Applied Acoustics 70 (2009) 626–635

Contents lists available at ScienceDirect

Applied Acoustics
journal homepage: www.elsevier.com/locate/apacoust

Evaluation of the acoustic performance of classrooms in public schools

Paulo Henrique Trombetta Zannin *, Daniele Petri Zanardo Zwirtes
Laboratório de Acústica Ambiental – Industrial e Conforto Acústico, Universidade Federal do Paraná, Centro Politécnico, Departamento de Engenharia Mecânica, Bairro Jardim
das Américas, CEP 81531-990 Curitiba – PR, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents the results of an evaluation of acoustic comfort of classrooms built according to a
Received 31 October 2007 standard design. Three constructive designs located in the metropolitan area of Curitiba (Brazil) have
Received in revised form 15 June 2008 been evaluated, two schools built under each of these three designs, in a total of six schools. The acoustic
Accepted 20 June 2008
quality of the classrooms have been analyzed based on measurements of the reverberation time, sound
Available online 11 September 2008
pressure level inside and outside the classrooms, and sound insulation. Measurements of ambient noise
(external and internal) followed the Brazilian Standards NBR 10151 and NBR 10152. Measurement of
Keywords:
reverberation time and sound insulation followed the international Standards ISO 140-4, ISO 140-5,
Acoustic comfort
Ambient noise
ISO 717-1, and ISO 3382. Results (sound insulation and reverberation time) have been compared with ref-
Classrooms erence values found in the Brazilian Standard NBR 1279, and in the Standards ANSI S12.60 and DIN
Reverberation 18041. Results reveal poor acoustical quality of the surveyed classrooms, for all 3 constructive designs
Sound insulation studied. The surveyed designs do not meet the guidelines of either the Brazilian Standards or of the Inter-
national Standards employed as references.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction ronment with pleasant or at least acceptable noise levels,

preserving the health of its users as a whole and improving the stu-
The education of every citizen is essential to all modern socie- dents’ scholastic performance.
ties. The most formal education takes place in the classroom, The subject of acoustical comfort (ambient noise, sound insula-
where the learning process involves intensive verbal communica- tion, reverberation time, speech intelligibility, auralization, acous-
tion between teachers and students and among students. The effi- tical materials) in classrooms of primary schools, in secondary
ciency of this communication, and hence, the efficiency of the schools, as well as in University classrooms has been the focus of
learning environment, is measured by the acoustic conditions of several studies around the world [5–22]. Another focus of studies
the classrooms [1]. High levels of noise in classrooms make stu- has been the perception of noise by students and teachers, and
dents prematurely tired, consuming their cognitive abilities which the influence of noise on those people [20,23–26].
could be better employed in paying attention to and understanding The goal of the present study was to evaluate the acoustical
the content of their classes [2]. conditions of three school constructive models from the State of
Many aspects that appear with the evolution of the modern era Paraná, southern Brazil. Thus, six schools have been chosen, two
serve to deteriorate the acoustic environment of classrooms [3]. In of each of the three different types of Standard construction. The
the past, classrooms were silent and pleasant. Today they are rela- acoustical parameters evaluated were the external and internal
tively more noisy and reverberative. The main reason for the exis- ambient noise, strictly following the Brazilian Standard on commu-
tence of acoustic problems in classrooms is not a lack of resources, nity noise, NBR 10151, and the Brazilian Standard for Acoustical
but rather, a lack of perception of the problem on the part of the Comfort in Buildings, NBR 10152. Reverberation time inside empty
professionals involved and a lack of solutions [4]. Seep et al. [4] sta- and furnished classrooms has also been measured. The weighted
ted that the best way to solve acoustic problems is to avoid them in apparent sound reduction index has also been measured between
the design phase. The ability to listen is known to be a duty of the classrooms and the corridor, as well as the sound insulation by faç-
student and not a pedagogic activity or an architectural challenge ades. These measurements have followed the Standards ISO 140-4,
[2]. The school environment should promote an atmosphere that ISO 140-5, and ISO 717-1. Measured data have been compared
induces everyone’s interest in listening and being involved in com- with reference values found in the Standards ANSI S12.60 and
munication. Thus, the social benefit of noise reduction is an envi- DIN 18041.
The evaluated schools (primary schools) belong to the public
* Corresponding author. Tel./fax: +55 4133613433. system of the State of Paraná, Southern Brazil, and are located in
E-mail address: [email protected] (P.H.T. Zannin). the city of Curitiba, capital of Paraná State.

