https://ntrs.nasa.gov/search.jsp?

R=19770005050 2019-10-01T15:02:48+00:00Z

NASA TECHNICAL NOTE NASA TN D-8243

-

M
u-
cy
w
n
z
c

AERODYNAMIC DESIGN GUIDELINES

AND COMPUTER PROGRAM FOR
ESTIMATION OF SUBSONIC
WIND TUNNEL PERFORMANCE

William T. Eckert, K e m e t h K? Mort,

and Jeau Jope
Ames Research Center
und U S . Army Air Mobility RGD Laboratory doL"T1o&

Moffett Field, Cali$ 94035 v

".,'>76 ,916 +is

N A T I O N A L A E R O N A U T I C S A N D SPACE A D M I N I S T R A T I O N WASHINGTON, 0. C. OCTOBER 1976

1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.
TN D-8243
4. Title and Subtitle 5. Report Date
AERODYNAMIC DESIGN GUIDELINES AND COMPUTER PROGRAM FOR ESTIMATION October 1976
OF SUBSONIC WIND TUNNEL PERFORMANCE 6. Performing Ormnizrtion Code

7. Author(s1 8. Performing Organization Report No.

William T. Eckert,* Kenneth W. Mort, and Jean Jope A-5944
10. Work Unit No.
9 Performing Organization Name and Address 505-06-31
NASA Ames Research Lenter
and 11. Contract or Grant No.
Ames Directorate,U.S. Army Air Mobility R&D Laboratory
Moffett Field, CA 94035
13. Type of Report and Period Covered
12 Sponsoring Agency Name and Address Technical Note
National Aeronautics and Space Administration
Washington, D. C. 20546 and 14. Sponsoring Agency Coda .
U.S. Army Air Mobility R&D Laboratory, Moffett Field, CA 94035
I 5 Supplementary Notes
*Ames Directorate, U.S. Army Air Mobility R&D Laboratory, Moffett Field, CA 94035

16. Abstract
This report brings together and refines the previously scattered and over-simplified
techniques for the aerodynamic design and l o s s prediction of the components of subsonic wind
tunnels. General guidelines are given for the design of diffusers, contractiohs, corners, and
the inlets and exits of non-return tunnels. A system of equations, reflecting the current tech-
nology, has been compiled and assembled into a computer program (a user's manual for this program
is included) for determining the total pressure losses. The formulation presented is applicable
to compressible flow through most closed- or open-throat, single-, double-, or non-return wind
tunnels. A comparison of estimated performance with that actually achieved by several existing
facilities produced generally good agreement.

17. Key Words (Suggested by Authorls)) 18. Distribution Statement

Wind tunnel design (subsonic) Unlimited
Wind tunnel (subsonic)
Duct flow Energy use analysis
Duct losses Power consumption
Wind tunnel performance analysis
Aircraft testing STAR Category - 02
19 Security Classif. (of this report) 20 Security Classif. (of this page) 21. No. of Pages 22. Rice'

Unclassified Unclassified 197 $7.00

'For sale by the National Technical Information Service. Springfield,'Virginia 22161

 TABLE OF CGYTENTS

Page
...
NOTATION (Engineering Symbols)
SUMMARY ... .... ....... V
1
.
INTRODUCTION . . . 1
....
e

