C/e

AEDC-TR-77-61

AERODYNAMICCHARACTERISTICSOF PERFORATED
WALLS FOR TRANSONICWIND TUNNELS

PROPULSION WIND TUNNEL FACILITY

ARNOLD ENGINEERING DEVELOPMENT CENTER
AIR FORCE SYSTEMS COMMAND
ARNOLD AIR FORCE STATION, TENNESSEE 37389

June 1977

Final Report for Period January 1,-1975 - ,lanuary 1, 1977

Approved for public release; distribution unlimited.

Prepared for
DIRECTORATE OF TECHNOLOGY
ARNOLD ENGINEERING DEVELOPMENT CENTER
AIR FORCE SYSTEMS COMMAND
ARNOLD AIR FORCE STATION, TENNESSEE 37389

NOTICES

When I). S. Government drawings specifications, or other data are used for
any purpose other than a definitely related Government procurement
operation, the Government thereby incurs no responsibility nor any
obligation whatsoever, and the fact that the Government may have
formulated, furnished, or in any way supplied the said drawings,
specifications, or other data, is not to be regarded by implication or
otherwise, or in any manner licensing the holder or any Other person or
corporation, or conveying any rights or permission to manufacture, use, or
sell any patented invention t h a t m a y in any w a y b e related thereto.
Qualified Users may obtain copies of t h i s report from the Defense
Documentation Center.
References to named commercial products in this report are not to be
considered in any sense as an endorsement of the product by the United
States Air Force or the Government.

This report has been reviewed by the Information Office (OI) and is releasable
to the National Technical Information Service (NTIS). At NTIS, it will be
available to the general public, including foreign nations.

APPROVAL STATEMENT

This technical report has been reviewed and is approved for publication.
FOR THE COMMANDER

ALEXANDER F. MONEY
Research and Development
Division
Directorate of Technology

"

ROBERT O. DIETZ
Director of Technology

UNCLASSIFIED.
READ INSTRUCTIONS
B E F O R E COMPLETING F O R M

R E P O R T D O C U M E N T A T I O N PAGE
1. R f ~ P O R T N U M B E R

2 GOVT ACCESSION NO

3. R E C I P I E N T ' $

CATALOG

NUMBER

AEDC-'YR-77-61
4.

TITLE

(and 5ubtIIlv)

$.

AERODYNAMIC C H A R A C T E R I S T I C S O F P E R F O R A T E D
W A L L S FOR TRANSONIC WIND T U N N E L S
7.

AUTHOR(l)

Type

OF REPORT & PERIOD COVERED

Final Report - January 1, 1975January 1, 1977

6:

PERFORMING ORG, REPORT NUMBER

8.

CONTRACT

OR GRANT NUMBER(a )

J. L. jdc'ocks, ARO, Inc.

9.

PERFORMING ORGANIZATqON NAME AND ADDRESS

10. P R O G R A M E L E M E N T . P R O J E C T .
A R E A & WORK U N I T N U M B E R S

Arnold Engineering Development Center (DY)

Air Force Systems C o m m a n d
Arnold Air Force Station~ Tennessee 37389
I ].

CONTROLI-ING

Program Element 65807F

OFFICE NAME AND ADDRESS

12. R E P O R T D A T E

~krn01drEngineering Development Center (DYFS)

AirlForce Systems C o m m a n d
~,rn01clAi~" Force Stetion~ Tennessee 37389

1~. N U M E I E R O F P A G E S

14. M O N I T O R I N G A G E N C Y N A M E & A D D R E S S ( I / ~J/ferenf t r o m C o n f r o ~ | l n l ~ O / l i c e )

IS.

TASK

June 1977
75'
SECURITy

C L A S S . ( o / t h l l report)

..

UNCLASSIFIED
15a.
16. D I S T R I B U T I O N

STATEMENT

OECLASS~FIC ATION ' DOWNGRADING

SCHEDULE

N/A

( o f fhlll Reporl)

Approved for public r e l e a s e , distribution unlimited.

17.

DISTRIBUTION

S T A T E M E N . T ( O I fhe mbatralct entered In B l o c k 20.

tS.

SUPPLEMENTARy

II

d l f f e r e n f from Report)

NOTES

Available in DDC
19.

K E Y WORDS ( C o n t i n u e on reverse e l d e i f n e c e e l = r y

and identify

by

bJock number)

transonic wind tunnels

perforation
walls
aerodynamic characteristics
20.

ABSTRACT

(CoOt|hue on reverlle oide I I neceseory e n d I d e n t i t y by b l o c k n u m b e r l

A combined experimental-theoretical approach is developed for determination of the

crossflow characteristics of ventilated walls. The technique is based on m e a s u r e m e n t
Of static p r e s s u r e s at the boundaries of a two-dimensional flow, calculation of the interior flow properties with Murman-Cole type finite difference techniques, and Calculation of the boundary-layer development on and m a s s flux distribution through the ventilated wall using Patankar-Spalding type solution techniques. Results verify the validity
FORM

D D , JAN ?3 1473

E D t T I O N O F 1 N O V 6S IS O B S O L E T E

UNCLASSIFIED

UNCLASSIFIED
ZO. ABS'rPA CT (Continued)
of assuming proportioliallty between local p r e s s u r e and flow angle as a boundary condition for wall interference e s t i m a t e s . However, it is also shown that the thickness of
the boundary layer represents a dominant influence on the perforated wall c h a r a c t e r i s tics, with beth the slope and intercept of the characteristic being dependent on the
boundary layer. The effects on the wall characteristics of suppressing the edgetone
noise by use of screen overlays or splitter plates within discrete holes are documeated.
,

AF$C
Ae~ld

AFI

"re~

UNCLASSIFIED

_+

",

!~

A E D C - T R-77-61

~i

~-

.:

PREFACE

The work reported herein was conducted by the Arnold Engineering

Development Center (AEDC), Air Force Systems Command (AFSC), under
Program Element 65807F.

The results were obtained by ARO, Inc., AEDC

Division (a Sverdrup Corporation Company), operating contractor for

the AEDC, AFSC, Arnold Air Force Station, Tennessee, under ARO Project
Numbers P32A-29A, P32A-COA, and P32A-J5A.
J. L. Jacocks, ARO, Inc.

The author of this report was

The manuscript (ARO Control No. ARO-PWT-TR-

77-30) was submitted for publicatlon on April 25, 1977.

,'AEDc-TR-77-61
CONTENTS

I .0

INTRODUCTION . . . . . . . . .

2.0

DEVELOPMENT OF THEORY

~5

3.0

4.0

-i

5.0
,J
/

6.0

2. I

Potential Flow

......

2.2

W a l l Boundary Layer

. . . . .

/6 ~

14

APPARATUS
3.1

Aerodynamic*Wind

Tunnel

(IT)

. .

3.2

P r e s s u r e D i s t u r b a n c e Generators

3.3

P e r f o r a t e d W a l l Geometry

3.4

Instrumentation

. .

17
.

18

.....

2i

. . . . . . . .

. . . . . .

,..

21

PROCEDURE
4.1

Test Procedure

. . . . . . . . .

4.2

P r e c i s i o n of M e a s u r e m e n t s

.......

. . . . . . .

26

. . . . . .

27

RESULTS AND DISCUSSION

5.1

Assessment of Accuracy

....

5.2

P e r f o r a t e d W a l l Characterist4cs

5.3

Effect of Noise Suppression Devices

. . . .

....

. .

. . . . .

CONCLUDING REMARKS . . . . . . . . . . .

30
.

. . . .

REFERENCES . . . . . . . . . . . . .

....

27

43
48

.:

50

ILLUSTRATIONS
,

Figure
.!

I.

Schematic of Physical Model and CorrespondingMath

Model

2. C o m p a r i s o n of the Inverse S m a l l - P e r t u r b a t i o n

Approach

~,

w i t h a n Exact Solution for F 1 0 w over a RightC i r c u l a r Cylinder

. . . . . . . . . . . .

3. C o m p a r i s o n of the Inverse S m a l l - P e r t u r b a t i o n

Approach

w i t h Exact Solutions for F l o w T h r o u g h O b l i q u e S h o c k s

4. I l l u s t r a t i o n

of the N o n l i n e a r R e l a t i o n s h i p

F l o w A n g l e and N o r m a l i z e d ~ W a l l

11

13

Between

Mass Flux . . . . . .

16

AE DC-TR-77-61

Figure
5. Characteristics of the Pressure Disturbance
Generators . . . . . . . . . . . . . . . .

19

6. Perforated Wall Geometry . . . . . . . . . .

22

7. Comparison of Calculated Flow Angle Distributions

with Laser Velocimetry Measurement . . . . . .

Z9

8. Comparison of Calculated and Measured Boundary-Layer

Displacement Thickness . . . . . . . . . . . .

31

9. Representative Variation of Pressure Coefficient, Flow

Inclination, Wall Mass Flux, and Boundary-Layer
Displacement Thickness with Tunnel Station . . . . . . .

