Abstract
Façade airborne sound insulation is crucial for protection of indoor environment from environmental noise. In Turkey, sound insulation in new buildings is bound by law, but 6 million pre-1980 dwellings with thin brick walls and single-glazed windows used in highly transparent façades should be retrofitted. In this study, sound reduction index of masonry and cavity exterior walls which consist of brick, mortar, gypsum board and mineral wool and of common window types is measured in sound insulation test rooms. The study compares and evaluates the effects of plaster, brick thickness, cavity depth, mineral wool thickness and mineral wool placement on sound reduction index values, using traditional materials and building techniques. Traditional brick wall façades and possible retrofitting of these façades are evaluated for sound insulation of bedrooms and living rooms in different noise zones, 55, 60, 65, 70 and 75 dBA, with various transparency ratios, 0%, 30%, 40%, 50% and 70%. The analysis shows that window types and single-layer walls are the deterministic factors in evaluating sound insulation in retrofitting projects and that it is not possible to provide proper aural comfort in high noise zones.
Introduction
‘Assessment and Management of Environmental Noise Directive’ (2002/49/EC), 1 published by Republic of Turkey Ministry of Environment and Forestry, aims to provide the necessary measures against the deterioration of mind and body health and peace and tranquillity of people due to exposure to environmental noise. The Directive stipulates that sound reduction index of the building elements which surround rooms sensitive to noise shall either be calculated by TS EN 12354 2 standard series or be measured according to TS EN ISO 10140 3 standard series to ensure limit background noise levels. According to the Directive, architectural projects of new buildings should assure Leq room limit noise values given in Annex-VII of the Directive. 1 The Directive also provides noise exposure categories of A, B, C and D, to be taken into consideration for planning of new residential areas. 1 The Directive is still in implementation process; noise maps are being formed to lay out noise zones, and sound insulation values of building elements are being measured in laboratories.
Façade airborne sound insulation is the most important factor for protection of indoor environment from environmental noise. Although building elements are bound by standards, their properties may differ according to the country they are manufactured in and their application conditions. That is why it is important to measure sound insulation of building elements in each country individually. Laboratory measurements give more reliable results than theoretical methods especially in hollow blocks and multilayer elements. The traditional building material of Turkish façades, brick, is chosen as the main wall element of this study. The earliest mud bricks were used in Western Turkey around 7500 BC. 4 Modern and standardized clay bricks are manufactured locally and used in 95% of residential buildings. 5
After the two major oil crises in 1970s, energy efficiency in building stock became a major concern around the world; thus, Thermal Insulation Directive 6 was introduced in Turkey at the beginning of 1980s. Following this Directive, the windows covering the whole dwelling façade became much smaller, starting a new architectural era in Turkey. 7 First double insulating glass in Turkey was manufactured in 1974 and it started to be commonly used at the beginning of 1980s. In Turkey, there are a total of 19 million 482 thousand households, only 65.3% of which were constructed after 1980 8 which leaves more than 6 million households with façades using single-glazed glass in large window proportions and thin walls, maximum 145 mm thick, to be retrofitted.
Traditional housing façades in Turkey are commonly made of hollow brick manufactured in Turkey and single-glazed windows. Retrofitting of these traditional façades is crucial in terms of improving sound insulation of the existing housing stock. In this study, sound reduction index of masonry and cavity exterior walls which consist of brick, mortar, gypsum board and mineral wool is measured in sound insulation test rooms, and sound insulation of combined façades is calculated for different noise zones.
Some scientific studies were made on sound insulation of brick walls over the years. Fringuellino and Smith 9 worked on sound transmission through hollow brick walls with varying thickness, materials and types of holes. The work by Guillen et al. 10 was on sound reduction index laboratory measurements of masonry–air cavity–brick walls and masonry–air cavity–gypsum board walls, masonry walls being clay blocks, concrete blocks and hollow bricks. Building Research Establishment (BRE) has published numerous research reports on sound insulation over the years. Sewell et al. 11 from BRE worked on the comparison of the results for plastered brick and dense blockwork party walls associated with plastered lightweight masonry inner leaves of external walls to earlier results from similar party walls associated with heavier masonry or lightweight panelling external wall leaves to find whether sound insulation is reduced. National Research Council of Canada (NRC) has conducted extensive research on sound insulation. Warnock 12 from NRC studied the sound transmission loss measurements through 190- and 140-mm blocks with added drywall and through cavity block walls.
The literature on sound insulation of gypsum board walls is quite extensive. Halliwell et al. 13 made laboratory measurements in NRC for 350 types of gypsum board wall systems and reported them. Some other significant works from NRC were by Nightingale et al. 14 on factors affecting sound insulation of gypsum board walls, such as electrical outlet boxes, on framing details 15 and by Bradley and Birta 16 on resilient supports. The influence of air layers and damping layers between gypsum boards was evaluated by Bravo et al. 17 The effect of mineral wool density of gypsum double walls was investigated by Uris et al. 18 Matsumoto et al. 19 researched on the development of lightweight gypsum drywalls with high sound insulation performance. Uris et al. 20 studied the effect of additional perforated boards on single-frame double-layer gypsum board partitions.