0003-682X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apacoust.2008.06.007
 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635 627

2. Method traffic was not significant, outside noise was generated by a loud-
speaker (ISO 140-5). To measure the sound insulation of façades, a
The aim of this work was to verify the acoustic quality of class- pink noise signal was generated using a BK 1405 noise generator.
rooms built according to standard designs for school buildings. To This noise was distributed along the façade by means of a 900 W
this end, the standard modular school models 010, 022 and 023 loudspeaker system tilted at 45° and placed at least 3.5 m away
implemented by the government of the state of Paraná, Brazil, have from the façade. Two BK 4190 microphones, one facing the façade
been evaluated. and the other inside the classroom, recorded the sound signals
Six schools, two of each constructive design, were evaluated in simultaneously, as recommended by the ISO 140-5 standard. Back-
the metropolitan region of Curitiba. To facilitate the identification ground noise and reverberation time measurements were taken in
of the schools of this research, the schools of design 010 were empty classrooms to make the corrections according to the acous-
dubbed S1 and S2, those of design 022 were called S3 and S4, tic characteristics of the reception environment.
and those of design 023 were called S5 and S6. The surveyed AR 0014 flat cable (Brüel and Kjaer) was used for measuring the
schools are attended by children from the 5th to the 8th grades sound insulation of façades and between classrooms and corridors.
(11–14 years old) of our fundamental education system, which Flat cable serves to connect the external microphone to the BK
corresponds to the primary school. The results of this work were 2260 analyzer without the need to leave gaps besides the existing
obtained by in situ measurements of the reverberation time RT, ones around doors or windows. The use of this equipment contrib-
sound insulation coefficients (sound insulation of façades and be- utes to the reliability of the measured data.
tween classrooms and corridors), and ambient noise (outside the After the field measurements, the data were transferred to the
classrooms and inside the classrooms), expressed by the continu- Qualifier software (BK 7830), which processes all the data collected
ous equivalent sound level, LAeq, in dB. and provides a single sound insulation value R0w – weighted appar-
ent sound reduction index, between rooms in a building and R045 ;w
– weighted apparent sound reduction index, sound insulation of
2.1. Measurement of reverberation time RT
façades. Data processing using this software program is specified
by the ISO 140-4, ISO 140-5 and ISO 717-1 [32] standards.
To measure the reverberation time RT, a two-channel BK 2260
(Brüel and Kjaer) modular real-time sound analyzer has been used,
which emits a pink noise signal to a BK 2716 power amplifier con- 2.3. Measurement of external and internal ambient noise
nected to the sound source. A BK 4296 omnidirectional dodecahe-
dron sound source was used. The generated sound was captured by Classrooms have been evaluated under the following conditions:
a microphone connected to the BK 2260 analyzer, which automat- (1) empty classrooms and (2) occupied classrooms. In all situations
ically calculated the reverberation time RT60 for each frequency of the evaluated schools were not empty, while routine activities were
the spectrum of interest. carried out. The sound pressure levels were obtained from mea-
The measurements were taken following the specifications of surements taken both inside the classrooms and outside the
the ISO 3382 standard [31]. Measurements were taken at three dif- schools. In both cases, the measurements were taken according to
ferent points in each classroom. Three readings were taken at each the Brazilian NBR 10151 standard [27], which regulates noise eval-
point. In the laboratory, measurements were then transferred to a uations in inhabited areas for purposes of community comfort.
computer using Qualifier 7830 software from Brüel and Kjaer, External measurements have been conducted in order to char-
which calculated the mean reverberation time and the respective acterize the area where the schools are located. Noise levels inside
standard deviation for each evaluated frequency. This procedure the classrooms were evaluated to verify the acoustical conditions
was repeated for all classrooms in which RT was measured. of the classrooms for the development of teaching–learning activ-
ities. The influence of noise produced in surrounding schoolyards
and sports courts on the noise levels measured inside the class-
2.2. Measurements of the sound insulation
rooms was also checked.
All the measurements were taken on normal school days (Mon-
The procedures for taking field measurements of the weighted
days to Fridays), without atypical noises such as rain, thunder, or
apparent sound reduction index R0w between rooms in a building strong winds, as recommended by the NBR 10151 standard [27].
are specified by the ISO 140-4 standard [29], while measurements
This standard establishes that external measurements should be
of the sound insulation of façades, namely weighted apparent sound
performed at the site of highest noise level. Accordingly, the sound
reduction index R045 ;w , are specified in the ISO 140-5 standard [30].
level meter was positioned on a tripod at a distance of 2 m from
A pink noise was generated by the BK 2260 sound analyzer dur-
walls or façades, and then the measurement was performed. The
ing the measurements of the sound insulation coefficients between
duration of each measurement was of 10 min. The continous
the classrooms and the corridors. This noise was amplified with a
equivalent sound level LAeq has been measured, along with its
BK 2716 power amplifier and then distributed through the class- range of variation, from LAmin to LAmax.
room using the BK 4296 omnidirectional dodecahedron sound
The noise inside each classroom was measured at a single spot
source. Two BK 4190 microphones, one in the corridor and the positioned at the center of the classroom. The duration of this mea-
other in the classroom, picked up the sound simultaneously.
surement was of 3 min. The continuous equivalent sound level LAeq
As outlined by the ISO 140-4 standard [29], the noise level at and its range of variation from LAmin to LAmax were thus obtained.
the back of the reception room and the reverberation times were
Values measured are displayed in Tables 2 and 3. Sound pressures
recorded in order to make the corrections as a function of the areas were measured using Brüel and Kjaer BK 2238 and BK 2237 sound
of absorption of the reception room, following the procedures of
level meters and the measured values were analyzed using Brüel
the ISO 3382 standard [30]. and Kjaer’s BK 7820 Evaluator software.
The number of points evaluated was determined according to
the dimensions of the spaces, observing a minimum distance of
0.5 m between the wall and the microphone and of 1.5 m between 3. Description of the evaluated designs
the microphone and the floor.
With regard to the measurement of sound insulation by façades, Public schools in the state of Paraná are designed in standard
given the observation of the schools’ surroundings, that noise from modules adjustable to the need for new schools, depending on
628 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635