32

10. Representative Locus of Pressure Coefficient and Flow

Inclination - the Wall Characteristic

.....

34

11. Variability of the Wall Characteristic with Changes

in the Imposed Pressure

Distribution

36

12. Effect of Boundary-Layer Displacement Thickness on the

Characteristic Slope . . . . . . . . . . . . .

37

13. Effect of Boundary-Layer Displacement Thickness on the

Characteristic Intercept . . . . . . . . . . . .

40

14. Comparison of the Varlable-Poroslty Wall Characteristics

for Upstream and Downstream Displacement of the
Cutoff Plate .

42

15. Effect of Noise Suppression Devices on the

Characteristics of C o n f l g u r a t l o n A

. . . . . . . .

44

16. Effect of Noise Suppression Devices on the Characteristics of Configuration D . . . . . . . . .

46

APPENDIX
A. COMPARATIVE CROSSFLOW CHARACTERISTIC DATA FOR EACH
PERFORATED WALL GEOMETRY . . . . . . . . . .

53

NOMENCLATURE . . . . . . . . . . . . . . .

74

AED C-TR-77~1

:~

1.0 INTRODUCTION

~:IAlthough t r a n s o n i c wind t u n n e l s w i t h v e n t i l a t e d
have!been i n use f o r o v e r 25 y e a r s (Ref.
aerodynamic p r o p e r t i e s

test

s e c t i o n walls~

1), an u n d e r s t a n d i n g o f t h e

o f the w a l l s has y e t to be developed.

of the wall crossflow characteristics

Knowledge:i

( p r e s s u r e - f l o w angle r e l a t i o n s h i p )

i s r e q u i r e d f o r s e v e r a l reasons, b u t an accurate, method has n o t been

available

to o b t a i n t h i s

information.

Previous. techniques

t h r o u g h 3) f o r d e t e r m i n a t i o n o f w a l l c h a r a c t e r i s t i c s
assumption o f

(Refs.

have r e l i e d

1"
on t h e

e q u i v a l e n c e between f l o w angle a t t h e w a l l and mass f l u x

through the wall.

However, Rae (as r e p o r t e d i n Ref. 4 ) d e m o n s t r a t e d

t h a t t h e b o u n d a r y - l a y e r development on the w a l l Created a n o n l i n e a r

i n t e r d e p e n d e n c e between mass f l u x and f l o w a n g l e .

,,~Direct measurement of the local static pressure and flow angle in

the vicinity of a ventilated wall can be accomplished~, as s h o w n b y
~t

Berndt (Ref. 5).

This approach is feasible for documentation o f t h e

characteristics of a given wall geometry~ but becomes inefficlent as the

number of wall configurations is increased.

~:

To bypass these difficulties , the present investigation was designed to develop a new test technique thatwould yield definitive
information on ventilated wall characteristics.

An inverse technique

was~selected wherein sufficient yet tractable static pressure measurer

ments made at the boundaries of a two-dimensional flow would a l l o w
calculation of the remaining flow variables.

The potential flow field

was calculated with the line relaxation method of Murman and Cole (Ref.
6) with the primary result being the flow angle distribution in the
vicinity of a ventilated wall.

The measured pressures and inferred flow

angles were then used to calculate the boundary-layer developmen ~ on and

mass-flux distribution through the Ventilated wall.

AEDC-TR-77-6|
The theoretical approach is described in Section 2.0, with the
experimental apparatus and procedure being described in Sections 3.0 and
4.0.

Section 5.0 presents the results, including independent measure-

ments made to verify the accuracy of the theoretical calculations.

2.0 DEVELOPMENT OF THEORY

2.1 POTENTIAL FLOW
The basic hypothesis of the present work is presented pictorially
in Fig. I.

It is assumed that a two-dimensional transonic flow fle;id /!

can be established within a region bounded by a contoured solid (bottom)

wall, solid plane sidewalls, and a ventilated (top) wall.

It fs further

assumed that the flow can he mathematically described by small perturbations from a uniform flow with boundary conditions derived from static
pressure measurements around a control volume.

The llne relaxation

technique of Murman and Cole (Ref. 6) provides the calculatlonal tool to

solve the resulting boundary-value problem.

The result is t h e dlstrl-

bution of flow angle in the vicinity of the ventilated wall.

The small perturbation, two-dlmenslonal, potential flow equation ~

descriptive of internal transonic flow is given (Ref. 7) by

[1 - M2 - ( y +

1)~

~ x] ~xx + ~xx = 0

where ~x and ~y are the p e r t u r b a t i o n v e l o c i t i e s

(1)

( d iv id e d by a reference

v e l o c i t y ) p a r a l l e l and normal to the tunnel a x i s and M i s a reference

Mach number.

To the degree of approximation inherent w i t h i n Eq. (1) the

local pressure coefflclent, Cp, and flow angle, 8, are given b y

Cp = -2 ~x

0 ---y

(2)

AE DC-TR-77-61

- ~ ,~,

Cp dx

I
x

,',., C I
"Pl

BOUNDARY-VALUE PROBLEM

I
I
I Cx" Cp
I
I

~,j~= Cp dx
PERFORATED WALL
FLOW
~__

37. 5 in.
2 5 in..1. I
13 in. ------~
~
I
SOLID CONTOURED WALL

Figure 1. Schematic of physical model and corresponding math model.

12 in.

AE DC-TR-77-61

Measurements of static pressureat the boundaries of a control volu~ne~as,

indicated in Fig. I can be used to specify sufficient boundary condly ,A~;,
tions of mixed type to obtain a solution of Eq. (I). Specifica!l~,~

~;i~ii~

natural boundary conditions at the upstream and downstream planes~of~the,

control volume are given by
4x (o,y) ffi-0.5 Cp (o,y)

(3)
Ox (L,y) = -0.5 C

(L,y)
.

c~

Boundary conditions of the Diri~chlet type are obtained b y Integratibn of

the measured pressures at the top and bottom boundaries by
"~'~ " ,

CI]{D

X
,

O(x,o)

-0.5 j" Cp (x,o)dx + ~o

0

(4)

0(x,h) = -0.5

Cp (x,h)dx + 0i

[;{;

i]!I; ~CI]:

O
m

where two constants of integration (0o and 01) rp~maln to be evaluated.

A physical interpretation

of t h e s i g n i f i c a n c e of t h e s e c o n s t a n t s i s .that.

their difference represents the average flow inclination at the upstream

boundary of the test section, or

m,-~

h
01 - 00 =

0y (o,y)dy

(5)

,O

Since the magnitude o f the potential is of no consequence, 0o was arbitrarily set to zero without loss of generality.

The remaining constant

of integration cannot be evaluated" from the Static pressure measurements

alone but requires an additional item of information relating to flow
inclination or geometry.

The selected PrOcedure for defining 01 was based on the concePt.of i:'
boundary-layer displacement thickness development on the bottom wall ,,f ~
with, of course, knowledge of the wall geometry.

For each flow condl-

tion, a preliminary solution of the flow field compatible with the

AEDC-TR-77-61

measured pressures was numerically obtained assuming ~I = O, say ~(x,y).

The elevation, Ys' of the streamline through the coordinate origin was
then computed from numerical differentiation and integration of this
solution in the form of
X

Ys(X) = f ~y(X,o)dx

(6)

0
7

In general, the resulting streamline did not agree with the effective
wall geometry because the average flow inclination at the test section
_I

r! L

entrance cannot be assumed a priori.

However, superposition of a uni-

form crossflow on the ~ solution does allow matching of both the effective bottom wall geometry and the boundary conditions derived from
static pressure measurements.

The required crossflow was determined

explicitly by forcing agreement between the computed streamline (with

rotation) and the effective wall elevation at two points, x = 0 and
x = x I. T h e resulting rotated streamline generally agreed with the
effective geometry for all values of x, with allowance for the expected
i

accuracy level because of the small perturbation assumption.

Given the

displacement thickness, ~I' and the wall geometry, yg, ~I was computed
from

~I = h___
x I [61(xI ) - 61(o ) + yg(Xl)

_ Ys(Xl)]

(7)

The displacement thickness at the test section entrance, ~i(o), was

measured and found to be weakly dependent on Mach number and, for
convenience, a nominal value of 0.07 in. was used for all calculations.
During the first portion of experiments, the displacement thickness was
measured at the test section exit for each flow condition and used in
Eq. (7), whereas for the later experiments, x I was fixed at the middle
of the test section with 61 = 0.04 in., which was representative of
measured values.

AE DC-TR-77-61

Superposition of the uniform crossflow does not require extenslve

recomputation of the flow field because Eq. (I) is linear in this r@Bpect,
and the final solution is given by

~(x,y)

= ~(x,y)

(8)
i

!
The numerical technique selected to solve for $ was that of Murman

and Cole (Ref. 6).

A computer code was written specifically for a

finite control volume with boundary conditions applicable to the present

problem.

