An application of a parametric transducer to measure acoustic

Preparing to load PDF file. please wait...

0 of 0
An application of a parametric transducer to measure acoustic

Transcript Of An application of a parametric transducer to measure acoustic

This is a repository copy of An application of a parametric transducer to measure acoustic absorption of a living green wall. White Rose Research Online URL for this paper: Version: Accepted Version
Article: Romanova, A., Horoshenkov, K.V. and Hurrell, A. (2019) An application of a parametric transducer to measure acoustic absorption of a living green wall. Applied Acoustics, 145. pp. 89-97. ISSN 0003-682X
Article available under the terms of the CC-BY-NC-ND licence (
Reuse This article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can’t change the article in any way or use it commercially. More information and the full terms of the licence here: Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [email protected] including the URL of the record and the reason for the withdrawal request.
[email protected]

1 An application of a parametric transducer to measure acoustic absorption 2 of a living green wall
3 Anna Romanova, Engineering & Science, University of Greenwich, Chatham, ME4 4TB, UK 4 Kirill V. Horoshenkov and Alistair Hurrell, Mechanical Engineering, University of Sheffield, 5 Sheffield, S1 3JD, UK 6 7 ABSTRACT 8 This work reports on a new method to measure the absorption coefficient of a Living Green 9 Wall (LGW) in-situ. A highly directional parametric transducer and acoustic intensity probe 10 are used to make this method robust against background noise and unwanted reflections. This 11 method is tested under controlled laboratory conditions and in-situ on a real green wall. The 12 methods is compared favourably against impedance tube data obtained for porous media which 13 properties are relatively easy to measure using a standard laboratory setup. The new method is 14 an alternative to the ISO354-2003 and CEN/TS 1793-5:2016 standard methods to measure 15 acoustic absorption of materials. 16 17 Keywords: urban noise, Living Green Wall (LGW), living plants, green wall, acoustic 18 absorption, acoustic measurement, parametric sound. 19 20 1. Introduction 21 There has been strong evidence that some living plants (foliage) are able to absorb a 22 considerable proportion of the energy in the incident sound wave. Some of this evidence were 23 obtained through the standard laboratory experiment [1], some were derived through the 24 application of a model (e.g. [2, 3]) and some were collected in-situ [4]. However, there is still 25 no valid theoretical model which is based on clear physics and which can explain the observed

26 absorption spectra in a sufficiently broad frequency range. The evidence assembled so far 27 suggest that three main mechanisms are responsible for the absorption of sound by living 28 plants. In the lower part of the audible frequency range (e.g. below 100-400 Hz) the thermal 29 dissipation mechanisms are important [5]. In the medium frequency (e.g. 400-2000 Hz) where 30 the acoustic wavelength is still much larger that the characteristic leaf dimension (e.g. 15 - 250 31 mm for typical plants [3]) the viscous dissipation is the prime absorption mechanism [2, 6]. In 32 the higher frequency range (e.g. above 1-2 kHz) where the acoustic wavelength becomes 33 comparable or smaller than the characteristic leaf dimension, the leaf vibration and multiple 34 scattering begin to contribute to the dissipation of the energy in the incident sound wave [3, 6]. 35 One obstacle to the development of a unified model for sound propagation through foliage is 36 the lack of reliable experimental data on the acoustic reflection/absorption coefficient spectra 37 for a representative range of acoustic frequencies and angles of incidence. These data can then 38 be related to the morphological characteristics of plants which can be directly measured so that 39 a robust model can be developed and tested through a reliable experiment. An apparent lack of 40 data on the acoustic reflection/absorption coefficient spectra for plants can be explained by the 41 difficulties in measuring the absorption by plants in the laboratory or in-situ. This difficulty in 42 laboratory conditions relates to the standard ISO 354 test [7] that requires 10 m2 area of living 43 plant or LGW specimen transported and installed in a reverberation chamber which is a rather 44 cumbersome and expensive procedure. The alternative, ISO 10534-2 test [8] does not allow for 45 a large enough LGW specimen to be tested in a broad enough frequency range. The difficulty 46 of measuring the absorption of LGW or individual living plants in-situ is a lack of reliable 47 standard methods for measuring the absorption of complex surface geometries such as plants 48 and the strong influence of the ground from which these plants are grown. The BS 1793-5:2016 49 [9] method relies on an omni-directional source and microphones. As a result, it suffers from 50 the interference between the sound reflected from the LGW, its edges and the ground. It is also