the forecast of the number of students and on the type of terrain

where they are to be installed.
The characteristics of the construction designs selected here
were as follows: (1) design 023, which consists of a proposal of
independent blocks with a central circulation area and classrooms
arranged on both sides of a corridor (Fig. 1), (2) design 022, com-
prising classroom blocks arranged in a row without a corridor be-
tween them (Fig. 2) and (3) design 010, similar to design 023,
formed of independents blocks of classrooms arranged on the
two sides of a central corridor (Fig. 1).
Designs 010 and 023 are similar in terms of block composition,
but they have considerable differences in terms of corridor height,
building components (e.g., windows) and coating materials.
Fig. 3 shows the corridor between classrooms in the Schools Al-
fredo Parodi and Luiza Ross, both Standard 010.
Figs. 4–6 show the location of the sports facilities and play-
Fig. 2. Classroom construction design 022.
grounds (school yards), with respect to the evaluated classrooms,
for each one of the construction patterns.

An investigation was made of the internal ambient noise in the

4. Results and discussion classrooms. To this end, measurements were conducted inside five
empty classrooms, with the detection of noise coming from classes
The schools selected for this study were constructed during held in other classrooms in the same block. The purpose of these
three different periods. Design 010 was built in 1977 (S1) and measurements was to ascertain if the noise produced in a class-
1978 (S2). The two schools built according to design 022 were room would affect other classrooms around it.
completed in 1998 while those of design 023 were built in 2001 To determine the daily reality, all the measurements were taken
(S5) and 2005 (S6). with the windows open, seeking to interfere as little as possible in
the schools’ daily routines. Some situations could therefore not be
4.1. External and internal ambient noise evaluated in the same way. This was the case of school S3, where it
was impossible to evaluate how the noise produced in each class-
To evaluate the acoustic composition of the school environ- room affected the other classrooms because the noise coming from
ment, measurements were taken of the continuous equivalent the schoolyard exceeded that of adjoining classrooms. Table 2
sound levels. Firstly, the external environment was evaluated shows the noise levels measured in five empty classrooms with
based on measurements taken on sidewalks around the schools, the adjoining classrooms engaged in normal activities.
at the distances established by the Brazilian NBR 10151 standard. The NBR 10152 standard [28] establishes 40 dB(A) as a comfort-
The purpose of this evaluation is to check if there is any influence able noise level in classrooms, although 50 dB(A) is an acceptable
of traffic noise in the classrooms. value for classroom purposes. As can be seen in the above table,
According to the Brazilian NBR 10151 standard, which estab- the noise levels far exceed the limit determined by this standard.
lishes the sound levels for external environments, the maximum Excessive noise levels affect not only the quality of verbal com-
LAeq admissible for school zones during daytime is 50 dB(A). Table munication but also lead to serious problems in the student’s intel-
1 displays the values of the sound levels (LAeq) measured in the lectual development, such as slow language learning, difficulties in
noisiest spots surrounding the school, as well as the range of the written and oral language, limitations in reading skills and in the
measured sound levels. composition of vocabulary [39].
The values listed in the above Table indicate that schools S5 and With regard to the levels listed in Table 2, it can be concluded
S6 (design 023) and school S4 (design 022) are installed in quieter that the classrooms affect each other negatively, generating high
zones than the other schools. The values measured around all the levels of ambient noise that are incompatible with the values
schools are higher than those established by the NBR 10151 stan- established by the Brazilian standard for acoustic comfort in
dard [27]. classrooms.

Fig. 1. Classroom construction designs 010 and 023.

