Purdue University

Purdue e-Pubs
Publications of the Ray W. Herrick Laboratories

School of Mechanical Engineering

8-1-2002

Modeling of Membrane Sound Absorbers

J Stuart Bolton
Purdue University, bolton@purdue.edu

Jinho Song

Follow this and additional works at: http://docs.lib.purdue.edu/herrick

Bolton, J Stuart and Song, Jinho, "Modeling of Membrane Sound Absorbers" (2002). Publications of the Ray W. Herrick Laboratories.
Paper 27.
http://docs.lib.purdue.edu/herrick/27

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu for
additional information.

Dearborn, Michigan, USA

August 19-21,
19-21 2002

Modeling of Membrane
Sound Absorbers
August 19th
19th, 2002

Jinho Song and J. Stuart Bolton

R W
Ray
W. Herrick
H i kL
Laboratories
b t i
Purdue University

Introduction - Background
Conventional Sound Absorbing Material

Open cell Foams/Glass fibers/Polymeric fibers

Sound energy dissipation by thermal and viscous
interaction of sound field and material fibers
Visco thermal
Visco-thermal
boundary layer

Sketch of
Fibrous Material
O ill t
Oscillatory
flow
fl

Purdue University

Herrick Laboratories

Introduction - Background
Conventional Nonfibrous Material Usages

Some environmental needs

- Healthy Surroundings/Ease of Maintenance
- Recycling
- Moisture-Resistance
Moisture Resistance
Fibrous materials can be used with impermeable membranes
to tune their performance
ms

Membrane

Fibrous Material

Fibrous Material
Fibrous material
+ Membrane
Frequency

- Conventionally, membrane does not dissipate any energy

Purdue University

Herrick Laboratories

Introduction Motivation
Recently, it has been observed that
stacked sheets of accordion-folded,
impermeable membranes (e.g.,
(e g mylar
sheets) offer substantial levels of low
frequency absorption even though
such arrays feature no obvious
p
elements.
dissipative
Alternating layers of folded mylar

- How does this sound absorption arise?

- How do you model this effect?

Purdue University

& Objectives

OBJECTIVES:
Identify the origin of the
sound absorption and
behavior capacity of such
a treatment
-

Develop models of those

processes
Use those models to optimize
the acoustical performance

HYPOTHESIS:
Dissipation results from
losses due to local flexing
of membranes stiffened
by curvature (i.e., by
folding) or tension

Herrick Laboratories

Theoretical Model Theoretical

Approach

System Model
I

II

REFLECTED WAVE

TRANSMITTED WAVE

INCIDENT
WAVE
TENSIONED MEMBRANE

Assumed Solutions (Modal Sums)

Region I : PI (r , z , t ) e

jkz

Bn J o (k rn r )e

jk z n z

Boundary Conditions
Continuity of Velocity :

Region II : PII (r , z , t ) Cn J o (k rn r )e
N

Membrane
Displacement : y ( r , t ) An J o (k 0n r )
N

jk z n z

PI
j o z
1

z 0

y
t

PII
j o z
1

z 0

y
t

Membrane Equation of Motion :

y
2

k 2f

P PII
y I
T

- Dissipation introduced by modeling T as complex : T To (1 j )

Purdue University

( cf

kf

cf

( : Loss Factor)

Herrick Laboratories

Theoretical Model
Solution Method
Apply two velocity continuity
boundary
bou
da y co
conditions
dto sa
and
d membrane
e ba e
equation of motion on a point-bypoint basis across the membrane

Solution Method
System Response
(Membrane Displacement)
Membrane Displacement

Particle
Position
II

r
Frequency[Hz]

Radius[m]

Membrane Displacement (Top View)

Number
of Point
at which
B.C.s
applied

Purdue University

A1

...

AN

B0

...

BN

C 0

...

C N

forcing

Vector

Frequ
uency[Hz]

Coefficien

Matrix

1600
1400
1200
1000
800
600
400
200
-0.04 -0.03 -0.02 -0.01

0.01

0.02

0.03

0.04

Radius[m]

Herrick Laboratories

Model Verification

Power Comparison

Power Comparison
Compare power calculated by using Acoustical Solution with power
calculated using Membrane
Membrane-based
based solution
- Acoustical Solution
a

1
WI Re{ PI (u I *)(2r )dr}
2
0

Power Dissipated :

1
WII Re{ PII (u II *)(2r )dr}
2
0

Wd ,a WI WII

if model

- Membrane
M b
B
Based
dS
Solution
l ti
a

Wd ,m

Should be equal

1
Re{ ( PI u I* PII u II* )(2r )dr}
2
0

works properly

1
Re{ T ( 2 y k 2f y )(iy )* (2r )dr}
2
0

Losses introduced by using

Purdue University

T To (1 j )

