1964 Effect of Fabric Covering on Sound Absorption

8/6/2019 1964 Effect of Fabric Covering on Sound Absorption

http://slidepdf.com/reader/full/1964-effect-of-fabric-covering-on-sound-absorption 1/16

RESEARCH DEPARTMENT

Effect of fabric coverinqs

the soundabsorptlon of

porous materials

RESEARCH REPORT N

1964/7

THE BRITISH BROADCASTING CORPORATION

ENG INEERING DIVISION





R E S E A R C H D E P A R T M E N T

EFFECT OF FABRIC COVERINGS ON THE SOUND ABSORPTION

OF POROUS MATERIALS

•

. A .N . Bar-d , B .Sc., A . l n s t . P .

R e s e a r c h R e p o r t N o . B - 0 8 1

(1964/7)

/r;r'7i,,,, c r # Y . hV< , 0'.,,

.---

_-( W. P r oc to r W il so n



.8.1 not be

wIthout the

Corporation.

This Report i. the property o r the

British Broadcasting Corporation and

reproduced in any form

written permission ot the



Report No. B-081


OF POROUS MATERIALS

Section Title P

SUMMARY .

1. INTRODUCTION. . . . . . . . . . . . . . . . . . . . .

2 . RESULTS .

2.1. Absorption Coefficients ~~th Stretched Fabrics

2.2. Absorption Coefficients with Draped Fabrics .

2.3. Absorption Coefficients of Fabrics with No Porous Backing Material

3. PREDICTION OF BEHAVIOUR OF FURTHER MATERIALS. . . . . . . . . .

4. DISCUSSION A.l\l"DCONCLUSIONS ... . . . . . . . . . . . . . . . . ..

5 . REFERENCES . . . . . . . . . ,





Research

February 1964 Report No. ~081

0964/7 )


OF POROUS MATERIALS

SilllMARY

Measurements have been carried ou t by the reverberation room method

determine the effect of fabric coverings on the sound absorption of porous bac

materials. It is hoped that it may prove possible in the future to predict

behaviour of materials from measurements on small samples.

1. INTRODUCTION

In recent years Tygan or perforated hardboard covers for absorbers

become unpopular with many of the users of studios and other types of fabric have

considered as possible substitutes for these materials. The Tygan coverings

manufactured from a plastic fibre and are generally of an open weave constructi

making them effecti vely transparent to sound. Perforated and slot ted hardboard co

are sometimes used to produce a resonance peak in the absorption characteristic o

reduce the absorption at high frequencies. Suitable combinations of absorbers

such coverings have been produced to give absorption characteristics of many des

forms.

Since it was not known how far these effects could be reproduced by o

fabrics, measurements were undertaken to determine the absorption of resist

materials wi th. fabrics draped or .stretched over their surfac~s. The two m

factors in the construction of a fabric considered to be of importance are the sur

density of the material and the flow resistance. The surface density of a mate

IS defined as the mass per uni t area of the-material (gm/m2) while the flow resist

is defined by the relationship

R = p / v

where R IS the flow resistance ( rayIs)

P IS a pressure differential maintained across the sample (dyne/em

v is the velocity of airflow through the sample produced by the

p re ss ur e d if fe re nt ia l ( cm /s ec ),



2

It was thought probable that the j fluffiness' of the material would affect

the high frequency absorption obtained wi th the sample, and though no simple method

of measuring this was apparent, this factor was borne in mind in assessing the effects

of surface density and flow resistance.