CAUTIONARY DESIGN GUIDELINES 2

Diffusers. .... ....... 2
Contractions .... ....... 3
Corners .. .... ....... 3
Non-Return Wind Tunnels ..... 4
PERFORMANCE ESTIMATION ....... 4
General Approach .. ....... 4
.......
Problem Restrictions 5
.......
Computation Formulas 5
.....
Flow-state parameters 5
........
Local conditions 5
....
Section pressure losses 6
....
COMPUTER PROGRAM DESCRIPTION 11
........
Method of Solution 11
...
Computing Equipment Required 12
Hardware and machine components 12
Software ............ 13
......
Programming Techniques 14
Source Code ........... 15
......
Operating Instructions 15
Input ............. 15
output ............. 16
..
Computer system restrictions 17
........
Optional inputs 18
......
Diagnostic messages 18
Test Case ............ 21
....
DISCUSSION AND APPLICATIONS 22
Results ............. 23
Evaluation............ 24
APPENDIX A .NON-STANDARD FUNCTIONAL FORMS 26
......
Local Flow-State Parameters 26
.............
Mach number 26
...........
Reynolds number 27
.........
Friction coefficient 27
........
Section Pressure Losses 27
.......
Corners (constant area) 27
.........
Corners (diffusing) 28
..............
Diffusers 29
Exit ................. 30
Flow straighteners: airfoil members (thick) 31
...
Internal flow obstruction: drag item 32
Vaned diffusers .............. 32
Loss value transferred to reference location 32
.........
Wall pressure differential 33
............
Input power required 34

iii
 TABLE OF CONTENTS .C o n c l u d e d
Page
APPENDIX B .NUMERICAL FUNCTION-APPROXTMATIONS ............. 35
Corners ............................... 35
Diffusers . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . 35
. . . . . . . . . . . . . . . ..............
Mesh S c r e e n 38
...........................
Vaned D i f f u s e r s 38
APPENDIX C .COMPUTER PROGRAM F0RTRAN CODES .............. 39
..........................
N o t a t i o n (F0RTRAN) 40
PERF0RM ............................... 52
DATACK . . . . . . . . . . . . . . . . . . .............. 79
SPEED . . . . . . ............. 97
FRICTN................................ 98
0UTPUT.................... . . . . . . . . . . . . . 99
PLgTIT. . . . . . . . . . . . . . . . . . .............. 104
.............
APPENDIX D .INPUT .AND OUTPUT FOR SAMPLE CASES 107
REFERENCES
TABLES
. .
....................
6 . .. ..............
.............
142
144
FIGURES . . . . . ............. 153

.t

iv
 NOTATION
(Engineering Symbols)

F0RTRAN
Symbol name Description
A A cross-sectional area of local section, m2 (ft2)

total cross-sectional flow area at drive

fan(s), m (ft2)

Anow AL cross-sectional area of local flow, m2 (ft2)

A0 A0 cross-sectional flow area of test section at

upstream end, m2 (ft2)

A1 A1 cross-sectional flow area of section at

upstream end, m2 (ft2)
3,
A2 A2 cross-sectional flow area of section at i-
downstream end, m2 (ft2)

ASTAR cross-sectional area for sonic flow at

specified flow conditions, m2 (ft2)

AR AR cross-sectional flow area ratio of upstream

and downstream ends of section

aT AT speed of sound in still gas, computed at total

(stagnation) conditions, m/sec (ft/sec)

AS0 speed of sound in moving flow at upstream end

of test section, m/sec (ft/sec)

B dummy constraint used in defining the friction

term of turning vane loss function

CD drag coefficient of flow obstructions:

drag/@

cV CHdRI) chord of turning vanes, m (ft) .

D D cross-sectional diameter of circular duct,

m (ft)

cross-section diameter at the upstream end of

Del
an equivalent circular duct with equal area,
m (ft)
'Note that in this section, as throughout the report, all letter 0's
occurring in FfiRTRAN names are shown with slashes, as 8 ; all number zeros are
shown without slashes.
v
 F~RTRAN
Symbol name Description
De2 cross-section diameter at the downstream end of
an equivalent circular duct with equal area,
(ft)
4 x (cross-sectional area)
Dh DH hydraulic diameter:
perimeter 9

ER ER energy ratio: ratio of energy of flow at the

test section to the output energy of the fans

f (d) FKTVl function defining turning vane loss parameter

FKTV2 KTV
K EK local total pressure l o s s coefficient of sec-

tion: -
*PT
9
KCONTRACTION EKCNTR local total pressure loss coefficient from con-
tracting portion of thick-airfoil flow
straighteners