Herrick Laboratories

Model Verification
-7

x 10

Using 4 modes
Acoustical Solution
Membrane Based Solution

Power Comparison
Using 4 membrane modes

-7

x 10

Acoustical Solution
Membrane Based Solution

Using 30 modes

10

10

4
-7

x 10

Using 10 modes
3
Acoustical Solution
Membrane Based Solution

1
5

10

10

Using 30 membrane modes

2

10

10

Usingg 10 membrane modes

Purdue University

Herrick Laboratories

Model Verification

Velocity Measurement

Power
Amplifier
Microphone

PreAmplifier

Signal
Analyzer
Amplifier

Membrane

Laser Sensor
Sound Source

Purdue University

Fi i Backing
Finite
B ki

Herrick Laboratories

Model Verification

Vibrational Modes

Theory

Experiment
0.5

0
0.05

4
2

0
0.05

0.05

1st

Phase

|v/p|/|v/p|max

|v/p|/|v/p|max

Absolute velocity of membrane - Experiment

0.05

y -0.05 -0.05 x

y -0.05 -0.05 x
Phase
0.05

Magnitude
0.05

-0.05
-0.05

0
x

0.05

-0.05
-0.05

0
x

1
0.5

0
0.05

2nd

Phase

|v/p|/|v/p|max

|v/p|/|v/p|max

Absolute velocity of membrane - Experiment

0.05

2
0

-2
0.05

Phase
0.05

Magnitude

Purdue University

y -0.05 -0.05 x

0.05

0
x

0.05

y -0.05 -0.05 x

-0.05
-0.05

0.05

0.05

-0.05
-0.05

0
x

0.05

Herrick Laboratories

Model Verification
Power
Amplifier

Experimental Set-up

PreAmplifier

Signal
Analyzer

Microphone
Anechoic Termination

Sound
Source
Sou
ce
Test Sample

B&K Pulse
S t
System

Speaker Amplifier
Microphones

Computer
B&K Standing
Wave Tube

Purdue University

Herrick Laboratories

Model Verification

Experimental Results

Sound Pressure from

microphones : P1 to P4

Acoustic Impedance:
Z=(1+R)/(1-R)
Transmission Loss:

Normal Incident Pressure

Transmission Coeff. &
Reflection Coeff.: T, R

TL=20 log10 (1/|T|)

Sound Absorption:
=1-|R|2

a quick and convenient method

f determining
for
d t
i i th
the ffundamental
d
t l
acoustical properties

Purdue University

Sound Dissipation:
d =1-|R|
1 |R|2 -|T|
|T|2

Herrick Laboratories

Model Verification

Model Optimization

TL
Given experimental
results as input, Find
appropriate material
properties (To , s , )
I

II
Membrane
Surface
Density

Membrane
Tension
T=To(1+i)

Tensioned Membrane

T 82 Pa

0 .0040

s 0 .0870 kg

m2

Note : Most absorption results from transmission through membrane in anechoic

termination case

Purdue University

Herrick Laboratories

Absorption Mechanism

Frequencies of Peak Absorption

Membrane with finite-depth backing space

Imaginary part of Impedance

Absorption Coefficient

Many
y resonances
because of
membrane dynamics

Z n Z n , Membrane Z n , Backing
Frequency [Hz]

T 75 Pa

0.0063

s 0.19 kg

Frequency [Hz]

m2

l 0.1192m

Absorption peaks when Im{ Zn } = 0

Significant sound absorption in narrow frequency regions produced by

p
in the membrane
dissipation

Purdue University

Herrick Laboratories

Absorption Mechanism

Frequencies of Peak Absorption

Limp Membrane with finite
depth backing space
(Conventional Picture)

Membranes with various Tensions

Absorption Coefficient

20
15

T 0 Pa

To=25
To=75
To=200

0.9

10

Im (Z)

0.8
5

0.7

0.6

-5

0.5

-10
Duct only
Membrane Only
Sum
Im(Z)=0

-15
-20

200

400

600

800
1000
Frequency[Hz]

1200

0.4
1400

1600

0.3
0.2

Absorption
n Coefficient

Theory

No loss mechanism
from the membrane

0.9
0.8
0.7

0.1
0

Only dissipation
comes from wall losses

0.6
0.5

200

400

600

800
1000
Frequency[Hz]

1200

1400

1600

Different
Diff
t kinds
ki d off soundd absorption
b
ti
characteristics with various tension values

0.4
0.3
0.2
0.1
0

200

400

600

800
1000
Frequency[Hz]

Purdue University

1200

1400

1600

0.0063

s 0.19 kg

m2

l 0.1192m

Herrick Laboratories

Conclusions & Future Work

Theoretical models for the various membrane systems were
developed which can reproduce the acoustic behavior of stiffened
developed,
membrane systems
A sound absorption mechanism for a tensioned membrane was
suggested and verified by using the relation between the acoustic
impedance and sound absorption
The effects of various parameters in the sound absorption
performance were presented, which can provide guidelines for
designing a membrane system to enhance its sound dissipation
Effect
ec o
of membrane
e b a e pe
permeability,
eab y, s
stiffening
e g by cu
curvature
auea
and
d use
of light weight dissipative material in backing space will be
considered in future

Purdue University

Herrick Laboratories