Measurements of weight and flow resistance were made on a large number of

fabrics to determine the range of the parameters involved and six fabrics were

selected for full-scale measurements of absorption coefficient in the reverberation

room (Table 1 ) , Sufficient fabric was purchased to enable 1 0 0 ft2 ( 9 · 5 m2) of porous

material to have twice its own area of ,fabric ,draped in front of it,

TABLE 1

F l o w R e s i s t a n c e a n d W e i g h t o f C h o s e n F a b r i c s

Fabric Surface Density (gm/m2) Flow Resistance (ray Is)

Cotil Fabrics 1 5 6 11

Kval 1 7 0 3

Cotil Fabrics 1 5 5 3 3

Kval 5 7 5 0

Cotton Everglaze 1 5 1 1 3 8

Spectrum Fabrics 5 1 8 1 0 - 5

Range 1 9 1 / A 9 5 8

Edinburgh Weavers 4 8 0 3 2

Tiana 0 3 2 1

Sundour Fabrics 4 7 0 2 3 0

Measurements were carried out with the fabrics alone and covering two

types of resistive backing material which were near the ends of the range of flow

resistance found with such materials. They were Bondacoust wadding (of flow resistance

about 5 rayls/an) and Stillite Therbloc (of flow resistance about 110 r ay l s/ cm},

The BBC's standard method of measurement of absorption in the reverberation

room was followed. Four frames, each of area 6 ft x 4 ft ( 1 · 8 m x 1 ' 2 m) were

mounted on the walls and floor of the large reverberation room, measurements were mad

of the reverberation time wi th and wi thout the sample, and the absorption was cal-

culated from the change of the average reverberation time.

2. RESULTS

The range of values of surface density and flow resistance which were

obtained with a . large number of samples is shown in Fig. 1; the results for the SlX

materials chosen for full-scale tests are ringed in this figure. The selection wa

made to allow the influence of flow resistance variation to be studied at each of two



1,000

c Llo , I !

,~I 0 I :

b 'O t0 (~ I i

500I - " - v"% (; v

1

18 1~~ 0 ~

4000

!0 u

';0 o >I

- - d - - r -,

0 0 I i

3

Ii

C, 9 I ! C,

0 0 'tP

o J j I200 q

d &I I

Ite 0

I@

r I100 o

r.o 0 0<'l ~ 10

oor.

F i g . 1-V a l u e s o f f l o w

r e s i s t a n c e a n d ' s u r f a c e

d e n s i t y o f f a b r i c s : ; -;;;ccp

o Fabrics used for preliminary U

measuremen ts ~o't:l. . .

~ Fabrics selected for rever-b er ati on roo m me as ure me nt s

flow resistance, Rayls

surface densities of materials. No selection of surface finish was possible

the heavy fabrics are all more fluffy than the light ones. The values of

resistance and surface density encompass a wide variety of materials and may en

prediction of behaviour of further materials to be made by interpolation.

2.1. Absorption Coefficients with Stretched Fabrics

Figs. 2 and 3 show the absorption coefficients obtained with the

selected fabrics stretched over a 2 inch (5 em) layer of Bondacoust. Figs. 4 a

are the results for the fabrics stretched over a 2 inch (5 em) layer of Therbl

It can be seen from these results that a wide range of absorption characterist

may be obtained, and certain trends emerge.

Compared with the absorption of an uncovered low-resistance backing mate

such as Bondacoust, all the covering materials produce an increase in absorption

frequencies below about 700 c i s , accompanied by a reduction of absorption at

higher frequencies. These trends become more evident as the flow resistance

the fabric is increased, but for the heavy fabrics which all have 'fluffy' surfthe high frequency absorption is largely recovered. The variation of absorpt

with surface density at a given flow resistance is masked at high frequencies

the surface absorption of the heavy materials but it does appear that at the

f re que nc ie s th e he av ier m at er ial ab sor bs m or e e ff ic ie ntl y.

Compared with the absorption of an uncovered high resistance back

material such as Therbloc, only slight increases of absorption are found at the l

frequencies, and the characteristic high frequency reductions are not so severe

with Bondacoust. It appears that the high flow r-esistance of the backing mater

1S sufficient to swamp some of the effects of the flow resistance of the fabrics.

The lightweight fabrics give rise to a greater total absorption

measured with a high resistance backing than with a low resistance backing.

peak value of absorption which is at or about 500 c i s is practically the same

both backings but elsewhere an increase of absorption is found. Wi t h the h

fabrics the backing does not greatly affect the absorption at medium and high

quencies but at the lower frequencies an improvement is found when using the

resistance backing for all except the high flow resistance fabric.



