- Email: [email protected]
Absorption characteristics of masjid carpets
Applied Acoustics 105 (2016) 143–155
Contents lists available at ScienceDirect
Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust
Absorption characteristics of masjid carpets Ahmed Elkhateeb ⇑, Adnan Adas, Maged Attia, Yasser Balila Department of Architecture, Faculty of Environmental Design, King Abdulaziz University, P.O. Box 80210, Jeddah 21589, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 1 July 2015 Received in revised form 7 December 2015 Accepted 11 December 2015
Keywords: Absorption coefficient Carpet Polyurethane foam Polyethylene foam Reverberation Knot density
a b s t r a c t This work investigated the absorption characteristics of eight types of carpets that are especially designed and manufactured for masjids and two types of carpets pads. Measurements were carried out in the reverberation room according to ISO 354. Each type was tested three times: first when it was installed directly on the floor, second when it was installed above 5.7 mm of polyurethane foam, and last when it was installed above 10 mm of polyethylene foam. The results showed that the absorption coefficient is directly proportional to frequency and knot density. The results also demonstrated that adding pads of polyurethane or polyethylene foam increased the absorption, principally in the mid-frequency range. The effect of polyurethane foam on absorption was higher than that of polyethylene foam. Finally, the absorption coefficients of the examined carpets were found to be close to those of Muslim worshippers when they are in the position known as ‘‘standing in rows”. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction In its essential form, the masjid (mosque) is a right prismatic room with a rectangular base. In contemporary masjids, worshippers and carpets constitute the two main acoustic absorbers. Other room surfaces are typically rendered with reflective materials such as marble, paints, glasses, or ceramics. Thus, knowledge of the absorption characteristics of both carpets and worshippers is essential for the acoustic design/assessment of new or existing masjids. A carpet is a special type of textile that is used as a floor covering. Carpets are manufactured by attaching pile tufts (or face yarns) to a backing (or backing yarns) either manually (handmade) or using a machine (machine-made). Masjid carpets are machine-made woven carpets. The main reason to use this type of carpet in masjids, according to manufacturers in Egypt and Saudi Arabia, is its durability in comparison to other machine-made carpets. Thus, the following sections will address only this type of woven carpets. Face yarns can be natural materials such as wool or cotton, or they can be synthetic materials such as acrylic, olefin, nylon, polypropylene, or polyester; blends of wool and nylon can also be used [1]. Backing consists of two threads, which are usually perpendicular, called wefts (or fillings) and warps (or warp chains) [2]. Wefts run widthwise and are made of jute in different diameters. Warps run lengthwise and are made of cotton threads. Warp ⇑ Corresponding author. Tel.: +966 5 3714 3684; fax: +966 6952756. E-mail address: [email protected] (A. Elkhateeb). http://dx.doi.org/10.1016/j.apacoust.2015.12.005 0003-682X/Ó 2015 Elsevier Ltd. All rights reserved.
chains are interlaced with filling and are used to fix face yarns in place and maintain the internal structure of the backing. Additionally, stuffer threads run parallel to the warp chains but are not interlaced with fillings. The main purpose of stuffers is to maintain the dimensional stability, the structure of the backing, and the appropriate backing density required for some types of carpets [1]. In practice, woven carpets are manufactured by a highly complicated technique in which face yarns are weaved through the backing, or simply bent around weft threads, to give the carpet its well-known appearance. The process of weaving uses different methods such as Axminster, Saxony or Wilton. Thus, the structure of the backing may differ widely according to machine and carpet style, and is just as important as the pile itself. Among the different parameters that indicate the quality of a carpet are knot density (calculated as knots per square inch KPSI, knots per square meter KPSM or knots per square cm KPSC), pile height and type, weight per square meter, pitch, total shots, and backing structure. Modern acoustic design requires calculating many parameters. One of the most important acoustic parameters is the reverberation time T. To calculate T, the absorption properties of the room boundaries must be known. Many researchers have commented on the lack of data that are especially oriented to masjid acoustics [3–5]. Consequently, the current practice either ignores the acoustics of masjids or depends on approximations that may lead to inaccurate results. The main purpose of this work is to present the measurement results of the absorption characteristics of 8 types of carpets that are specifically designed and manufactured for masjids. Measurements were carried out in the reverberation
144
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
chamber of the Acoustic Research and Tests Unit (ARTU) in the Faculty of Environmental Design, King Abdulaziz University, Jeddah, Saudi Arabia. It is also of interest to relate their absorption characteristics to their technical specifications such as knot density, pile material, pile height, weight per square meter, and backing structure. Finally, the absorption characteristics of masjid carpets will be compared to the absorption characteristics of Muslim worshippers as previously measured [6]. Because carpets are widely applied as floor coverings in different rooms and have a noticeable acoustic effect, their absorption characteristics have been investigated in many prior research studies. Harris [7] measured the normal absorption coefficient and flow resistance for several hundred carpet samples in order to investigate which variables in carpet construction have a direct effect on the acoustic absorption. Based on the data obtained, certain samples were examined in the reverberation chamber at the National Bureau of Standards for comparison with the tube measurement. Results showed that pile density is one of the fundamental variables that can affect carpet absorption. Shoshani and Rosenhouse [8], investigated the relationship between the absorption coefficients of a cover made of woven fabric and its main parameters, including fibre content, yarn count, cover factor, the air gap behind the fabric, the frequency of the impinging sound wave, and the influence of washing on sound absorption. Results showed that Noise Reduction Coefficients (NRCs, the arithmetic average, rounded to the nearest multiple of 0.05, of the absorption coefficients at 250, 500, 1000 and 2000 Hz [9]) for the examined samples ranged between 0.09 and 0.22; the absorption coefficient was much higher at higher frequencies. Shoshani [10] measured absorption coefficients of tufted carpets backed by several layers of stitch-bonded nonwovens using an impedance tube. Results
(a) Acrylic fibres
indicated that the absorption coefficients of backed carpets were significantly higher than those of un-backed carpets between 250 and 1000 Hz. The increase in the absorption of backed carpets depends primarily on the thickness of the backing, whereas the effect of fibre content is only marginal. Finally, Shoshani and Wilding [11] examined the effect of pile characteristics on the absorption of tufted carpets in the frequency range 125–4000 Hz using an impedance tube. The parameters considered were fibre content and denier, pile density, and average pile height. During measurements, only one pile parameter in the examined samples was varied. One of the important results was the significant increase in the absorption coefficient in the low and medium frequency range, 250–1000 Hz, when there was an air gap behind tufted carpet used as a wall covering. 1.1. Description of the selected samples After a comprehensive survey, masjid carpet manufacturers in the west of Saudi Arabia were identified. Visits and meetings with those manufacturers revealed that there are 8 types of masjid carpets in production. These carpets can be classified into 3 categories on the basis of their face yarn materials: 3 are made of acrylic, 4 are made of heat-stabilised polypropylene (PPHS) and one is made of an 80/20 blend of wool and acrylic. However, the first two are the most common because of their low prices. Masjid carpets have a standard width of about 3.99 m and a length of 25 m. This width fits 3 rows of worshippers, which is the customary grouping during prayer according to the Islamic faith [12,13]. In Saudi Arabia, it is a common practice to install masjid carpets above paddings of either polyurethane foam (PUF) or polyethylene foam (PEF). The former is an open-cell foam (synthetic rubber) adhered to thin solid backing,
(b) Polypropylene fibres
(c) Wool fibres
Fig. 1. Tuft fibres as seen by the electron microscope.
Table 1 Technical specifications of the tested samples. #
a
c d
Pile material
Total shots
Weight
Thickness
Per m2
Per pile
Pile height
kg
mg
mm
Backing
Overall
# of fibres per tuft
Backing structure
Knot density KPSM
KPSC
KPSI
Pitch
Row
Filling
1 2 3
PAZ/1 DIA/7 KAN/2
Acrylic
Single Single Single
4.275 2.60 3.27
5.5 7.0 6.3
12 10.5 11
2.17 2.65 2.68
14.17 13.15 13.68
200–210 200–210 165–180
Wilton TRa STAGb
235,800 250,000 275,100
24d 25 28
152 161 177
11 13 11
14 12 16
Jute Jute Jute
4 5 6 7
PRI/6 LOU/3 CRY/5 MAD/4
Polypropylene HS
Double Single Double Double
2.19 3.84 2.33 2.75
6.8 7.2 5.6 6.4
10.5 10.5 9.5 10.5
2.92 2.45 2.00 2.07
13.42 12.95 11.50 12.57
180–185 105 175–180 170–175
STAGb STAGb STAGb Wilton
137,550 176,850 189,000 235,800
14 17d 19 24
89 114 122 152
11 11 9 11
8 10 14 14
Jute Jute Jute Jute
8
XMIN7/9
Wool
Triple
2.845
17.3
8.5
2.64
11.14
Uncountable
AXMIc
97,520
10
63
7
9
Jute
PUF PEF
Polyurethane foam Polyethylene foam
1.50 0.28
– –
– –
– –
5.70 10.00
– –
– –
– –
– –
– –
– –
– –
– –
9 10 b
Sample code
Triple row. Double row staggered. Axminster. Each knot contains double pile.
145
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Wes
Face Yarns
Wes
Face Yarns
Stuffer Warp Chains
Warp Chains
(a) Double row, in-line structure (Wilton)
Wes
Face Yarns
(b) Double row, staggered structure
Wes
Face Yarns
Stuffer Warp Chains
Warp Chains
(d) Axminster structure
(c) Triple row structure
Fig. 2. Internal structure of the backing of the tested samples, 3d and section perpendicular to wefts.
13.00 12.00
AMPA A1
Room Absorpon m2
11.00 10.00 9.00 8.00 7.00 6.00 5.00 4.00 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000
Fr. (Hz) Fig. 3. The maximum permissible sound absorption AMPA in the empty room vs. the measured equivalent sound absorption of the empty room A1.
and has an overall thickness of 5.7 mm. The latter is a closed-cell (crosslink) foam covered on one face with aluminium foil; the overall thickness is 10 mm.