KDIFFUSION EKD local total pressure loss coefficient from

diffusing portion of multi-loss-type sections

KEXP EKEXP net expansion loss coefficient for diffusers

EKADD additional diffuser expansion l o s s coefficient

KEXPAdditional due only to more diffusion in one plane than
the other

EKBASE basic diffuser expansion l o s s factor coeffi-

cient for three-dimensional diffusion

EKC expansion l o s s coefficient for conical

diffusers

EK2DR expansion loss coefficient for a two-

dimensional, rectangular cross-section
diffuser

EKS expansion l o s s coefficient for three-

KEaSquare dimensional expansion in square cross-section
diffusers

EK2DCS estimated expansion loss value for a two-

dimensional diffuser (one with expansion in
only one plane) with cross-section shape of
some square/circular hybrid

vi
 FQRTRAN
Symbo1 name Description
EKZDC estimated expansion loss value for a hypotheti-
KExp2DCircular cal two-dimensional diffuser with circular
sides:

K~~ c ircular
KEXP
2DRec tangular

EKCSAV estimated expansion loss coefficient for three-

dimensional, combination circular and square
cross-section diffuser

KEXPANSION diffuser loss coefficient due to expansion:

'>2
(ARi
KFRICT ION turning vane loss due to friction

KFRICTION (CONICALI diffuser loss due to friction for the equivk-

lent conical diffuser

KMESH EKMESH mesh screen-type loss parameter

KRef. 9 diffuser loss factor presented in reference 9:

Ap/q
[(AR - 1)/ARI2

KRN mesh screen Reynolds number sensitivity factor

KROTATION turning vane l o s s coefficient due to rotation

KTV EKTV turning vane loss coefficient

EKTV90 turning vane l o s s parameter for given vanes at

KTV90
a 90' turn

K, EKV local total pressure loss coefficient for vaned

diffusers

KVANED DIFFUSER local total pressure loss coefficient €or vaned

2
diffuser, K, I)

KO EKO section total pressure loss coefficient

referred to test section conditions:
APT
-
QO

vi i
 FdRTRAN
Symbol name Description
flow obstruction (drag item) total pressure
loss coefficient referred to test section
conditions

L EL centerline length of section, m (ft)

II characteristic dimension on which Reynolds

number is based

M AMACH local Mach number

Mo EM0 Mach number at upstream end of test section

N N section assigned sequence number for order of

occurrewe in circuit

P~~~ power loss due to drag of flow obstruction,

w (hP)
PINPUT PWRIP tunnel drive power required to be input to flow
by the fans, W (hp)

PIN PUT^^^ power input required to overcome drag of flow

obstruction, W (hp)

PREQUIRED PmdP total fan motor output power required to drive

wind tunnel at specified speed, W (hp)

P local static pressure, N/m2 (lb/ft2)

PT tunnel total (stagnation) pressure, 'N/m2

PT (lb/ft2)

P PATM atmospheric (barometric) pressure, N/m2

TATM '(lb/ft2)

total (stagnation) pressure in the circuit

PTSC settling chamber, N/m2 (lb/ft2)

9 local dynamic pressure: -2

pv2 , N/m2 (lb/ft2)

povo2
QO test section dynamic pressure: -
2 , N/m2
(lb/ft2)

R R gas constant, m2/sec2 OK (ft2/sec2 OR)

-
Re equivalent' radius: dA/n ,m (ft)

viii
 F@RTRAN
Symbol name Description
pVR .
RN RN Reynolds number: -
!J

RNREF reference Reynolds number at which turning vane

90" loss parameter, K T V ~ ~was
, determined

S drag area of flow obstruction (i.e., area for

which CD is determined), m2 (ft2)

S distance along diffuser wall, m (ft)

s? length o f diffuser, taken along wall, m (ft)

T tunnel temperature in moving flow, OK (OR)

TT TT tunnel total (stagnation) temperature, OK (OR)

V V local flow velocity, m/sec (ft/sec) I :a

vF flow velocity at the drive fan(s), mysec

(ft Isec)

V~~~~~~ flow velocity in a multiple-duct section, mlsec

(ft/sec)

vo test sec'tion upstream-end flow velocity, m/sec

(ft/sec)

X location of inflection point in contraction.

