19 th INTERNATIONAL CONGRESS ON ACOUSTICS MADRID, 2-7 SEPTEMBER 2007 A METHODOLOGY TO MEASURE THE ACOUSTIC PERFORMANCE OF ACCESS FLOORS

Transcription

1 19 th INTERNATIONAL CONGRESS ON ACOUSTICS MADRID, 2-7 SEPTEMBER 2007 A METHODOLOGY TO MEASURE THE ACOUSTIC PERFORMANCE OF ACCESS FLOORS PACS: e Asdrubali, Francesco 1 ; D Alessandro, Francesco 2 1 University of Perugia, Industrial Engineering Department, Via Duranti, 67 Perugia - Italy; fasdruba@unipg.it 2 University of Perugia, Industrial Engineering Department, Via Duranti, 67 Perugia - Italy; dalessandro.unipg@ciriaf.it ABSTRACT The European standard EN ISO :2000 gives a methodology to measure the acoustic performance of access floors in terms of both airborne and impact sound insulation. Access floors are often used in office buildings and working environments as they create a hollow space below the floor surface for network cabling and electrical wiring: in this way they are useful from the functional (ease of inspection, repair and changing of system elements), the aesthetic (no exposed wires and elements) and acoustic point of view. This paper presents the measurement procedures used at the Acoustic Laboratory of the University of Perugia Italy, in order to perform the tests defined by the EN ISO The standard requires the measurement of two parameters, L n,f e D n,f, that qualify respectively the airborne and the impact sound insulation trough an access floor, and the calculation of the two corresponding single number indexes L n,f,w e D n,f,w. A sample of access floor has been tested in the purposely prepared reverberating rooms of the Laboratory, with and without covering. Particular attention has been paid to test rooms set up and to sample installation because of the complexity of these phases. INTRODUCTION An access floor is a raised floor system consisting of modular panels supported by steel posts, placed at a certain height to create a gap below the floor surface. This hollow space can be efficiently used for data transmission, automation, processing, telephone network cabling and electrical wiring, avoiding unattractive exposed installations and offering great flexibility for operations involving the inspection, repair, changing or adding of system elements. Moreover a wide range of finishes can be applied over the panels (tiles, carpet, parquet, lino, etc ). For these reasons access floors are becoming widely used especially in the building of office centres and working environments. A measurement methodology and a test facility were set up for this purpose at the Acoustic Laboratory of the University of Perugia in order to evaluate the acoustical properties of access floors following the procedures given by the European Standard EN ISO [1]. This standard requires the measurement of the two parameters L n,f e D n,f, defined as the normalized lateral impact sound pressure level and the normalized lateral airborne sound insulation; the corresponding single number ratings L n,f,w e D n,f,w are then calculated by means of the EN ISO e 2 standards [2], [3]. These two quantities allow to qualify the acoustical performance of the floor as far as the airborne and the impact noise; more precisely, they define the efficiency of the floor in stopping the propagation of noise between two adjacent rooms having the same access floor and divided by a wall that is usually more insulating than the floor (Figure 1). Separating walls are indeed commonly installed directly over the raised floor, so creating an easy path for sound through the gap placed below the floor surface that links without obstacles the two rooms. Hence inadequate structures or materials and uncorrected designing can cause conditions of high acoustical discomfort.

2 Wall Figure 1: Two adjacent rooms with the same access floor. MEASUREMENT METHODOLOGY The European Standard EN ISO gives a laboratory method to measure the airborne sound insulation and the impact sound insulation of an access floor with an air gap of defined height, mounted below a wall separating two closed rooms. The propagation of sound through paths different from the air gap below the floor must be negligible. In this way it is possible to simulate the behaviour of two typical offices or adjacent rooms, having the floor in common, with the underlying gap and the dividing partition (Figure 2). 7 Figure 2: Requirements of the test chamber [1]: 1) elastic separation; 2) flexible material; 3) floor height; 4) access floor; 5) elastic separation; 6) dividing wall; 7) sound absorbing material. The normalized lateral airborne sound insulation D n,f (in one-third octave frequency bands) is defined as the difference between the space-time averages of the sound pressure levels generated in two chambers by a loudspeaker placed in one of them (emitting room); the difference is then normalized with respect to a reference value of the sound absorption area in the receiving room: A D n,f =L1-L2-10log (db) A 0 (1) where: L 1 is the average sound pressure level in the emitting room; L 2 is the average sound pressure level in the receiving room; A is the equivalent sound absorption area of the receiving room; A 0 is the reference value of sound absorption area (in laboratory A 0 = 10 m 2 ). The normalized lateral impact sound pressure level L n,f (in one-third octave frequency bands) is defined as the mean value of the sound pressure levels measured in different positions of the receiving room; the impact sound is generated by a normalized tapping machine placed in several positions on the floor of the receiving room; the value is finally normalized with respect to a reference value of the sound absorption area in the receiving room: 19 th INTERNATIONAL CONGRESS ON ACOUSTICS ICA2007MADRID 2