 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635 629

Fig. 3. Corridor between the classrooms from school Alfredo Parodi – standard 010 e school Luiza Ross – standard 010.

levels in the 630, 800 and 1600 Hz frequencies, indicated by the

gray columns. These noise levels are related with loud talking
and shouting, which are a normal part of sports activities.
Knowing that the frequencies of 500–2000 Hz are considered
the main speaking frequencies, the noise levels generated during
the physical-education classes were found not only to disturb
classroom activities but also to compete directly with the teacher’s
voice.
Because the sound levels proved incompatible with the condi-
tion of acoustic comfort in the empty classroom, the sound levels
generated during a Portuguese language class at School S3 and a
Mathematics class at School S6 have been measured, and are pre-
sented in Table 4, below.
These sound levels are related predominantly to the teacher’s
voice during class, with the students only listening. For these
measurements, the sound level meter BK 2238 has been placed
in front of the teacher. The sound level meter was placed on a tri-
pod. The microphone was positioned at a distance of 1 m from the
mouth of the teacher. Measurements have been conducted during
daytime, and have lasted 1 min. The measurements were taken
during the normal class period without interference from physi-
Fig. 4. Plan of the school Alfredo Parodi, built under standard 010, showing location cal-education classes. As the Table 4 indicates, the LAeq values
of classrooms and sports facilities. are high, demonstrating the vocal effort which the teachers are
forced to make. This effort is even greater when there are physi-
cal-education activities going on in the schoolyard, since they
Moreover, another disturbing noise comes from physical-edu- raise the ambient noise inside the classroom, as shown in Table
cation activities. The schools of designs 010 and 022 showed an 3. Noise levels above 60 dB(A) exceed the normal speaking voice
inappropriate distance between the classrooms and schoolyards levels by 5 dB(A), impairing the listener’s attention and concen-
and sports courts. Fig. 7 illustrates this proximity at schools S1 (de- tration [40].
sign 010) and S3 (design 022). The figure clearly depicts the class-
room windows in both schools facing the schoolyard, which 4.2. Reverberation time RT
contributes to increase the ambient noise inside the classrooms.
Table 3 lists the values measured while recreational activities Understanding speech is hindered by the combined effects of
were carried out at the school yards, but with the evaluated class- excessive noise and reverberation in the classroom, which tends
rooms empty. to interfere in the learning process. The combination of noise and
The noise levels produced during physical-education classes are reverberation exerts a stronger negative effect on speech recogni-
high. The close proximity of classrooms and the schoolyard where tion than the sum of their separate effects [41,42].
these activities take place is extremely detrimental to the teach- Several acoustical standards (national or international) provide
ing–learning process, not only because of the noise levels that im- reference values for RT that should be observed when a certain
pair understanding of speech but also due to the students’ room is built for a certain purpose. In the case of classrooms, spe-
distraction and diminished concentration resulting from the visual cific standards can be cited, from countries such as Germany, Ja-
stimuli provided by physical-education activities right beside the pan, United Kingdom, and USA [34,35,43,44]. In Brazil, the
classroom windows. The noise levels shown in Table 3 and the standard that specifies RTs for closed spaces (furnished and unoc-
photographs in Fig. 7 reveal a serious error in the conception of cupied) is the NBR 12179 [33]. Although it does not specifically
the architectural design with regard to the position of these spaces. mention classrooms, it shows a curve for optimum RT in the fre-
Fig. 8 presents a frequency analysis conducted during a physi- quency of 500 Hz for a conference room, which will be applied here
cal-education class at School S1. The frequency analysis was per- for classrooms. In Japan [43], RT values are for the mean in 2-oc-
formed during 3 min, and the measurement was conducted at a tave bands including 500 Hz and 1000 Hz, and RT is measured with
central spot inside the classroom. The figure reveals higher sound furnished but empty classrooms. In the United Kingdom the Build-
630 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635

Fig. 5. Plan of the school Anibal Khuri Neto, built under standard 022, showing location of classrooms, school yard, and sports facilities.

ing Bulletin BB-93 [44] indicates that RT is given in terms of the Table 6 displays RT values for two classrooms of each constructive
maximum mid frequency reverberation time Tmf, the average RT design evaluated.
in the 500 Hz, 1000 Hz and 2000 Hz octave bands, and RT should RTs of all the schools evaluated here were higher than the 0.6 s
be measured with the classroom empty and unfurnished. In the limit established by the ANSI S12.60 standard [34] (Tables 5 and 6).
USA [34] RT is given as the maximum RT in octave bands with mid- When compared with the recommendation of the World Health
band frequencies of 500 Hz, 1000 Hz and 2000 Hz, and RT is mea- Organization [39], only School S2 (Design 010) falls within the
sured with the classroom empty and furnished. In Germany the established range of 0.4–0.8 s. The RTs of the schools S3, S4, S5
DIN E 18041 [35–38]–Hörsamkeit in kleinen bis mittelgrossen Räu- and S6 are well above the range established by WHO. Considering
men (Audibilidade em salas de tamanho pequeno a médio) from the Brazilian Standard NBR 12179, none of the classrooms meets
2003, establishes that RT values are for the mean in 2-octave bands the limit for RT in the frequency of 500 Hz, according to Table 5.
500 Hz and 1000 Hz, and RT is measured with the classroom empty Considering the Standard DIN E 18041 [35–38], again only School
and furnished. The DIN E 18041 establishes that for occupied S2 presents reasonable RT values (Table 5).
rooms, one should subtract 0.2 s from the RT values in Table 5. The differences in the RTs of the constructive designs are due to
For the analysis of RT, the recommendations of the WHO [39] have the different building materials employed. In the classrooms of de-
also been considered, based on the French regulation of January 9, sign 010, the floors are parquet and the ceiling is wood paneled.
1995. According to the French regulation cited by WHO [39] the RT The 022 design also has parquet floors, while the 023 design has
in classrooms with a volume of less than 250 m3 should be of be- ceramic tile floors. The ceilings of designs 022 and 023 are not pan-
tween 0.4 and 0.8 s, RT being the arithmetic mean of the RT mea- eled, but simply plastered and painted. The walls of all the con-
sured in the frequencies 500, 1000, and 2000 Hz. Table 5 shows the structive designs are of bricks overlaid with plaster and painted,
recommendation for RT. but in school S6, the walls are finished up to half-height with wall
Reverberation time was measured in furnished unoccupied tiles. The schools of design 010 were constructed about 20 years
classrooms, which have a volume of about 139 m3 (designs 010 before those of designs 022 and 023, and their interior finishing
and 022) and 156 m3 (design 023) and seat up to 40 students each. (walls, floors and ceilings) and reverberation times offer better
 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635 631