wall (distance from upstream end), m (ft)

Y G specific heat ratio of gas

A RUFNES surface roughness in honeycomb cells, m (ft)

AER difference between estimated and true circuit .

energy ratios; i.e., error in energy ratio
est imate

APF static pressure rise across the fan(s), N/m2

( l b / ft2)

APT total pressure drop through a section, N/m2

(lb/ft2)

total pressure drop through a single duct of a

multiple-duct section, N/m2 (lb/ft2)
total pressure rise across the fan(s), N/m2
(lb/ft2)

ix
 F~RTRAN
Symbol name Description
average total pressure rise across a single fan,
APTFDUCT N/m2 (lb/ft2)
total pressure drop through a multiple-duct
section, N/m2 (lb/ft2)

summation of all total pressure drops through

APTTOTAL the wind tunnel circuit, N/m2 (lb/ft2)

DELP local pressure difference across wind tunnel

wall, N/m2 (lb/ft2)

A~REQUIRED difference between true and estimated required

drive power levels for given levels of oper-
ating velocity and fan efficiency; i.e.,
error in required power estimate, W (hp)

AVO difference between estimated and true test sec-

tion operating velocity for given power and
fan efficiency levels; i.e., error in oper-
ating velocity estimate, m/sec (ft/sec)

AE increment of flow-obstruction downstream influ-

ence factor greater than unity: E - 1
(greater than or equal to zero)

SLR diffuser side length ratio: ratio of change in

height to change in width from upstream to
downstream end, or its inverse, whichever is
l e s s than o r equal to unity

E EP S flow-obstruction downstream influence coeffi-

cient (greater than or equal to unity)

drive motor electrical efficiency, percent

ETAFAN fan aerodynamic efficiency, percent

TH diffuser half-angle, rad

h SLAMDA friction coefficient for smooth pipes

u EMU flow viscosity, N sec/m2 (lb sec/ft2)

Ustd EMUSTD standard-day value of viscosity, N sec/m2

( l b sec/ft2)

EMUT reference viscosity at a known temperature, com-

puted for still gas (stagnation conditions),
N sec/m2 ( l b sec/ft2)
X
 F8KT?&!?
Symbol name Description
V ENU kinetic viscosity of gas, m2/sec (ft2/sec)

P RH~S local static density, N sec2/m4 (lb sec2/ft4)

PF RHdSF
x
static densit at the fan(s),
(lb sec2/ft )
N sec2/m4

RH@T density computed for total (stagnation) condi-

tions, N sec2/m4 (lb sec2/ft4)

PO
RHOSO static density at upstream end of test section,
N sec2/m4 (lb sec2/ft4)

N
SUMEKO summation of section total pressure losses
i=1 referenced to test section conditions

SUMEL summation of section centerline lengths, m (ft)

9 PHI corner flow turning angle, deg

28 TH2 diffuser equivalent cone angle:

xi
 AERODYNKiXXC DESXGX GUIEELINES fi'E COMFLTER PROGRAM FOR ESTIMATION

OF SUBSONIC WIND TUNNEL PERFORMANCE

William T. Eckert, Kenneth W. Mort, and Jean Jope

Ames Research Center

and
Ames Directorate, U.S. Army Air Mobility R&D Laboratory

SUMMARY

This report brings together and refines the previously scattered and over-
simplified techniques for the aerodynamic design and loss prediction of the
components of subsonic wind tunnels. General guidelines are given for the
design of diffusers, contractions, corners, and the inlets and exits of non-
return tunnels. A system of equations, reflecting the current technology, has
been compiled and assembled into a computer program (a user's manual for this
program is included) for determining the total pressure losses. The formula-
tion presented is applicable to compressible flow through most closed- or open-
throat, single-, double-, or non-return wind tunnels. A comparison of
estimated performance with that actually achieved by several existing
facilities produced generally good agreement.