The flnite-dlfference representation of Eq. (I) was coded With

a mesh of uniform spacing in each of the coordinate directions.

The

difference operator was varied among elliptic, parabolic, hyperbolic,

and a special shock-polnt differencing (Ref. 8) according to the local
Mach number and velocity gradient.

Boundary conditions were applied at

the boundaries of the mesh.

"

Several idealized numerical experiments were conducted to verify

the accuracy of the coding and the practicality of the solution.

These

studies included verification of stability, ability to rapidly converge

with imbedded shocks, self-conslstency between inverse and direct solutions, and comparisons with other exact and approximate solutlons.

One of the most illustrative examples of the accuracy Of the smallperturbation approach is provided by comparison with an exact solution
for flow over a right-clrcular cylinder (Ref. 9).

As indicated i~ Fig.

2, a control volume was selected in the vicinity o f a cylinder immersed

in uniform incompressible potential flow.

Both the exact pressure

coefficient distribution on the streamline representing the bottom wall

and the distributlon on the other three boundaries were used to solve
for the inclusive flow field.

The calculated flow angles at the upper

and lower boundaries are compared wlth the exact solution in Fig. 2.
Two salient points of this comparison are that the inverse Solution
appears to be nominally ten percent in error, a consequence of assuming

I0

- m

EXACT
--- - ,SMALL
INVERSE

SOLUTION

PE R T U R B A T I O N

I
I

SOLUTION

0.2

"

C
0

~ m .

lY =4

I
I
~ y = 2

___--------

=2

0.!

-4
m

0
tk=.

ui
~

-2

-I

0
LI

""

'

,"

-0.ii i

-0.2L
3D

Figure 2. Comparison of the inverse small-pe~urbati'on approach with

an exact solution for flow over a right-circular cylinder.

,,j

AE DC-T R-77-61

small perturbations,
calculational

and that, beglnning with symmetric inputs,

scheme yields symmetric results.

M = 0.1 was assumed for these computations,

characteristic

It should be noted that

but the nonlinear transonic

of Eq. (I) was retained.

It can be argued that the use of experimentally

pressure coefficients

as boundary conditions

degree of approximation

implicit in Eq. (I).

determined .(exact)

is inconsistent with the

This hypothesis was

examined by using as boundary conditions pressure coefficients

by subtracting
butions.

the

the higher-order

velocity

computed

terms from the exact distri-

The resulting calculated flow angles were compared wit6 the

exact solution for flow over a cylinder and showed errors up to three
times that indicated in Fig. 2.

It was concluded that the inverse

solution approach in combination with tangential rather than normal

boundary conditions were in concert and would yield results c o m p a t i b l e
with the real flow.

i.

A second class of illustrative

T_

examples of the usefulness of the

/I

technique is based on uniform flow through an oblique planar shock.

Fig. 3, two examples are given that were constructed

from the exact

solution to the Euler equations.

flows at Mach

In these examples,

In

numbers of 1.1 and 1.2 were turned by a shock at angles of 75 and 60

deg, respectively.

The major difference between the two'is that the

stronger shock results in subsonic flow, whereas

supersonic

throughout.

differencing

For the supersonic

is consistently utilized;

the other flow is

fl0w, backward or upstream

hence errors are accumulated

within the field that are incompatible with the exact, imposed boundary
conditions,

and a reflection back into the field occurs at the boundaries.

For th& subsonic downstream flow, central differencing

permitting communication

of the imposed boundary conditions

field with correspondingly

not unexpected,

is utilized,

more accurate results.

into the

These findings were

since it was known (Ref. 6) that the hyperbolic differ ~'

encing operator was only accurate to the first-order',

accuracy was characteristic

and second-order

of the elliptic differencing operator.

12

AE DC-T R-77-61

EXACT

SOLUTION

SMALL
PERTURBATION
INVERSE SOLUTION

"

0 03 I
0.02

M2>I

001
o

0.03
0.02
M2<I

M I =1.1
0.01

MI
,

,i

M2

;r

Figure 3. Comparison of the inverse small-perturbation approach with

exact solutions for flow through oblique shocks.

13

AEDC-TR-77-61

From these numerical experiments,

it was concluded that the calcu-

lated flow angles in the vicinity of a ventilated wall would be of,

usable accuracy for entirely subcritlcal

:;

flows, but that the adequ@cy of

the procedure would require close reassessment

large region of supercritical

for conditions with ,a

flow.

2.2 W A L L B O U N D A R Y L A Y E R

As discussed by Goethert
boundary-layer
correlations

(displacement)

(Ref. 6) and Lukasiewicz

(Ref. I0), t h e

thickness is an important parameter i f

of ventilated wall characteristics

are attempted.

The

;,:

presence of a boundary layer also results in a nonlinear relationship

between flow inclination and wall mass flux as illustrated by Rae (Ref.
4).

This effect can be readily appreciated by considering

continuity equation for two-dimenslonal

thickness,

flow written as

d61
6 - 61 d
d--x-- ; ~ % ~ dx ( p = u )
where 61 is the displacement

the Integral

'

,,~.

(9)

+ O - I = 0

is a constant thickness

inclusive of the boundary layer where the streamwlse mass flux p u

is evaluated,

8 the flow inclination at 6 , and I is the mass flux

through the wall normalized by p u .

adopted for e and I is such

constant displacement

Note that the sign c o n v e n t i o n

that suction (outflow) is positive.

the effects of pressure gradients,

Ignoring

equality of e and I would require

thickness, which generally does not occur. :

The conventional a p p r o a c h t o

the calculation of boundary-layer.

development over a porous wall requires specification of the wal ! mass~

flux as an independent parameter.

Since flow angle outside of the,

boundary layer was known for the present approach,

, :

a method for using , .

Eq. (9) to solve for the mass flux was developed by G. H. Saunders of
ARO, Inc.

The two-dimenslonal,

code of Whitfleld

turbulent,

boundary-layer

prediction

(Ref. 11), based on that of Patankar and Spalding

14

~
:

AEDC.TR-77.61

'r

(Ref~.Jr12i;~,was modified to incorporate the integral continuity equation

" "

to relite'wall mass flux to flow angle. ~ For each streamwlse i n c r e m e n t . '

Withih~the boundary-layer calculation,

"

the wall mass flux was'~itera -

tively s~ecified until the calculated flow angl e from Eq.. (9) matched
the potential flow results.

The end result of these"calc~lations Was ~

',~"]

--:

the wall mass-flux distribution, boundary-layer development, and the

distribution of other parameters such as skin friction and boundary-. ''~'.',
layer shape factor.

.,

The results must be.interpreted with caution

becaus'e , in addition to the usual boundary-layer approximations,

it was

implf~itly assumed that the finite-size perforations.could be'repre sented'-as an equivalent, homogeneous porous wall and"that the no-slip ""
condition was valid (in spite of having inclined holes).

~:~Another functional relationship between mass flux and flow a n g l e

can be derived (Ref. 4) from a combination of the integral continuity
and momentum equations and is given by

/ .

:
....

I
{
dH
Cf
% = +-----H
I
I0 + ~2 d~x ~ H 2 "

dC [61(I +.H) + $ (I - M2)]

+71 ~dx

(io)

'

The parameters of significance are the shape factor, H, and the skin L
friction coefficient, Cf.

For large suction, the'shape factor approaches

unity and the skln-frlctlon coefficient is of comparable magnitude t0~

the mass flux which results in X ~ 0.

Conversely,

for large 51owing;

the Skin friction approaches zero and the shape factor becomes large
.such that dX/de->0. At moderate suction orl blowing rates with a repre L
sentative shape factor of H = 1.5 at the Mach numbers of interest, Eq.
(10) indicates X ~ 0/2.5 would be appropriate.

Some representative

comparisons of the relationship between mass flux and flow angle as

. "" '

computed from Whitfield's code are presented in Figl 4.

: i ,

These results

are i n ~ e e o r d with Eq. (10) and clearly illustrate the'noniinearity.

between flow inclination and normalized wall mass fiux.~

15

mh

~i

"~
5

3~
m

.BI

0=0

D ~

~.

0
04

.o

0
PO

r" o

o
r0

.o

4~

-m

m
3

IrO

.o

=J.=

=11

I0

CO

om
~>z
r-~
c
c~
[m
-

Z~

ro
3>-0
-Im

'

'

NORMALIZED WALL MASS FLUX,

9!

/)
(=
C)
.-I

19"LL-I:J l-OO

q>

II

>,,

9V

0
0
04

AEDC-TR'77-61

3.0 APPARATUS
3.1 AERODYNAMIC WIND TUNNEL (1T)
The experimentswere conducted in the Aerodynamic Wind Tunnel (IT),
which is a continuous-flow transonic tunnel with atmospheric intake and
exhaust.. The test section is of square cross section, nominally 12 by
12 in. and 37.5 in. longm and is enclosed within a plenum chamber.
.

. Stagnation pressure is fixed at approximately 2,850 psfa wlth slight

v~rlations attributable to tunnel resistance and ambient pressure.