51 recommended only for flat, homogeneous samples so that its application to volumetric 52 absorbers such as living plants is questionable. 53 The aim of this work is to apply and validate a method which is able to measure the acoustic 54 absorption of a large specimen of a vertical placed living plant in a broad frequency range 55 which is representative of the spectrum of noise emitted by traffic and other common sources 56 of noise. This method requires a parametric transducer and intensity probe which sensitivities 57 are highly directional. In comparison with the BS 1793-5:2016 method [9], the method 58 proposed in this paper is less prone to the effect of the ground reflection or to the edge effects 59 and it can be used either in a laboratory conditions or in-situ. Laboratory applications of 60 parametric transducer have been reported before to measure the complex reflection coefficient 61 of flat material samples of limited dimensions [10,11] and sonic crystals [12] at normal and 62 oblique angles of incidence. In this respect, the novelty of the parametric transducer method 63 used in this paper is three-fold. Firstly, this method is applied to measure the absorption of a 64 green wall which surface is far more complicated. Secondly, we use the sound intensity probe 65 and signal deconvolution which makes this method particularly resistant to environmental 66 noise. Thirdly, this method is now applied outdoors to a realistic section of a green wall which 67 is typical to the conditions under which the acoustic absorption of green walls need to be 68 measured. 69 The paper is organised in the following manner. Section 2 describes the design of the Living 70 Green Wall’s (LGW) used in the experiments. Section 3 describes the experimental setup and 71 specimen characterisation procedures. Section 4 presents the results. Section 5 draws 72 conclusions. 73 74

75 2. Green wall arrangement 76 LGW module system for this work was provided by ANS Group Global Ltd - Living Wall & 77 Green Roof Specialist company. The wall is arranged in the form of a rectangular heavy duty 78 plastic modules which measure 100 mm deep, 250 mm wide and 500 mm tall with 14 79 compartments for plants (7 compartments tall and 2 compartments wide as shown in Figure 1 80 - right). All modules are identical and have a special hook catchment at the back which allows 81 the modules to be hung on the wall. There is a hood at the back to allow for water pipeline 82 installation and to provide a click-in system for the module placed on top. There are trenches 83 at both sides of the module to allow for firm fixing with screws. In total 8 modules and 96 84 plants are required to form a 1 m² of the wall. On average, when watered 1 m² of green wall 85 section weighs 72 kg. 86 The modules are cladded on to the wooden rails that are firmly attached to the wall and/or 87 facade. In between of the wall and the rail a specialised waterproof membrane is stitched to 88 protect the building wall from excess water and damp (see Figure 2). Advanced green wall 89 options offer wireless wall moisture control with automated on/off water supply systems. When 90 constructing the wall, the modules may come on site pre-planted, or alternatively planting can 91 be done on site. The choice of plants for the Living Green Wall (LGW) is down to the 92 designer’s preference. However, factors such as the south or north side facing building wall, 93 average temperature, humidity, average rainfall and wind are normally taken into account.
94 95 Figure 1. Living Green Wall module, with plants and empty (ANS Group Global Ltd).













97 98

99 Figure 2. Living Green Wall module installation front and side view (ANS Group Global Ltd).


101 3. Experimental setup

102 3.1. Acoustic equipment

103 An intensity probe, Brüel & Kjær, type 4197 [13] with Brüel & Kjær NEXUS conditioning

104 amplifier type 2690 and parametric transducer, a directional loudspeaker HSS-3000 Emitter

105 [14] with HSS-3000 amplifier were used in the reported experiments. The intensity probe was

106 firmly attached to a telescopic tripod and placed at a height of 0.9 m and 1.7 m away from the

107 measured surface. The orientation of the intensity probe with respect to the wall was

108 perpendicular as shown in Figure 3. The directional loudspeaker was also attached to a tripod

109 and it was placed 4 m away from the wall. The line connecting the centre point of the directional

110 loudspeaker and the middle of the intensity probe was set perpendicular to the wall as shown

111 in Figure 3. The size of the loudspeaker was 180 mm wide and 300 mm long and 30 mm thick.