INTRODUCTION

In the past, most of the work on the design of ducts and wind tunnels and
on the determination of their pressure (and power) losses has been either
highly specialized, considering only one type of component, or over-simplified,
covering several types of components but giving only a superficial idea of
what parameters are important. However, for the recent NASA studies directed
toward new and modified wind tunnel facilities, it has been necessary to do a
careful job of estimating, easily and quickly, the performance of all circuit
components. This report brings together, revises, and updates the techniques
for the aerodynamic design and performance prediction of subsonic wind tunnels.

The basic procedures and guidelines for the aerodynamic design of

critical wind tunnel components, as presented in references 1 through 3, have
been revised and updated, as required. The diffuser and contraction design
curves developed and suggested herein show the relative design points for
several existing facilities. Also provided are recommendations derived from
recent NASA studies on end treatments for non-return wind tunnels.

The method of loss analysis presented is a synthesis of theoretical and

empirical techniques. Generally, the algorithms used were those substantiated
by experimental results. The methods of references 4 through 11 for predic-
ting componen; losses have been refined and incorporated. The performance
calculations, based on user-selected flow conditions at the test section,
assume that the circuit geometry has been predetermined.

The comparison of the actual and predicted performance for several

existing wind tunnel facilities shows generally good agreement.

CAUTIONARY DESIGN GUIDELINES

This report presents the means for rapidly estimating the performance of
a wind tunnel circuit after its geometry has been determined. However, an
improper design of any of its several components (diffusers, contractions, or
corners, for example) could result in performance penalties caused by inter-
action with the flow in other components; such penalties cannot be predicted.
In addition, improper design could cause poor test-section flow quality which
would not be indicated by the performance analysis. Therefore, the purposes
of this section are to point out critical areas of concern in wind tunnel
design and to attempt to establish proper design criteria.

Diffusers

Diffusers, especially those just downstream of high-speed sections, are

very sensitive to design errors which may cause flow separation. The equiva-
lent cone angle and area ratio must be properly selected to avoid steady-state
or intermittent separation of the flow from the diffuser walls. (This separa-
tion can cause vibration, oscillatory fan loading, oscillations in test sec-
tion velocity, and higher losses in downstream components.) Generally, proper
diffuser design requires that, for a given area ratio, the equivalent cone
angle be constrained below a certain value. ("Equivalent" denotes an imagi-
nary conical section with length and with inlet and exit areas identical to
the actual section.) This cone angle should probably be held 0.5" to 1' lower
for diffusers with sharp corners than for those with a rounded cross section.

Since the portion of the wind tunnel between the test section and the
fans is usually the higher-loss segment, it is the most critical in affecting
circuit performance. Therefore, it was used as a basis for establishing
recommended design limits as a guide to diffuser selection. It was assumed
that the fans serve to reenergize the boundary layer of downstream sections
and that the fans and the upstream and downstream components have no inter-
action that affects their losses; this may or may not be true (see ref. 12).
The overall area ratio and cone angle between the test section and fan contrac-
tion were examined for several wind tunnels. This analysis used the centerline
lengths o f all intervening components, including corners. (The actual effect
of corners is unknown: they may alter the onset of separation somewhat.)
Figure l(a) compares curves for the first appreciable stall for flows with
thin inlet boundary layers, from references 1 and 2, with the design points of
selected existing wind tunnels. These. curves were used to aid in defining the
separation trend; good correlation with the symbols is not necessarily
expected. Figure l(a) shows that most of these wind tunnels were designed
beyond (above) the two-dimensional stall curve but below the conical stall

2
curve. (Snme of these diffusers are f a r from conical.) The recommeiided
design region, shown in figure l(b), was positioned with the prior knowledge
that the NASA-Ames 7- by 10-Foot Wind Tunnel has a partially separated dif-
fuser just downstream of the test section, and that the NASA-Ames 4G- by
8O-Foot Wind Tunnel has some local separation in the corners of the primary
diffuser. The upper portion of the design region is recommended for diffusers
with rounded corners, and the lower portion for diffusers with sharp corners.

Contractions

Contracting sections are subject to separation in the same manner as

diffusers; however, the penalties are usually much less severe in the con-
tracting sections. Separation of the flow can occur if the contraction is too
short for the amount of area reduction. Figure 2(a) presents the general wall
shapes suggested in reference 3 and figure 2(b) shows the design boundary for
these shapes in comparison with the designs of several selected wind tunnel
facilities.

