Vous êtes sur la page 1sur 9

Textile Research Journal Article

Sound Absorption Behavior of Knitted Spacer Fabrics

Yanping Liu, Hong Hu1
Abstract This paper presents an experimental Institute of Textiles and Clothing, The Hong Kong
investigation on the sound absorption behavior of Polytechnic University, Hung Hom, Kowloon, Hong Kong,
knitted spacer fabrics. Both weft-knitted and warp- China
knitted fabrics were used in the study due to their
different structure features. The weft-knitted spacer
fabric used was composed of two varied plain knit-
ted outer layers and one textured polyester multifil-
ament spacer layer, and it was considered as a
porous sound absorber. The warp-knitted spacer
fabric used was made with mesh structure on the
outer layers and monofilament yarn in the spacer
layer, and it was considered as microperforated
panel sound absorber. The noise absorption coeffi-
cients (NACs) of these two kinds of fabrics under
both single and multilayer forms as well as their
combinations were tested using a two-microphone
impedance measurement tube. The results show
that the fabric surface structure and thickness,
spacer yarn type and their connecting ways, fabric
combinations and their arrangement methods have
significant effects on the sound absorbability. The
results also demonstrate that good sound absorba-
bility could be achieved by using knitted spacer fab-
rics if suitable fabric structures and combinations
are used.

Key words spacer fabric, sound absorption,

thickness, porous material, perforated panel

Introduction frequencies, an adiabatic process takes place when the

sound waves cross through the irregular pores of porous
material [4], which produces heat loss due to friction. How-
Noise has caused a serious environmental problem in our
ever, the process is an isothermal one at low frequencies.
daily life and has become an increasing public health con-
In this case, poroelastic materials absorb sound due to
cern [1]. A variety of ways are available to reduce noise and energy loss caused by heat exchange [5]. In general, poroe-
they can be basically grouped by passive and active medi-
lastic efficiency is limited to high frequencies.
ums [2]. While passive mediums reduce noise by dissemi-
A resonator is another kind of passive medium for noise
nating energy and turning it into heat, active mediums
reduction by transforming the resonator itself into a vibra-
need the application of external energy in the noise reduc- tion. A noise resonator absorbs sound by means of vibra-
ing process [3].
Porous material is a typical passive medium widely used
for sound absorption. Sound absorbability of this kind of
material depends on the sound wave frequency. At high 1
Corresponding author: e-mail: tchuhong@polyu.edu.hk

Textile Research Journal Vol 80(18): 1949–1957 DOI: 10.1177/0040517510373639 © The Author(s), 2010. Reprints and permissions:
Figures 1–13 appear in color online: http://trj.sagepub.com http://www.sagepub.co.uk/journalsPermissions.nav
TRJ 1950 Textile Research Journal 80(18)