3 Ln,f =L f -10log A A0 (2) (db) where: Lf is the average sound pressure level in the receiving room generated by the normalized tapping machine; The equivalent sound absorption area A of the receiving room can be calculated from the reverberation time measured according to EN ISO 354 standard, using the well known Sabine relation [4]: A= 0.16 V T (m2 ) (3) where: V is the receiving room volume; T is the receiving room reverberation time. The single number ratings Ln,f,W e Dn,f,W are calculated through the procedures given by the EN ISO and 2 standards; the calculations are equivalent to those used for the weighted sound reduction index Rw and for the weighted impact sound pressure level Ln,W. SET-UP OF TEST ROOM AND DESCRIPTION OF THE SAMPLES The Acoustic Laboratory of the University of Perugia has two coupled reverberation rooms: an opening between them allows the mounting of the samples that has to be tested during the sound insulation measurements [5]. The geometrical dimensions (total length = 9.1 m; width = 5.4 m; height of the receiving room = 3.5 m; height of the emitting room = 2.9 m) and the structural characteristics of the test rooms satisfy the requirements of EN ISO Description of the tested access floor samples The tested access floor is constituted by modular panels (model S-Floor CA 36 produced by Edil Beton Perugia S.p.A.) installed on the entire surface of the coupled reverberation rooms; the panels are placed on the support posts at 25 cm from the floor of the rooms and fixed with steel plates. The characteristics of the sample are (Figure 3): Composition: light cement mixture reinforced with a metal mesh and mineral fibres; Dimension of the panel: 600 x 600 x 36 mm; Surface mass: 63,9 Kg/m2 (mass of the single panel: 23 Kg); Finish: smooth upper face; Support posts: constituted by a steel square base, a threaded bar (diameter = 14 mm), PVC spacer seals inserted between the top of the support and the panel base and a square steel plate placed on the top surface of the floor where the panels meet; the plates serve to fasten the panels mechanically by means of a screw that reaches the head of the support posts. Figure 3: Sample mounting (left); particular of the panels and the support posts (right). 3 19th INTERNATIONAL CONGRESS ON ACOUSTICS ICA2007MADRID

4 Test room set-up and sample assembly Test room set-up is particularly laborious. The test sample is assembled in the following way: the posts are fixed on the floor of the coupled reverberation rooms and then the panels are placed upon the posts; since the posts are telescopic, it is possible to level off the floor and to obtain the desired height from the floor of the room (in our tests an height of 25 cm is used); finally the panels are fixed between them by means of the top steel plates of the posts. A layer of antivibration material (thickness = 10 mm) is placed between the access floor and the lateral walls of the test room. A lateral wall and both the bottom walls of the air gap beneath the access floor are covered with a sound-absorbing mat made of rock wool (thickness = 80 mm; density = 70 kg/m3) in accordance with EN ISO (Figure 4). The wall that separates the test room is installed directly on the access floor. It has to be strongly insulating (at least 10 db more than the access floor under test) in order to be sure that sound passes only through the air gap beneath the access floor. To that end a certified lightweight structure (RW = 63 db) is used; the stratigraphy of the dividing wall is (Figure 4): double gypsum board (thickness of the single board = 12.5 mm); metal support structure for the rock wool insulating layer (thickness = 50 mm; density = 50 Kg/m3); gypsum board (thickness = 12.5 mm) placed inside the internal air gap; metal support structure for the rock wool insulating layer (thickness = 50 mm; density = 50 Kg/m3); double gypsum board (thickness of the single board = 12.5 mm). A 10 mm thick antivibration layer is put between the access floor and the dividing wall. After the first measurement campaign, a carpet layer (thickness = 6 mm; surface mass = 4.5 Kg/m2) has been laid on the floor without using glue and a second measurement campaign has been carried on, in order to evaluate the improvement of the acoustical performances of the floor, especially as far as impact noise. Wall mounted absorbing panels have been installed in both the test rooms in order to reach the reverberation times specified in the EN ISO standard. Suspended plane diffusers hanging from the ceiling allow to obtain an optimal diffuse sound field inside the test room [5]. Figure 4: Particular of the sound absorbing mat placed on the lateral walls of the gap (left); stratigraphy of the separating wall used for the tests (right). Measurement chain The instrumentations used for the tests are: Omnidirectional dodecaedric source (mounting loudspeakers model GHX 165 Audio Design) emitting a white noise; Normalized tapping machine (length = 0.80 m; width = 0.3 m; height = 0.28 m) provided with 5 hammers made of tempered steel; 2-channel amplifier model SU-A900 Technics; ½ inch microphones model 40 AR GRAS; Preamplifiers model PRE12H 01dB; Signal conditioning amplifier model OPUS 01dB; 2-channel signal acquisition system model Symphonie 01dB; Analysis software model dbbati32 01dB. 4 19th INTERNATIONAL CONGRESS ON ACOUSTICS ICA2007MADRID