Fig. 6. Plan of the school Walde Rosi Galvão, built under standard 023, showing location of classrooms, school yard, and sports facilities.

Table 1
Traffic noise in the proximities of the schools

School Construction design LAeq (dB) LAmax (dB) LAmin (dB)

S1 010 66.3 80.1 45.7
S2 010 66.2 91.2 41.7
S3 022 68.4 87.6 44.0
S4 022 60.5 75.1 38.3
S5 023 59.2 76.8 38.5
S6 023 51.8 73.9 39.5

Table 2
Noise levels in five empty classroom with the other rooms engaged in normal
activities

School Construction design LAeq (dB) LAmax (dB) LAmin (dB)

S1 010 59.4 76.1 46.4
S2 010 63.2 76.0 51.9
S4 022 51.1 71.1 40.2
S5 023 59.1 70.3 52.1
S6 023 60.7 75.1 47.9

acoustic conditions than do the classrooms in the newer buildings.

This is attributed to the low sound absorption coefficients of the Fig. 7. Proximity between classrooms and school yards.
interior finishes currently in use.
The RTs measured in all the schools showed the lack of acoustic impairs communication between students and teachers, since high
comfort in the classrooms. The acoustic deficiency in these spaces reverberation times diminish the intelligibility of speech.
632 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635

Table 3 Table 4
Noise levels in empty classrooms whose (open) windows look out on the school yard Noise levels in the classrooms during class

School Construction design LAeq (dB) LAmax (dB) LAmin (dB) School Construction design LAeq (dB) LAmax (dB) LAmin (dB)
S1 010 66.7 81.3 54.4 S3 022 74.0 85.6 53.8
S2 010 66.0 80.5 54.8 S6 023 73.7 94.5 54.8
S3 022 74.6 88.7 52.6
S4 022 62.5 73.5 50.6

Table 5
Recommendations for reverberation time RT
4.3. Sound insulation
Country Reverberation time (s)

Although several authors [4,45,46] state that ambient noise lev- Brazil 0.4 < RT 6 0.6 – classrooms with 120 < V 6 300 m3
French (cited in 0.4 < RT < 0.8 – classrooms with V 6 250 m3
els and reverberation time are the most important parameters that
WHO [39]) 0.6 < RT < 1.2 – classrooms with V > 250 m3
affect the acoustic quality of classrooms, sound insulation should Germany RT between 0.8 and 1.0 – classrooms with V bis 250 m3
not be disregarded. RT between 0.9 and 1.1 – classrooms with V bis 500 m3
Sound insulation should be a priority in school environments RT between 1.1 and 1.2 – classrooms with V bis 750 m3
Japan RT = 0.6 – classrooms with V  200 m3
where the sources of noise cannot be altered, especially in schools
RT = 0.7 – classrooms with V  300 m3
affected by high levels of noise from road, air and railroad traffic. UK RT = Tmfa < 0.6
Another important factor is the sound insulation between quiet RT = Tmfb < 0.8
and very noisy spaces, as in the case of the schools of designs USA RT = 0.6 – classrooms with V < 283 m3
010 and 022, where physical-education classes are held in school- RT = 0.7 – classrooms with 283 m3 < V 6 566 m3
Classrooms with V > 566 m3 see ANSI S12.60 for special
yards located very close to the classrooms.
recommendations on the control of RT in these spaces
Due to the complexity of the measuring process in terms of the
a
quantity of equipment and number of people involved, sound insu- For primary schools: classrooms, class bases, general teaching areas, small
groups rooms.
lation measurements were taken in only one school of each design. b
For secondary schools: classrooms, general teaching areas, seminar rooms,
After the field measurements, the data were transferred to the tutorial rooms, language laboratories.
Qualifier software (BK 7830), which processes all the data collected
and provides a single sound insulation value R0w – weighted appar-
ent sound reduction index, between rooms in a building and R045 ;w
– weighted apparent sound reduction index, sound insulation of and Table 7 show the single value of sound insulation of the façade
façades. The Figures below display the result of data processing for classrooms S1, S3 and S6.
for the determination of sound insulation between the classrooms The sound insulation of façades recommended by the ANSI
and the corridor R0w , and the sound insulation of the façades R045 ;w . S12.60 standard [34] is 50 dB. As indicated in Table 7, the values
The determination of ‘‘Rating of sound insulation in buildings and obtained for R045 ;w are far below the desirable level. The façades
of buildings elements” was done according to ISO 717-1–Airborne of designs 010 and 023 have a large area of window, which contrib-
sound insulation, using the software Qualifier BK 7830. Figs. 9–11 utes to the reduction of the sound insulation. The classrooms of