tion via heat loss due to sound pressure variations. machine of gauge 14, using the nylon/spandex 70D/20D
Perforated panels in combination with air-back cavity and yarn as the outer layer yarns and 150Den/64F textured pol-
rigid backing, which are considered as a kind of Helmholtz yester multifilament as the spacer yarns. The yarn path
resonator, have been used widely in various noise control notation of the fabric structure is shown in Figure 1. This
applications [6]. Microperforated panels (MPPs) with the was a weft-knitted spacer fabric specially designed for
hole diameter under millimeter (0.5–1 mm) have been rec- sound absorption. In this fabric structure, both the top and
ognized as one of the most promising alternatives for next- bottom layers are produced with varied plain knitted struc-
generation sound absorbing materials [7]. ture and they are interconnected together with six separate
As a kind of porous material, textiles such as nonwoven, spacer yarns through tuck stitches. The fabric sample after
woven, and knitted fabrics have recently attracted great a steaming treatment is shown in Figure 2. It can be seen
attention for sound absorption application due to their that lots of void pores between two outer layers are formed
low-cost and low environment impact [8–10]. A number of due to the interconnection of textured polyester multifila-
investigations on nonwoven fiber webs in terms of the ment yarns. Thus, this weft-knitted spacer fabric could be
noise-absorption properties and theoretical analyses have considered as a kind of porous sound absorber. In addi-
been conducted in this area [11]. However, despite their tion, the slits formed between the adjacent spacer yarns are
promising noise absorption properties and low cost, it is good for sound waves to penetrate into the fabric.
difficult to produce a textured surface for nonwovens with The warp-knitted spacer fabric used in this study was
an aesthetically pleasing appearance. Thus nonwoven fiber denoted as sample B, and it was produced on a KARL
webs are usually draped with a woven fabric [12]. Plain MAYER RD 6 N E24 double-needle bar Raschel machine
weft-knitted fabric was also proposed for application in of gauge 24, which is equipped with six yarn guide bars.
sound absorption, but its noise absorption performance is The chain notation and material used for each bar are
poor [10]. In order to improve the noise absorption ability listed in Table 1. Different from the weft-knitted spacer
of knitted fabrics, spacer structures were first introduced fabric sample, the spacer yarn used in the warp spacer fab-
by Dias et al. [13, 14]. Their studies focused on the sound ric was monofilament yarn. The fabric sample after dyeing
absorption properties and theoretical modeling of weft- is shown in Figure 3. It can be seen that the holes in the
knitted spacer fabrics, which are composed of two plain elliptical form are regularly distributed on the fabric sur-
knitted surface layers and a spacer layer made of multifila- face. The perforated ratio, which is defined as the ratio
ment yarns through tuck stitches. They found that the between the surface area of the holes and the total area,
sound absorbency of the weft-knitted spacer fabric is effec- was 5.4%. As the minor axis of each hole is smaller than
tive only from 2000 Hz onwards. Furthermore, they also one millimeter, the fabric could be modeled as a microper-
studied the sound absorbency of the weft-knitted spacer forated panel (MPP) absorber. Besides, the void spacer
fabric made of monofilament yarn as a spacer yarn, in com- between the two outer layers could be considered as an air
bination with a uniform pattern of micropores on the sur- gap.
faces. Their work showed that this kind of fabric could All the fabric samples were conditioned for 24 hours at
provide reasonable absorbability at mid-high frequencies, 20 °C and 65% relative humidity before testing. The struc-
but with a narrower absorption frequency range. Though tural details of two kinds of fabrics after being conditioned
knitted spacer fabrics are more expensive than nonwovens,
their appearances and structures are designable and this
characteristic can raise their added values.
This paper reports an investigation on the sound
absorption behavior of both weft and warp-knitted spacer
fabrics and their combinations. The objective is to find out
a sound absorber with improved sound absorbency at
lower frequencies with a relatively wider absorption fre-
quency range.

Both the weft and warp knitted spacer fabrics and their
combinations were used in this study. The weft-knitted
spacer fabric used was denoted as sample A, and it was Figure 1 Yarn path notation of weft-knitted spacer fabric.
knitted on a Stoll CMS 822 E7.2 computerized flat knitting
Sound Absorption Behavior of Knitted Spacer Fabrics Y. Liu and H. Hu 1951 TRJ

Figure 2 Weft spacer fabric sam-


Figure 3 Warp spacer fabric sam-

ple with hole size and distribution.

Table 1 Chain notation and yarns used for warp-knitted spacer fabric.
Chain notation Yarn
GB1 1-0-0-0/1-2-2-2/2-3-3-3/3-4-4-4/4-5-5-5/4-3-3-3/3-2-2-2/2-1-1-1// 150D/48F polyester
GB2 4-5-5-5/4-3-3-3/3-2-2-2/2-1-1-1/1-0-0-0/1-2-2-2/2-3-3-3/3-4-4-4// 150D/48F polyester
GB3 1-0-1-2/1-2-1-0// 30D/1F polyester
GB4 2-3-4-5/3-2-1-0// 30D/1F polyester
GB5 4-4-4-5/4-4-4-3/3-3-3-2/2-2-2-1/1-1-1-0/1-1-1-2/2-2-2-3/3-3-3-4// 150D/48F polyester
GB6 1-1-1-0/1-1-1-2/2-2-2-3/3-3-3-4/4-4-4-5/4-4-4-3/3-3-3-2/2-2-2-1// 150D/48F polyester
TRJ 1952 Textile Research Journal 80(18)

Table 2 Structural details of spacer fabrics.