5 RESULTS OF THE MEASUREMENT CAMPAIGNS Measurement of normalized lateral airborne sound insulation D n,f Figure 5 reports the third-octave bands trends of the normalized lateral airborne sound insulation D n,f, measured with and without the carpet laid on the concrete panels. 70,00 60,00 50,00 without carpet with carpet difference Dn,f,w = 51 db Dn,f [db] 40,00 30,00 Dn,f,w = 47 db 20,00 10,00 0, frequency [Hz] Figure 5: Normalized lateral airborne sound insulation D n,f vs. frequency for the two test configuration. The application of the surface finish increases the insulating properties of the structure especially at low ( Hz) and high ( Hz) frequencies, while its effect is lower at medium frequencies; in terms of the single number index D n,f,w there is an increase of 4 db. The effect of the surface finish with respect to airborne sound insulation is to cover the slits that may occur between the panels of the floor; moreover the carpet has intrinsic sound absorbing properties that influence the sound field inside the chambers. However the effect of the carpet is much more evident for the impact noise measurements, as it can be seen in the next paragraph. Measurement of normalized lateral impact sound pressure level L n,f Figure 6 reports the third-octave bands trends of the normalized lateral impact sound pressure level L n,f, measured with and without the carpet laid on the concrete panels. 80,00 70,00 Ln,f,w = 71 db 60,00 Ln,f [db] 50,00 40,00 30,00 20,00 without carpet with carpet difference Ln,f,w = 55 db 10,00 0, frequency [Hz] Figure 6: Normalized lateral impact sound pressure level L n,f vs. frequency for the two test configuration 19 th INTERNATIONAL CONGRESS ON ACOUSTICS ICA2007MADRID 5

6 The effect of the carpet becomes much more important for impact sound level measurements: with the application of the finish sound pressure levels decrease significantly in the entire considered frequency range and the single number index L n,f,w has a reduction of 16 db. Therefore the use of a resilient surface finish (as carpet, lino, rubber, etc ) is strongly suggested in access floor in order to achieve adequate values of impact noise attenuation. CONCLUSIONS The measurement procedures provided by the European Standard EN ISO have proven to be complex and laborious, especially with regards to the test room set-up; the execution of the measurements is indeed quite simple, not being too different from those needed to evaluate the airborne sound insulation of vertical walls R and the normalized impact noise pressure level L n for floors. The construction of a high insulating dividing wall and the installation of high absorbing material on the lateral walls of the air gap make this procedure quite long and expensive but it is essential to simulate properly the acoustical behavior of two typical offices or rooms having in common an access floor with the underlying air gap and the separation wall. The results obtained in the measurement sessions highlight on one hand the accuracy of the procedure used for the execution of the test and on the other hand the good properties of the tested access floor as far as both the airborne and impact sound insulation. Access floors seem to combine great functionality and flexibility (ease of inspection, repair and modification of the underlying systems) with adequate conditions of acoustical comfort. Finally the use of a resilient surface finish (carpet in this case) over the concrete panels is strongly recommended in order to achieve adequate values of impact noise attenuation. ACKNOWLEDGEMENTS The Authors wish to thank EDIL BETON Perugia S.p.A. for having supplied the test samples and Dr. Augusto Ambrosi for his support during the experimental campaign. References: [1]. EN ISO , Acoustics Measurement of sound insulation in buildings and of building elements Laboratory measurement of room-to-room airborne and impact sound insulation of an access floor, [2]. EN ISO 717-1, Acoustics - Rating of sound insulation in buildings and of building elements - Part 1: Airborne sound insulation, [3]. EN ISO 717-2, Acoustics - Rating of sound insulation in buildings and of building elements - Part 2: Impact sound insulation, [4]. EN ISO 354, Acoustics - Measurement of sound absorption in a reverberation room, [5]. F. D Alessandro, G. Pispola, Sound absorption properties of sustainable fibrous materials in an enhanced reverberation room, Proc. Inter-noise 2005, Rio de Janeiro, Brazil, th INTERNATIONAL CONGRESS ON ACOUSTICS ICA2007MADRID 6