Although the Directive 1 works on architectural projects of new buildings, its sound insulation principles may also be used in retrofitting projects. Retrofitting the existing buildings is usually more cost-effective and sustainable than building new facilities. Guigou-Carter et al. 21 worked on comparing the measured and predicted sound insulation for a thermal retrofitted building.
This study is aimed to evaluate laboratory sound insulation measurements of Turkish brick walls and to determine brick wall façades to be used in different noise zones according to the Directive. 1 This study compares and evaluates the effects of plaster, brick thickness, cavity depth, mineral wool thickness and mineral wool placement on sound reduction index values, using traditional materials and building techniques. Traditional brick wall façades and possible retrofitting of these façades are evaluated for sound insulation in different noise zones with various transparency ratios.
This study guides the retrofitting process of old buildings and building of new dwellings to provide proper sound insulation against the most common type of environmental noise, road traffic noise. It provides numerous brick-based wall types possible to use with different window types and various transparency ratios. It considers the façade elements individually as well as the façade as a whole, in terms of sound insulation. It would help provide aural comfort conditions in retrofitted buildings and new dwellings.
Measurement technique and building element properties
Sound reduction index measurements are carried out at Istanbul Technical University, Faculty of Architecture, Building Physics and Environmental Control Laboratory, Acoustics Unit, sound insulation test rooms, constructed in accordance with ISO 10140-5. 22 Two adjacent reverberant rooms with suppressed flanking transmission are constructed as source (50 m 3 ) and receiver (105 m 3 ) rooms. The dimensions of the rooms are not similar and the walls are not parallel. Both rooms have sound-diffuser panels with various sizes and angles, hanging from the ceiling. The two rooms are connected through a test opening of 10 m 2 (2.3 m × 4.34 m) where the wall specimens are measured. The measurement technique, properties of materials and types of walls measured are explained in this section.
Measurement technique
For the measurements, B&K Pulse analyser platform, a desktop computer, B&K 2716 power amplifier, B&K 4296 omnisource loudspeaker and two B&K 4165 1/4 in microphones are used. Laboratory airborne sound insulation of wall building elements is measured according to ISO 10140-2 23 and ISO 10140-4. 24 Two microphones are placed, one in each room, to record sound pressure levels in source and receiver rooms, simultaneously. Meanwhile, loudspeaker generates the sound field in the source room. The measurements are taken for 20 s. The loudspeaker has two positions in the source room and the microphones have five positions in their rooms. The background noise is measured for checking receiver room sound pressure level. Reverberation time is measured in the receiver room with two loudspeaker positions and three microphone positions. Sound reduction index values, R, in 1/3 octave frequency bands between 100 and 3150 Hz, are derived as a result of these measurements.
The specifications for measuring sound reduction index for windows are given in ISO 10140-1 25 and ISO 10140-5 22 standards. The dimensions of the windows are 1250 mm × 1500 mm. The rest of the test opening is a filler wall, which is a double-brick cavity wall: 135- and 85-mm plastered bricks separated by a 100-mm cavity filled with mineral wool. The niches on both sides of the window have different depths, about 90 and 180 mm. The installation of the windows is similar to the method used in practice. The windows are both side hung and bottom hung. The sides of the windows are made airtight using an elastic sealant on both sides.
According to Annex-II, Article 4 of Assessment and Management of Environmental Noise Directive (2002/49/EC), 1 the calculation and measurement results need to be given in 1/3 octave bands, as sound reduction index, in decibels. The single-number quantity of sound reduction index, R, in 1/3 octave frequency bands, is expressed as weighted sound reduction index, Rw. 26
ISO 717-1 26 also takes into consideration the different sound level spectra of various noise sources such as noise sources inside a building and traffic outside a building, using Rw with spectrum adaptation terms C and Ctr. Even though building acoustics work between 100 and 3150 Hz, due to light construction techniques, some problems might occur in lower frequencies, such as music and television sounds in neighbouring rooms. To deal with this issue, spectrum adaptation terms for different frequencies’ weightings are applied. Spectrum no. 1 is C, and it is for A weighted pink noise. Spectrum no. 2 is Ctr, and it is for A weighted traffic noise. Spectrum adaptation terms’ values are added to the single-number rating, Rw, to take into account the characteristics of particular sound spectra and to characterize sound reduction for these noise types.
Properties of measured masonry brick walls
The vertically perforated brick blocks with thicknesses of 145, 190 and 240 mm are used in this study. The decision was made in collaboration with the manufacturer as these are the most commonly used products. The types of bricks used are categorized in EN 771-1 27 as clay masonry units (Category I) with low gross density (LD units). The material properties and acoustical properties of brick blocks used are given in Table 1.