Samples Air resistance (KPa·s/m) Thickness (mm) Porosity (%) Density (kg·m–3) WPC CPC
A 0.513 7.480 89.068 150.868 9.2 24.0
B 0.013 4.324 90.958 124.774 10.0 16.5
WPC, wales per cm; CPC, courses per cm.

are given in Table 2. The air resistance was measured with

the KES-F8-AP1 air permeability tester and the fabric
Results and discussion
thickness was measured with a KES-FB3-A compression
tester at a pressure 0.5 gf/cm2. Here WPC is the number of Sound absorption behavior of single layer
wales per cm and CPC is the number of courses per cm. spacer fabrics
The noise absorption coefficients (NACs) of single layer
spacer fabrics without an air-back layer are shown in Fig-
Sound absorbency tests ure 5. It is found that the NACs of both samples A and B
The sound absorbency tests were carried out by measuring increase with increase of the frequency. The NACs of sam-
the noise absorption coefficient (NAC) using a Brüel & ple A are 0.06, 0.14 and 0.35 at 500 Hz, 1000 Hz and 2000
Kjær two-microphone impedance measurement tube 4206 Hz, respectively, which exhibits a typical sound absorption
according to the ISO 10534-2 standard. As shown in Fig- behavior of porous material. Compared with sample A,
ure 4, the tube, of which the diameter is 29 mm, was sample B has lower NACs for all the frequencies. The
designed to measure the noise absorptions of normal inci- NACs of sample B are 0.035, 0.05 and 0.07 at 500 Hz, 1000
dent sounds within 0–6400 Hz. Hz and 2000 Hz, respectively. Furthermore, the curve of
All the samples for testing were firstly cut into the circular sample B has two slight peaks. This is the typical behavior
form of 29 mm in diameter, and then placed into the imped- of MPP absorbers originating from combining different
ance tube either being close to the metal plunger or leaving a sizes of perforated holes. Due to the existence of slits like
distance as air-back cavity depending on the test conditions. perforated holes, sample A can also be considered as a
The sample placement shown in Figure 4 is a combination of perforated panel absorber if the air-back cavity exists. The
A-B without an air-back cavity. For testing multilayered sam- NACs of A and B with two different thicknesses (8 and 16
ples, the samples should be closed and conjoined together. cm) of air-back cavity are shown in Figure 6. It can be seen
The single layer and multilayer weft-knitted and warp-knit- that all the sound absorption curves are in the form of fre-
ted spacer fabrics, which were laminated with the same or quency spectra, because the existence of the air-back cavity
different kinds of fabrics, were tested in this study. causes frequency-selected sound absorptions due to the

Figure 4 Sound absorbency meas-

urement system.
Sound Absorption Behavior of Knitted Spacer Fabrics Y. Liu and H. Hu 1953 TRJ

frequencies is still very low without the use of air-back


Sound absorption behavior of multilayered

spacer fabrics
The above analysis has shown that the single layer spacer
fabric possesses low NACs at lower frequencies. Though
the use of the air-back cavity could enhance the NACs, the
frequency range became very narrow. In addition, the use
of the air-back cavity is not convenient in the practical
application for spacer fabrics. In order to improve the
NACs at lower frequencies, the use of laminated absorbers
with the increased thickness is the most commonly used
method. In this section, the NACs of the sound absorbers
laminated with the same kind of spacer fabric are pre-
Figure 5 NACs of single layer spacer fabrics without air- For porous sound absorbers, the NACs depend on their
back cavity. thickness, porosity, airflow resistivity and appearance. In
general, increasing the thickness can considerably enhance
the NACs at low frequencies. However, the increase of the
thickness does not have the obvious effects on NACs at
high frequencies and may result in slight reduction in NAC
sometimes. Also, the effect of increasing the thickness to
enhance the NACs at low frequencies is limited, since the
NACs no longer increase after the thickness reaches a crit-
ical value. This phenomenon can be confirmed by Figure 7,
where the NACs of the porous sound absorbers laminated
together with different layers of the same weft-knitted
spacer fabric are shown. It can be found that the two-lay-
ered spacer fabrics already have a very obvious improve-
ment in the sound absorption capacity of the spacer fabric
at low frequencies. However, the improvement becomes