4

Influence of flow resistance on the absorption of fabrics stretched

over 2 inch (5 c m) B on da co us t

Fig. 2 - Lightweight fabrics Fig. 3 - Heavy fabrics

~---=c§1~'

1 = = = = ~=;t :~,

_~6_

= =r===~~~ - , , = = " = '

= c c _ =

~~~t l18$28{\J C'1 L O f'. S f .

frequency. cis

Fig. 4 - L ig ht we ig ht f ab ri cs

0000000~0g_ag882N"'.no:i~

Fig. 5 - He av y fa bri cs

Influence of flow resistance on the absorption of fabrics stretched

over 2 inch (5 c m) T he rb lo c

--)(-- low flow resistance---0--- medium flow resistance

--e-- high flow resistance-- no fabric

2.2. Absorption Coefficients with Draped Fabrics

Since it is frequently preferred, for aesthetic reasons, to use fabrics

draped rather than stretched over a surface, the measurements were carried out for

each fabric over each backing with twice the area of fabric hung In folds over the

o ri gi na l s am ple a rea .



Figs. 6 and 7 show a comparison of the results obtained in two extre

cases with fabrics draped or stretched. It is not necessary to reproduce all t

results obtained but the following tendencies were found:

(a) With low flow resistance fabrics, stretching is more effective than draping

when a low resistance backing is used, and it still remains almost as

effecti v e wi th a high resis t ance backing.

(b) With medium flow resistance fabrics stretching is as effective as drapingwhen a low resistance backing is employed but less effective with a high

r es is ta nc e b ac ki ng .

(c) With high flow resistance fabric~ draping is always more effective than

stretching.

2.3. Absorption Coefficients of Fabrics with No Porous Backing Material

Measurements were made with the fabrics stretched or draped in front of

2 inch (5 em) air space with no porous backi rr g material. Figs. 8 and io show t

a bsorption characteristi cs of stre tched heav y and light fabrics respect ively.

In the case of heavy fabrics an increase of flow resistance leads

improved absorption at the lower frequencies coupled with a sligh t increase of t

peak values. Two of the lightweight fabrics show the same effect but the third,

high flow resistance fabric, upsets the pattern; this may be due to a membran

resonance of the fabric. This resonance effect has been found also in impedance tU

measurements, the frequency of resonance. being proportional to the square root of t

tension with which the fabric was stretched across the tube.

1·4

1'3

1·2

.... 1·1

~o·8°'8cO,

il0'0a.

g O ' 5 ==jc:

.00'4e

0'3

Compar i so n of t he e ff e ct s o f dr a pi~g or s t r e t c h i n g a f ab r ic o v er a

r es ist ive ba cki ng ma ter ia l

= . ~ . c . . z : :; . ' I -:a _ ' ; l . -= r = 1==

.C i..~ ~

=

·i==r = = ' E - - .'F=

= = r = = = = r ~ =~ ~ § ~ § ~ ~ ~ g ~ g ~ Mfrequency, cIs

=

F i g . 6 - L ow fl o w r es i st a nc e f a br i c

---0--- draped fabric

-- no fabric

--e-- stretched fabric

= c-,;~

.. ~. =t=;".£~

=

F i g . 7 - H ig h f lo w re s is t an c e fa br i c



6

Influence of flow resistance on the absorption of hedvy fabrics

over a 2 inch (5 cm) ai r space

Fig. 8 - Fabrics stretched Fig. 9 - F abrics dra ped

Fig. 10· - Fabrics stretched Fig. 11 - F abrics draped

Influence of flow resistance on the absorption of lightweight fabrics

over a 2 inch (5 cm) air space

~-.-- high flow resistance ----0-- mediumflow resistance

--x-- low flow resistance

Draping of heavy fabrics is generally not advantageous and, in fact, Some