Properties of vertically perforated bricks.
LD: low gross density.
For each type of brick, as the first step, only the exterior surface is treated with plaster; in the second step, the inner surface is also treated with plaster. Three centimetres of cement plaster blended with quicklime is applied on the exterior, whereas 2 cm of gypsum plaster with an aggregate of expanded perlite is applied on the interior. As the third step, 4-cm rigid mineral wool insulation board is installed on plastered exterior wall surface and 3-mm cement-based moisture permeable plaster is applied for finishing. This a common practice in Turkey for the mandatory thermal insulation of buildings. Table 2 gives a detailed list and drawings of the masonry brick walls used in this part of the study.
Description and drawings for measured masonry brick walls, weighted sound reduction index and spectrum adaptation terms, Rw (C; Ctr).
As shown in Table 2, A1 wall consists of a 14.5-cm-thick vertically perforated brick wall with 3 cm of cement plaster blended with quicklime on the exterior surface. A2 wall is A1 wall treated with 2-cm gypsum plaster with an aggregate of expanded perlite on the inner surface. A3 wall is A2 wall installed with 4-cm rigid mineral wool insulation board on exterior wall surface and finished with 3 mm of cement-based moisture permeable plaster. Walls B1 and C1 are essentially the same as A1; only the thicknesses of bricks used are 19 and 24 cm, respectively. The same principle applies between A2, B2, C2 walls and A3, B3, C3 walls.
Properties of measured cavity walls
Sound reduction index measurements are made for cavity walls with an exterior leaf of vertically perforated masonry brick wall and an inner leaf of gypsum board on steel frame. The properties of the gypsum boards used in the study are given in Table 3.
Properties of gypsum board.
Brick walls of thicknesses 14.5, 19 and 24 cm are treated with 30-mm-thick cement plaster blended with quicklime on the exterior surface and nothing on the interior surface. An independent frame of 30-mm U profiles is installed, ending at 50 or 100 mm from the wall. At every 600 mm, 30-mm C profiles are used as steel studs and placed inside the frame. All profiles are enclosed with elastic band, and steel stud profiles have been attached to the wall with brackets. A drawing and two photographs of these brackets can be found in Figure 1. The photograph on the left is a bracket attached to a wall, and the one in the middle shows straight and bent versions of the bracket. The two tips of the bent version in the photograph are to be folded again according to the cavity thickness and then mounted to stud profile. The drawing on the right shows a section of wall, with bracket and board relationship.
Drawing and photographs of brackets.
A detailed list and drawings of the cavity brick walls used in this part of the study are given in Table 4.
Description and drawings for measured cavity brick walls with attached gypsum board, weighted sound reduction index and spectrum adaptation terms, Rw (C; Ctr).
Properties of measured windows
The windows measured in this study have polyvinyl chloride (PVC) frames, which is the most common frame in Turkey. The first window is a double-glazed window, identified as 4 + 16 + 4, with two 4-mm glass panes and 16-mm cavity. This type of window was chosen in collaboration with the manufacturer as these are the most commonly used windows. The second window is also double-glazed, but one glazing is laminated. It is identified as 4 + 12 + 4 + 0.76 + 4, double 4-mm panes laminated with a filter, and another 4-mm-thick glass used with 12-mm air cavity. This type of window is used for spaces with noise insulation problems.
Measurement results and discussion
In this section, the results of the measured masonry brick walls, cavity brick walls with attached gypsum board and windows are discussed.
Measurement results and discussion for masonry brick walls
Figure 2 shows the sound reduction index values of walls A1, A2, A3, B1, B2, B3, C1, C2 and C3 in 1/3 octave bands between 100 and 3150 Hz frequencies. As the walls are mainly single-leaf walls, analysis is based on the three regions of behaviour for a homogeneous wall; region 1 stiffness-controlled, region 2 mass-controlled and region 3 damping-controlled, separated by resonance frequency and critical frequency, respectively. 28 The sound insulation characteristics of hollow brick walls are found to be quite complex. Because of the perforations along the body of the brick block, there may be several different stiffness values in different directions. Therefore, critical frequency dip may be broad. 9 In high-frequency range, sound reduction index may be affected by critical frequency of hollow block’s thin surface. This will result in a critical frequency dip in high frequencies. 12 For thick plates, lowest thickness resonance is effective in high frequencies and is followed by a plateau. Unfortunately, thickness resonance frequency is difficult to predict for hollow plates. 29 This effect can significantly lower sound reduction index curve for hollow walls within the frequency range of 100–5000 Hz. 9
Comparison of all masonry walls in the study.
Figure 2 may be analysed in parts, the first part being all walls including 14.5-cm brick walls A1 (one side plastered), A2 (both sides plastered) and A3 (with exterior insulation board). As 14.5-cm brick has a critical frequency of 74 Hz, the system is generally in the damping region and damping is effective on sound reduction index values. Below 315 Hz, A3 wall has the lowest sound reduction index values and A1 wall has the highest. Above 315 Hz, the results are just the opposite. Rw + Ctr of all these three walls are 40 dB, which means the traffic noise resistance is the same.