To

.%

prevent water vapor condensation in the test section,.stagnation temperature is normally varied within the range of 150 to 190"F as required.

.z

Supersonlc flows are established in the blach number range of 1.1 to

i

1.5 with a two-dimensionalflexiblenozzle.

Subsonic and transonic

flows~'are obtained with a sonic nozzle contour in conjunction with

adjustments in the tunnel backpressure and with plenum suction through
t~J

an auxillary evacuation system.

The tunnel test section configuration consisted of solid, planar

sldewalls with several contoured, solid bottom walis to generate dlff~rlng pressure distributions within the testreglon.

The bottom wall

(floor) was attached to the nozzle exit with a flexure, and the downstream end'of the wall was suported by a remotely controllable Jackscrew.

Using the Jack-screw, variations in wall angle, 8w, of 1 deg

reiatlve to the'tunnel centerllne were set (convergence is considered

positive).

The ventilated wall specimens were installed at the top

(ceiling) of the test section, parallel to the tunnel centerline.

As applied

to

thetunnel geometry, the coordinate system of'Fig. I

is referenced to the nozzle exit, bottom wall, with t h e x-axis parallel

to the tunnel centerllne.
units of inches, with a x i a l

All length dlmensions where cited are in

location usually phrased as tunnel station,

that is, the distance downstream from the nozzle exit.

17

A E D C - T R-77-61

3.2 PRESSURE DISTURBANCE GENERATORS

Lifting models in conventional ventilated wind tunnels generate
far-field disturbances which, after interacting with the tunnel boundary
as a whole, can be treated as simple pressure distributions imposed on
the ventilated walls.

To simulate the resulting pressure distributions

in a manner amenable to analysis, several two-dimensional bottom-wall

bumps were fabricated with rather arbitrary contours as disturbance
generators.

The profiles of the contours are presented in Fig. 5a.

Each bump (nominally 12 in. in length) including a flat plate (Contour

F) was installed with the leading edge at tunnel station 13.

Upstream

and downstream of the contours, flat-plate extensions were used.

Unless

indicated otherwise, data presented were obtained using the thickest

bump (Contour A) with the bottom wall parallel to the tunnel centerline.

The dashed lines in Fig. 5a represent the boundary-layer displacement thickness development over the various contours.

The calculations

were based on the potential flow solution, with integration of the flow
inclination at the bottom wall compatible with the measured pressures at
M = 0~8.

The resulting effective aerodynamic contours changed as

functions of Mach number, wall angle, and ventilated wall geometry.

Boundary-layer separation evidently occurred on all of the bottom wall
contours (except the flat plate) so that the only viable method of
solving for the interior potential flow field was the inverse technique.

An illustration of the types of pressure distributions achieved at

the ventilated wall with the various disturbance generators is given in
Fig. 5b.

Again, it should be noted that variations in Mach number,

bottom wall angle, or ventilatedwall geometry significantly affected

the resulting pressure distributions.

]8

-i

AEOC-TR-77-61

INVISCID
S..TREAMLINE 7

FLOW

CONTOUR A

" " :.'"

1
CONTOUR

.~'~. ,.~ .'-.: .'.::7".~i,:~.; '

'.~ ,~i...~;:

~: "/~

.=:: :,

CONTOUR

CONTOUR

CONTOUR

- "../::";~.'I.

,';~

~.",

a. Geometric and effective contours

Figure 5. Characteristicsof the pressuredisturbance generators.

19

AEDC-TR-77-6t

0.2

I
I

I I~"1

CONTOUR A
(" j

~0.1
Z
ILl

LI.
LI.
ILl
0

LIJ
n*
U)
(n
hi
G:
a.

-0. I

-o.2 I
0

I0
15
20
.25
TUNNEL STATION, inches

b. Pressure distributions at the ventilated wall

Figure 5. Concluded.

20

30

35

AEDC-TR~77-61
i

3.3

PERFORATED

WALL GEOMETRY

'

"

"

Four basic wall configurations were tested, denoted A through D,

With each being based on the 60-deg i n c l i n e d h o l e p e r f o r a t e d w a l l d e i

vel~ped at AEDC (Ref. I).

i,

in Fig. 6.

Pertinent dimensions of each wail are give n

Configuration D (Fig~ 6c) was a varlable-poroslty wail

conistlng" of two match-drilled plates with the airside plate held

stationary and the backside or cutoff plate translated streamwise to
achieve variations in porosity.

For convenience, upstream movement of

theicutoff plate for decreasing porosity is labeled positive porosity

and:downstream movement negative.

Perforated walls in transonic wind tunnels generate noise, termed

edg~tones, that is thought to degrade the quality of model test data.
Two methods of suppressing the edgetoneshave been developed (and configurations A and D weretested with each).

One modification consisted

of inserting a splitter Plate , SPL (Ref. 13) in each hole, longitudinally bisecting the hole, w l t h t h e splitter-plate dimensions being 0.012
in.lwide and 0.063 in. deep.

The second modification Consisted of a

. screen attached to the airside plate surface; in this instance the

screen was of 40 by 60 mesh with 0.006 in. wire diameter.
]

I
I
I

i To distinguish among the differing wall geometries the configuration code is followed by wall porosity and,~if appropriate, either SPL
or SCR to denote the noise suppression'device present (for example, D1.0!SCR).

Data are presented for 20 distinct ventilated w a l l

geometries.

3.4 | N S T R U M E N T A T | O N

Primary tunnel parameters and the respective instrumentation used

to sense and measurethe parameters included: plenum pressure measured
by a servo-driven precision mercury manometer; stagnation and diffuser
e x i t pressures measured wit~ strain-gage transducers referenced to the

2]

AEDC-TR -77-61

16.1 DEG

~
,s.9

1.752--

oEG_~...<~

.-

0 0 0 0
O0
o
3.034

O0

O0
o
O0
o
0
O0
o
0
0 o
--0
O0
FLOW
"-

0.125D

_J_
0.125
a. Configuration A
Figure 6. Perforated wall geometry.

22

AEDC-TR-77-61

0.713

(2)

0
FLOW
"

0.213 D

~r-

o.125
b.~ ConfigurationB
Figure 6. Continued.

23

AEDC-TR-77-61

2 0 . 0 6 DEG
~--I. 240---.,]
20.06

-"---- 0

0
0

0
0

FLOW

0.166 D

FIXED POROSITY

"

-70.125 TYP

r = I 0 . 0

PLATE
MOTION

VARIABLE POROSITY (UPSTREAM)

VARIABLE POROSITY (DOWNSTREAM)

c. ConfigurationC and D
Figure 6. Concluded.

24

AEDC-TR-77-61

plenum pressure; and stagnation temperature measured with an ironconstantan thermocouple.

Other pressure measurements were accomplished

with five 48-port Scanivalves@using strain-gage transducers referenced

to plenum pressure.
system.

These data were recorded with an on-llne computer

Raw data were recorded on punched paper tape, and results were

tabulated in engineering units to aid in conducting the tests.

The

paper tape was subsequently processed to obtain the results presented

herein.

Flow angularity measurements were obtained in the proximity of

selected ventilated walls using laser velocimetry (LV) with a system
described in Ref. 15.

This system is a two-component, dual-scatter,

moving frlnge-type system operated in the off-axis, backscatter co1lection mode.

On-line indication of the two velocity components was

available from the data processor, but all results presented were processed off-line from digital magnetic tape recordings of the raw data.

Static pressure measurements in the vicinity of the ventilated walls

were first made uslng a 0.5-1n.-diam static pipe extending from the
stagnation chamber through the nozzle and test section.

With the pipe

centerline nominally 1.25 in. below a perforated wall, it was discovered

that the measured pressures were slightly dependent on the orientation
of the orifice.

The crossflow velocity componen t induced a variation in

static pressure around the circumference of the pipe.

Further com-

plications, such as o r i f i c e - e d g e a b e r r a t i o n s and n o n r e p a l r a b l e l e a k s ,

forced abandonment of the statlc-pipe concept, although some data are
presented.

The test section sidewall was selected as the location for

the all-lmportant static pressure measurements near the ventilated

walls.

The measurements required for upstream and downstream boundary

conditions were also made at the tunnel sidewall; however, pressure

measurements for the bottom w a l l w e r e obtained on'the centerline.

25

AEDC-TR-77-61

A limited number of boundary-layer surveys was made with multipler ~

tube pitot pressure rakes.

These data were reduced to the conventional

parameters of displacement and momentum thicknesses assuming isoenergetic

flow with Prandtl number equal to one.

4.0 PROCEDURE

l;r

4.1 TEST PROCEDURE

,~,J

Tunnel test section conditions were unconventional in that there

was, in general, no region of uniform flow.

The disturbance field of

the contoured bottom Wails extended into the nozzle sothat,

glance, there is no free-streamMach number for use in

at first

Eq. (I).