dB
80

70

60

50

40

30

20
20 31,50 63 125 250 500 1000 2000 4000 8000 A Hz
Leq
Cursor:1600 HzLeq=69,5 dBLFMax=74,5 dBLFMin=52,0 dB

Fig. 8. Frequency analysis of the noise produced by physical-education activities.

 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635 633

Table 6
RT measurement results (mean and standard deviation – SD) for two classrooms of each constructive design evaluated

Frequency School Standard School Standard School Standard School Standard School Standard School Standard
(Hz) S1 010 S2 010 S3 022 S4 022 S5 023 S6 023
RT mean SD RT mean SD RT mean SD RT mean SD RT mean SD RT mean SD
125 1.34 0.03 1.45 0.27 2.84 0.00 2.71 0.35 3.31 0.39 3.29 0.18
250 1.10 0.16 0.89 0.04 1.99 0.04 1.70 0.08 2.12 0.23 2.64 0.15
500 1.02 0.03 0.72 0.02 2.04 0.09 1.77 0.19 1.44 0.05 2.10 0.04
1000 1.04 0.01 0.89 0.14 2.20 0.07 1.59 0.07 1.34 0.03 1.99 0.08
2000 0.93 0.05 0.70 0.00 2.09 0.02 1.76 0.23 1.27 0.02 1.91 0.03
4000 0.78 0.01 0.64 0.00 1.62 0.02 1.44 0.13 1.10 0.00 1.55 0.02

dB R'45° R'45°w = 21 dB
Table 7
Sound insulation of façades 30

School Construction design R045 ;w (dB)

25
S1 010 15
S3 022 26
S6 023 21 20

15

dB R'45° R'45°w = 15 dB 10
20

5
15

0
10
125 250 500 1k 2k 4k Hz

Fig. 11. Weighted apparent sound reduction index – sound insulation of façade –
5
school S6 – Luarlindo dos Reis Borges – standard 023.

0
dB R' R'w = 22 dB
60
-5

-10
50
125 250 500 1k 2k 4k Hz

Fig. 9. Weighted apparent sound reduction index – sound insulation of façade –

school S1 – Alfredo Parodi – standard 010. 40

30

dB R'45° R'45°w = 26 dB
35
20

30

10
25

20 0
125 250 500 1k 2k 4k Hz
15
Fig. 12. Weighted apparent sound reduction index – sound insulation between the
classroom and the corridor – school S1 – Alfredo Parodi – standard 010.

10

5 design 022 have two different façades, one composed of windows

and the other of doors. The sound insulation of this design is supe-
125 250 500 1k 2k 4k Hz
rior to that of the others, since the measurements were taken of the
Fig. 10. Weighted apparent sound reduction index – sound insulation of façade – façade with doors. This fact indicates that the maximum sound
school S3 – Aníbal K. Neto – standard 022. insulation supplied by the façades of design 022 is 26 dB.
634 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635

dB R' R'w = 13 dB 5. Conclusions

60
The acoustic measurements and physical evaluation of the three
construction designs revealed design errors in all the schools of
50
this study. The errors involve both the architectural design and
the materials used in the interior finish of the schools.
40
The main error in the schools of designs 010 and 022 is the po-
sition of the schoolyards and recreation spaces. In design 023, the
use of slotted glass vents in the walls separating the classrooms
30 from the halls allows noise from the other classrooms and from
the hall to penetrate into the classroom, thus increasing the ambi-
ent noise.
20 As for reverberation times, it was found that the old classrooms
(design 010) offer better acoustic conditions than the more modern
ones (designs 022 and 023), due to the finishing materials em-
10 ployed, especially for the floors and ceilings. This finding indicates
the need for classrooms to be modified in order to increase their
areas of sound absorption.
0
The levels of ambient noise measured in the classrooms were
higher than recommended by the NBR 10152 standard. The inade-
quate noise levels in these spaces are in large part due to the sound
-10
125 250 500 1k 2k 4k Hz insulation of both façades and dividing walls, which were found to
be far below the level specified by the ANSI S12.60 standard.
Fig. 13. Weighted apparent sound reduction index – sound insulation between the The findings of this study reveal a lack of acoustic comfort in
classroom and the corridor – school S6 – Luarlindo dos Reis Borges – standard 023.
classrooms and point to the need for interventions. Such renova-
tions would involve not only the schools of this study, but all the
schools built according to the three designs evaluated here. It
should be kept in mind that the architectural designs discussed
Table 8
here are also built in many other regions of the state of Paraná.
Sound insulation between classrooms and corridors
Thus, the errors or acoustic deficiencies in one school are very
School Construction design R0w (dB) likely repeated in all the others, impairing the learning ability of
S1 10 22 hundreds of students throughout the state.
S6 23 13

