system in the wooden house, more detailed experiments are needed on vibration response characteristics of each building element, especially double window system. Several studies of vibration response characteristics of building components are described in the literature. For lightweight wooden houses, low-frequency field measurements and simulations were performed in a typical Norwegian wooden house by aircraft noise, which investigating vibration characteristics of each building element and mechanisms for low-frequency fluid-structure interaction [3]. Then a series of countermeasures were tested by laboratory measurement, and full- scale field measurements were performed by aircraft noise and loudspeaker source [4]. Moreover, comparing with field measurements in a room with brick structure by sonic boom as sound source, it is shown that there could be different vibration characteristics between the European countries depending on building structure [5]. Regarding the vibration characteristics of window systems, field vibration measurements of single window in a test wooden house were conducted by impulsive low frequency sound source, and the window vibration tends to be significant at around 10 Hz due to the resonance with the air spring of the room [6]. Besides, regarding the mass-air-mass resonance phenomena of double window systems, vibration mode behaviors of the system were observed through a laboratory experiment [7]. However, the coupling effect of window vibration and room acoustic modes has not been fully discussed. In this paper, field vibration measurements are performed in a test wooden house under the cases with single and double window systems, where sound insulation measurements were already done [1]. As outdoor sound source, low-frequency band noise from 25 Hz to 160 Hz is emitted from a subwoofer. To obtain vibration distribution characteristics of façades, a number of vibration accelerators are arranged on each interior wall, and window elements including interior/exterior glazing for double window, and vibration transfer functions are measured from sound pressure at a reference point. From the analysis of measured results, the contributions of each outer wall and window to façade sound insulation are compared, and the vibration mechanism of double window system is also discussed. 2. EXPERIMENTAL SETUP 2.1. Mock Wooden House In the test wooden house as described in [8], field measurements of sound and vibration are conducted under two conditions with a single or double window system. The house was constructed by Japanese traditional timber framework method, and additional soundproof measures were conducted in the test room. For the inner walls and ceiling, two layers of hard gypsum board (12.5 mm thick) were put on the existing gypsum board (9.5 mm thick), which increasing the total surface density of the façade to about 55 kg/m 2 . The wooden floor structure is supported by steel studs at an interval of 0.9 m over the underfloor space with a height of about 0.6 m. The room has a sliding terrace window with aluminum sashes ( w = 1.6 m, h = 1.8 m). In Case 1, an ordinary window system with 5 mm thick single glazing is installed. In Case 2, a double window system is installed for soundproofing, which is composed of an exterior window (double glazing: 3 mm + A6 mm + 3 mm), an air layer with a thickness of 190 mm, and an interior window (single glazing: 5mm). 2.2. Measurement Setup The measurement system is shown in Figure 1(a). A subwoofer (Electro-Voice EKX18SP) is set on the ground at a distance of 8 m away from the center of the window in the direction of 45 degrees, emitting the pink noise in the one-third octave bands from 25 Hz to 160 Hz with an averaging time of 20 seconds. As reference, sound pressures are measured at two points outdoor and indoor, and vibration accelerations are simultaneously measured at four points while moving the pickups on the surfaces in the room. Partition wall Side wall (Near Src .) 0.4m d d 4 17 1 20 C d d d Corner door Acceleration pickup sensors 5 8 12 15 Inner wall Input y Side wall (Far Src .) 