How-

ever, the presence of one ventilated wall allows the introduction of a

pseudo Math number, the so-called plenum Math n,-,ber, which is simply a
Math number computed on the basis of an Isentroplc expansion from the
stagnation pressure to the plenum pressure.

It is emphasized that the

flow within the plenum chamber actually was at a very low velocity.

The

pressure differences across ventilated walls are normally small; hence,

the plenum Math number is a good approximation to the nonexistent farfield or free-stream Mach number.

Test conditions were established by adjusting the plenum suction

flow rate and the diffuser exit pressure in such a way that the desired,
Math number was set while simultaneously maintaining reasonably uniform
pressure gradients at the downstream end of the test s e c t i o n . Data were
obtained, for each configuration at Math numbers between 0.5 and 0.85;
Limited data were acquired between Math numbers of 0.9 and 1.2.

"

For some ventilated wallsp the bottom wall angle was adjusted to
achieve changes in the mean boundary-layer thickness over the ventilated
wall.

Convergence of the bottom wall forced more flow into the plenum,

thereby thinning the boundary layer.

yielded the opposite effect.

25

Divergence of the bottom wall

AEDC-TR-77-6t

4.2 PRECISION OF MEASUREMENTS

Uncertainties in the pressure measurements have been estimated at
the 95-percent confidence level considering the effects of the precision
and repeatability of the instrumentation.

For most of the pressure

measurements, three or more individual data points were acquired in

sequence and averaged to minimize the influence of low-frequency, smaliamplitude oscillations in the tunnel flow.

These effects were combined

using the Taylor series method of error propagation.

The resulting

uncertainties in the basic parameters are:

Mach number, M

0.005

Pressure coefficient, C

0.006

Wall angle, 8w

0.02

Displacement thickness, 61

0.03

The laser velocimetry data were derived from an average of approximately 1,000 samples/point, yielding negligible random error.

Bias

errors introduced within the system or from particle lag were of unknown
magnitude.

5.0 RESULTS AND DISCUSSION

5.1 ASSESSMENT OF ACCURACY
Calculation of t h e flow angularity distribution rather t h a n making
direct measurements leads to results of unknown accuracy.

A highly

complicated relationship exists between the precision of pressure coefficient measurements and the precision of the calculated flow angles.
Additional uncertainties arise from the assumptions of small perturbations and two-dimensional flow.

27

AEDC-TH-77-61

To obtain a qualitative appraisal of accuracy, independent measurements of the flow angles were made for selected ventilated walls I in.
from the walls and compared with calculations made for the same conditions.

These results are presented in Fig. 7 in the form of flow angle

as a function of tunnel station for wall configurations A6.0, BT.0, and

CI0.0.

Limited transverse surveys with the LV system indicated the flow

in the vicinity of the perforated walls was two-dimensional with no

perturbations from discrete holes detectable I in. from the wall.

The general agreement between the two independent estimates of flow

angle is considered excellent.

It is evident that the inverse technique

yields results of good accuracy and is experimentally far less demanding

thandirect measurements.

For the larger crossflow velocities (flow

angle) the comparisons in Fig. 7 indicate possible contamination of the

results by boundary-layer growth on the sidewalls since the LV data are
generally of larger magnitude.

Enforcement of continuity on a two-

dimensional basis and matching of a streamline at the bottom wall resulted in artiflcially increased outflow and decreased inflow at t h e '
ventilated wall.

To o b t a i n a n i n d i c a t i o n

of the boundary-layer

types of experiments were conducted.

The f i r s t

program accuracy,

setup consisted

two

of

replacing the ventilated wall with a solid wall and performing all
measurements and ca~culatlons as if there were mass flux through the
wall.

The resulting magnitude of the computed mass flux was approxi-

mately 10 percent of that obtained with ventilated walls of moderate

porosity.

Again, the.sidewall boundary-layer growth is suspected as the

major cause of this discrepancy.

Since the extremes of wall mass flux

decreased with decreasing wall porosity, it is evident that the'relative

accuracy of the resuAts is dependent on wall porosity.

28

AEDC-TR-77-61

0.08

LV DATA
CALCULATED

0.04

"

M = O.8

-0.04

#w'O
-0.08

12
16
20
24
TUNNEL STATION, inches

28

Figure 7. Comparison of calculated flow angle distributions with

laser velocimetry measurement.

29

32

AE DC-TR-77-61

The second experiment consisted of measuring the boundary-layer

profile at tunnel station 30 on configuration A6.0 with each of the
bottom wall contours, with wall angle and Mach number as test variables ~.
The resulting comparison of measured and calculated displacement thick"
nesses is presented in Fig. 8.

The agreement of the two was considered

excellent, and it was therefore assumed that the boundary-layer calculations were equally valid at other tunnel stations.

Boundary-layer '~

measurements were not made on any other ventilated wall geometry, which
required the assumption that the effects of changes in wall geometry=

were fully reflected in the resulting distribution of flow inclination.

Since the calculated displacement thickness evidenced signlficant

variation over the ventilated wall extent, the selected approach for ~
analysis was to obtain the average thickness directly above the bottom
wall contours.

' if

5.2 PERFORATED WALL CHARACTERISTICS

~
i/

,i

5.2.1 Defini~on of the Wall Characteris~c

As a means o f i n t r o d u c t i o n ,
figuration

A6.0 is

presented

a set of results

as a f u n c t i o n

obtained

of tunnel

with

station

;~

con-

i n Ftg~ 9~

Note that the static pressure distribution is the only experimental r~

result with flow angle, wall mass flux, and boundary-layer displacement
thickness being derived quantities.

The measured pressure dlstrlbution

was in general not sufficiently regular for the numerical computational

procedure, and a moderate amount of data smoothing was employed.
given fraction of irregularity in pressure coefficient,

For a

the resulting

irregularity in flow angle was magnified with mass flux irregularities

magnified again.

This amplification of errors introduces fluctuations

in the crossflow characteristic that have no physical meaning and should

be ignored.

As a consequence, the calculated wall mass flux was con-

sidered to be of insufficient accuracy to allow correlation of the wall

characteristics and the flow angle was used for this purpose.

30

AEDC-TR-77-61

4)

0.3
O
J:
U
C
, D

LINEOF PERFECT

(n
(/)
bJ

AGREEMENT
(
~

0 "0 O /
.

-,- 0.2

I-

~ 0.1

o/

a
ul
p.

/00.

_1
.J
j

olZo
O
0

0.1
0.2
0.3
MEASURED DISPLACEMENT THICKNESS, inches

Figure 8. Comparison of calculated and measured boundary-layer

displacement thickness.

3"1

AE DC-TR-77-61

0.05

-0.01
-0.10
0.02' I

FLOW ANGLE, 'odians

MASS FLUX

-0.02

**~

8w= 0.5
- 0.04

CONFIG

A6.O

10
15
ZO
25
TUNNEL STATION, inchos

0.2'

BI
0.1

30

Figure 9. Repro$ontativovariation of pressureoo~icient, flow

inclination, wall mass flux, and bounda~/-I~/or displacement thickness with tunnel station.

32

35

AEDC-TR-77-61

Techniques are available (Refs. 16 and 17) for wall interference

correction of model test data, but little information is available
concerning the proper boundary condition to represent the wind tunnel
walls.

The classical perforated boundary condition typically employed

for subsonic wall interference calculations may be written as

~x +

~ ~y = 0

(11)

where 8 is the Prandtl-Glauert factor and R represents an unknown porosity factor or wall permeability.

Equation (11) is a linearized

approximation to viscous flow through a porous medium where the average

velocity normal to the wall is assumed proportional to the pressure drop
through the wall, and the pressure outside the wall is assumed equal to
the free-stream pressure.
wall characteristics,

In terms of experimental data for perforated

the locus of pressure coefficient and flow angle

provides the required boundary condition in the form of

dC
= I
~
R
2 dO

(12)

Thus, for utilization of the present results within existing theoretical wall interference prediction methods, there should exist a
linear relationship between C

and e. A representative crossplot of

P
pressure coefficient as a function of flow angle is presented in Fig. 10
wherein the streamwlse path is denoted by directional arrows and the
leading- (LE) and trailing-edge (TE) tunnel stations of the bottom wall
contour were as indicated.

The characteristics were obviously not

linear at the upstream portion of the wall, although reasonable llnearity existed over and downstream of the bottom wall contour.

The up-

stream portion of the perforated wall was used to establish flow conditions, whereas normally four ventilated walls would be utilized at
significantly less model blockage ratios; therefore, it is suggested
that the data obtained directly above the contour were more representative of conventional conditions.

33

[2
(3
:D
',4

0.02
Cp

I
0

-0.0 Z

,e,,:-o.5
o.,

~ I L~ ~- -

CONFIG A6.0

0.04

0.06

IN FLOW
-o08
-0.028

OUTFLOW
l

-0.024

- 0 ~ 2 0 -0.016

-O. OIZ -0. 008

-0.004

0.004

0.008
e

Figure 10. Representativelocus of pressure coefficient and flow

inclination - the wall characteristic.