The basic unit in the structure of carpet piles is the fibre. For the samples that were studied, in the case of acrylic piles, the fibres have a diameter of about 46 lm. Each acrylic pile consists of 3 tufts
146
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
(a) Sample PAZ/1
(b) Sample KAN/2
(d) Sample LOU/3
(e) Sample MAD/4
(c) Sample PRI/6
(f) Polyurethane foam PUF
Fig. 4. Samples during measurements.
Table 2 The standard error d in (s). Setup Sample
Empty room (ER) Frequency range
Directly on floor Frequency range
On PUF Frequency range
On PEF Frequency range
Low
Mid
High
Low
Mid
High
Low
Mid
High
Low
Mid
High
ER
±0.38
±0.14
±0.03
–
–
–
–
–
–
–
–
–
DIA/7 KAN/2 PAZ/1
– – –
– – –
– – –
±0.45 ±0.51 ±0.54
±0.15 ±0.17 ±0.16
0.03 0.04 0.03
±0.42 ±0.37 ±0.45
±0.10 ±0.12 ±0.12
±0.03 ±0.04 ±0.03
±0.50 ±0.52 ±0.47
±0.17 ±0.16 ±0.13
±0.03 ±0.05 ±0.03
PRI/6 CRY/5 MAD/4 LOU/3 XMIN-7/9
– – – – –
– – – – –
– – – – –
±0.53 ±0.51 ±0.47 ±0.52 ±0.49
±0.18 ±0.17 ±0.17 ±0.15 ±0.15
0.04 0.03 0.04 0.03 0.04
±0.48 ±0.48 ±0.51 ±0.43 ±0.42
±0.12 ±0.13 ±0.14 ±0.15 ±0.15
±0.03 ±0.03 ±0.03 ±0.03 ±0.03
±0.46 ±0.45 ±0.53 ±0.46 ±0.54
±0.15 ±0.16 ±0.15 ±0.13 ±0.19
±0.04 ±0.04 ±0.03 ±0.04 ±0.08
PUF PEF
– –
– –
– –
±0.66 ±0.58
±0.18 ±0.21
0.04 0.08
– –
– –
– –
– –
– –
– –
plaited (twisted) together (Fig. 1a), each of which contains about 55–70 acrylic fibres according to pile weight. Each pile is raised up by bending the tufts around weft threads, thus the total number of fibres per pile ranges from 330 to 420. Pile height in acrylic carpet ranges from 10.5 to 12 mm according to carpet type and design. In contrast to the acrylic case, the tufts in polypropylene piles are not plaited; instead, pile fibres are loose (Fig. 1b); each pile contains about 170–185 fibres, each of which has a diameter of about 51 lm. Polypropylene piles are also raised up by bending the piles around weft threads, so the total number of fibres per pile ranges from 340 to 370. Pile height in polypropylene carpet ranges from 9.5 to 10.5 mm according to carpet type. Under the electron microscope, wool fibres appear dishevelled and very loose (Fig. 1c). Wool fibres have a diameter of about 45 lm whereas acrylic fibres have a range of diameters from 21 to 37 lm. Pile height in the tested wool carpet is 8.5 mm. Table 1 lists the codes of the selected samples and their main technical specifications. Knot density in the selected samples has a wide range from about 100,000 to above a quarter million KPSM (10–28 KPSC). Pile height varies from 9.5 to 12 mm, and weight from about 2 kg/m2 to more than 4 kg/m2. In these samples, backing thickness ranges from 2 to 3 mm and the net weight of the backing (without face yarns) ranges from 30% to 65% of the total weight per square meter.
Filling diameters range from 0.60 to 1.30 mm according to carpet style. Backing structure depends on the way wefts are arranged and warp chains are interlaced with the wefts. For the samples under consideration, 2 structures have been recognised and named according to their forms (Fig. 2): Double row structure: Wefts are arranged in two rows either inline (Wilton [2], Fig. 2a) or staggered (Fig. 2b). The former arrangement was used in samples PAZ/1 and MAD/4; the latter, in samples KAN/2, LOU/3, CRY/5, and PRI/6. Triple row structure: Wefts are arranged in three rows. This has been used only in sample DIA/7 (Fig. 2c) and sample XMIN-7/9 (Axminster [2], Fig. 2d). 2. Materials and methodology For the purpose of this work, samples of each of the 8 mentioned types along with their main technical specifications were collected, categorised, coded, and shipped to the ARTU; each sample was 10 m2. Moreover, samples of both paddings were obtained. Each carpet sample was tested 3 times according to ISO 354 and ISO 9613-1 standards [14,15]. For the first test, the sample was installed directly on the Lab floor (DOF); this will be called setup
Table 3
a values, all samples DOF. Sample
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.09 0.09 0.04 0.06 0.03 0.05 0.02 0.06
0.03 0.02 0.04 0.02 0.01 0.02 0.03 0.00
0.03 0.01 0.05 0.02 0.02 0.02 0.05 0.04
0.05 0.03 0.09 0.04 0.05 0.05 0.05 0.04
0.04 0.06 0.13 0.03 0.05 0.06 0.07 0.06
0.10 0.04 0.16 0.09 0.13 0.08 0.09 0.07
0.10 0.11 0.24 0.12 0.12 0.11 0.13 0.10
0.17 0.20 0.34 0.23 0.24 0.21 0.21 0.18
0.28 0.27 0.47 0.34 0.32 0.28 0.31 0.27
0.31 0.32 0.51 0.36 0.35 0.30 0.30 0.28
0.46 0.49 0.59 0.49 0.48 0.44 0.43 0.39
0.57 0.58 0.70 0.57 0.55 0.50 0.49 0.49
0.70 0.69 0.72 0.69 0.66 0.63 0.63 0.58
0.76 0.78 0.82 0.68 0.64 0.69 0.66 0.62
0.78 0.78 0.81 0.73 0.68 0.71 0.74 0.64
0.83 0.83 0.80 0.79 0.69 0.72 0.76 0.70
0.85 0.85 0.84 0.80 0.78 0.79 0.84 0.75
0.95 0.93 0.94 0.86 0.87 0.86 0.88 0.87
0.95 0.96 0.89 0.94 0.86 0.86 0.94 0.89
0.96 0.98 0.92 0.92 0.96 0.86 0.87 0.91
1.00 1.06 1.01 1.02 1.02 1.06 0.95 0.98
Table 4
a values, all samples on PUF padding. Sample
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
One-third octave centre frequency (Hz) 100
One-third octave centre frequency (Hz) 100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.03 0.02 0.04 0.00 0.02 0.00 0.01 0.01
0.04 0.03 0.02 0.06 0.02 0.02 0.00 0.02
0.02 0.01 0.04 0.04 0.06 0.02 0.06 0.00
0.10 0.10 0.13 0.09 0.11 0.07 0.11 0.06
0.16 0.18 0.21 0.15 0.16 0.13 0.15 0.10
0.19 0.31 0.31 0.26 0.30 0.18 0.21 0.14
0.32 0.48 0.38 0.30 0.44 0.28 0.29 0.17
0.56 0.62 0.63 0.59 0.60 0.58 0.51 0.35
0.65 0.59 0.72 0.70 0.63 0.70 0.60 0.45
0.59 0.56 0.74 0.62 0.63 0.65 0.70 0.46
0.72 0.65 0.83 0.76 0.66 0.77 0.82 0.66
0.79 0.67 0.82 0.80 0.71 0.79 0.88 0.72
0.81 0.74 0.88 0.78 0.73 0.83 0.92 0.85
0.84 0.79 0.85 0.82 0.73 0.83 0.90 0.90
0.82 0.80 0.81 0.78 0.73 0.77 0.93 0.91
0.83 0.83 0.83 0.79 0.73 0.82 0.85 0.94
0.90 0.92 0.86 0.86 0.78 0.84 0.89 0.94
0.90 0.92 0.91 0.84 0.80 0.79 0.82 0.89
0.92 0.97 0.84 0.79 0.81 0.80 0.80 0.91
0.91 0.94 0.86 0.87 0.88 0.83 0.78 0.85
0.86 0.96 0.90 0.87 0.87 0.80 0.93 0.87
147
148
Sample
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
One-third octave centre frequency (Hz) 100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.01 0.00 0.03 0.02 0.05 0.02 0.00 0.02
0.03 0.01 0.04 0.02 0.01 0.00 0.00 0.03
0.07 0.09 0.15 0.07 0.03 0.07 0.09 0.08
0.07 0.06 0.07 0.06 0.07 0.05 0.07 0.05
0.10 0.13 0.18 0.11 0.13 0.14 0.13 0.11
0.18 0.19 0.25 0.15 0.12 0.15 0.13 0.12
0.19 0.29 0.29 0.18 0.23 0.19 0.20 0.14
0.36 0.45 0.50 0.43 0.42 0.33 0.35 0.30
0.45 0.52 0.54 0.52 0.53 0.48 0.43 0.33
0.51 0.51 0.58 0.53 0.50 0.55 0.47 0.36
0.68 0.60 0.75 0.64 0.62 0.66 0.64 0.50
0.71 0.69 0.78 0.73 0.69 0.74 0.73 0.61
0.81 0.79 0.81 0.75 0.71 0.82 0.84 0.79
0.85 0.77 0.88 0.80 0.69 0.80 0.86 0.88
0.79 0.78 0.81 0.80 0.74 0.76 0.89 0.92
0.81 0.86 0.83 0.78 0.66 0.80 0.82 0.96
0.93 0.88 0.83 0.88 0.77 0.74 0.87 0.95
0.91 0.96 0.86 0.81 0.83 0.80 0.83 0.94
0.88 0.99 0.87 0.87 0.82 0.83 0.91 1.02
0.86 0.99 0.80 0.86 0.80 0.87 0.87 1.05
0.97 1.03 0.87 0.89 0.85 0.84 0.88 1.21
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Table 5
a values, all samples on PEF padding.
149
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Table 6 NRC values, all samples and setups in ascending order according to their knot density. #
Code
Pile type
Knot density
Weight (kg/m2)
Backing str.