3 Room A Data recorder 18 10 11 x 45° 7 14 Window Front wall Wooden stud. 8m Src. 2 6 9 13 16 19 z d S Sash (Exterior) Sash (interior) x (a) (b) d =22.5 cm Src. d d Corner Corner Central Wooden stud. Wooden stud. d z z d =22.5 cm x (c) (d) y d =22.5 cm Figure 1: Measurement system (a) and arrangement of accelerometers: (b) on the front wall, (c) on the side walls; (d) on the partition wall. As shown in Figure 1(a), there are two 1/2-inch microphones, one of which is measured at position S with a distance of 1 m away from sound source as reference, the other one is measured at corner position C indoor with a distance of 0.4 m from each inner wall boundary. In vibration measurements, four built-in preamplifier type acceleration pickup sensors were placed on measuring surface by double-sided tape. The direction of sensors is perpendicular to measuring surface and facing inward, whereas for those positions measured on exterior window of double window system, the directions are facing outward. For each measurement, the six-channel signals of sound and vibration are recorded simultaneously. Vibration response measurement are performed on each room element, such as front wall, side walls, partition wall, ceiling, and floor. The measuring positions on the front wall are described in Figure 1(b). In order to clarify complicated vibration distribution characteristics of window system, ten points (No.5~No.9, No.12~No.16) on window glazing and two points (No. 10, No.11) on connection part of two sash are placed. In the same way, the measuring points are also placed on both exterior and interior windows in double window system. In addition, for measuring points on the room elements, such as side walls, ceiling and floor, the central and four corner positions are determined and the corner position is placed with a distance of 22.5 cm from the inner wall boundaries. For example, the arrangement of measuring points on the side walls is shown in Figure 1(c). For the partition wall, except four corner positions, four more positions are placed on wooden studs, as shown in Figure 1(d). 2.3. Data Analysis Method Transfer function complex value of sound and vibration are calculated as functions of the frequency f , according to Equation 1 and Equation 2. " !" ($) 𝑇𝐹 ! (f) = " #$% ($) (1) '($) 𝑇𝐹 & (f) = " #$% ($) (2) where, 𝑝 () (f) is the complex sound pressure at C point indoor and a(f) is the complex vibration acceleration at each measuring position. As reference, 𝑝 *+, (f) is the complex sound pressure at S point outdoor. For each time of the six-channel signals recorded, the total length 20 s of data is divided into 20 parts, and for each part with a duration time of 1 s, the six-channel signals are performed by Short- time Fourier transform with Hanning window and overlap ratio as 0.9. Then, complex transfer functions for each part are calculated. Finally, transfer functions 𝑇𝐹 & (𝑓) and 𝑇𝐹 ! (𝑓) are obtained by the arithmetically averaging of all 20 parts results. Relative sound and vibration levels are calculated from the complex amplitude of transfer functions, respectively. Moreover, phase characteristics are calculated from the angle of the vibration transfer function. 3. RESULTS AND DISCUSSION 3.1. Vibration Response Characteristics of Each Façade The energy-average value of relative vibration level on each room element in Case 1 is calculated and the results are shown in Figure 2. Moreover, the value on front wall in Case 2 is also calculated and the measured vibrations on interior window are used. It is seen that compared with those in other part, the vibration levels of f r ont w al l a nd side wall (Near Src. ), especially front wall, are more remarkable at all frequency bands. Around 50 Hz, the vibration level of front wall rises in Case 2, which is corresponding with the r esonance frequency ( f rmd = 52.7 Hz) of double window system. Besides, for floor vibration ch aracteristics, there are low vibration response at low frequency ranges below 60 Hz. It is considered to be due to whole bending stiffness of floor is increased by the installation of steel studs under floor. -10 31.5 Hz 40 Hz 50 Hz 63 Hz 80 Hz 100 Hz Relative Vibration Level [dB] -20 -30 -40 -50 -60 -70 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 Side wall (Near Src. ) Side wall (Far Src. ) Ceiling Floor Partition wall Front wall Figure 2: The energy-average relative vibration levels on each room element for Case 1 and on Front wall in Case 2 front wall for Case 2. Figure 3 shows that relative vibration levels at the positions where large vibration appears under two cases. Generally, it is seen that below 40 Hz, the vibration levels of windows are considered to be dominant. Moreover, it is shown that there are more complicated vibration level distributions for single window according to the results at central and corner positions. For vibration of interior wall, there are similarly vibration response characteristics under two window cases. It is worth noting that, vibration at the central position of side wall (Near Src. ) tend