For the second part of analysis, all walls including 19-cm brick walls B1, B2 and B3 are compared. In the region below resonance frequency, 127 Hz, stiffness is effective and insulated wall, B3, has lower values than B1 and B2. Although sound reduction index values of B2 wall are higher than B1 wall, they decrease in the 250–315 Hz frequency interval.
When 24-cm brick walls, C1, C2 and C3, are compared, it is found that C2 wall has higher sound reduction index values than C1 above 630 Hz. Frequencies below resonance frequency, 159 Hz, are stiffness-controlled, so like the other walls without gypsum plaster, C1 has higher insulation values in this region. For frequencies higher than resonance frequency, because of region 3, damping is effective and the insulated wall, C3, has higher insulation values. Rw (C; Ctr) value of C2 wall is the same as C1 wall, even though gypsum plaster is applied, while the value increases in a similar situation for 14.5- and 19-cm-thick walls.
For all walls, dips around calculated critical frequencies are broad, because of different stiffness values of perforated brick blocks. Dips in high frequencies are also clearly visible for all wall types. These dips are compatible with the thin surface thickness of the perforated bricks. The 14.5-, 19- and 24-cm brick blocks have thin clay surfaces that are 8.5, 10 and 13.5 mm thick, respectively. Critical frequencies for these surfaces are calculated to be 1800, 1530 and 1130 Hz, respectively. The dips around these frequencies can be seen in Figure 2.
General results of the comparison between walls with the same brick thickness are that low frequencies are controlled by stiffness, so walls with no gypsum plaster have higher insulation values in this region, whereas in higher frequencies where damping is effective insulated walls have much higher values.
When walls which have cement plaster but do not have gypsum plaster, A1, B1 and C1 walls, are compared, it is found that A1 has lower value than C1 up to 500 Hz and higher value than B1 between 500 and 2500 Hz. Sound reduction index values of 24-cm-thick brick wall, C1, are in general higher than 14.5 and 19 cm walls, A1 and B1, in all the frequency range.
In Figure 2, if walls A2, B2 and C2 (A1, B1 and C1 walls treated with gypsum plaster) are studied, it is seen that all three walls have the same sound insulation value at 500 Hz. A2 wall has lower sound reduction index values than walls B2 and C2 in 100–315 Hz range, but it has higher values above 500 Hz, especially higher than B2. B2 has higher sound insulation values in 125–500 Hz range but has lower values than A2 and C2 walls in frequencies higher than 500 Hz. The resistance against traffic noise, Rw + Ctr, is the same for A2 and B2, which is 40 dB. For pink noise, 14.5-cm-thick brick wall is more effective, whereas 24-cm-thick wall is more effective for traffic noise. Rw values for walls A2 and C2 are the same, 43 dB.
Finally, A3, B3 and C3 walls which have 4-cm rigid mineral wool insulation board installed on exterior wall surface are compared. The 14.5-cm-thick brick wall has lower sound insulation values than others in low frequencies and has higher values after 250 Hz. Comparing Rw (C; Ctr) values shows that B3 wall has a 1 dB higher sound insulation value against traffic noise than A3 wall, although 14.5- and 19-cm-thick brick walls have the same value, 45 dB, for pink noise. The wall which includes 24-cm-thick brick has higher Rw (C; Ctr) values than all the others.
As an overall result, adding plaster on one side increases Rw values up to 1 dB, and Rw + C and Rw + Ctr values up to 2 dB; adding exterior mineral wool insulation board increases Rw values by 4–6 dB, Rw + C by 2–5 dB and Rw + Ctr up to 2 dB.
Measurement results and discussion for cavity walls
For two panels separated by an airspace, sound insulation is mainly affected by the depth of cavity and sound insulation performances of each panel. The behaviour of sound insulation over the frequency spectrum may be divided into three regimes, A, B and C. Regime A occurs in low frequencies, below f0, where airspace has a negligible effect and panels act as one. The total mass of the panels determines the sound reduction in this regime. f0 is the resonant frequency of the two panels coupled by airspace, also defined as double-wall resonance, which is determined by cavity depth and panel masses. Regime B occurs in frequencies higher than f0, where standing waves are set up in the airspace. Transmission loss values of each panel and cavity depth play role in the total sound insulation performance in this regime. In Regime C, panels act independently and the airspace in between acts as a small room. Sound reduction in this regime is influenced by transmission loss values of each separate panel and surface absorption coefficient for the panels. Regimes B and C are separated by the frequency defined as ‘c/2πd’ which is affected by the depth of cavity. 28
Figure 3 shows the sound reduction index values of walls including 14.5-cm-thick brick blocks, A1, A4, A5, A6, A7 and A8, in 1/3 octave bands between 100 and 3150 Hz frequencies.