8 4 5 6 7 1 2 3
XMIN-7/9 PRI/6 LOU/3 CRY/5 MAD/4 PAZ/1 DIA/7 KAN/2
WO PPHS PPHS PPHS PPHS ACR ACR ACR
10 14 17 19 24 24 25 28
2.535 2.19 3.84 2.33 2.75 4.275 2.6 3.27
AXMIN STAG STAG STAG Wilton Wilton TR STAG
Thickness (mm) Filling
Pile
Overall
2.64 2.92 2.45 2.00 2.07 2.17 2.65 2.68
8.5 10.5 10.5 9.5 10.5 12 10.5 11
11.14 13.42 12.95 11.5 12.57 14.17 13.15 13.68
1. For the second test, it was installed above PUF padding; this will be called setup 2. For the last test, it was installed above PEF padding; this will be called setup 3. In addition, the absorption characteristics of the pads by themselves were also examined. The reverberation room in the ARTU is a rectangular room that has a net floor area SF of 33.74 m2 and net volume V of 163.84 m3. The length of the longest straight line Imax in the room (in this case, the longest diagonal of the room) is 10.22 m, which (according to room volume) satisfies condition (1) of the standards [14]. Since V differs from the 200 m3 stated in the standards, the maximum permissible sound absorption of the empty room, AMPA, were obtained by multiplying the values given in Table 1 of the standards by the value (V/200)2/3 [14]. Then, the calculated AMPA was compared to the measured equivalent sound absorption area of the empty room A1. Fig. 3 presents A1 versus AMPA according to ISO 354 [14], as can be seen, A1 satisfies the standards. In practice, the empty room absorption A1 and the absorption of the room with the tested specimen A2 were obtained by measuring the empty room reverberation time T1 and the reverberation of the room with the tested specimen T2, as shown in Fig. 4. Thus, the absorption of the tested specimen AT was calculated as the difference between A2 and A1 according to Equation (8) of the ISO standards [14]. Measurements of reverberation time were carried out using the Interrupted noise method. The instrumentation was based on products by Brüel & Kjær (B&K); these include inter alia Dirac software ver. 6.0, sound level meter type 2250 as a microphone, Omnisound source type 4292-L and power amplifier type 2734. During measurements, a maximum length sequence (MLS) signal of length 10.9 s was used as a stimulus, and the signal was pre-averaged 5 times for each position. For each specimen measurements, 12 decay curves were collected, 4 positions for the microphone and 3 for the sound source. Positions of both microphone and sound source conformed to the specifications [14]. During measurements, ambient atmospheric temperature TA, relative humidity ht and ambient atmospheric pressure Pa were monitored and recorded to calculate both c (the propagation speed of sound in air, m/s) and 4mV (Air absorption, where m is the sound attenuation coefficient in m1). Values of m were calculated at onethird octave intervals based on apt (pure-tone sound attenuation coefficient in decibels per meter) where apt was calculated by applying equations (3, 4, 5, B.1, B.2 and B.3) of the standard ISO 9613-1 [15]. Finally, the absorption coefficient a of the tested specimen was calculated using the formula [14]:
a¼
AT S
ð1Þ
where S is the area of the tested specimen in m2. Table 2 lists the standard error d of the 95% confidence level that was calculated for each group of the measured decay curves according to the formula:
Pitch
Row
# of fibres/tuft
7 11 11 9 11 11 13 11
9 8 10 14 14 14 12 16
Uncountable 180–185 105 ⁄ 4 175–180 170–175 200–210 200–210 165–180
STDEV d ¼ tð0:025;n1Þ pffiffiffi n
NRC (DOF)
(On PUF)
(On PEF)
0.31 0.34 0.36 0.35 0.36 0.38 0.35 0.47
0.50 0.59 0.57 0.58 0.58 0.56 0.54 0.63
0.45 0.50 0.50 0.48 0.49 0.49 0.47 0.58
ð2Þ
where t: t-distribution curve. n: Total number of decay curves in each measurement. STDEV: Standard deviation of the measured T for each specimen. The standard error d for the empty room was found to be ±0.38 s in the low frequency bands, ±0.14 s in the mid frequency bands and ±0.03 s in the high frequency bands. The mean of d in the room with tested specimens had almost the same values with only slight differences. In the low frequency bands, it was ±0.50 s for setup 1, ±0.45 s for setup 2 and ±0.49 s for setup 3; in the mid frequency bands, ±0.16 s for setup 1, ±0.13 s for setup 2 and ±0.15 s for setup 3; and in the high frequency bands, ±0.03 s for setup 1, ±0.04 s for setup 2 and ±0.03 s for setup 3. The relatively high d in the low frequency bands may be a consequence of the small room volume, while the very small d in the mid and high frequency bands provides a high level of confidence in, and reliability of, the measurements in those bands. 3. Results Tables 3–7 list the absorption coefficient a for the examined carpets and pads calculated at one-third octave intervals. It also shows the Noise Reduction Coefficient NRC for the examined samples. For ease of comparison, values of a for each type of face yarn (acrylic and polypropylene) and for each setup (1, DOF), (2, on PUF), and (3, on PEF) were averaged. As can be concluded from the results, a is an increasing function of frequency and knot density, while pile materials and backing structure have almost no effect. Carpets have a limited effect as absorbers in the low frequency range, but the effect increases in the mid frequency range and becomes very significant in the high frequency range. As a general rule, the higher the knot density, the higher the absorption coefficient a, but some exceptions appear at high frequencies. This conclusion can be seen clearly in Fig. 5a where the averaged values of a are compared for the samples in setup 1. It is clear that the average values of a in the case of acrylic carpet (knot density average 26 KPSC and pile height average 11.17 mm) are higher than those of polypropylene carpet (knot density average 18 KPSC, pile height average 10.25 mm) and wool carpet (knot density 10 KPSC, pile height 8.5 mm). The change in a between different carpet types is small and sometimes can be neglected when their knot densities are close. For example, a values are close in PAZ/1 and DIA/7 (knot density 24 & 25 KPSC respectively, Fig. 5b). The same conclusion applies to LOU/3 and CRY/5 (knot density 17 & 19 KPSC respectively, Fig. 5c). However, an obvious change appears when there is a clear difference in knot density; for example, an obvious change can be observed between PAZ/1 and KAN/2 (knot density
0.88 0.16 0.85 0.13 0.83 0.13 0.72 0.28 0.60 0.39 0.53 0.44 0.46 0.42 0.37 0.29 0.25 0.14 0.19 0.09 0.13 0.07 0.10 0.09 0.12 0.09 0.03 0.02 0.02 0.01 0.07 0.04 0.01 0.03 0.01 0.03 PUF PEF 1 2
0.00 0.05
0.02 0.02
8000 6300 5000 4000 3150 2500 2000 1600 1250 1000 800 630 500 400 315 250 200 160 125
One-third octave centre frequency (Hz)
100 Code #
Sample
Table 7
a values for pads.
0.99 0.26
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
10,000
150
24 & 28 KPSC respectively) in Fig. 5b. Fig. 6 presents a graphical comparison between the a values of the two samples KAN/2 (knot density 28 KPSC, pile height 11 mm) and XMIN-7/9 (knot density 10 KPSC, pile height 8.5 mm); the differences are quite obvious. There is some evidence that pile height and weight per square meter of the carpet improve the absorption. This may clarify why the NRC of LOU/3 (0.36 NRC, 17 KPSC and 10.5 mm pile height) is higher than that of CRY/5 (0.35 NRC, 19 KPSC and 9.5 mm pile height). Moreover, the NRC of PAZ/1 (0.38 NRC, 24 KPSC and 12 mm pile height) is higher than that of DIA/7 (0.35 NRC, 25 KPSC and 10.5 mm pile height), as shown in Table 6. The weight per square meter of the carpet has a similar effect, for example PAZ/1 (0.38 NRC, 24 KPSC and 4.275 kg/m2) is higher than that of DIA/7 (0.35 NRC, 25 KPSC and 2.6 kg/m2). Nevertheless, some results do not match well with this assumption; for example, the NRC of PAZ/1 (0.38 NRC, 24 KPSC and 4.275 kg/m2) is obviously less than that of KAN/2 (0.47 NRC, 28 KPSC and 3.27 kg/ m2) because the latter sample has a higher knot density. We could not independently examine the effect of these two parameters, pile height and weight per square meter, on a as this would require keeping the other carpet variables (such as knot density) constant, which is not currently possible. Pads improve the absorption of carpets in the mid frequency range. Some effects appear also in the low and high frequency ranges, as shown in Fig. 7a and b. The effect of PUF on a in the mid-frequency range is higher than that of PEF; this can be seen clearly in values of NRC for the three examined setups, as shown in Fig. 8 and Table 6. The effect of knot density and pile height on a is still similar to the case of setup 1 although some fluctuations or dips may occur at certain frequencies depend on padding and carpet types; see Fig. 9a and b. The absorption characteristics of the padding itself, both PUF and PEF are shown in Fig. 10. It can be seen that, for PUF, a increases monotonically in all frequency bands although small dips appear at 200, 315, 400, and 630 Hz. Above 630 Hz, the values of a increase rapidly and become comparable with the absorption of some types of carpets, see Table 7. Values of a exceed 0.5 for frequencies above 2500 Hz and reach almost 0.9 at 8 kHz. Moreover, the a of PUF is higher than that of some of the carpets. Conversely, the absorption characteristics of PEF are typical of a resonant absorber with a peak at 2500 Hz (a = 0.44), after which it decreases rapidly, becomes almost constant between 5 and 8 kHz (0.13), and then increases again slightly to reach 0.26 at 10 kHz. To check the applicability of the results of this work and to see if the results can be used to predict the absorption of other woven carpets based on their knot density, a blind sample was collected arbitrarily from a masjid. The knot density was counted (20 KPSC) and the total height was measured (12.5 mm). This blind sample was tested systematically applying the methodology described in Section 2. Later, the measured values of a were compared with the most similar sample (CRY/5, knot density 19 KPSC and total height 11.5 mm), as shown in Fig. 11. Although some differences appear because of the small differences in knot density and total height, the values are still comparable and can be used to obtain results of acceptable accuracy.