Acknowledgements

The authors gratefully acknowledge CNPq (Brazil) for the finan-

As for the sound insulation between classrooms and corridors, cial aid granted through a Master’s scholarship, the Government of
the recommendation of ANSI S12.60 standard [34] is 45 dB. Figs. the Federal Republic of Germany (DAAD – Deutscher Akademi-
12 and 13 and Table 8 lists the values obtained for designs 010 scher Austauschdienst) for the donation of equipments and soft-
and 023, since the 022 design has no corridors. wares. The authors are indebted to the directors, teachers and
The values of sound insulation between classrooms and corri- students of the state schools in the cities of Curitiba and Pinhais
dors shown in the above table are far below those specified by for their cooperation with and participation in this research.
the standard. The difference between the values obtained for the
two designs is related to the type of vent at the top of the walls References
separating the two spaces. In design 010, the ventilation slits con-
sist of adjustable metal-framed glass louvers, while in design 023 [1] Lubman D, Sutherland LC. Good classroom acoustics is a good investment. In:
International congress on acoustics ICA, Rome, Italy; 2001. Proceedings.
they consist of slotted glass vents which let noise directly into [2] Hagen M, Huber L, Kahlert J. Acoustic school desining. In: Forum acusticum,
the classrooms, as illustrated in Fig. 14. Seville, Spain; 2002. Proceedings.

Fig. 14. Detail of the vents in the walls between classrooms and corridors.
 P.H.T. Zannin, D.P.Z. Zwirtes / Applied Acoustics 70 (2009) 626–635 635