Comparison of 14-cm brick masonry wall and all 14-cm brick cavity walls with attached gypsum board.
Comparison of A1 masonry wall to the wall with attached gypsum with 5-cm cavity, A4, and the wall formed by increasing the cavity to 10 cm, A6, in Figure 3 shows the effect of adding a gypsum board with cavity to a solid wall, especially in high frequencies. Although the walls with attached gypsum board have high acoustical performance in high frequencies, masonry wall has higher values in some of the lower frequencies. Adding a gypsum board with cavity to the brick wall increases stiffness of the system and sound reduction index decreases in low-frequency range.
Resonant frequency coupled by airspace is 155 Hz for the 5-cm-thick cavity and drops to 110 Hz for the 10-cm-thick cavity. In frequencies lower than these, the total mass is effective on sound insulation values. In Regime B, cavity depth is in effect up to 1100 and 550 Hz, for 5- and 10-cm-thick cavity walls, respectively. Higher than these frequencies is Regime C and absorption takes effect. The figure clearly shows that the effect of cavities on sound reduction index values is strong in high frequencies.
In Figure 3, comparing A4 wall and A5 wall (the same wall filled with 4 cm of mineral wool) shows that installing mineral wool into cavity increases sound reduction index values especially in low- and medium-frequency regions. Below 155 Hz, these two walls have similar sound insulation values because the total mass of the system is in effect. Above 1100 Hz, the sound reduction values of the two walls are not far apart.
A6 wall is a 14.5-cm-thick brick wall with 10-cm cavity and attached gypsum board; for A7 and A8 walls, 4 and 8 cm of mineral wool are added into the cavity, respectively. The sound reduction index value of A6 wall drops to almost 32 dB in 125 Hz region, near where resonant frequency coupled by airspace occurs. Filling 8/10 of the cavity rather than 4/10 of the cavity increases the sound reduction index values in low- and medium-frequency regions and decreases the effect resonant frequency coupled by airspace. Above 550 Hz, there is not much difference in the sound insulation values of A6, A7 and A8 walls.
In Figure 4, 19-cm-thick brick wall with cement plaster on one surface and attached gypsum board on the other is used to compare three combinations: 4-cm mineral wool in 5-cm-thick cavity, 4-cm mineral wool in 10-cm cavity and 8-cm mineral wool in 10-cm cavity. In low-frequency region, resonant frequency coupled by airspace around 155 Hz is distinctive. Although cavity depth is important in Regime B, for this case in low and medium frequencies, doubling the cavity depth from 5 to 10 cm does not seem to have a positive effect on sound reduction indices. Comparing B5 and B7, doubling the cavity depth increases sound reduction indices in high-frequency region, because of increased absorption of air and system. Region C, high-frequency region, is affected by absorption, but after about 550 Hz, B7 and B8 walls have almost the same sound insulation values, so doubling mineral wool thickness does not have that effect in this case.
Comparison of all 19-cm brick cavity walls with attached gypsum board.
In Figure 5, 24-cm-thick brick wall with cement plaster on one surface and attached gypsum board on the other is used to compare three combinations: 4-cm mineral wool in 5-cm-thick cavity, 4-cm mineral wool in 10-cm cavity and 8-cm mineral wool in 10-cm cavity.
Comparison of all 24-cm brick cavity walls with attached gypsum board.
Increasing the cavity depth from 5 to 10 cm, in wall C7, causes sound insulation increase in Regime B, low- and medium-frequency regions, as cavity depth is effective. The values in high-frequency region increase about 15 dB because of the increase in absorption of air and of the system. At C8 wall where mineral wool is increased to 8 cm, sound reduction index at resonant frequency coupled by airspace of 110 Hz is increased from 41 to 45 dB; although the overall increase in the whole frequency spectrum is about 5 dB.
Figure 6 is used to compare walls A5, B5 and C5, walls with brick thicknesses of 14.5, 19 and 24 cm and with attached gypsum boards 5 cm away from the wall, including 4-cm mineral wool inside the cavity. Sound reduction index of 14.5-cm brick wall, A5 wall, is lower than others below resonant frequency coupled by airspace of 155 Hz. A similar pattern occurs with 19- and 24-cm-thick brick walls as well. Although 24-cm brick wall has higher sound reduction index values than others in low frequencies, it has lower values than others in higher frequencies even though perforation ratios of all are similar. Rw + C values regarding pink noise and Rw + Ctr values against traffic noise for 14.5-cm brick wall are 52 and 49 dB, for 19-cm brick wall are 54 and 50 dB and for 24-cm brick wall are 54 and 51 dB, respectively. Even though 24-cm brick wall has higher values of insulation against traffic noise than 14.5- and 19-cm brick walls, for pink noise, it is similar to 19-cm brick wall.