Figure 6 NACs of spacer fabrics with air-back cavity.

resonance of the system. Moreover, the existence of the

air-back cavity results in narrower absorption frequency
ranges, but with higher NACs at lower frequencies if com-
pared to the case where no air-back cavity is used. The
thickness of the air-back cavity also has the influence on
the NACs. The increase of the air-back cavity makes the
absorption frequency move toward the lower frequency
From Figure 5 and 6, it can also be found that sample
A has better sound absorbability than sample B for both Figure 7 NACs of weft-knitted spacer fabrics laminated
cases of using and not using the air-back cavity due to with different layers.
higher thickness. However, its sound absorbency at lower
TRJ 1954 Textile Research Journal 80(18)

slower after two layers, and the effect of increasing the Based on the sound absorption principle of MPPs, it is
thickness on the NACs is no longer observed after five lay- possible to fabricate different kinds of warp-knitted spacer
ers. To further improve the NACs, other methods such as fabrics with different frequency characteristics. The
increasing the porosity, airflow resistivity and fabric sur- lamination with different warp-knitted spacer fabrics or
face smoothness or a combination of different kinds of fab- the use of different mesh size on the same fabric can be
rics, could be considered. The peaks around 1600 Hz found potential ways to achieve a broader absorption frequency
in Figure 7 for the absorbers laminated from two to eight range.
layers of spacer fabrics could be the resonance of the sys-
For MPP absorbers, the NACs depend on the perfora- Sound absorption behavior of combined
tion ratio, dimension of the perforated holes, thickness of multilayered spacer fabrics
the panel and thickness of the air-back cavity. The MPP In industrial applications, multilayered sound absorbers
absorbers provide high absorbability at mid-high frequen- composed of both perforated panels and porous materials
cies. However, their absorption frequency ranges are nor- are widely adopted for broadband noise absorptions [15].
mally limited because of their nature as a resonator. To In this regard, the sound absorption behavior of multilay-
obtain a wider frequency range for sound absorption, the ered spacer fabrics laminated with different combinations
combination of multilayered MPP absorbers with different of sample A (porous material) and sample B (perforated
frequency characteristics or the use of a MPP absorber panel) were studied. Two cases, where sample A was
with different size of perforated holes are always adopted. placed on the front or the back of sample B, were consid-
The NACs of the MPP absorbers formed with different ered.
layers of warp-knitted spacer fabrics are shown in Figure 8. The NACs of different layers of sample B backed with
It can be found that the NACs significantly increase with one layer of sample A are shown in Figure 9. Compared
the increase of fabric layers. In addition, the resonance with Figure 8, it can be noticed that the sound absorbability
phenomena are observed for all the fabric layers and their of the warp-knitted fabrics can be considerably improved by
resonance region shifts towards the lower frequency side backing one layer of sample A. Moreover, it can also be
with an increase of the fabric layers. By examining the found that the resonance regions shift towards the lower fre-
curve of the thickest absorbers formed with eight layers of quency side when the layers of sample B increase. This
sample B, two peaks respectively at 2128 Hz and 6184 Hz phenomenon has already been observed in Figure 8. How-
are found due to resonance. According to the curves in ever, the NACs for the low frequencies under 500Hz are
Figure 8, it can be predicted that the NAC will continu- still less than 0.5, which is a critical value normally used to
ously increase with increase of the fabric layers. Thus, it is assess whether a sound absorber is good or not. As two-lay-
expected that an absorber with better sound absorbability ered weft-knitted spacer fabrics already have a very obvi-
can be obtained if enough layers are used. This behavior is ous improvement of the sound absorption capacity at lower
different from that of the sound absorbers formed with dif-
ferent layers of weft-knitted spacer fabric, as explained

Figure 8 NACs of warp-knitted spacer fabrics laminated Figure 9 NACs of different layers of B backed with one
with different layers. layer of A.
Sound Absorption Behavior of Knitted Spacer Fabrics Y. Liu and H. Hu 1955 TRJ

Figure 10 NACs of different layers of B backed with two Figure 11 NACs of one layer of A backed with different
layers of A. layers of B.