4. Discussion One of the important applications of the absorptive materials is to control the reverberation in an enclosure by decreasing the amplitude of the impinging sound waves [16]. Porous materials among the most common absorptive materials; these include cellular and fibrous materials, both of which are presented in this
151
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Average (ACR - DOF)
Average (PPHS - DOF)
XMIN-7/9 DOF
1.20
1.00
0.80
α
0.60
0.40
0.20
0.00
-0.20
Fr. (Hz)
(a) Setup 1, directly on floor PAZ/1 DIA/7 KAN/2
1
1
0.6
α
α
0.6
PRI/6 LOU/3 CRY/5 MAD/4
0.2
0.2
-0.2
-0.2
Fr. (Hz)
Fr. (Hz)
(c) Polypropylene carpets
(b) Acrylic carpets
Fig. 5. Values of a, setup 1, directly on floor.
KAN/2
XMIN-7/9
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz) Fig. 6. Effect of knot density on a, comparison between KAN/2 and XMIN-7/9.
work. In general, a porous material is a solid substance that contains cavities, channels or holes within its body [17]. These cavities allow the acoustic energy that is transmitted through air to penetrate and interact with the inner solid structure of the material [18]. Fibrous materials, such as carpets, consist of a series of tunnel-like pores that are formed because of the nature of fibrous materials [17]. In such materials, the absorption is generally due to the viscosity of air that affects the particle vibration in the confined space of the pore. Thus, it is found that, for a simple fibrous absorber, the absorption characteristics are affected by fibre radii and
the porosity within a solid material of fibres that effectively provides a matrix. For this reason, the absorption characteristics of fibrous samples of different materials will be the same if they have the same fibre radii and bulk density. Close-up views under the electron microscope, as shown in Fig. 12a–c, indicate how the pores confined between the fibres of carpets ensure an acceptable range of porosity so that sound waves can penetrate the carpet to allow multiple interactions with its structure. On the other hand, cellular materials contain pores, or simply cells, that can be either open or closed. An open-cell material, such
152
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Average (ACR + DOF)
Average (ACR + PUF)
Average (ACR + PEF)
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz)
(a) Acrylic Average (PPHS + DOF)
Average (PPHS + PUF)
Average (PPHS + PEF)
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz)
(b) Polypropylene Fig. 7. Comparison between a averages in the three setups used in this study.
NRC DOF
NRC PUF
NRC PEF
4
5
0.70 0.60
NRC
0.50 0.40 0.30 0.20 0.10 0.00 1
2
3
6
7
8
Sample # Fig. 8. NRC for the examined carpets in the three setups (see Table 1 for the names of samples).
as polyurethane foam, has a continuous channels of communication with the external surface of the material; such pores effectively increase the absorption of sound [17]. In Fig. 12d and e, close top and cross-sectional views show the variable open-cell cavities of PUF; such pores allow absorption over a wide frequency range despite the small thickness and very light weight of this material.
In contrast, pores in closed cell materials, such as polyethylene foam, are totally isolated from their neighbours; thus their efficiency in absorbing sound waves is considerably less than that of open pores [17]. The closed-cell structure of polyethylene foam (Fig. 12f) prevents this essential penetration, thereby limiting the ability of PEF to act as an absorber despite the fact that it contains more pores than PUF, as can be concluded by comparing the weight per square meter of the two materials (1.5 kg/m2 for PUF and 0.28 kg/m2 for PEF). Data about the absorption coefficient of carpets with either tufted or woven design can be found in many resources such as websites and acoustic handbooks. However, most of these data fail to mention the technical specifications of the carpets. Consequently, using these data for other carpet types could result in an inaccurate estimation. The results of this work establish a clear scientific basis on which a good estimation for the absorption coefficient of other woven carpets can be performed to an acceptable degree of accuracy if their technical specifications are known. Fig. 13 shows the average absorption curve of acrylic carpets in the three examined setups (DOF, on top of PUF and on top of PEF padding) along with the absorption curves of worshippers in each of the two positions: standing in well-defined rows or sitting down at random locations [6]. In the first position, the worshippers stand in orderly rows (usually on carpeted floor) and listen to the imam (leader in group prayer) reciting certain verses of the holy Quran. In the second position (sitting down at random locations on a carpeted floor), the worshippers listen to the imam who delivers
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 KAN/2 + PUF
LOU/3 + PUF
153
XMIN-7/9 + PUF
1
0.8
α
0.6
0.4
0.2
0
-0.2
(a) Setup 2, carpets on PUF padding KAN/2 + PEF
Fr. (Hz)
LOU/3 + PEF
XMIN-7/9 + PEF
1.4 1.2 1
α
0.8 0.6 0.4 0.2 0 -0.2
(b) Setup 3, carpets on PEF padding
Fr. (Hz)
Fig. 9. Effect of padding on a, comparison between 3 types of carpets examined in setups 2 and 3.
PEF
PUF
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz) Fig. 10. Absorption coefficient a of the padding, PUF and PEF.
his Friday speech. These two positions are the most important from the acoustic point of view. As shown in Fig. 13, the absorption caused by worshippers in the standing position was very close to
the absorption of the carpet at frequencies up to 2500 Hz, and thereafter it became slightly higher. Thus, the acoustic environment does not differ significantly between empty and fully occu-
154
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
α (Blind Sample DOF)
α (Esmated, DOF)
1.20
1.00
α
0.80
0.60
0.40
0.20
0.00
-0.20
(a) Setup 1, directly on floor
Fr. (Hz)
1.00
1.20
0.80
1.00 0.80
0.60
0.60 0.40 α (Blind Sample + PUF)
0.20
α (Esmated, on PUF)
0.40
0.00
0.00
-0.20
-0.20
(b) Setup 2, on PUF
α (Blind Sample + PEF)
0.20
α (Esmated, on PEF)
(c) Setup 3, on PEF
Fig. 11. Measured values of a for the blind sample vs. the expected values.
(a) Acrylic fibres, sample LOU/3
(d) PUF, top view
(b) Polypropylene fibres, sample MAD/4
(e) PUF, cross-seconal view
(c) Wool fibres, sample XMIN7/9
(f) PEF, cross-seconal view
Fig. 12. Voids inside the tested samples as seen by the electron microscope.
pied masjids in the case of standing worshippers. In case of worshippers sitting down at random locations, the absorption of worshippers and carpets tracked each other closely up to 1250 Hz, and then the absorption caused by the worshippers became almost constant at about 0.60 whereas the absorption attributable to the carpet increased gradually.
It is important to mention here that the optimum reverberation time of masjids – and consequently the absorption required – is still under discussion and has not yet been established. A recent study [13] concluded that masjids must be considered a special type of speech rooms that need a relatively long reverberation time. Accordingly, the carpets must be carefully chosen if we want to increase the existing reverbera-
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Average (ACR + DOF)
Average (ACR + PUF)
Average (ACR + PEF)
α Standing on Carpet
155
1.2
α Random Sing Down on Carpet 1 0.8
α
0.6 0.4 0.2 0 -0.2
Fr. (Hz) Fig. 13. Average absorption coefficient a of acrylic carpets vs. absorption of Muslim worshippers (standing in well-defined rows and sitting down at random locations).
tion. One of the suggested solutions is to use carpets that have a lower knot density and are installed directly on the floor (without pads).
(12-BUI12759-03). The authors gratefully acknowledge the Science and Technology Unit, King Abdulaziz University for technical support. Thanks also to Dr. Hend Ahmed, Dr. Salem Alhamedee and Arch Zinub Najeeb for their help and continuous support.
5. Conclusions References This work has presented the results of measurement of the absorption coefficient of 8 types of carpets that were especially designed for masjids; in addition, 2 types of paddings, polyurethane foam and polyethylene foam, were characterised. Each carpet sample was tested according to ISO 354 and ISO 9613-1 standards in 3 setups: the first when the sample was installed directly on the Lab floor DOF, the second when it was installed above PUF padding, and the last when it was installed above PEF padding. The absorption of the paddings themselves was also examined. Measurements were conducted in the reverberation chamber of the ARTU in the Faculty of Environmental Design, King Abdulaziz University. Results showed that carpets are good absorbers in the mid- and high-frequency ranges. The absorption coefficient a of a carpet is directly proportional to frequency and to knot density, but the pile materials and backing structure have almost no effect. There is some evidence that the pile height and the weight per m2 of a carpet improve its absorption. Pads improve the absorption of carpets in the mid frequency range, and some effects also appear in the low and high frequency ranges. The effect of PUF on a in the mid-frequency range is higher than that of PEF. The results of this work establish a clear scientific basis on which a good estimation for the absorption coefficient of other woven carpets can be performed to an acceptable degree of accuracy, given their technical specifications. Acknowledgments This work was supported by the NSTIP strategic technologies program in the Kingdom of Saudi Arabia, Project Number
[1] Amirkhan E, Groseclose A. A comprehensive guide to oriental and specialty rug cleaning. 1st ed. Dallas: Oriental and Rug Cleaning Company Inc.; 2006. [2] Complete textile glossary; 2001.. [accessed 10 May 2015]. [3] Karabiber Z. Acoustical problems in mosques. In: ASA/EAA/DAGA meeting, Berlin, Germany, 15–19 March, 1999. [4] Fausti P, Pompoli R, Prodi N. Comparing the acoustics of mosques and Byzantine churches. In: CIPA 19th international symposium, Antalya, Turkey, 30 September–3 October, 2003. [5] Karabiber Z, Erdogan S. Comparison of the acoustical properties of an ancient and a recent mosque. In: Forum Acusticum, Sevilla, Spain, September 2002. [6] Oldham D, Elkhateeb A. The absorption characteristics of Muslim worshippers. Build Acoust 2008;15(4):335–48. [7] Harris C. Acoustical properties of carpet. J Acoust Soc Am 1955;27(6):1077–82. [8] Shoshani Y, Rosenhouse G. Noise absorption by woven fabrics. Appl Acoust 1990;30(4):321–33. [9] Egan MD. Architectural acoustics. J. Ross Publishing; 2007. p. 50. [10] Shoshani YZ. Effect of nonwoven backings on the noise absorption capacity of tufted carpets. Text Res J 1990;60(8):452–6. [11] Shoshani YZ, Wilding MA. Effect of pile parameters on the noise absorption capacity of tufted carpets. Text Res J 1991;61(12):736–42. [12] Elkhateeb A, Refat M. Sounds from the past: the acoustics of Sultan Hassan Mosque and Madrasa. Build Acoust 2007;14(2):109–32. [13] Elkhateeb A, Adas A, Attia M, Balila Y. The acoustics of masjids, why they differ from the classical speech rooms. In: The 22th international congress on sound and vibration ICSV22, Florence, Italy, 12–16 July, 2015. [14] British Standard Institute. ISO 354:2003(05-15), Acoustics – measurement of sound absorption in a reverberation room; 2003. [15] British Standard Institute. ISO, 9613-l, Acoustics – Attenuation of sound during propagation outdoors – Part 1: Calculation of the absorption of sound by the atmosphere, 1st ed., 1993-06-01. [16] Seddeq H. Factors influencing acoustic performance of sound absorptive materials. Aust J Basic Appl Sci 2009;3(4):4610–7. [17] Arenas J, Crocker M. Recent trends in porous sound absorbing materials. Sound Vib 2010;44(7):12–7. [18] Kuczmarski M, Johnston J. Acoustic absorption in porous materials, Cleveland, Ohio, March 2011. .