112 According to the original theories developed by Westervelt for a parametric acoustic array in

113 the form of a semi-permeable screen [15] and by Lockwood for a parametric acoustic disk [16]

114 the process of generation of the difference wave is primarily confined to the vicinity of the

115 transducer. This means that the amplitude and behaviour of the differential (low-frequency)

116 sound wave away from this transducer is mainly controlled by the source strength density of 117 the primary high-frequency sound field near the transducer’s surface (see eqs. (1), (2) and (4) 118 in ref. [16]). In the far field, i.e. where our measurements were taken, this differential wave 119 propagates like a spherical wave radiated by a highly directional transducer. Because the source 120 strength density of this wave is proportional to the squared sound pressure in the primary (high121 frequency) wave (see eq. (5) in ref. [16]), the whole process of audible sound generation by a 122 parametric transducer is biased towards the areas where this primary pressure is particularly 123 high. The primary frequency of the parametric transducer used in this work was 44 kHz. The 124 peak sound pressure of this primary wave was 440 Pa at 0.3 m from the transducer’s center. 125 This was sufficient to develop strong non-linear effects causing the emission of the differential 126 wave. The sound pressure in the primary wave reduced to approximately 35 Pa at 4 m away 127 from the transducer. At this position the non-linear effects were relatively weak so that the 128 presence of either a green wall or another surface would be unlikely to affect noticeably the 129 parametric sound generation process in the reported experiments. 130

Building Wall 0.9m

Green Wall
Intensity probe 1.7m
131 132 Figure 3. Experimental set-up schematics. 133

Parametric transducer

134 For each of the experiments, the intensity probe was shifted left or right and up or down to 135 measure the directivity of the incident and reflected sound waves. The horizontal offset values 136 were: 0; 50; 70; 100; 150; 250; 500 and 750 mm. The vertical offset values were: ±60 mm. The 137 exact locations of the loudspeaker and intensity probe were measured by means of measuring 138 tapes and a set of lasers with level indicators. The choice of these offsets was based on the 139 transducer directivity and typical scattering pattern measured at 1.7 m. The maximum values 140 of the horizontal offset corresponded approximately to ±24 deg in terms of the azimuth angles. 141 The horizontal transducer directivity and the horizontal scattering pattern of a brick wall are 142 shown on Figures 4 and 5, respectively. These results suggest that the 90% of the emitted 143 acoustic energy in the horizontal plane is contained within ±10-12 deg segment. The horizontal 144 directivity of the reflected sound is broader, but the bulk of energy is contained within the ±24 145 deg segment. The vertical directivity of the transducer was not measured in the reported 146 experiments. It was assumed that the vertical directivity pattern is sufficiently narrow to neglect 147 the ground interference and wall edges reflection and scattering effects. Given the fact that the 148 vertical dimension of the parametric transducer was 60% wider than its horizontal dimension, 149 one can assume that the directivity would broaden proportionally. Extrapolating the results 150 shown in Figure 4 into the vertical direction suggests that the 90% of the acoustic energy 151 emitted in the vertical direction should be contained within ±16-19 deg segment. For the 152 experimental setup shown in Figure 3 it is possible to estimate that no more than 3% of the 153 emitted acoustic energy would fall on the ground at the foot of the green wall we measured. 154 The procedures for signal processing used to generate the data shown in Figure 4 and 5 are 155 described in section 3.3.

156 157 Figure 4. The horizontal transducer directivity pattern measured at 1.7 m from the transducer 158 centre.
159 160 Figure 5. The horizontal pattern of the acoustic intensity scattered by a brick wall being 4 m 161 away from the transducer centre and measured at 1.7 m away from the transducer centre. 162 163 3.2 Material specimens 164 The absorption properties of five different material specimens were studied. These were: (i) 165 brick wall (Figure 6); (ii) 100 mm thick, hard-backed melamine foam (Figure 7(a)); (iii) green

166 wall filled with 100 mm slightly moist soil without any plants (Figure 7(b)); (iv) green wall 167 planted with Hedera helix (Figure 7(c)); and (v) green wall planted with Bergenia cassifolia 168 (Figures 7(d)). The basic morphological characteristics of the two plants are summarised in 169 Table 1. The soil and the two types of plants were planted in the nursery in a green wall which 170 dimensions were 2.5 m wide and 1.8 m high. The soil without the plants had 5 litres of water 171 per 1 m2 and in all of the experiments with the plants the soil had 32 litres of water per 1 m2. 172 Table 1 presents basic morphological characteristics for the two plants studied in this work. 173 The values presented in Table 1 are taken as the average values for the selected plants used in 174 the experiments on the day.