).012

AEDC-TR-77-61
To quantify the wall characteristics, a least-squares, linear fit
of the Cp-8 locus was obtained; this fitting was limited to the wall
region directly above the bottom wall contour.

The resulting slope and

intercept were taken to be the primary descriptors of the ventilated

wall characteristics.

Representative Cp-8 loci for each wa1 configuration are given in

Appendix A.

The origin of each curve is displaced fo9 clarity.

Again,

it should be noted that these data were obtained with the bottom wall
parallel to the tunnel centerline with the circular arc (Contour A)
installed.

5.2.2 Influence of Boundary-Layer Thickness on the Characteristic Slope

L

The effect of changes in the imposed pressure distribution as

achlevedlwith the differing bottom wall contours and wall angle is
presented in Fig. 11.
A6.0 at M = 0.6.

These results were obtained with configuration

It is clear that a unique wall characteristic does not

exist in the sense of previous investigations. The shape of the C -8

f

:-.

locus is dependent On the pressure distribution.

Furthermore, the mean

slope decreases as the boundary layer is thickened because of either

bottom wall divergence or reduced contour height.

A quantitative description of the Wall characteristic dependence on

the boundary-layer displacement thickness is presented in Fig. 12.
These data were obtained with variations of the bottom wall geometry.
If the characteristic is reasonably linear (see Appendix A), then the
slope is correlated by the ratio of the averag e displacement thickness
to the hole diameter w i t h the exception of a few (inexplicable) outlying
points, the correlation indicating decreased slope with increased
thickness.

This approach was consistent in that both the eharacterlstlc

slope and the boundary-layer thicknes s were averages over the wall
extent directly above the bottom wall contours.

35

AEDC-TR-77-61

0.04
0
o.
(.t

LLJ
b.
dJ
O
(J
Ld
n-

0
E

(n
bJ

r,-

CL

0
CONFIGURATION

A6.0

M=0.6

+1"

0
-0.04
-0.04

-0.02

0
0
0
0
FLOW ANGLE, rodions

Figure 11. Variability of the wall characteristic with changes

in the imposed pressure distribution.

36

0.02

AEoc.TR.~,'I:S,

I0~0

0
[]
m

CONFIG
A. 6 . 0
8 7;0
C IO:Q

< 2.0
zu
j

P,
~. :1.0

(/)

0.5

qp

0.2
0.4
0.6
0,8
1.0
AVERAGE DISPLACEMENT THICKNESS/HOLE DIAMETER
a. Fixed porosity walls
Figure 12. Effect of boundary-layer displacement thickness
on the characteristic dope.

,/

3"/

A E DC-TR-77-61

I0.0

~o

~,~

5.0

~ .

"'X.o

_'~-._A _ _
.

u_

,_o

Z.O

"

0"-,8.

o i

---O.----

,.=,

\ )"~o..
A

'

X0, %.

"

5.0

o,

"

1.0

U)

--

r I 0 .

0.5
0

O k\

"K,,_

,,

REF 19

I
I
I
I
I
0.2
0.4
0.6
0.8
1.0
AVERAGE DISPLACEMENT THICKNESS/HOLE DIAMETER
r

b. ConfigurationD
F!gure 12. Concluded.

38
i,

AEDC-TR-77-61

It is evident that the slope of the wall characteristic, and hence

the boundary condition for wall interference estimates, is dependent on
the relative thickness of the wall boundary layer.

Anything that changes

the boundary-layer thickness will change the wall characteristic.

Since

the tunnel wall boundary layer generally becomes thinner with increasing
Reynolds number or with increased model size, the representative wall
interference boundary condition is variable.

Furthermore, a lifting

model in a ventilated wind tunnel would generate disturbances at the

wail such that, in general, the ceiling boundary layer would be thicker
than that on the floor with resulting disparity in floor/ceiling boundary conditions.

This conclusion is consistent with the flndingsof

Mokry, et al. (Ref. 18) who attributed the difference in wall boundary
conditions to a nonlinear wall characteristic.

Reference 18 also de-

monstrated that increased Math number increased the wall characteristic

slope.

This trend may now be interpreted to be the result of decreased

boundary-layer thickness caused by an enlargement of the model disturbance field and additional suction at the wall.

The characteristics of configuration D with upstream movement of

the cutoff plate are summarized in Fig. 12b. The variable porosity
feature provides good control over the wall boundary condition.

Also

indicated in Fig. 12b are the results from a previous attempt (Ref. 19)
at quantifying the characteristics of this wall geometry.

Reasonable

agreement is evident (same order of magnitude) except at five-percent

porosity where no explanation of this discrepancy is apparent.

5.2.3 Influence of the Boundary-Layer Thickness on the Tunnel Calibration

The i n t e r c e p t
the best available

o f t h e C - 8 l o c u s a p p r o x i m a t i o n was c o n s i d e r e d t o be
P
e s t i m a t e of the p r e s s u r e drop a c r o s s the w a l l s c o r -

r e s p o n d i n g to uniform flow c o n d i t i o n s w i t h i n a
v a l u e s a r e found to be c o r r e l a t a b l e

test

section.

These

wfth the average boundary-layer

displacement thicknesses as presented in Fig. 13.

39

These data indicate

AE DC-TR-77-61

0.04

0.02
o
=

_z

-0.02

z_

D 5.0

\c,o.o

-o.o4

_u

-0.06

"= -0.08
~

-0.10

'

O I0.0

0
0.2
0,4
0,6
0,8
1.0
AVERAGE DISPLACEMENT THICKNESS/ HOLE DIAMETER
'Figure 13. Effect of boundary-layer displacement thickness
on the characteristic intercept.

4O

AEDC-TR-77-61

significant variation of the wall pressure differential with changes in

boundary-layer displacement thickness, which in turn'affects the Mach
number precision of conventional transonic wind tunnels.

Current tech-

niques of wind tunnel practice include calibration of the tunnel centerline Mach number against the plenum Mach number and subsequent use of
that calibration to infer a free-stream Mach number from the plenum Mach
number, regardless of the model blockage.

Since the model installation

tends to reduce the wall boundary-laYer thickness relative to that

occurrlng in the empty tunnel, the calibration becomes inapplicable.
Admittedly, the error in Mach number would be small, but nonetheless
present.

A further source of imprecision in Mach number results from

calibration at one Reynolds number and thence applying that calibration

for tests at all available Reynolds numbers.

The sensitivity of test sectlon Mach number to wall boundarylayer thickness suggests that the present practice of using plenum
pressure to define a free-streamMach number should be re-examlned.

5.2A Variable-PorosityWall Geometry

As discussed in Section 3.3, the varlable-porosfty wall (configuration D) was tested with both upstream and downstream displacement of
the cutoff plate.

Examination of the C -8 locl in Fig. A-4 (Appendix

P
A) clearly indicates distinct differences in the wall characteristics as
a result of the cutoff plate movement direction.

This difference is

illustrated in more detail in Fig. 14, along wlth the theoretically

required wall characteristics for alleviation of subsonic wall interference or supersonic wave reflection.

The subsonic interference-free

curve was calculated with the computer code described in Section 2.1,
and the supersonic llne is the result of small disturbance theory (Ref.
19).

Configuration D with upstream cutoff plate movement reasonably

matches the linear characteristic required for supersonic wave cancellation, and the downstream movement characteristic approximates the

41

AE DC-T R -77-61

Z"

cp

UPSTREAM
CUTOFF
DS.0

DATA FOR
M = 0.6

DESIRED
SUBSONIC
CHARACTERISTIC - - ~

" ~ D E S I RED
SUPERSONIC
" " ~. ~
CHARACTER,
~RISTIC

/!
I

DOWNSTREAM
CUTOFF
D -5.O

Figure 14. Comparison of the variable-porosity wall characteristics

for upstream and downstream displacementof the cutoff plate.

42

AEDC-TR-77-61

cardiod-shaped Cp-8 locus desired for subsonic flow (note the Cp 6rigin
is shifted to compensate for the different plenum Maeh numbers).
Therefore, consideration should be given to using this variable-poroslty
wall with both upstream and downstream displacement of the cutoff plate
as a function of Mach number~

5.3 EFFECT OF NOISE SUPPRESSION DEVICES

Several techniques (Refs. 13 and 14) have been developed for suppression of the edgetones, but it was not known what effect these wall
modifications had upon the crossflow characteristic.

As discussed in

Section 3.3, two types of noise suppression devices were investigated:

splitter plates (SPL) bisecting each hole and a screen overlay (SCR) on
the airside surface.

The crossflow characteristics obtained are pre'

sentedlin Appendix A.

For purposes of discussion, some data for wall configuration A

are reproduced as Fig. 15.

The changes in the characteristic resulting

from the splitter-plate modification were an increased pressure drop

across the wall and an enlarged spread of double-valuedness in C
~P
for fixed 8. The multiple-value pressure for a given flow inclination,
particularly inflow, is indicative of sensitivity to boundary-layer
thickness and is thought to be evidence that the wave cancellation
properties of the wall would be adversely affected.