Contents lists available at ScienceDirect
Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust
Absorption characteristics of masjid carpets Ahmed Elkhateeb ⇑, Adnan Adas, Maged Attia, Yasser Balila Department of Architecture, Faculty of Environmental Design, King Abdulaziz University, P.O. Box 80210, Jeddah 21589, Saudi Arabia
a r t i c l e
i n f o
Article history: Received 1 July 2015 Received in revised form 7 December 2015 Accepted 11 December 2015
Keywords: Absorption coefficient Carpet Polyurethane foam Polyethylene foam Reverberation Knot density
a b s t r a c t This work investigated the absorption characteristics of eight types of carpets that are especially designed and manufactured for masjids and two types of carpets pads. Measurements were carried out in the reverberation room according to ISO 354. Each type was tested three times: first when it was installed directly on the floor, second when it was installed above 5.7 mm of polyurethane foam, and last when it was installed above 10 mm of polyethylene foam. The results showed that the absorption coefficient is directly proportional to frequency and knot density. The results also demonstrated that adding pads of polyurethane or polyethylene foam increased the absorption, principally in the mid-frequency range. The effect of polyurethane foam on absorption was higher than that of polyethylene foam. Finally, the absorption coefficients of the examined carpets were found to be close to those of Muslim worshippers when they are in the position known as ‘‘standing in rows”. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction In its essential form, the masjid (mosque) is a right prismatic room with a rectangular base. In contemporary masjids, worshippers and carpets constitute the two main acoustic absorbers. Other room surfaces are typically rendered with reflective materials such as marble, paints, glasses, or ceramics. Thus, knowledge of the absorption characteristics of both carpets and worshippers is essential for the acoustic design/assessment of new or existing masjids. A carpet is a special type of textile that is used as a floor covering. Carpets are manufactured by attaching pile tufts (or face yarns) to a backing (or backing yarns) either manually (handmade) or using a machine (machine-made). Masjid carpets are machine-made woven carpets. The main reason to use this type of carpet in masjids, according to manufacturers in Egypt and Saudi Arabia, is its durability in comparison to other machine-made carpets. Thus, the following sections will address only this type of woven carpets. Face yarns can be natural materials such as wool or cotton, or they can be synthetic materials such as acrylic, olefin, nylon, polypropylene, or polyester; blends of wool and nylon can also be used [1]. Backing consists of two threads, which are usually perpendicular, called wefts (or fillings) and warps (or warp chains) [2]. Wefts run widthwise and are made of jute in different diameters. Warps run lengthwise and are made of cotton threads. Warp ⇑ Corresponding author. Tel.: +966 5 3714 3684; fax: +966 6952756. E-mail address: [email protected] (A. Elkhateeb). http://dx.doi.org/10.1016/j.apacoust.2015.12.005 0003-682X/Ó 2015 Elsevier Ltd. All rights reserved.
chains are interlaced with filling and are used to fix face yarns in place and maintain the internal structure of the backing. Additionally, stuffer threads run parallel to the warp chains but are not interlaced with fillings. The main purpose of stuffers is to maintain the dimensional stability, the structure of the backing, and the appropriate backing density required for some types of carpets [1]. In practice, woven carpets are manufactured by a highly complicated technique in which face yarns are weaved through the backing, or simply bent around weft threads, to give the carpet its well-known appearance. The process of weaving uses different methods such as Axminster, Saxony or Wilton. Thus, the structure of the backing may differ widely according to machine and carpet style, and is just as important as the pile itself. Among the different parameters that indicate the quality of a carpet are knot density (calculated as knots per square inch KPSI, knots per square meter KPSM or knots per square cm KPSC), pile height and type, weight per square meter, pitch, total shots, and backing structure. Modern acoustic design requires calculating many parameters. One of the most important acoustic parameters is the reverberation time T. To calculate T, the absorption properties of the room boundaries must be known. Many researchers have commented on the lack of data that are especially oriented to masjid acoustics [3–5]. Consequently, the current practice either ignores the acoustics of masjids or depends on approximations that may lead to inaccurate results. The main purpose of this work is to present the measurement results of the absorption characteristics of 8 types of carpets that are specifically designed and manufactured for masjids. Measurements were carried out in the reverberation
144
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
chamber of the Acoustic Research and Tests Unit (ARTU) in the Faculty of Environmental Design, King Abdulaziz University, Jeddah, Saudi Arabia. It is also of interest to relate their absorption characteristics to their technical specifications such as knot density, pile material, pile height, weight per square meter, and backing structure. Finally, the absorption characteristics of masjid carpets will be compared to the absorption characteristics of Muslim worshippers as previously measured [6]. Because carpets are widely applied as floor coverings in different rooms and have a noticeable acoustic effect, their absorption characteristics have been investigated in many prior research studies. Harris [7] measured the normal absorption coefficient and flow resistance for several hundred carpet samples in order to investigate which variables in carpet construction have a direct effect on the acoustic absorption. Based on the data obtained, certain samples were examined in the reverberation chamber at the National Bureau of Standards for comparison with the tube measurement. Results showed that pile density is one of the fundamental variables that can affect carpet absorption. Shoshani and Rosenhouse [8], investigated the relationship between the absorption coefficients of a cover made of woven fabric and its main parameters, including fibre content, yarn count, cover factor, the air gap behind the fabric, the frequency of the impinging sound wave, and the influence of washing on sound absorption. Results showed that Noise Reduction Coefficients (NRCs, the arithmetic average, rounded to the nearest multiple of 0.05, of the absorption coefficients at 250, 500, 1000 and 2000 Hz [9]) for the examined samples ranged between 0.09 and 0.22; the absorption coefficient was much higher at higher frequencies. Shoshani [10] measured absorption coefficients of tufted carpets backed by several layers of stitch-bonded nonwovens using an impedance tube. Results
(a) Acrylic fibres
indicated that the absorption coefficients of backed carpets were significantly higher than those of un-backed carpets between 250 and 1000 Hz. The increase in the absorption of backed carpets depends primarily on the thickness of the backing, whereas the effect of fibre content is only marginal. Finally, Shoshani and Wilding [11] examined the effect of pile characteristics on the absorption of tufted carpets in the frequency range 125–4000 Hz using an impedance tube. The parameters considered were fibre content and denier, pile density, and average pile height. During measurements, only one pile parameter in the examined samples was varied. One of the important results was the significant increase in the absorption coefficient in the low and medium frequency range, 250–1000 Hz, when there was an air gap behind tufted carpet used as a wall covering. 1.1. Description of the selected samples After a comprehensive survey, masjid carpet manufacturers in the west of Saudi Arabia were identified. Visits and meetings with those manufacturers revealed that there are 8 types of masjid carpets in production. These carpets can be classified into 3 categories on the basis of their face yarn materials: 3 are made of acrylic, 4 are made of heat-stabilised polypropylene (PPHS) and one is made of an 80/20 blend of wool and acrylic. However, the first two are the most common because of their low prices. Masjid carpets have a standard width of about 3.99 m and a length of 25 m. This width fits 3 rows of worshippers, which is the customary grouping during prayer according to the Islamic faith [12,13]. In Saudi Arabia, it is a common practice to install masjid carpets above paddings of either polyurethane foam (PUF) or polyethylene foam (PEF). The former is an open-cell foam (synthetic rubber) adhered to thin solid backing,
(b) Polypropylene fibres
(c) Wool fibres
Fig. 1. Tuft fibres as seen by the electron microscope.
Table 1 Technical specifications of the tested samples. #
a
c d
Pile material
Total shots
Weight
Thickness
Per m2
Per pile
Pile height
kg
mg
mm
Backing
Overall
# of fibres per tuft
Backing structure
Knot density KPSM
KPSC
KPSI
Pitch
Row
Filling
1 2 3
PAZ/1 DIA/7 KAN/2
Acrylic
Single Single Single
4.275 2.60 3.27
5.5 7.0 6.3
12 10.5 11
2.17 2.65 2.68
14.17 13.15 13.68
200–210 200–210 165–180
Wilton TRa STAGb
235,800 250,000 275,100
24d 25 28
152 161 177
11 13 11
14 12 16
Jute Jute Jute
4 5 6 7
PRI/6 LOU/3 CRY/5 MAD/4
Polypropylene HS
Double Single Double Double
2.19 3.84 2.33 2.75
6.8 7.2 5.6 6.4
10.5 10.5 9.5 10.5
2.92 2.45 2.00 2.07
13.42 12.95 11.50 12.57
180–185 105 175–180 170–175
STAGb STAGb STAGb Wilton
137,550 176,850 189,000 235,800
14 17d 19 24
89 114 122 152
11 11 9 11
8 10 14 14
Jute Jute Jute Jute
8
XMIN7/9
Wool
Triple
2.845
17.3
8.5
2.64
11.14
Uncountable
AXMIc
97,520
10
63
7
9
Jute
PUF PEF
Polyurethane foam Polyethylene foam
1.50 0.28
– –
– –
– –
5.70 10.00
– –
– –
– –
– –
– –
– –
– –
– –
9 10 b
Sample code
Triple row. Double row staggered. Axminster. Each knot contains double pile.