Comparison of 14.5-, 19- and 24-cm-thick brick cavity walls with attached gypsum board, 5-cm cavity filled with 4 cm of mineral wool.
Figure 7 demonstrates sound reduction index values of 14.5-, 19- and 24-cm brick walls with 4 cm of mineral wool in 10-cm cavity. The 14.5-cm brick wall has a lower mass than the other two, but they all have the same perforation ratio. The drop effect of resonant frequency coupled by airspace at 110 Hz can be seen for walls A7 and B7. The wall including 19-cm-thick brick has higher sound reduction index values than the one including 14.5-cm brick, due to thickness and mass. The same effect is noticed between 24- and 19-cm brick walls.
Comparison of 14.5-, 19- and 24-cm-thick brick cavity walls with attached gypsum board, 10-cm cavity filled with 4 cm of mineral wool.
Figure 8 is used to compare walls A8, B8 and C8, walls with brick thicknesses of 14.5, 19 and 24 cm and with attached gypsum boards 10 cm away from the wall, including 8-cm mineral wool inside the cavity. The wall including 24-cm brick has significantly higher sound reduction index values than the other two because of mass. As the mass of brick wall increases and sufficient cavity depth and sufficient thickness of mineral wool are supplied, the sound reduction index values increase in all frequencies.
Comparison of 14.5-, 19- and 24-cm-thick brick cavity walls with attached gypsum board, 10-cm cavity filled with 8 cm of mineral wool.
Figure 9 compares the sound reduction index values of A3, A5, B3, B5, C3 and C5 walls. A3 wall is a 14.5-cm-thick brick wall installed with 4-cm rigid mineral wool insulation board and plastered on exterior wall surface. A5 wall is a 14.5-cm-thick brick wall with attached gypsum board leaving a 5-cm cavity filled with 4-cm-thick mineral wool. The 19- and 24-cm brick walls have a similar pattern. Although the components of the walls are very similar, sound reduction index of the cavity walls is much higher than the others until about 700 Hz. Rw + C values regarding pink noise for 14.5-, 19- and 24-cm brick walls including exterior insulation are 45, 45 and 47 dB, and including cavity are 52, 54 and 54 dB, respectively. Rw + Ctr values regarding traffic noise for 14.5-, 19- and 24-cm brick walls including exterior insulation are 40, 41 and 43 dB, and including cavity are 49, 50 and 51 dB, respectively.
Comparison of walls with 4-cm exterior insulation board and cavity walls with 5-cm cavity filled with 4 cm of mineral wool.
An alternative comparison may be made between building elements with the same single-numbered quantities. A8 wall is a 14.5-cm-thick brick wall with attached gypsum board including 10-cm deep cavity filled with 8-cm-thick mineral wool. B7 wall is a 19-cm-thick brick wall with attached gypsum board including 10-cm deep cavity filled with 4-cm-thick mineral wool. Although Rw for these walls are 57 and 58 dB, respectively, Rw + C value is 56 dB and Rw + Ctr value is 52 dB for both walls. Increasing brick thickness by 4.5 cm and decreasing mineral wool thickness by 4 cm has resulted in the same pink noise and traffic noise resistance in this case.
A7 wall is a 14.5-cm-thick brick wall with attached gypsum board including 10-cm deep cavity filled with 4-cm-thick mineral wool. B5 wall is a 19-cm-thick brick wall with attached gypsum board including 5-cm deep cavity filled with 4-cm-thick mineral wool. Rw (C; Ctr) values for both these walls are 56 (−2; −6) dB. The total thickness of the walls is also the same. Increasing brick thickness by 4.5 cm and decreasing cavity depth by 5 cm has resulted in the same single-numbered quantities in this case.
Measurement results and discussion for windows
The measured sound reduction index values for the two windows, double-glazed and laminated double-glazed, are compared in Figure 10. The 4-mm-thick glass is in mass-controlled region up until very high frequencies. Resonant frequency of two panels coupled by airspace, f0, is at 210 Hz for both windows and the dip is apparent. Total transmission loss of both panes is effective in sound reduction index values. Laminated double-glazed window has higher values than double-glazed window throughout the frequency range. Rw (C; Ctr) values are 33 (−1; −4) for W1 and 38 (−1; −4) for W2.
Comparison of measured windows, double-glazed W1 and laminated double-glazed W2.
Brick façades for different noise zones
According to the Directive, architectural projects of new buildings should assure LAeq room limit noise values given in Annex-VII of the Directive. 1 For planning of new residential areas, noise exposure categories of A, B, C and D should be taken into consideration. 1 In this section, traditional brick wall façades and possible retrofitting solutions of these façades are evaluated for sound insulation in different noise zones, for two rooms, with various transparency ratios.