loss of low frequency absorption is found (compare Figs. 8 and 9). With the light-

weight fabrics, however, a general improvement is found at high frequencies, particu-

larly in the case of the high flow resistance fabric (Figs. 10 and 11).

3 . PREDICTION OF BEHAVIOUR OF FURTHER M ATERIALS

It was expected at the outset of these measurements that an estimate of the

behaviour of additional materials could be made by interpolation or limited extrapola-

tion from the measured results. A determination of the surface density and flow

resistance coupled with the feel of the surface should enable a reasonable estimate

of the absorption to be made.

One occasion has in fact arisen in which verification of an estimate was

possible. A lightweight Duracour of flow resistance 170 rayls and surface density

160 gm/m2

would be expected to behave in the same way as the Cotton Everglaze measured

previously. A comparison of the results is shown in Figs. 12 and 13. The general

trend is similar but lower values are found at 500 c/s for the Duracour tKan for the

C o tt o n E v er g la z e.



CORparis on of ab sorp tion c oeff icie nts ob tain ed wi th ligh twe ight Durac our

and Cotto n Ever glaze over t wo resi stiv e ba cking mat eria ls

o0·1

o~' "

, . . . - ,@'l

"<. D ? J ~ g : B Q S § t s g 8 g

.. N C'l ~ ~r ~ 2 ~.~

frequency. cis fl"'equency, cis

Fig. 12 - B on da co us t b ac ki ng Fig. 13 - 7h er bl oc b ac ki ng

---0--_ Cotton Everglaze--no fabric

--.-- lightweight Duracour

More recently it has been shown that it is possible to calculate

absorption at random incidence for a finite sample areal from tube admittance meas

ments. Such measurements2 are made on a small piece of the material which is mou

at one end of a cylindrical tube. A loudspeaker at the other end generates p

sound waves in the tube and a'probe microphone explores the standing wave pat

which is set up. From a determination of the position and depth ·of the interfer

minima it is possible to calculate the acoustic impedance or admittance of the sam

The calculation based on Northwood's results is shown in Figs. 14, I S

and 17 to be capable of a good approximation to the measured values for both low

high resistance backing materials. However, in the case of the lightweight

flow resistance fabrics resonance effects occur as mentioned previoqsly.

calculated curves are shown in Fig. 16 for this case ~nd the difference between t

two sets of results is representative of the uncertainty that might exist in fig

derived f rom meas urements on sma ll sa mples.

4. DISCUSSION AND CONCLUSIONS

The measurements described have confirmed that a wide range of absorp

characteristics may be obtained with different fabric coverings. If the max

absorption over the widest frequency range is required, materials of high sur

density and low flow resistance should be used over a high resistance backing.

is also possible 'to obtain an absorption characteristic similar to that obtained

perforated materials; many of the lower or medium flow resista~ce materials

absorption characteristics very similar to those observed with 25% perforated h

board. The lightweight high flow resistance materials have properties very sim

to those observed for 5% perforated hardboard.



8

Comparison of measured and calculated absorption coefficients of fabrics

stretched over 2 inch (5 cm) Bondacous t