The change in

pressure drop across the wall at zero erossflow velocity would change a
tunnel calibration but have no other significant effect.

Results obtained with the screen overlay evidence an opposite trend

in the inflow region of the wall characteristic relative to the splitter
plate.

The multiple-valued area was enlarged but resulted in a negative

pressure shift with increased boundary-layer thickness.

The effect of

this change on the wave cancellation properties" of the wall is unknown.

The screen solidity was 30 percent, and one would intuitively expect

43

AEDC-TR-77-61

.+

0,16

BASIC WALL

0.12

SPLITTER PLATE

I
r

SCREEN OVERLAY " - - I " " ' t

Q.

u
I--"

t
t

0.08

I
.]

tAJ

O: 0 4
la.
t~J

o(.1

fj~ '
SJ / SI~

t~J
n(n
uJ
r~
G.

,
/

-0.04

4- .
-0.08

-0. 12

-0. 16
-0.06

-0.04

-0.02
0
FLOW ANGLE, rodions

0.02

).04

Figure 15. Effect of noise suppression devices on the characteristics

of configuration A.
b,

44

A E DC-TR:77-61

that the wall characteristic slope should have increasedaccordlngly,

but the least-squares slope was increased only 10 percent Or less" relative to that of the basic wall.

The effect of the noise suppression devices on the characteristics

of wali con flguration D are presented in Fig. 16 for selected porosities.
Again, the splitter plate tended to open'up the characteristic, which is
belleved to adversely affect the Wave cancellation propertles'of the
wall.

On the other hand, the screen overlay tended to close the charac-

terlstlc and yielded practically slngle-valued curves.

These result@

indicate that a screen overlay on the variable-porosity wall would

improve the supersonic wave cancellation properties.

However, referrlng

to Fig. 16, the screen overlay also straightened the characterlstlc With
downstream movement of the cutoff plate which would remove any chance of
achieving a reduced subsonic wall interference test environment.

In most instances, the screen overlay on configuration D tended to

[

reduce the characteristic slope.

The slope change does not affect the

utility of the wall because the varlable-poroslty feature retains control

of the wall characteristic.

45

AEDC-TR-77-61

0.16

BASIC WALL

0.12

SPLITTER

PLATE

SCREEN OVERLAY
I

Q,

u
I-:

0.08

Z
hi

0.04
h
tl.
b.I
0

j-~1"*

hi
n*.,=
(/)

(n

. ;/ ' I /

yfJ

-0.04

W
n.,
(1.

-0.08

-0.12

-0.16
-0.06

-G04

-0.02

0.02

FLOW ANGLE, rodions

a. ~" = 2.5 (upstream)
Figure 16. Effect of noise suppression devices on the characteristics
of configuration D.

46

0.04

AEDC-T

i 0.16,

BASIC WALL
~

O. 12

R-77-61

- .

SPLITTER PLATE

....

SCREEN OVERLAY

0.08
u I

ioo,
~,.

I"'-; /"7
_~.-F''j J / /

~i .0. 0 4

~i'~!- 0 . 0 8

-0.12
;,!?

-0 16
-0~06

-0.04 "

-0.02
0
FLOW ANGLE, radians
I~- ~ = 5.0 (upstream)
Figure 16. Continued.

4?

0.02

004

A E D C - T R-77-61

0.12

BASIC WALL
O.

0.08
m n l l

Z
laJ

--

U
m
h

SPLITTER

%)
I

PLATE

SCREEN O V E R L A Y
l

0.04

..o J
0

0
,,I
rim

-0.04

ILl
O.

-0.08

-0.12

-0.16

"0' 2.~).06

-0.04

-0.02

0.02

rr

0.04-

FLOW ANGLE, rodions

c. ~"= -5.0 (downstream)
Figure 16. Concluded.
6.0 CONCLUDING REMARKS

A method has been (~eveloped that allows sufficiently accurate

determination of the crossflow characteristics of ventilated walls.

The

procedure enables direct comparison of the characteristics of different

walls and requires the experimental measurement of static pressure 0niy,
the flow inclination being calculated from those measurements.

The

static pressure and flow inclination streamwlse distribution in the

vicinity of the wall then enables calculation of the boundary-layer
development on, and mass flux through, the ventilated wall.

48

AEDC-TR-77-61

These techniques were employed to document the characteristics of

two basic perforated wall geometries, fixed porosity and variable porosity with 60-deg inclined holes, with the following results:

I.

The perforated wall crossflow characteristic, defined as the

locus of pressure coefficient (Cp) against flow angle (8),
is not unique and shows dependence on the longitudinal pressure distribution.

Increased boundary-layer thickness or increased porosity

decrease the slope, dCp/dS, of the wall characteristic.

The pressure drop across the wall at zero crossflow velocity

is dependent on boundary-layer thickness.

The effects on the wall characteristic of two types of noise suppression devices, splitter plates bisecting each hole and a screen
overlaid on the airslde surface, were documented with the following
results being obtained:

I.

The splitter plates increase the pressure drop across the

walls and open up (decreased single-valuedness) the wall
characteristic.

The screen overlay decreases the pressure drop across the

walls and tends to yield more linear characteristics.

Analysis of these results and their relationship with current wind

tunnel operating procedures and practice has indicated the following
conclusions:

49

AE DC-TR-77-61

The crossflew characteristic of most (not all

perforated

walls can be assumed linear for purposes of calculating subsonic wall interference effects.

However, each wall of the

wind tunnel test section may require a different characteristic representation to accommodate differences in mean wall i
boundary-layer thicknesses.
\
.

If plenam pressure is sensed to indicate the free-stream Maeh

number via empty-tunne! calibration, the resulting

test

sec-

tion Maeh number will probably vary as a function of Reynolds

number, model blockage, and posslbly model attitude.

REFERENCES

I.

Goethert, B. H. Transonic Wind Tunnel Testing.

Pergamon Press, New

York, 1961.

Chew, W. L., Jr.

2 ,

,Experimental and Theoretical Studies on Three-

Dimensional Wave Reflection in Transonic TestSections, Part

III: Characteristics of Perforated Test-Sectlon Walls with
Differential Resistance to CrossFlow."

AEDC-TN-55-44

(AD84158), March 1956.

S.

Chew, W. L., Jr.

"Characteristics of Perforated Plates with Con-

ventional and Differential Resistance to Cross-Flow and Airflow Parallel to the Plates."

Proceedings of, the Propulsion

Wind Tunnel Transonic Seminar, AEDC, July 1956.

. Vidal, R. J., Erlekson, J. C., Jr., and Catlin, P . A.

with a Self-Correcting Wind Tunnel."

"Experiments

Advisory Group for~

Aerospace Research and Development AGARD CP-174, October 1975.

50

AEDC-TR-77-6t

5.

Berndt, S. B. and Sorensen, H.

"Flow Properties of Slotted Walls

. . . . . . . for Transonic Test Sections."

Advisory Group for Aerospace

Research and Development AGARD CP-174, October 1975.

6 . Murman, E. M. and Cole, J. D.

sonic Flows."

7.

"Calculation of Plane Steady Tran-

AIAA Journal, Vol. 9, No. I, January 1971.

Liepmann, H. W. and Roshko, A.

Elements of Gasdynamlcs.

John

Wiley and Sons, New York, 1958.

8.

Murman, E. M.

"Analysis of Embedded Shock Waves Calculated by

Relaxation Methods."

Proceedings of the AIAA Computational

Fluid Dynamics Conference, Palm Springs, California, July

1973.
9. M/ine-Thomson, L. M~

Theoretical Hydrodynamics.

The Macmillan

Company, 1968.

10. ~Lukaslewicz, J.

"Effects of Boundary Layer and Geometry on Char-

acterlstlcs of Perforated Walls for Transonic Wind Tunnels."

Aerospace Engineerln~, April 1961.

11.

Whitfield, D. L. "Analytical, Numerical, and Experimental Results

on Turbulent Boundary Layers."

AEDCrTR-76-62 (AD-A027588),

J u l y 1976.

12.

Patankar, S. V. and Spalding, D. B.

Boundary Layers.

13.

Heat and Mass Transfer in

Morgan-Grampian, London, 1967.

Dougherty, N. S., Jr., Anderson, C. F., and Parker, R. L., Jr.

"An

Experimental Investigation of Techniques to Suppress Edgetones

from Perforated Wind Tunnel Walls."
August 1 9 7 5 .

5]

AEDC-TR-75-88 (AD-A013728),

AEDC-TR-77-61

14.

Schutzenhofer, L. A. and Howard, P. W.

"Suppression of Background

Noise in a Transonic Wind-Tunnel Test Section."

AIAA Journal,

Vol. 13, No. 11, November 1975.

15.

Cline, V. A. and

Lo, C.' F.

"Application of the Dual-Scatter'Laser

Velocimeter in Transonic Flow Research."