Noise exposure categories to be taken into consideration for planning of new residential areas are A, B, C and D. In Category A (Lday < 55 dBA), noise is not a deterministic factor in planning. In Category B (Lday 55–64 dBA), noise control measures should be determined for planning decisions. In Category C (Lday 65–74 dBA), planning can only be made if there is no quieter space, and if it is in the public interest. Noise control measures should be determined in these areas. In Category D (Lday > 75 dBA), no planning decisions can be made. 1 Traffic noise levels selected for this study are LAeq 75, 70, 65, 60 and 55 dBA. Frequency spectra for these levels were determined by on-site measurements in Istanbul city and traffic noise simulations in SoundPlan software for the same site.
Annex-VII of the Directive 1 provides LAeq room limit noise values for bedrooms and living rooms. For closed window conditions, the bedroom limit noise value is 35 dBA, which is equal to balanced noise criterion (NCB) 26 curve. For closed window conditions, the living room limit noise value is 45 dBA, which is equal to NCB 37 curve. In this study, NCB curves provide frequency spectra for maximum noise in rooms.
Considering the common design strategies of Turkish housing, a bedroom and a living room are designed to take into account room dimensions and absorption. Designed bedroom has a 3 × 4 m and living room has a 5 × 6 m floor area; both rooms have 2.7 m height. Crocker 30 suggested that equivalent sound absorption area of rooms such as living rooms can be calculated by multiplying 0.8 by the floor area. This approach is used in this study. Using traffic noise, room noise limit value, façade area and equivalent sound absorption area of room, needed sound reduction index values and Rw were calculated for two rooms in five different noise zones. Frequency spectrum graphs for traffic noise levels (LAeq 75, 70, 65, 60 and 55 dBA) and room noise limit values (bedroom LAeq 35 dBA, NCB 26 and living room LAeq 45 dBA, NCB 37) are given in Figure 11. Frequency spectrum graphs for required sound reduction index values are given in Figure 12.
Traffic noise levels (LAeq 75, 70, 65, 60 and 55 dBA) and room noise limit values (bedroom LAeq 35 dBA, NCB 26 and living room LAeq 45 dBA, NCB 37).
Required sound reduction index values for bedrooms (BR) and living rooms (LR) according to various traffic (Tr) noise levels (LAeq 75, 70, 65, 60 and 55 dBA).
The façade transparency ratios determined for this study are 0% (opaque façade), 30%, 40%, 50% and 70%, and they are all used with three types of windows, W0, W1 and W2. W0 is a single-glazed window type, which is typical of the existing old housing. The sound insulation values for this window are taken from the literature. 31 W1 and W2 are the double-glazed and laminated double-glazed windows with PVC frame, which are measured in laboratory in this study.
The brick walls evaluated in this study are used for forming façades with various transparency ratios. Double-brick cavity walls 32 given in Table 5 and measured in the same research are also used in forming the retrofitted façades by adding a layer of brick and comparing these retrofitted façades to different noise zones. Rw values of façades with various brick walls and windows are calculated and compared to needed Rw values in different noise zones (Table 6).
Description and drawings for measured double masonry brick cavity walls, weighted sound reduction indices and spectrum adaptation terms, Rw (C; −).
Sound insulation check of the façades for different noise zones.
The highest needed Rw is for bedroom in the 75-dBA traffic noise zone. Opaque walls, walls with no transparency, have Rw values high enough to fulfil even the highest needed Rw value. The only exceptions are the brick walls without the second layer of plaster.
The results showed that with single-glazed windows (W0) and 70% transparency, which was common practice in the 6 million pre-1980 dwellings, it is not possible to control environmental noise in 75-, 70- and 65-dBA noise zones for bedrooms, as well as 75 dBA for living rooms. Laminated double-glazed window (W2) is almost never used, because it is extremely expensive for household use. Even using W2 in case of 75-dBA noise zone, no 70% transparent façade provides the needed sound insulation. In retrofitting old dwellings, double-glazed window (W1) is commonly used, which is insufficient in controlling 75- and 70-dBA environmental noise in bedrooms with 70% transparent façade.
In case of 50% and 40% transparent façades with W0, it is not possible to control environmental noise in 75-, 70- and 65-dBA noise zones for bedrooms and 75 dBA for living rooms. W1, commonly used for retrofitting, does not provide the necessary sound insulation in 75-dBA noise zones for bedrooms, and even in 70-dBA noise zone some of the brick walls would also need retrofitting. Even using W2, the seldom used expensive option, in 75 dBA, all the single-layer masonry brick walls would need additional layers such as exterior insulation board, attached gypsum board or another layer of brick.
In façades with W0 and 30% transparency, 75- and 70-dBA environmental noise cannot be controlled for bedrooms. W1, commonly used for retrofitting, does not provide the necessary sound insulation in 75-dB noise zones for bedrooms. Even using W2, the seldom used expensive option, in 75 dBA, some of the single-layer masonry brick walls would need additional layers.
