Mixing materials in false ceilings to increase sound di ff usion in education spaces

Giulia Fratoni 1 , Dario D’Orazio, Luca Barbaresi, Massimo Garai University of Bologna, Department of Industrial Engineering (DIN) Viale del Risorgimento, 2, 40136 Bologna, Italy

Luca Cappellini SAINT-GOBAIN ITALIA S.P.A. Via Giovanni Bensi, 8, 20152 Milano, Italy

ABSTRACT According to the International standards on education spaces, in situ acoustic tests should achieve specific reverberation time targets. Since the match between predictive formulas and measurements increase with the sound field di ff useness, it is extremely important to pursue such condition through proper design choices. For example, false ceilings can play a key role in controlling room acoustic features; however, the use of such elements is generally considered only for absorption purposes. For this reason, the present work concerns mixing materials in false ceilings within a group of teaching spaces here taken as case studies. The alternation of materials with di ff erent acoustic impedances has been investigated through experimental acoustic measurements and numerical models in order to assess the related sound di ff usion increase. The ceiling treatment here proposed proved to be an e ffi cient and smart method to exploit di ff raction e ff ects along material discontinuities junctions. The match between early design predictive formulas and the results of the acoustic measurements enhanced the reliability of the acoustic design process and set up a potential new guideline for indoor acoustic treatments.

1. INTRODUCTION

Acoustic comfort of education spaces is generally determined by target values of reverberation time. Predictive methods proposed by International standards typically involve the Sabine’s formula, and thus the di ff use sound field assumption [1–4]. Therefore, achieving such condition in teaching environments as much as possible becomes essential for the match between predictive outcomes and the results of in-situ measurements. With this purpose, the attention is here focused on the passive acoustic treatments of false ceilings, which are the indoor surfaces that mostly a ff ect room acoustic criteria thanks to their extension and location in the built environments [5].

1 giulia.fratoni2@unibo.it

a slaty. inter.noise 21-24 AUGUST SCOTTISH EVENT CAMPUS O ¥, ? GLASGOW

While a widespread tendency is covering false ceilings with uniform sound absorbing surfaces, the present work outlines a method for specific materials’ mix to increase the di ff useness of sound field. In fact, a passive treatment false ceilings with the same material could hinder the fulfillment of the classic di ff use field theory’s conditions, due to the non-uniform distribution of the overall absorbing materials in the room [6]. Conversely, inhomogeneous surfaces with di ff erent specific acoustic impedances (see Figure 1) are more likely to cause di ff raction e ff ects along material discontinuities junctions [7], increasing the global sound field di ff usion throughout the space [8].

Figure 1: Inhomogeneous surface with di ff erent specific acoustic impedances (image taken from [7]).

2. METHOD

This paper addresses the e ff ect of di ff erent materials placement in false ceilings for the acoustical conditions in education spaces. Patterns with specific impedances alternation have been designed, installed, and measured in two teaching environments to test such a design choice. Mixing perforated gypsum board and plasterboard modules has been considered the adequate solution for large volumes, whilst mixing perforated gypsum board and rock wool modules at high density has been considered the adequate solution for mid-sized volumes. The present study presents the experimental results of in situ acoustic measurements compared with the predicted target values (Sabine’s formula). Also, numerical models have been employed to preliminarly assess the related sound di ff usion increase.

2.1. Case studies The case studies considered in the present study are:

– Hall 1, a large university lecture hall (V = 1200 m 3 ); – Hall 2, a medium-sized classroom (V = 150 m 3 ).

The installation of three materials with di ff erent acoustic impedances has been arranged as it is shown in Figure 2. The placement and the choice of false ceilings modules depends on the size of the hall. Generally, it is better to have a central region of the ceiling with reflecting properties at medium-high frequencies - where most of the speech signal energy is - and to keep any absorbing

(a) Hall 1

(b) Hall 2

Figure 2: Education spaces taken as case studies. Materials mix in false ceilings treatments is highlighted: standard plasterboard modules (yellow), perforated gypsum modules (blue), and rock wool modules at high density (red).

material in the perimeter areas of the false ceiling. This solution grants a proper balance between the absorption of the occupancy and the absorption of the halls, enhancing the speech intelligibility conditions. Acoustic measurements were carried out according to [9–11] as acceptance test of the passive treatment design. In Hall 2 it has been possible to perform the measurements in two di ff erent conditions: with a uniform ceiling treatment and with the materials mix shown in Figure 2.