145
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Wes
Face Yarns
Wes
Face Yarns
Stuffer Warp Chains
Warp Chains
(a) Double row, in-line structure (Wilton)
Wes
Face Yarns
(b) Double row, staggered structure
Wes
Face Yarns
Stuffer Warp Chains
Warp Chains
(d) Axminster structure
(c) Triple row structure
Fig. 2. Internal structure of the backing of the tested samples, 3d and section perpendicular to wefts.
13.00 12.00
AMPA A1
Room Absorpon m2
11.00 10.00 9.00 8.00 7.00 6.00 5.00 4.00 100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500 3150 4000 5000
Fr. (Hz) Fig. 3. The maximum permissible sound absorption AMPA in the empty room vs. the measured equivalent sound absorption of the empty room A1.
and has an overall thickness of 5.7 mm. The latter is a closed-cell (crosslink) foam covered on one face with aluminium foil; the overall thickness is 10 mm.
The basic unit in the structure of carpet piles is the fibre. For the samples that were studied, in the case of acrylic piles, the fibres have a diameter of about 46 lm. Each acrylic pile consists of 3 tufts
146
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
(a) Sample PAZ/1
(b) Sample KAN/2
(d) Sample LOU/3
(e) Sample MAD/4
(c) Sample PRI/6
(f) Polyurethane foam PUF
Fig. 4. Samples during measurements.
Table 2 The standard error d in (s). Setup Sample
Empty room (ER) Frequency range
Directly on floor Frequency range
On PUF Frequency range
On PEF Frequency range
Low
Mid
High
Low
Mid
High
Low
Mid
High
Low
Mid
High
ER
±0.38
±0.14
±0.03
–
–
–
–
–
–
–
–
–
DIA/7 KAN/2 PAZ/1
– – –
– – –
– – –
±0.45 ±0.51 ±0.54
±0.15 ±0.17 ±0.16
0.03 0.04 0.03
±0.42 ±0.37 ±0.45
±0.10 ±0.12 ±0.12
±0.03 ±0.04 ±0.03
±0.50 ±0.52 ±0.47
±0.17 ±0.16 ±0.13
±0.03 ±0.05 ±0.03
PRI/6 CRY/5 MAD/4 LOU/3 XMIN-7/9
– – – – –
– – – – –
– – – – –
±0.53 ±0.51 ±0.47 ±0.52 ±0.49
±0.18 ±0.17 ±0.17 ±0.15 ±0.15
0.04 0.03 0.04 0.03 0.04
±0.48 ±0.48 ±0.51 ±0.43 ±0.42
±0.12 ±0.13 ±0.14 ±0.15 ±0.15
±0.03 ±0.03 ±0.03 ±0.03 ±0.03
±0.46 ±0.45 ±0.53 ±0.46 ±0.54
±0.15 ±0.16 ±0.15 ±0.13 ±0.19
±0.04 ±0.04 ±0.03 ±0.04 ±0.08
PUF PEF
– –
– –
– –
±0.66 ±0.58
±0.18 ±0.21
0.04 0.08
– –
– –
– –
– –
– –
– –
plaited (twisted) together (Fig. 1a), each of which contains about 55–70 acrylic fibres according to pile weight. Each pile is raised up by bending the tufts around weft threads, thus the total number of fibres per pile ranges from 330 to 420. Pile height in acrylic carpet ranges from 10.5 to 12 mm according to carpet type and design. In contrast to the acrylic case, the tufts in polypropylene piles are not plaited; instead, pile fibres are loose (Fig. 1b); each pile contains about 170–185 fibres, each of which has a diameter of about 51 lm. Polypropylene piles are also raised up by bending the piles around weft threads, so the total number of fibres per pile ranges from 340 to 370. Pile height in polypropylene carpet ranges from 9.5 to 10.5 mm according to carpet type. Under the electron microscope, wool fibres appear dishevelled and very loose (Fig. 1c). Wool fibres have a diameter of about 45 lm whereas acrylic fibres have a range of diameters from 21 to 37 lm. Pile height in the tested wool carpet is 8.5 mm. Table 1 lists the codes of the selected samples and their main technical specifications. Knot density in the selected samples has a wide range from about 100,000 to above a quarter million KPSM (10–28 KPSC). Pile height varies from 9.5 to 12 mm, and weight from about 2 kg/m2 to more than 4 kg/m2. In these samples, backing thickness ranges from 2 to 3 mm and the net weight of the backing (without face yarns) ranges from 30% to 65% of the total weight per square meter.
Filling diameters range from 0.60 to 1.30 mm according to carpet style. Backing structure depends on the way wefts are arranged and warp chains are interlaced with the wefts. For the samples under consideration, 2 structures have been recognised and named according to their forms (Fig. 2): Double row structure: Wefts are arranged in two rows either inline (Wilton [2], Fig. 2a) or staggered (Fig. 2b). The former arrangement was used in samples PAZ/1 and MAD/4; the latter, in samples KAN/2, LOU/3, CRY/5, and PRI/6. Triple row structure: Wefts are arranged in three rows. This has been used only in sample DIA/7 (Fig. 2c) and sample XMIN-7/9 (Axminster [2], Fig. 2d). 2. Materials and methodology For the purpose of this work, samples of each of the 8 mentioned types along with their main technical specifications were collected, categorised, coded, and shipped to the ARTU; each sample was 10 m2. Moreover, samples of both paddings were obtained. Each carpet sample was tested 3 times according to ISO 354 and ISO 9613-1 standards [14,15]. For the first test, the sample was installed directly on the Lab floor (DOF); this will be called setup
Table 3
a values, all samples DOF. Sample
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.09 0.09 0.04 0.06 0.03 0.05 0.02 0.06
0.03 0.02 0.04 0.02 0.01 0.02 0.03 0.00
0.03 0.01 0.05 0.02 0.02 0.02 0.05 0.04
0.05 0.03 0.09 0.04 0.05 0.05 0.05 0.04
0.04 0.06 0.13 0.03 0.05 0.06 0.07 0.06
0.10 0.04 0.16 0.09 0.13 0.08 0.09 0.07
0.10 0.11 0.24 0.12 0.12 0.11 0.13 0.10
0.17 0.20 0.34 0.23 0.24 0.21 0.21 0.18
0.28 0.27 0.47 0.34 0.32 0.28 0.31 0.27
0.31 0.32 0.51 0.36 0.35 0.30 0.30 0.28
0.46 0.49 0.59 0.49 0.48 0.44 0.43 0.39
0.57 0.58 0.70 0.57 0.55 0.50 0.49 0.49
0.70 0.69 0.72 0.69 0.66 0.63 0.63 0.58
0.76 0.78 0.82 0.68 0.64 0.69 0.66 0.62
0.78 0.78 0.81 0.73 0.68 0.71 0.74 0.64
0.83 0.83 0.80 0.79 0.69 0.72 0.76 0.70
0.85 0.85 0.84 0.80 0.78 0.79 0.84 0.75
0.95 0.93 0.94 0.86 0.87 0.86 0.88 0.87
0.95 0.96 0.89 0.94 0.86 0.86 0.94 0.89
0.96 0.98 0.92 0.92 0.96 0.86 0.87 0.91
1.00 1.06 1.01 1.02 1.02 1.06 0.95 0.98
Table 4
a values, all samples on PUF padding. Sample
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
One-third octave centre frequency (Hz) 100
One-third octave centre frequency (Hz) 100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.03 0.02 0.04 0.00 0.02 0.00 0.01 0.01
0.04 0.03 0.02 0.06 0.02 0.02 0.00 0.02
0.02 0.01 0.04 0.04 0.06 0.02 0.06 0.00
0.10 0.10 0.13 0.09 0.11 0.07 0.11 0.06
0.16 0.18 0.21 0.15 0.16 0.13 0.15 0.10
0.19 0.31 0.31 0.26 0.30 0.18 0.21 0.14
0.32 0.48 0.38 0.30 0.44 0.28 0.29 0.17
0.56 0.62 0.63 0.59 0.60 0.58 0.51 0.35
0.65 0.59 0.72 0.70 0.63 0.70 0.60 0.45
0.59 0.56 0.74 0.62 0.63 0.65 0.70 0.46
0.72 0.65 0.83 0.76 0.66 0.77 0.82 0.66
0.79 0.67 0.82 0.80 0.71 0.79 0.88 0.72
0.81 0.74 0.88 0.78 0.73 0.83 0.92 0.85
0.84 0.79 0.85 0.82 0.73 0.83 0.90 0.90
0.82 0.80 0.81 0.78 0.73 0.77 0.93 0.91
0.83 0.83 0.83 0.79 0.73 0.82 0.85 0.94
0.90 0.92 0.86 0.86 0.78 0.84 0.89 0.94
0.90 0.92 0.91 0.84 0.80 0.79 0.82 0.89
0.92 0.97 0.84 0.79 0.81 0.80 0.80 0.91
0.91 0.94 0.86 0.87 0.88 0.83 0.78 0.85
0.86 0.96 0.90 0.87 0.87 0.80 0.93 0.87
147
148
Sample
PAZ/1 DIA/7 KAN/2 PRI/6 LOU/3 CRY/5 MAD/4 XMIN-7/9
One-third octave centre frequency (Hz) 100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10,000
0.01 0.00 0.03 0.02 0.05 0.02 0.00 0.02
0.03 0.01 0.04 0.02 0.01 0.00 0.00 0.03
0.07 0.09 0.15 0.07 0.03 0.07 0.09 0.08
0.07 0.06 0.07 0.06 0.07 0.05 0.07 0.05
0.10 0.13 0.18 0.11 0.13 0.14 0.13 0.11
0.18 0.19 0.25 0.15 0.12 0.15 0.13 0.12
0.19 0.29 0.29 0.18 0.23 0.19 0.20 0.14
0.36 0.45 0.50 0.43 0.42 0.33 0.35 0.30
0.45 0.52 0.54 0.52 0.53 0.48 0.43 0.33
0.51 0.51 0.58 0.53 0.50 0.55 0.47 0.36
0.68 0.60 0.75 0.64 0.62 0.66 0.64 0.50
0.71 0.69 0.78 0.73 0.69 0.74 0.73 0.61
0.81 0.79 0.81 0.75 0.71 0.82 0.84 0.79
0.85 0.77 0.88 0.80 0.69 0.80 0.86 0.88
0.79 0.78 0.81 0.80 0.74 0.76 0.89 0.92
0.81 0.86 0.83 0.78 0.66 0.80 0.82 0.96
0.93 0.88 0.83 0.88 0.77 0.74 0.87 0.95
0.91 0.96 0.86 0.81 0.83 0.80 0.83 0.94
0.88 0.99 0.87 0.87 0.82 0.83 0.91 1.02
0.86 0.99 0.80 0.86 0.80 0.87 0.87 1.05
0.97 1.03 0.87 0.89 0.85 0.84 0.88 1.21
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Table 5
a values, all samples on PEF padding.