Proceedings of the

AGARD Symposium on Application of N6n-lntrusive Instrumenta"

tion in Fluid Flow Research, France, May 1976.

16.

Garner , H. C., Rogers, E. W. E., Acum, W. E. A., and Maskell, E. C.

"Subsonic Wind Tunnel Wall Corrections."

Advisory Group I~
for

Aerospace Research and Development, AGARDograph 109, O~tober

1966.

17.

Pindzola, M. and Lo, C. F.

"Boundary Interference at Subsonic I

Speedsin Wind Tunnels with Ventilated Walls."

AEDC-TR-69-

4 7 (AD687440), May 1969.

18.

Mokry, M., Peake , D. J., and Bowker, A. J.

"Wall Interference on :

Two-Dimensional Supercritical Airfoils, Using Wall Pressure

Measurements to Determine the Porosity Factors for Tunnel
Floor and Ceiling."

National Research Council NRC No. 13894,

February 1974.

19.

Jacocks, J. L.

"Evaluation of Interference Effects on a Lifting

Model in the AEDC-PWT 4-Ft Transonic Tunnel."

AEDC-TR-70-72

(AD868290), April 1970.

W~

52

AEDC-TR-77-61

hri: :~ , : .."
APPENDIX A
, !i~.~::~'.. , . C O M P A R A T I V E C R O S S F L O W C H A R A C T E R I S T I C D A T A
FOR EACH P E R F O R A T E D W A L L G E O M E T R Y

Comparison of the characteristics of one ventilated wall geometry

with~another geometry is fully realistic only if all other variables are
fixed, specifically the imposed pressure distribution and wallboundarylayer thickness.

Since this situation could not be achieved, an alter-

nate presentation of the results is given herein with fixed pressure

disturbance geometry.

The characteristics of each perforated wall were

measured with the circular arc contour installed with the bottom wall
parallel to the tunnel centerline.

These data are presented for Mach

numbers of 0.5, 0.6, 0.7, and 0.8 with offset origins in Figs. A-I
through A-8.

Comparisons among the data are meaningful in the sense of

wall performance in a conventional wind tunnel with fixed modelgeometry.

..

i~

53

AE DC-TR-77-61

%~,~

0,,25

.o.o.,--

_o.o,

i/o.o,

o.o oo, o6.,

-0.2

-0.3
Figure A-1. Characteristics of configuration A6.0.

54

AEDC-TR-77-6i

,,.,~

0.213 D

30o~
~. "
'>

>

......

0.125._L

i >

~
a = 0.~I/j-l/~

Cp 0"3 - "

~'/

'

'

+<-~ o.2 - "

0"7:i

'

I" '
I

004

0.06
e

-0.3
Figure A-2., Characteristics of configuration B7.0.

SS

A E OC-T R -77-61

r-.-o.

166

1,
~

.=

03 r

06
t

07

._~_~ ~

-O.2
-O.3
Figure A-3. Characteristios of configuration C10.0.

56

AEDC-TR-77-61

0.125

X " ~ , ~

166 D

".': t "~, "-,~

:.~>i~i-~.~~:;~./,.. ~,

'"~::i!-~.:.~!'~:.,:~i~
I

3r

- y F/
=0.2

=0.3
a. T = 1.0 (upstream)
Figure A-4. Characteristics of configuration D.

5?
/

0.125

AEDC-TR-77-61

__~0.166

-0.06

-o.'o~ '-o.ov--~

o.'o~

-0.2

-0.3
b. T = 2.5 (upstream)
Figure A4, Continued.

58

o:o. o'.o

A EDC-TR-77-61

~f

0.166 O

J_
-f-

................. ~

..............

-1-

0125
Cp

-0.06

0.04

+ _ _ / ,

-0.02

=0.

0.02

-0.2

=0.3
c. r = 5.0 (upstream)
Figure A-4. Continued.

59

0.04

0.06

AEDC-TR-77-61

~0.166

.~'" " :~':; ~': ' " :,"::" ,'" ;"'~: ~/, , m

~:o.o~-~

-0.06

" "'"

"~:'= ::

'

--f-

0.3 r

Co.L

oo,o . a

~I

-0.2

-0.3
d. ~-- 10.0

Figure A-4. Continued.

60

o~--~,

AE DC-TR-77-61

0.125
_L

-~- 0.3

Mc~)o.2

.
0 . 8 ~
-0.2 I
-0.3
e. r = 5.0 (downstream)
Figure A 4 . Continued.

'

6]

AEDC-TR -77-6!

0.12~"

.166 D

__L ~ o ~ ~

o.,~

0.3
Cp

4-

o.2

/!
,

-0.06

-0.04

0.06

0.04

-0.2

-0.3
f. T = 2.5 (downstream)
Figure Ao4. Concluded.

'

62

AEDC-T R-77-61

0.3
M =0.5

Cp

+
o16

+
0.7
-0.06

+.

.o.o2

0.02

0.04

0.06

-0.2

-0.3
Figure A-5. Characteristics of confi~Jration A6.0 with splitter plates.

63

AEDC-TR-77-61
r

M=0.5

0.3
0,6

0.7f f _

-iv
/

-0.06

'

'

- 0 . 0 4 ~

/-t-

0.02

0.04

0.06

....

-0.1
-0.2

-0.3
Figure A-6, Characteristics Of configuration A6.0 with screen overlay.

AEDC-TR-77-61

"

-f-

0.125

:_".,~.'.'.':~.:...~:..~:,, / L".~:.~...-,,.

' .... ':'':: "~

Cp
0.6

I.

-O06

-0.04

4-/

~--

I/

'

0.02

0.04

0.06
e

0.8 ~ j ~ ' S .

-0.2

-0.3
a. r = 2.5 (upstream)
Figure A-7. Characteristics of configuration D with splitter plates.

65

AEDC-TR-77-61
1

t
0.125

-o.o,-o.o,~

o.o,e
'.i:o:,o~...

-0.2

"

-0.:5
b. 7" = 5.0 (upstream)
Figure A-7. Continued.

66

AEDc'TR'77"6i

0.125

__k

. -{-

Cp 0"3 I

M =0.

0.2
q

-0.~6

:~ ~ /

-0.02

:I

0.02

10.2

C" r~ = "5"0 (downstream)

Figur e AI7. Continued.

67

.../

0.04

0.06

AEDC-TR-77-61

0.125

T
~+

o.~r

Cp

M=0.5

0.2
0.6

-0.06

o.7/"
,f

)/~

/.--.~---

-0.04

-0.02

- -

/
~

0.06

0.04

0.8

-0.3
d. T = -2~5 (downstream)
Figure A-7. Concluded.
L*!

68

AEDC-T

~""'~

.....
.....
N.

R-77-61

M=0.5

I1'
v-

-0.06

-o.o4

~/
,

/+

0.02

0.04

0.06
0

~'/

-0.I
-0,2

-0,3
a. T = 1.0 (upstream)
Figure A-8. Characteristics of configuration D with screen oveday.

69

AEDC-TR-77-61

~ ~ = 0.166

0.125

0.125

_L
-t-

%,

-01.06

).04

08

006;

,"'

8
-0. I

-0.2
j~

-0.3

b. T = 2.5 (upstream)
Figure A-8. Continued.

e.

70

AEDC-T,R-77-61

~~~~0.166

0 125

T
M
0'5~
=~ ~ ~

0.3

0.7
l

";

-0.0"6 ~"

I
-0.04

0.02

0104

0.06

0.8
-0. I

-0.2
-0.3
c. = 5.0 (upstream)
Figure A-8. Continued.

71

AEDC-TR-77-61

~F0.166

O'Oo'

0:% o'o
-0.1

-0.2

-0.3
d. T = 10.0
I=igure A-8. Continued.

72

AEDC-TR-77-61

0.1~5 "
"i:

0.125

ID

~~ .

~i= .............~,

0.3

c~
M = 0.5"f--"*'~

-Q.'~06 . ' ~ -

o.;~

0.06

0.04

0.8
-0. I

-0.2

-0.3
e. ~"= -5.0 (downstream)
Figure A-8. Concluded.

73

AE Dc-TR-77~61

NOMENCLATURE
Cf

Boundary-layer skln-friction coefficient

Pressure coefficient

~k

Perforated wall hole diameter

Boundary-layer shape factor,

Tunnel h e i g h t

Tunnel length

Nominal Math number

Wall permeability factor

Wall thickness

Boundary-layer edge velocity

Streamwise coordinate

Transverse coordinate

yg

Bottom wall geometry

Ys

Streamline shape

Prandtl-Glauert

Specific heat ratio, 7 ffi 1.4

H = 61/~ 2

f ~l.l

,W

factor, B ffi (I - ~ ) I / 2

74

AEDC-TR-77-61

61

Boundary-layer displacement thickness

~2

Boundary-layer momentum thickness

Boundary-layer control-volume height

Flow angle

Normalized wall mass flux

p=

Boundary-layer edge density

Wall porosity
Potential function
Particular solution for potential function

75