2.2. Numerical models In order to assess the e ff ect of ceiling mix on the sound field di ff useness, specific numerical models were built with COMSOL Multiphysics v. 5.1. Figure 3 shows the two simplest configurations that have been explored: an homogeneous surface (uniform placement of rock wool modules) and an inhomogeneous surface (alternance of rock wool and gypsum board modules). The Acoustics and Pressure acoustics, frequency domain modules were employed. A plane wave radiation condition was set at the lower boundary of the control volume (3 x 1 x 1m), approximating the system excitement, i.e. the sound energy coming from the room towards the ceiling. The output surface is transversal to the control volume to detect the di ff erent sound pressure fluctuations.

(a) Homogeneous surface

(b) Inhomogeneous surface

Figure 3: Ceiling configurations set in COMSOL .

3. RESULTS

The first set of results is provided for Hall 2 in terms of measured reverberation time values compared to the target values required by standards ( T 30 = 0 . 5 − 0 . 8 s). Figure 4 reports the predicted and measured reverberation time with a uniform ceiling treatment and with the ceiling mix installed (permanent solution). It is possible to notice that the match between the two ranges of values is higher in case of inhomogeneous surfaces at the ceilings. In fact, in this case the discrepancies are within the 5 - 10 % of the measured value, compared to the di ff erences up to 30 % of the measured value for the configuration with uniform treatments.

140 g120 060 oso 020 2 a4 octave bands) —Predited —Measvced

— Predicted Measured

(a) Homogeneous surface

(b) Inhomogeneous surface

Figure 4: Hall 2: comparison between predicted and measured reverberation time with a single material (left) and with the ceiling mix (right).

(a) Homogeneous surface

(b) Inhomogeneous surface

(c) Homogeneous surface

(d) Inhomogeneous surface

Figure 5: Di ff raction e ff ects caused by material discontinuities.

The second set of results concerns the preliminary FEM analysis carried out for the two configurations assessed in this work. Figure 5 shows the di ff erent results of the two configurations described in the previous section. The results are provided in terms of sound pressure variations along the trasversal section considered as output surface (results provided for the frequency 1563 Hz). Future steps of the FEM analysis could involve the use of control points (microphones) to quantify the sound di ff usion increase.

4. CONCLUSIONS

The sound di ff using e ff ects caused by the mixed materials solutions have been explored with acoustic measurements and with numerical analysis. The match between target ranges and the results of the acoustic measurements enhanced the reliability of the design choices, entailing at the same time new potential guidelines for acoustic treatments in education spaces.

REFERENCES

[1] ISO 12354-6:2004 Building acoustics. Estimation of acoustic performance of buildings from the performance of elements Sound absorption in enclosed spaces . [2] Tetsuya Sakuma. Approximate theory of reverberation in rectangular rooms with specular and di ff use reflections. The Journal of the Acoustical Society of America , 132(4):2325–2336, 2012. [3] Sylvio R Bistafa and John S Bradley. Predicting reverberation times in a simulated classroom. The Journal of the Acoustical Society of America , 108(4):1721–1731, 2000. [4] Erling Nilsson and Emma Arvidsson. An energy model for the calculation of room acoustic parameters in rectangular rooms with absorbent ceilings. Applied Sciences , 11(14):6607, 2021. [5] Peter D’Antonio. Optimizing the signal to noise ratio in classrooms using passive acoustics. The Journal of the Acoustical Society of America , 139(4):1980–1980, 2016. [6] Erling Nilsson. The e ff ect of di ff erent acoustical treatment in a classroom. The Journal of the Acoustical Society of America , 141(5):3456–3456, 2017. [7] SN Chandler-Wilde and DC Hothersall. Sound propagation above an inhomogeneous impedance plane. Journal of Sound and Vibration , 98(4):475–491, 1985. [8] Yiu Wai Lam and Mohammad Reza Monazzam. On the modeling of sound propagation over multi-impedance discontinuities using a semiempirical di ff raction formulation. The Journal of the Acoustical Society of America , 120(2):686–698, 2006. [9] ISO 3382-1:2008 Acoustics — Measurement of room acoustic parameters . [10] EN 60268-16:2001 Sound system equipment . [11] UNI 11532-2:2020 Internal acoustical characteristics of confined spaces - Design methods and evaluation techniques .