Conclusion
In this study, sound reduction index values of exterior wall elements consisting of brick, attached gypsum board and mineral wool were measured in test rooms and discussed. The thicknesses of the vertically perforated bricks used were 14.5, 19 and 24 cm. Masonry brick walls with only exterior plaster, with both interior and exterior plaster and with exterior insulation board were measured. Cavity brick walls with a brick core and attached gypsum were measured with mineral wool placed inside the cavity. Cavity depths used were 5 and 10 cm, whereas mineral wool was not used or used with thicknesses of 4 or 8 cm. The main findings may be summarized as follows:
Comparison of sound reduction index values of walls with only one side plastered and with both sides plastered showed that wall with only one side plastered had higher values in low-frequency range, whereas it had lower values in medium- and high-frequency ranges. Having both sides plastered increased Rw value by 1 dB for 14.5- and 19-cm brick walls, but values stayed the same for 24-cm brick wall.
In the case of using exterior insulation board, sound reduction index values in low frequencies were lower than other masonry walls, but they were much higher in medium and high frequencies. Rw values increased by 4–6 dB, whereas Rw + Ctr values did not change or increased by 1 or 2 dB.
Comparing single-number quantities of building elements showed some similarities. The 14.5-cm-thick brick with one side plastered, with both sides plastered and with insulation board all had the same Rw + Ctr value. Adding the second layer of plaster to 24-cm-thick brick did not change Rw (C; Ctr) value. The 14.5- and 19-cm-thick brick walls with both sides plastered had the same Rw + Ctr value. The 14.5- and 24-cm-thick brick walls with both sides plastered had the same Rw value. The 14.5- and 19-cm-thick brick walls with exterior insulation boards had the same Rw + C value.
Using the mineral wool on the interior surface with gypsum board and 1 -m deep cavity, instead of using it on the exterior surface, increased the sound reduction index values and Rw values drastically, by up to 10 dB for single-number quantities.
Attaching gypsum board with cavity to brick wall and filling the cavity with mineral wool resulted in increases up to 26 dB for Rw, up to 25 dB for Rw + C and up to 21 dB for Rw + Ctr values. These maximum values occurred for 24-cm-thick brick wall and by adding 10-cm deep cavity with 8-cm-thick mineral wool. The minimum values occurred for 14.5-cm-thick brick wall and by adding 5-cm deep cavity with 4-cm-thick mineral wool, giving 12, 11 and 9 dB increase, respectively.
Placing mineral wool (4 or 8 cm thick) into an empty cavity (5 or 10 cm deep), for 14.5-cm-thick brick walls with attached gypsum board, increased sound reduction index values in low- and medium-frequency ranges. The increases of Rw values were 4 dB for 4-cm-thick mineral wool in 5- or 10-cm deep cavity and 5 dB increase for 8-cm-thick mineral wool in 10-cm deep cavity.
Doubling cavity depth (5–10 cm), for cavity brick walls with attached gypsum board, increased sound reduction index values especially in high-frequency range, resulting in 2 dB increase in single-number quantities for 14.5 and 19 cm and up to 9 dB for 24-cm-thick brick walls.
Doubling of mineral wool thickness from 4 to 8 cm, for cavity brick walls with attached gypsum board, resulted in higher sound reduction index values in low-frequency range, increasing Rw values by 1 dB for 14.5 and 19 cm, and by 5 dB for 24-cm-thick brick walls.
In the second part of the study, masonry and cavity brick walls were combined with windows to form façades with various transparency ratios. Sound insulation values of these façades were assessed for their efficiency against road traffic noise. In Turkey, about 6 million pre-1980 housing buildings were built with façades which have very high transparency ratios, single-glazed windows and thin brick walls. Retrofitting of these façades is essential for health and comfort. The assessment showed that windows are the deterministic factor in sound insulation efficiency of retrofitting brick façades. Pre-1980 façades with single-glazed windows and 70%, 50% or 40% transparency were insufficient in 75, 70 and 65 dBA noise zones, even if brick walls were retrofitted using added layers. Retrofitting with double-glazed windows, which is the most common practice, is not at all helpful in 75-dBA noise zones. It is also not sufficient in 70-dBA zones for 70% transparency, whereas for 50% and 40% transparency some of the brick walls would need retrofitting. Even using the expensive laminated double-glazed window, for 70% it is not possible to insulate against 75-dBA noise, and in other transparency ratios the walls need additional layers such as exterior insulation board, attached gypsum board or another layer of brick.
The results of this study helped show the sound insulation characteristics of traditional Turkish façade wall elements and discuss the factors affecting sound insulation values of masonry brick and brick cavity walls with attached gypsum board, as well as the effect they have on the whole façade.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research project was funded by Istanbul Technical University Scientific Research Projects Unit (grant no. 36647). The materials were provided and wall building elements were built by Kilsan – Kil Sanayi Ticaret A.S., Alcider – Association of Turkish Gypsum Producers, Izocam Ticaret ve Sanayi A.S. and Pimaş Plastik Insaat Malzemeleri A.S.