frequencies (Figure 7), the NACs of different layers of

sample B backed with two layers of sample A were also
investigated. The results are shown in Figure 10. It can be
seen that in this case the NACs for both lower and higher
frequencies are further improved and a broader absorption
frequency range has been achieved. It is also found that the
NAC of 8B-2A at 500Hz reaches 0.5. By observing the
curves in Figure 10, it is also noticed that the curved wave
forms for the middle and high frequencies become
straighter, and thus the sound absorption behavior of the
fabrics becomes more stable. If the layer number of sample
A continues to be increased, the NACs at low frequencies
could be improved further, but this improvement should be
limited since the results shown in Figure 7 have already
confirmed that the NACs no longer increase when the
thickness reaches a critical value. Besides, increasing the
layers of the fabrics will result in an increase of material
use and cost. At the same time, too high thickness of the Figure 12 NACs of two layers of A backed with different
sound absorbers is not convenient in practical application. layers of B.
By changing the arrangement sequence, i.e., by swap-
ping the positions of weft-knitted and warp-knitted spacer
fabrics, the sound absorption behavior gets quite different.
The NACs of one layer of A backed with different layers of low frequencies are further improved. At the same time,
B are shown in Figure 11. Compared with Figure 9, it can the curved wave forms for the middle and high frequencies
be found that in this case the sound absorption capacity of become straighter. This situation is the same as shown in
the multilayered spacer fabrics at low frequencies are Figure 10, but with lower values. The small peaks around
much better improved with increase of sample B layers, 1500Hz due to resonance are still observed in this case.
but the NACs for higher frequencies are much lower, The above results show that the arrangement sequence
though their values are bigger than 0.5. The same phenom- has an obvious effect on the sound absorption. This is
enon, i.e., the shift of the resonance regions towards the because the different positions of sample B result in dif-
lower frequency side, is also observed. The small peaks ferent sound absorption effects. When sample B is placed
around 1500 Hz are found due to resonance. The NACs of before sample A, sample B performs as a resonator
two layers of sample A backed with different layers of sam- absorber, which it is helpful for improving the NACs for
ple B are shown in Figure 12. It is found that the NACs at all the frequencies. Conversely, when sample B is placed
TRJ 1956 Textile Research Journal 80(18)