149
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Table 6 NRC values, all samples and setups in ascending order according to their knot density. #
Code
Pile type
Knot density
Weight (kg/m2)
Backing str.
8 4 5 6 7 1 2 3
XMIN-7/9 PRI/6 LOU/3 CRY/5 MAD/4 PAZ/1 DIA/7 KAN/2
WO PPHS PPHS PPHS PPHS ACR ACR ACR
10 14 17 19 24 24 25 28
2.535 2.19 3.84 2.33 2.75 4.275 2.6 3.27
AXMIN STAG STAG STAG Wilton Wilton TR STAG
Thickness (mm) Filling
Pile
Overall
2.64 2.92 2.45 2.00 2.07 2.17 2.65 2.68
8.5 10.5 10.5 9.5 10.5 12 10.5 11
11.14 13.42 12.95 11.5 12.57 14.17 13.15 13.68
1. For the second test, it was installed above PUF padding; this will be called setup 2. For the last test, it was installed above PEF padding; this will be called setup 3. In addition, the absorption characteristics of the pads by themselves were also examined. The reverberation room in the ARTU is a rectangular room that has a net floor area SF of 33.74 m2 and net volume V of 163.84 m3. The length of the longest straight line Imax in the room (in this case, the longest diagonal of the room) is 10.22 m, which (according to room volume) satisfies condition (1) of the standards [14]. Since V differs from the 200 m3 stated in the standards, the maximum permissible sound absorption of the empty room, AMPA, were obtained by multiplying the values given in Table 1 of the standards by the value (V/200)2/3 [14]. Then, the calculated AMPA was compared to the measured equivalent sound absorption area of the empty room A1. Fig. 3 presents A1 versus AMPA according to ISO 354 [14], as can be seen, A1 satisfies the standards. In practice, the empty room absorption A1 and the absorption of the room with the tested specimen A2 were obtained by measuring the empty room reverberation time T1 and the reverberation of the room with the tested specimen T2, as shown in Fig. 4. Thus, the absorption of the tested specimen AT was calculated as the difference between A2 and A1 according to Equation (8) of the ISO standards [14]. Measurements of reverberation time were carried out using the Interrupted noise method. The instrumentation was based on products by Brüel & Kjær (B&K); these include inter alia Dirac software ver. 6.0, sound level meter type 2250 as a microphone, Omnisound source type 4292-L and power amplifier type 2734. During measurements, a maximum length sequence (MLS) signal of length 10.9 s was used as a stimulus, and the signal was pre-averaged 5 times for each position. For each specimen measurements, 12 decay curves were collected, 4 positions for the microphone and 3 for the sound source. Positions of both microphone and sound source conformed to the specifications [14]. During measurements, ambient atmospheric temperature TA, relative humidity ht and ambient atmospheric pressure Pa were monitored and recorded to calculate both c (the propagation speed of sound in air, m/s) and 4mV (Air absorption, where m is the sound attenuation coefficient in m1). Values of m were calculated at onethird octave intervals based on apt (pure-tone sound attenuation coefficient in decibels per meter) where apt was calculated by applying equations (3, 4, 5, B.1, B.2 and B.3) of the standard ISO 9613-1 [15]. Finally, the absorption coefficient a of the tested specimen was calculated using the formula [14]:
a¼
AT S
ð1Þ
where S is the area of the tested specimen in m2. Table 2 lists the standard error d of the 95% confidence level that was calculated for each group of the measured decay curves according to the formula:
Pitch
Row
# of fibres/tuft
7 11 11 9 11 11 13 11
9 8 10 14 14 14 12 16
Uncountable 180–185 105 ⁄ 4 175–180 170–175 200–210 200–210 165–180
STDEV d ¼ tð0:025;n1Þ pffiffiffi n
NRC (DOF)
(On PUF)
(On PEF)
0.31 0.34 0.36 0.35 0.36 0.38 0.35 0.47
0.50 0.59 0.57 0.58 0.58 0.56 0.54 0.63
0.45 0.50 0.50 0.48 0.49 0.49 0.47 0.58
ð2Þ
where t: t-distribution curve. n: Total number of decay curves in each measurement. STDEV: Standard deviation of the measured T for each specimen. The standard error d for the empty room was found to be ±0.38 s in the low frequency bands, ±0.14 s in the mid frequency bands and ±0.03 s in the high frequency bands. The mean of d in the room with tested specimens had almost the same values with only slight differences. In the low frequency bands, it was ±0.50 s for setup 1, ±0.45 s for setup 2 and ±0.49 s for setup 3; in the mid frequency bands, ±0.16 s for setup 1, ±0.13 s for setup 2 and ±0.15 s for setup 3; and in the high frequency bands, ±0.03 s for setup 1, ±0.04 s for setup 2 and ±0.03 s for setup 3. The relatively high d in the low frequency bands may be a consequence of the small room volume, while the very small d in the mid and high frequency bands provides a high level of confidence in, and reliability of, the measurements in those bands. 3. Results Tables 3–7 list the absorption coefficient a for the examined carpets and pads calculated at one-third octave intervals. It also shows the Noise Reduction Coefficient NRC for the examined samples. For ease of comparison, values of a for each type of face yarn (acrylic and polypropylene) and for each setup (1, DOF), (2, on PUF), and (3, on PEF) were averaged. As can be concluded from the results, a is an increasing function of frequency and knot density, while pile materials and backing structure have almost no effect. Carpets have a limited effect as absorbers in the low frequency range, but the effect increases in the mid frequency range and becomes very significant in the high frequency range. As a general rule, the higher the knot density, the higher the absorption coefficient a, but some exceptions appear at high frequencies. This conclusion can be seen clearly in Fig. 5a where the averaged values of a are compared for the samples in setup 1. It is clear that the average values of a in the case of acrylic carpet (knot density average 26 KPSC and pile height average 11.17 mm) are higher than those of polypropylene carpet (knot density average 18 KPSC, pile height average 10.25 mm) and wool carpet (knot density 10 KPSC, pile height 8.5 mm). The change in a between different carpet types is small and sometimes can be neglected when their knot densities are close. For example, a values are close in PAZ/1 and DIA/7 (knot density 24 & 25 KPSC respectively, Fig. 5b). The same conclusion applies to LOU/3 and CRY/5 (knot density 17 & 19 KPSC respectively, Fig. 5c). However, an obvious change appears when there is a clear difference in knot density; for example, an obvious change can be observed between PAZ/1 and KAN/2 (knot density
0.88 0.16 0.85 0.13 0.83 0.13 0.72 0.28 0.60 0.39 0.53 0.44 0.46 0.42 0.37 0.29 0.25 0.14 0.19 0.09 0.13 0.07 0.10 0.09 0.12 0.09 0.03 0.02 0.02 0.01 0.07 0.04 0.01 0.03 0.01 0.03 PUF PEF 1 2
0.00 0.05
0.02 0.02
8000 6300 5000 4000 3150 2500 2000 1600 1250 1000 800 630 500 400 315 250 200 160 125
One-third octave centre frequency (Hz)
100 Code #
Sample
Table 7
a values for pads.
0.99 0.26
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
10,000
150
24 & 28 KPSC respectively) in Fig. 5b. Fig. 6 presents a graphical comparison between the a values of the two samples KAN/2 (knot density 28 KPSC, pile height 11 mm) and XMIN-7/9 (knot density 10 KPSC, pile height 8.5 mm); the differences are quite obvious. There is some evidence that pile height and weight per square meter of the carpet improve the absorption. This may clarify why the NRC of LOU/3 (0.36 NRC, 17 KPSC and 10.5 mm pile height) is higher than that of CRY/5 (0.35 NRC, 19 KPSC and 9.5 mm pile height). Moreover, the NRC of PAZ/1 (0.38 NRC, 24 KPSC and 12 mm pile height) is higher than that of DIA/7 (0.35 NRC, 25 KPSC and 10.5 mm pile height), as shown in Table 6. The weight per square meter of the carpet has a similar effect, for example PAZ/1 (0.38 NRC, 24 KPSC and 4.275 kg/m2) is higher than that of DIA/7 (0.35 NRC, 25 KPSC and 2.6 kg/m2). Nevertheless, some results do not match well with this assumption; for example, the NRC of PAZ/1 (0.38 NRC, 24 KPSC and 4.275 kg/m2) is obviously less than that of KAN/2 (0.47 NRC, 28 KPSC and 3.27 kg/ m2) because the latter sample has a higher knot density. We could not independently examine the effect of these two parameters, pile height and weight per square meter, on a as this would require keeping the other carpet variables (such as knot density) constant, which is not currently possible. Pads improve the absorption of carpets in the mid frequency range. Some effects appear also in the low and high frequency ranges, as shown in Fig. 7a and b. The effect of PUF on a in the mid-frequency range is higher than that of PEF; this can be seen clearly in values of NRC for the three examined setups, as shown in Fig. 8 and Table 6. The effect of knot density and pile height on a is still similar to the case of setup 1 although some fluctuations or dips may occur at certain frequencies depend on padding and carpet types; see Fig. 9a and b. The absorption characteristics of the padding itself, both PUF and PEF are shown in Fig. 10. It can be seen that, for PUF, a increases monotonically in all frequency bands although small dips appear at 200, 315, 400, and 630 Hz. Above 630 Hz, the values of a increase rapidly and become comparable with the absorption of some types of carpets, see Table 7. Values of a exceed 0.5 for frequencies above 2500 Hz and reach almost 0.9 at 8 kHz. Moreover, the a of PUF is higher than that of some of the carpets. Conversely, the absorption characteristics of PEF are typical of a resonant absorber with a peak at 2500 Hz (a = 0.44), after which it decreases rapidly, becomes almost constant between 5 and 8 kHz (0.13), and then increases again slightly to reach 0.26 at 10 kHz. To check the applicability of the results of this work and to see if the results can be used to predict the absorption of other woven carpets based on their knot density, a blind sample was collected arbitrarily from a masjid. The knot density was counted (20 KPSC) and the total height was measured (12.5 mm). This blind sample was tested systematically applying the methodology described in Section 2. Later, the measured values of a were compared with the most similar sample (CRY/5, knot density 19 KPSC and total height 11.5 mm), as shown in Fig. 11. Although some differences appear because of the small differences in knot density and total height, the values are still comparable and can be used to obtain results of acceptable accuracy.