r-~~-rt-'~'!H!41~--+-~'i :1

•..

'4I

Y·3 I·2

'1',f

·1' j ,I

I 1

7I:

\-:- f--f-.~

I

6

. . \/1 [-<'.

il I 1 \. .

.I ' I i

2tI h9 h fio,Y resestcrce

I I I ruI 1 1 1 1 I

0

,,!;-09

8 ( } 8O o .

l~D(}4ooo

~I )<-;-.,. / / , / 1 I /4

,: If; i ! I J I _ L J

12I J, , .i ! ,I I I !

}1I,. I \' ! I I I

'I ? I :~ ! i I iII I I i 1 ' 1 0 ' - ' I . .

I ,O·

III I I I! II \ ··0__ec _ '

1/ i I 1' " "i. ' . . l0

I, '

, II ! ! u u 1\,/ I

(}6 "

L! ! i 1) 1 I,

_j(}5

o / I i r I I

./ ! i I:i 1 !0

Hg1 f.:w re.;istcnce ,o·o· I , I I!! II I I i

_ l 1 i I! I III

JI

, .~4

I I I II, . 1(1)/ I 1

! I: I). I I

" ! .r '\ i',1

I r ' ! I 1 \V I i I

i -t . I09

.fjII

i

,, .I ! I' I./'\.. i

06'I I 1 x~

.il i I'"

o -,'--r-~ - , I

rI I I

0ttigh tow r-esrstnr-ce !

O·

01I I II III

0,

I 1 1 1 1 I I, ., ., .l'

~ 1·

~}•1/::'0'

~0-8

.60

i~0-

o

4 I i

I I

2 I I ,

1 ".-~o-L

0I r I I i

I I -], ., A· . . . '

- . '. ,i I

G : ..- +-[4..-5 ,4

&

2- _!1gh flow .res.sto-ce

1 - 3 . . = - + _ 1 1 III -ra - . ' I! III§ ~~§ ~ § § § §

--- -)(--- c al cu l eted

II : I! '1., 'I

/ ; ;I JJ _ Ii I., il Fig. i4 ~ Lightweight fabricsr-~T, . - ' , , { ' c f t ·1;,;'

;' II!I

/i

tI i ! I I" II Ii I I II i

",,,our. f<:m resst.n:e i!I J I i ~ I I I 1::1 _j__j_j

I ; I j Ii!I Tl f------i--l-· - 7 , ' , ' j - 7 , " " , !.t-- i I I

Q8800Q 800°$<8 8888><8° QOO~ M "S~ < : 5 < : 5 ' " ",,,.n~. o.~8 ..... ".iri'lf Nn~

fr'eqoer>:y. CJ s

Comparison of measured and calculated absorption coefficients of fabrics

stretched over 2 inch (5 cm) Therbloc

,

I , i

i I I 'I Ii

// ",.. , . ,

i f' !! Ii f\ ,/

!: 11

I ! i

l

I' I I I! Iii I

1 i I

I

:

! v"-...!

/'x " j ! 1 i iJ I iL ill ~

" !

o :

I I I ' II I

I I I !! Ii! I L I II'

r

: _ L I I ' .

~r.cY. cl>

I I II Ii I I I. 1 1 1 1 1 L _ L IJ! ! l.h.. \ I '; l/ 1 : -" ~ , I ', r0 /1 I, \>+~.•~,.-7!-IJ

i.e I I' '>~ii !~ I III! !'vf I

'i f ! I i.I' I

/, I I l I:/! i I ! I:'

Fig. s s

Heavy fabrics

Fig. i 6 - Lightweight fabrics

I I I II' I

I II,! I) , I! I: J,_

I /1 i - - " - " ~ 0••

I 1 .1 I, "-".:" "1/ I

Ii I j

. 1 li 1I'! I: II

-0-- I1 , I i

IQv.I f~ res.~5tor.ce I I

_ . - + - H - H t r I, j,j

S § ~ ~ ~ .~ § . § . 8 .'" "'''

~ - - - - - - o - - - - - - - measured

Fig. 17

Heavy fabrics



Some of the synthetic washable materials considered since these measurem

were made have been found to exhibit very wide variations of flow resistance

.sample to sample and in these cases a system of test measurements on each hatch

had to be instituted. For conventional fabrics, of more consistent constructi

an estimate of the behaviour of a specific type should be possible on the basis of

measurement of their surface density and flow resistance. Confirmation of t

estimates can be obtained from impedance tube measurements in cases ~here the deta

mounting of the fabric is not a major factor in its behaviour.

5. REFERENCES

1. Northwood, Grisaru and Medcof, 'Absorption of Sound by a Strip of Absorp-

tive material in a Diffuse Sound Field' ,J.A.S.A. 1959*, Vol. 31, No.5,

pp. 595 - 599.*

2. See, for example, Beranek, L. I ' Ac ou st ic M ea su re me nt s' , ( Jo hn W il ey & Sons,

Inc.), 1950, pp. 317 - 336.

*The results contained in this paper have been greatly extended in range In an unpublishedrecen tl y obtained pri vatel y from Dr. North wood.

BRH



Documents

1964 Effect of Fabric Covering on Sound Absorption