(MPP) absorber. The noise absorption coefficient

(NAC) increases with increase of the frequency for
both kinds of fabrics.
2. Both the weft-knitted and the warp-knitted spacer
fabrics backed with the air-back cavity exhibit fre-
quency-selected sound absorption with a resonance
3. The sound absorbability can be improved by lami-
nating different layers of fabrics. For the weft-knit-
ted spacer fabric, the NACs significantly increase
from one layer to four layers and thereafter more lay-
ers are no longer effective. However, for the warp-
knitted spacer fabric, the NACs can continuously be
improved with increase of the fabric layers, but with
a shift of the resonance region towards the lower fre-
quency side.
4. The combinations of weft-knitted and warp-knitted
Figure 13 Comparison of the NACs for sample A backed spacer fabrics can significantly improve their sound
with air layer or eight-layered sample B. absorbability, but their arrangement sequence has
an obvious effect. At higher frequencies, the NACs
of the warp-knitted spacer fabrics backed with weft-
knitted fabrics are much higher than those of the
behind sample A, it performs as a air-back cavity and leads weft-knitted spacer fabrics backed with warp-knitted
to an increase of the thickness of the system. In this case, fabrics. However, at lower frequencies, the NACs of
the NACs of the system can be significantly increased at the warp-knitted spacer fabrics backed with weft-
low frequencies, but with a cost of NAC reduction at high knitted fabrics are much lower than those of the
frequencies. The NACs of one or two layers of sample A warp-knitted spacer fabrics backed with weft-knitted
respectively backed with 4 cm air layer and eight layers of fabrics.
sample B are shown in Figure 13. Here, the thickness of 5. The air-back cavity can be replaced by multilayered
the air-back cavity (4 cm) is close to that of eight layers of warp-knitted spacer fabrics to achieve high NACs at
sample B (3.46 cm). It can be found that the curves low and middle frequencies.
between the same layered sample A backed with the air-
back cavity and eight layered sample B are nearly coinci-
dent when the frequencies are below 3000 Hz. The differ- Acknowledgement
ences of the NACs above 3000 Hz are due to the minor
The authors would like to acknowledge the funding sup-
thickness difference between eight layers of sample B and
port from the Block Grant except CRG of The Hong Kong
the air-back cavity as well as the space occupied by sample
Polytechnic University (Grant No. 1-ZV0M).
B. These results indicate that air-back cavity can be
replaced by multilayered warp-knitted spacer fabrics to
achieve high NACs at low and middle frequencies. This
can be a convenient way for the practical application. References
1. Lou, C. W., Lin, J. H., and Su, K. H., Recycling Polyester and
Polypropylene Nonwoven Selvages to Produce Functional
Conclusions Sound Absorption Composites, Textile Res. J. 75(5), 390–394
The sound absorption behavior of knitted spacer fabrics 2. Sagartzazu, X., Hervella-Nieto, L., and Pagalday, J. M.,
was studied. The effects of different fabric layers and Review in Sound Absorbing Materials, Arch. Comput. Meth-
arrangement sequences on the noise absorption coefficient ods Eng. 15(3), 311–342 (2008).
3. Kuo, S. M., Morgan, D. R., “Active Noise Control Systems,”
were analyzed and compared. The following conclusions
Wiley, New York, 1996.
can be obtained from this study:
4. Cummings, A., and Chang, I. J., Acoustic propagation in
porous media with internal mean flow, J. Sound Vib. 114(3),
1. While the weft-knitted spacer fabric exhibits the typ- 565–581 (1987).
ical sound absorption behavior of porous absorber, 5. Zhang, B., and Chen, T. N., Calculation of sound absorption
the warp-knitted spacer fabric exhibits the typical characteristics of porous sintered fiber metal, Appl. Acoust.
sound absorption behavior of microperforated panel 70(2), 337–346 (2009).
Sound Absorption Behavior of Knitted Spacer Fabrics Y. Liu and H. Hu 1957 TRJ

6. Sakagami, K., Morimoto, M., and Yairi, M., A Note on the 11. Shoshani, Y., and Yakubov, Y., A Model for Calculating the
Relationship between the Sound Absorption by Microperfo- Noise Absorption Capacity of Nonwoven Fiber Webs, Textile
rated Panels and Panel/Membrane-type Absorbers, Appl. Res. J. 69(7), 519–526 (1999).
Acoust. 70(8), 1131–1136 (2009). 12. Shoshani, Y. K., Noise Absorption by a Combination of
7. Maa, D. Y., Potential of microperforated panel absorber, J. Woven and Nonwoven Fabrics, J. Text. Inst. 82(4), 500–503
Acoust. Soc. Am. 104(5), 2861–2866 (1998). (1991).
8. Tascan, M., and Vaughn, E. A., Effects of Total Surface Area 13. Dias, T., Monaragala, R., and Lay, E., Analysis of Thick
and Fabric Density on the Acoustical Behavior of Nee- Spacer Fabrics to Reduce Automobile Interior Noise, Meas.
dlepunched Nonwoven Fabrics, Textile Res. J. 78(4), 289–296 Sci. Technol. 18(7), 1979–91 (2007).
(2008). 14. Dias, T., Monaragala, R., Needham, P., and Lay, E., Analysis
9. Shoshani, Y., and Rosenhouse, G., Noise absorption by woven of Sound Absorption of Tuck Spacer Fabrics to Reduce Auto-
fabrics, Appl. Acoust. 30(4), 321–333 (1990). motive Noise, Meas. Sci. Technol. 18(8), 2657–2666 (2007).
10. Dias, T., and Monaragala, R., Sound Absorbtion in Knitted 15. Zhu C. Y., and Huang, Q. B., A Method for Calculating the
Structures for Interior Noise Reduction in Automobiles, Absorption Coefficient of A Multi-layer Absorbent Using the
Meas. Sci. Technol. 17(9), 2499–2505 (2006). Electro-acoustic Analogy, Appl. Acoust. 66(7), 879–887 (2005).