4. Discussion One of the important applications of the absorptive materials is to control the reverberation in an enclosure by decreasing the amplitude of the impinging sound waves [16]. Porous materials among the most common absorptive materials; these include cellular and fibrous materials, both of which are presented in this
151
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
Average (ACR - DOF)
Average (PPHS - DOF)
XMIN-7/9 DOF
1.20
1.00
0.80
α
0.60
0.40
0.20
0.00
-0.20
Fr. (Hz)
(a) Setup 1, directly on floor PAZ/1 DIA/7 KAN/2
1
1
0.6
α
α
0.6
PRI/6 LOU/3 CRY/5 MAD/4
0.2
0.2
-0.2
-0.2
Fr. (Hz)
Fr. (Hz)
(c) Polypropylene carpets
(b) Acrylic carpets
Fig. 5. Values of a, setup 1, directly on floor.
KAN/2
XMIN-7/9
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz) Fig. 6. Effect of knot density on a, comparison between KAN/2 and XMIN-7/9.
work. In general, a porous material is a solid substance that contains cavities, channels or holes within its body [17]. These cavities allow the acoustic energy that is transmitted through air to penetrate and interact with the inner solid structure of the material [18]. Fibrous materials, such as carpets, consist of a series of tunnel-like pores that are formed because of the nature of fibrous materials [17]. In such materials, the absorption is generally due to the viscosity of air that affects the particle vibration in the confined space of the pore. Thus, it is found that, for a simple fibrous absorber, the absorption characteristics are affected by fibre radii and
the porosity within a solid material of fibres that effectively provides a matrix. For this reason, the absorption characteristics of fibrous samples of different materials will be the same if they have the same fibre radii and bulk density. Close-up views under the electron microscope, as shown in Fig. 12a–c, indicate how the pores confined between the fibres of carpets ensure an acceptable range of porosity so that sound waves can penetrate the carpet to allow multiple interactions with its structure. On the other hand, cellular materials contain pores, or simply cells, that can be either open or closed. An open-cell material, such
152
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Average (ACR + DOF)
Average (ACR + PUF)
Average (ACR + PEF)
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz)
(a) Acrylic Average (PPHS + DOF)
Average (PPHS + PUF)
Average (PPHS + PEF)
1.2
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz)
(b) Polypropylene Fig. 7. Comparison between a averages in the three setups used in this study.
NRC DOF
NRC PUF
NRC PEF
4
5
0.70 0.60
NRC
0.50 0.40 0.30 0.20 0.10 0.00 1
2
3
6
7
8
Sample # Fig. 8. NRC for the examined carpets in the three setups (see Table 1 for the names of samples).
as polyurethane foam, has a continuous channels of communication with the external surface of the material; such pores effectively increase the absorption of sound [17]. In Fig. 12d and e, close top and cross-sectional views show the variable open-cell cavities of PUF; such pores allow absorption over a wide frequency range despite the small thickness and very light weight of this material.
In contrast, pores in closed cell materials, such as polyethylene foam, are totally isolated from their neighbours; thus their efficiency in absorbing sound waves is considerably less than that of open pores [17]. The closed-cell structure of polyethylene foam (Fig. 12f) prevents this essential penetration, thereby limiting the ability of PEF to act as an absorber despite the fact that it contains more pores than PUF, as can be concluded by comparing the weight per square meter of the two materials (1.5 kg/m2 for PUF and 0.28 kg/m2 for PEF). Data about the absorption coefficient of carpets with either tufted or woven design can be found in many resources such as websites and acoustic handbooks. However, most of these data fail to mention the technical specifications of the carpets. Consequently, using these data for other carpet types could result in an inaccurate estimation. The results of this work establish a clear scientific basis on which a good estimation for the absorption coefficient of other woven carpets can be performed to an acceptable degree of accuracy if their technical specifications are known. Fig. 13 shows the average absorption curve of acrylic carpets in the three examined setups (DOF, on top of PUF and on top of PEF padding) along with the absorption curves of worshippers in each of the two positions: standing in well-defined rows or sitting down at random locations [6]. In the first position, the worshippers stand in orderly rows (usually on carpeted floor) and listen to the imam (leader in group prayer) reciting certain verses of the holy Quran. In the second position (sitting down at random locations on a carpeted floor), the worshippers listen to the imam who delivers
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 KAN/2 + PUF
LOU/3 + PUF
153
XMIN-7/9 + PUF
1
0.8
α
0.6
0.4
0.2
0
-0.2
(a) Setup 2, carpets on PUF padding KAN/2 + PEF
Fr. (Hz)
LOU/3 + PEF
XMIN-7/9 + PEF
1.4 1.2 1
α
0.8 0.6 0.4 0.2 0 -0.2
(b) Setup 3, carpets on PEF padding
Fr. (Hz)
Fig. 9. Effect of padding on a, comparison between 3 types of carpets examined in setups 2 and 3.
PEF
PUF
1
0.8
α
0.6
0.4
0.2
0
-0.2
Fr. (Hz) Fig. 10. Absorption coefficient a of the padding, PUF and PEF.
his Friday speech. These two positions are the most important from the acoustic point of view. As shown in Fig. 13, the absorption caused by worshippers in the standing position was very close to
the absorption of the carpet at frequencies up to 2500 Hz, and thereafter it became slightly higher. Thus, the acoustic environment does not differ significantly between empty and fully occu-
154
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155
α (Blind Sample DOF)
α (Esmated, DOF)
1.20
1.00
α
0.80
0.60
0.40
0.20
0.00
-0.20
(a) Setup 1, directly on floor
Fr. (Hz)
1.00
1.20
0.80
1.00 0.80
0.60
0.60 0.40 α (Blind Sample + PUF)
0.20
α (Esmated, on PUF)
0.40
0.00
0.00
-0.20
-0.20
(b) Setup 2, on PUF
α (Blind Sample + PEF)
0.20
α (Esmated, on PEF)
(c) Setup 3, on PEF
Fig. 11. Measured values of a for the blind sample vs. the expected values.
(a) Acrylic fibres, sample LOU/3
(d) PUF, top view
(b) Polypropylene fibres, sample MAD/4
(e) PUF, cross-seconal view
(c) Wool fibres, sample XMIN7/9
(f) PEF, cross-seconal view
Fig. 12. Voids inside the tested samples as seen by the electron microscope.
pied masjids in the case of standing worshippers. In case of worshippers sitting down at random locations, the absorption of worshippers and carpets tracked each other closely up to 1250 Hz, and then the absorption caused by the worshippers became almost constant at about 0.60 whereas the absorption attributable to the carpet increased gradually.
It is important to mention here that the optimum reverberation time of masjids – and consequently the absorption required – is still under discussion and has not yet been established. A recent study [13] concluded that masjids must be considered a special type of speech rooms that need a relatively long reverberation time. Accordingly, the carpets must be carefully chosen if we want to increase the existing reverbera-
A. Elkhateeb et al. / Applied Acoustics 105 (2016) 143–155 Average (ACR + DOF)
Average (ACR + PUF)
Average (ACR + PEF)
α Standing on Carpet
155
1.2
α Random Sing Down on Carpet 1 0.8
α
0.6 0.4 0.2 0 -0.2
Fr. (Hz) Fig. 13. Average absorption coefficient a of acrylic carpets vs. absorption of Muslim worshippers (standing in well-defined rows and sitting down at random locations).
tion. One of the suggested solutions is to use carpets that have a lower knot density and are installed directly on the floor (without pads).
(12-BUI12759-03). The authors gratefully acknowledge the Science and Technology Unit, King Abdulaziz University for technical support. Thanks also to Dr. Hend Ahmed, Dr. Salem Alhamedee and Arch Zinub Najeeb for their help and continuous support.
5. Conclusions References This work has presented the results of measurement of the absorption coefficient of 8 types of carpets that were especially designed for masjids; in addition, 2 types of paddings, polyurethane foam and polyethylene foam, were characterised. Each carpet sample was tested according to ISO 354 and ISO 9613-1 standards in 3 setups: the first when the sample was installed directly on the Lab floor DOF, the second when it was installed above PUF padding, and the last when it was installed above PEF padding. The absorption of the paddings themselves was also examined. Measurements were conducted in the reverberation chamber of the ARTU in the Faculty of Environmental Design, King Abdulaziz University. Results showed that carpets are good absorbers in the mid- and high-frequency ranges. The absorption coefficient a of a carpet is directly proportional to frequency and to knot density, but the pile materials and backing structure have almost no effect. There is some evidence that the pile height and the weight per m2 of a carpet improve its absorption. Pads improve the absorption of carpets in the mid frequency range, and some effects also appear in the low and high frequency ranges. The effect of PUF on a in the mid-frequency range is higher than that of PEF. The results of this work establish a clear scientific basis on which a good estimation for the absorption coefficient of other woven carpets can be performed to an acceptable degree of accuracy, given their technical specifications. Acknowledgments This work was supported by the NSTIP strategic technologies program in the Kingdom of Saudi Arabia, Project Number
[1] Amirkhan E, Groseclose A. A comprehensive guide to oriental and specialty rug cleaning. 1st ed. Dallas: Oriental and Rug Cleaning Company Inc.; 2006. [2] Complete textile glossary; 2001.