[3] Loro CLP. Avaliação acústica de salas de aula – Estudo de caso em salas de aula [26] Dockrell JE, Shield B. Children’s perceptions of their acoustic environment at
Padrão 023 da rede pública. Dissertation (Master’s), Universidade Federal do school and at home. J Acoust Soc Am 2004;115(6):2964–73.
Paraná, Curitiba; 2003. [27] Associação Brasileira de Normas Técnicas (ABNT). NBR 10151: Acústica–
[4] Seep B, Glosemeyer R, Hulce E, Linn M, Aytar P. Acústica de salas de aula. Avaliação do ruído em áreas habitadas, visando o conforto da comunidade –
Revista de Acústica e Vibrações, n 29; 2002. Procedimento. Rio de Janeiro; 2000.
[5] Sala E, Viljanen V. Improvement of acoustic conditions for speech [28] ABNT. Brazilian standard for the Acoustical Comfort in Buildings NBR 10152
communication in classrooms. Appl Acoust 1995;45:81–91. (in Portuguese); 1987.
[6] Ercoli L, Azzurro A. Case study: the acoustical characteristics of typical [29] International Organization for Standardization. ISO 140-4: acoustic –
argentinean classrooms. Build Acoust 2001;8(4):301–10. measurement of sound insulation in buildings and of building elements –
[7] Hodgson M. Empirical prediction of speech levels and reverberation in Part 4: field measurements of airborne sound insulation between rooms.
classrooms. Build Acoust 2001;8(1):1–14. Switzerland; 1978.
[8] Shield B, Dockrell JE. External and internal noise surveys of London primary [30] International Organization for Standardization. ISO 140-5: acoustic –
schools. J Acoust Soc Am 2004;115(2):730–8. measurement of sound insulation in building and of building elements –
[9] Hodgson M. Case-study evaluations of the acoustical designs of renovated Part 5: field measurements of airborne sound insulation of façade elements
university classrooms. Appl Acoust 2004;65:69–89. and façades. Switzerland; 1998.
[10] Hagen M, Kahlert J, Hemmer-Schanze C, Huber L, Meis M. Developing an [31] International Organization for Standardization. ISO 3382: acoustics –
acoustic school design: steps to improve hearing and listening at schools. Build measurement of the reverberation time of rooms with reference to other
Acoust 2004;11(4):294–307. acoustical parameters. Switzerland; 1997.
[11] Maffei L, Dragonetti R, Lembo P, Romano R. Assessment of large-scale action [32] International Organization for Standardization. ISO 717-1: acoustics – rating of
plans to reduce external background noise in school buildings. Build Acoust sound insulation in buildings and of building elements – Part 1: airborne
2004;11(4):259–69. sound insulation. Switzerland; 1996.
[12] Krüger EL, Zannin PHT. Acoustic, thermal and luminous comfort in classrooms. [33] NBR12179. Brazilian standard for acoustical treatment of closed rooms. Rio de
Buildi Environ 2004;39:1055–63. Janeiro, Brazil (in Portuguese); 1992.
[13] Yang WY, Hodgson M. Acoustical evaluation of preschool classrooms. Noise [34] American National Standard. ANSI S12.60: acoustical performance criteria,
Control Eng J 2005;53(2):43–52. design requirements, and guidelines for schools. Melville; 2002.
[14] Kennedy SM, Hodgson M, Edgett LD, et al. Subjective assessment of listening [35] Ernesti W. DIN 4109. Schallschutz im Hochbau. Verlag Ernst Vögel Gmbh (in
environments in university classrooms: perceptions of students. J Acoust Soc German); 1999.
Am 2006;119(1):299–309. [36] Bobran HW. Handbuch der Bauphysik 7. völlig neubearbeitete Auflage,
[15] Yang WY, Hodgson M. Auralization study of optimum reverberation times for BauVerlag, Stuttgart (in German); 1995.
speech intelligibility for normal and hearing-impaired listeners in classrooms [37] Fasold W, Veres E. Schallschutz + Raumakustik in der Praxis. Verlag Bauwesen
with diffuse sound fields. J Acoust Soc Am 2006;120(2):801–7. (in German); 2003.
[16] Hodgson M, Scherebnyj K. Estimation of the absorption coefficients of the [38] Hohmann, Setzer, Wehling. Bauphysikalische Formeln und Tabellen. Werner
surfaces of classrooms. Appl Acoust 2006;67(9):936–44. Verlag (in German); 2004.
[17] Yang W, Hodgson M. Validation of the auralization technique: comparative [39] World Health Organization – WHO. Noise in schools. Geneva; 2001.
speech-intelligibility tests in real and virtual classrooms. Acta Acust United [40] Pimentel SF. Efeito do ruído no homem dormindo e acordado. In: Encontro da
Acust 2007;93:991–9. Sociedade Brasileira de Acústica em Simpósio Internacional – SOBRAC,
[18] Yang WY, Hodgson M. Ceiling baffles and reflectors for controlling lecture- vol. 19. Proceedings. Available at: <http://www.icb.ufmg.br/lpf/pimentel,
room sound for speech intelligibility. J Acoust Soc Am 2007;121(6):3517–26. sobrac2000.html>; 2000 [accessed 27.12.04].
[19] Yang W, Hodgson M. Optimum reverberation for speech intelligibility for [41] Bradley J, Reich R, Norcross S. On the combined effects of signal-to-noise ratio
normal and hearing-impaired listeners in realistic classrooms using and room acoustics on speech intelligibility. J Acoust Soc Am USA
auralization. Build Acoust 2007;14(3):163–77. 1999;106(4):1820.
[20] Zannin PHT, Marcon CR. Objective and subjective evaluation of the acoustical [42] Candrell CC, Smaldino JJ. Classroom Acoustics for children with normal hearing
comfort in classrooms. Appl Ergonom 2007;38:675–80. and with hearing impairment. Lang Speech Hear Ser School USA
[21] Zannin PHT, Loro CP. Measurement of the ambient noise level, reverberation 2000;31:362–70.
time and transmission loss for classrooms in a public school. Noise Control Eng [43] Fukuchi T, Ueno K. Guidelines on acoustic treatments for school buildings
2007;55(3):327–33. proposed by the Architectural Institute of Japan. In: ICA – international
[22] Astolfi A, Pellerey F. Subjective and objective assessment of acoustical and conference on acoustic, Kyoto, Japan; 2004.
overall environmental quality in secondary school classrooms. J Acoust Soc Am [44] Building Bulletin 93, Acoustic design of schools – a design guide. London,
2008;123(1):163–73. 2003. (Retrieved from http://www.cie-ltd.co.uk/iceilings/Soundfield/bb93_
[23] Evans GW, Maxwell LE. The effects of noise on pre-school children’s pre- downloads.htm on July 2008).
reading skills. J Environ Psychol 2000;20:91–7. [45] Shield BM, Jeffery RL. A survey of noise levels in and around Primary Schools in
[24] Lercher P, Evans GW, Meis M. Ambient noise and cognitive processes among London. In: Proceedings of international congress on acoustics, vol. 17, Rome;
primary schoolchildren. Environ Behav 2003;35(6):725–35. 2001. Proceedings.
[25] Shield BM, Dockrell JE. The effects of noise on children at school: a review. [46] Karabiber Z, Vallet M. Classroom acoustics policies – an overview. In:
Build Acoust 2003;10(2):97–116. Euronoise, Naples, Italy; 2003.