to be significant near 36 Hz, even more than the window vibrations in Case 2, which could be considered as the eigenfrequency of the wall plate. For front wall, vibration level at corner position is much lower than those results at wooden stud. positions. It is considered that it could be due to the restraint effect of interior wall boundaries. -10 Relative Vibration Level [dB] Glazing -20 -30 -40 -50 Front wall Side wall -60 (a) -70 -10 Relative Vibration Level [dB] Glazing (b) Side wall -20 -30 -40 -50 -60 Front wall -70 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 Side wall (Near Src. ) Central Wall wooden stud. (No.18) Win. Central (No.14) Win. Corner (No.15) Wall Corner (No.20) Figure 3: The relative vibration levels on front wall and side wall (Near Src. ):(a) Case 1; (b) Case 2. 3.2. Relative Sound Pressure Level Figure 4 shows the relative sound pressure levels in two cases. It is seen that the indoor sound pressure levels increase, at those frequencies roughly corresponding with room natural frequencies. Below 35 Hz, the levels in Case 1 are significantly larger than in Case 2, which is due to be strongly affected by window vibrations. Around 58 Hz, the levels in Case 2 increase, which is due to resonance phenomenon of double window system. Besides, at the frequency ranges near f (0,1,0) and near f (0,1,1) , it is shown that there is a significant difference between two cases, the mechanisms for which could not be explained so far. f (1,0,0) f (0,1,0) f (0,0,1) f (1,1,0) f (1,0,1) f (2,0,0) f (0,1,1) f (1,1,1) Relative Sound Pressure Level [dB] -10 -20 -30 -40 -50 Case 1 Case 2 -60 -70 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 Figure 4: The relative sound pressure level results in Case 1 and Case 2. 3.3. Vibration Response Characteristics Distribution of Single/Double Window System Figure 5 shows that the energy-average value of relative vibration levels and standard deviation results calculated by all ten points of the window glazing under two cases. In Case 2, the results both exterior and interior window in double window system are calculated respectively. At the frequency ranges from 31.5 Hz to 40 Hz, it is seen that the relative vibration levels in Case 1 are remarkable, accompanied with large fluctuated standard deviation up to 10 dB. In Case 2, vibration levels both exterior and interior windows rise near the f rmd . Moreover, the vibration deviation results in Case 2 are small, less than around 4 dB at the frequencies from 55 Hz to 80 Hz. -10 Relative Vibration Level [dB] Case 1 Int. window in Case 2 Ext. window in Case 2 -20 -30 -40 40 Hz 50 Hz 63 Hz 80 Hz 100 Hz 31.5 Hz -50 10 Standard Deviation [dB] 8 6 4 2 0 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 Figure 5: The vibration response characteristics distribution for different window systems: (a) energy-average value; (b) standard deviation. Relative Vibration 39 Hz 35 Hz -10 -20 -30 -40 -50 -60 -70 Level [dB] Wall Wall Window Window Wall Window Window Relative Phase [pi] 1 0.5 0 -0.5 -1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Point number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Point number -10 Relative Vibration 58 Hz 33 Hz -20 -30 -40 -50 -60 -70 Level [dB] Relative Phase [pi] 1 0.5 0 -0.5 -1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Point number 1 2 Point number 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Figure 6: The vibration response characteristics distribution on front wall for different window systems: green line represents Case 1; blue/red line represents the interior/exterior window in Case 2. The point number represents measuring positions as depicted in Figure 1(b). In order to investigate phase characteristics distribution of different window systems, two representative frequencies are selected for each case, which are 35 Hz and 39 Hz in Case 1 and 33 Hz and 58 Hz in Case 2. The results are shown in Figure 6. Moreover, the vibration characteristics distributions of all measuring positions on window glazing through all frequency ranges from 31.5 Hz to 100 Hz are described in Figure 7 for Case 1 and in Figure 8 for Case 2. In Figure 6, generally relative vibration levels on window element are more significant than those on interior wall. Regarding window vibration characteristics, in Case 1, the same trend appears in the two glazing of single window at both 35 Hz and 39 Hz. Especially, at 39 Hz the phase difference is 𝜋 between eight corners and central positions on window glazing. In Case 2, according to the facing direction of sensors on interior and exterior window are opposite, it is shown that at 33 Hz, phase difference between exterior and interior window is almost zero, whereas for each window, the phase difference is 𝜋 between the corner and central positions. At 58 Hz, which is close to the f rmd , the phase difference at all points on interior or exterior window is almost zero, whereas the phase difference between interior and exterior windows is 𝜋 . Moreover, in Figure 8, it is worth noting that for Case 2, at the frequency ranges from 58 Hz to 80 Hz, the same phase distribution as that of 58 Hz appears. It may indicate that indoor sound wave would give surface excitation on window at around frequency ranges near the room normal mode f (0,1,0) . 4. CONCLUSIONS For the field measurement results in our previous work, in order to clarify the mechanism of sound and vibration coupled system of the house, experimental studies were performed on vibration response characteristics of whole building elements. Through the measured sound and vibration transfer function results, vibration response characteristics of the whole room, especially the double window system, were basically discussed. The energy-average level of each element vibration is calculated, such as front wall, side walls, ceiling, and floor. It is found that relative vibration levels on front wall and the side wall near sound source are more remarkable, compared with other building elements. Moreover, through relative vibration levels of window glazing under single/double window systems, it is shown that single window vibration is remarkable below 40 Hz, and at frequencies near f rmd , vibration of double window tends to rise significantly. Furthermore, vibration phase characteristics of window system, especially double window system are clarified. Especially, at f rmd of double window, it is found that there is almost no phase difference at all measuring points on the interior or exterior window, whereas the phase difference is 𝜋 between interior and exterior windows. 5. REFERENCES 1. Liu, J. & Inoue, N. & Sakuma, T. Experimental study on low-frequency sound insulation of wooden house façades. Proceedings of INTER-NOISE 2020 , Seoul, Korea, 23-26 August 2020. 2. ISO 16283-3:2016, Field measurement of sound insulation in buildings and of building elements, Part 3: Façade sound insulation. 3. Norén-Cosgriff, K. & Løvholt, F. & Brekke, A., et al . Countermeasures against noise and vibrations in lightweight wooden buildings caused by outdoor sources with strong low frequency components. Noise Control Engineering Journal, 64(6) , 737–752 (2016). 4. Løvholt, F. & Madshus C. & Norén-Cosgriff, K. Analysis of low frequency sound and sound induced vibration in a Norwegian wooden building. Noise Control Engineering Journal, 59(4) , 383–396 (2011). 5. Norén-Cosgriff, K. & Belyaev, I. & Løvholt, F. Building vibration induced by sonic boom – field test in Russia. Journal of Applied acoustics, 185 (2022). 6. Doi, T. & Iwanaga, K. & Jimbo, M. Experimental approach on natural frequency of window vibration induced by low frequency sounds. Proceedings of INTER-NOISE 2016 , Hamburg, Germany, 21-24 August 2016. 7. Pietrzko, S. J. Vibrations of double wall structures around mass-air-mass resonance. Proceedings of the 137 th Meeting of the Acoustical Society of America and the 2 nd Convention of the European Acoustics Association, Berlin, Germany, 14-19 March 1999. 8. Liu, J. & Inoue, N. & Sakuma, T. Validation of the low-frequency procedure for field measurement of façade sound insulation. Buildings, 11(547) , 15p. (2021). -10 Relative Vibration Level [dB] -20 -30 -40 -50 -60 40 Hz 50 Hz 63 Hz 80 Hz 100 Hz 31.5 Hz -70 1 Relative Phase [pi] 0.5 0 -0.5 -1 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 d =22.5 cm d d d 5 8 12 15 5 12 7 14 6 13 7 14 15 8 16 Src. 9 16 6 9 13 z d x Sash (Ext.) Sash (Int.) Figure 7: The measured vibration response characteristics distribution on window glazing in Case 1. -10 Relative Vibration Level [dB] -20 -30 -40 -50 -60 40 Hz 50 Hz 63 Hz 80 Hz 100 Hz 31.5 Hz -70 1 Relative Phase [pi] 0.5 0 -0.5 -1 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 (a) -10 Relative Vibration Level [dB] -20 -30 -40 -50 -60 40 Hz 50 Hz 63 Hz 80 Hz 100 Hz 31.5 Hz -70 1 Relative Phase [pi] 0.5 0 -0.5 -1 25 35 45 55 65 75 85 95 105 Frequency [Hz] 115 d =22.5 cm d d d (b) 5 8 12 15 12 5 7 14 6 13 14 7 15 8 16 Src. 16 9 6 9 13 z d x Sash (Ext.) Sash (Int.) Figure 8: The measured vibration response characteristics distribution on window glazing in Case 2: (a) exterior window; (b) interior window. Previous Paper 523 of 769 Next