From this comparison it is evident that, while some facilities were

designed more conservatively than others, no design severely exceeds the
design boundary. Since none of the facilities considered has shown signifi-
cant contraction-caused flow problems, the design boundary may be considered
empirically reasonable. Further, reference 1 3 generally tends to support the
positioning of the suggested design curve. However, the criteria of refer-
ence 13 are more conservative due to consideration of viscous effects which
were neglected in the study of reference 3 .

Corners

The corner losses in a wind tunnel can be large. To minimize them, turn-
ing vanes should be used for more efficient turning. Also, as with any other
high-loss item corners should, where possible, be located in a large-area sec-
tion where the flow speed is low. Corner vane losses can be minimized in two
additional ways: (1) by selecting an efficient vane cross-sectional shape and
adjusting it for proper alignment with the flow, and (2) by choosing the best
chord-to-gap spacing.

With reference to item (l), turning vane shapes can vary from bent plates
to highly-cambered airfoils. Some sources favor airfoil vanes as being more
efficient (ref. 4 , p. 63) while others claim that thin vanes can have lower
losses (ref. 5, p. 9 3 ) . But airfoil vanes with blunted leading edges may be
more forgiving of misalignment with the flow. The thicker vanes may,
therefore, hold some advantage.

When considering item (2), the best chord-to-gap ratio depends on the
vane type. For thick vanes, a ratio of about 2.5-to-1 is recommended (ref. 4 ,
p . 62) and for thin vanes a ratio of about 4-to-1 is suggested (ref. 5, p. 92).

3
 Non-Return Wind Tunnels

Non-return wind tunnels have presented some interesting problems in

tunnel design. This type of wind tunnel has the advantages of less structure
(and therefore lower construction costs) and of no exhaust-gas-purging or air-
exchange requirement. Careful design can make the non-return circuit oper-
ating power competitive with that of closed-return wind tunnels (the corner
losses can be traded for inlet and exit losses). However, an area of concern
for the non-return tunnel is its potential sensitivity to external winds which
could affect both the required power and the test section flow quality.

A recent series of NASA studies, which dealt with wind s.nsitivity prob-
lems, showed that a non-return wind tunnel should have three 2atures: (1) a
vertical exit system, (2) a horizontal inlet, and ( 3 ) an enclused area of pro-
tection, with a solid roof, at the inlet. References 14 and 15 detail the
development work for the end treatment considered in those studies.

Reference 16 describes an inlet geometry that was developed to reduce the

effects of w.ind. (This reference also presents a set of test-section flow-
quality requirements by which the characteristics of any inlet treatment may
be evaluated.) Although the end treatment designs shown in references 14
through 16 could be revised-or refined for additional wind protection, any
additional inlet treatment would increase the structural cost and could
increase the power requirement.

I PERFORMANCE ESTIMATION

Although the performance analysis presented in this report was systema-

tized and automated for rapid calculation of numerous cases or iterations (by
the computer program described in the following sections), the equations pre-
sented are equally amenable to manual calculation methods.

General Approach

The equations were derived in forms that use the most common and conveni-
ent defining parameters. The equations are listed and explained below and may
be used for component after component, each in turn.

The total pressure losses (proportional to power losses) of each compo-

nent are calculated and summed to give the total circuit loss and operating
power required. The computation technique is applicable to either closed- or
non-return circuit types made up of any combination of standard wind tunnel
components in any order. The flow conditions in the test section (velocity,
and 'stagnation temperature and pressure) and the external atmospheric pressure
are variable as required.

4
 Problsin Restrictions

Three restrictions were found to be necessary in order to allow rapid

solution of most cases with a minimum amount of effort. First, the cross-
sectional geometries were limited to the most common shapes: circular, rec-
tangular, and flat-oval (semi-circular side walls with flat floor and ceiling).
Second, air exchangers were omitted from this analysis due to lack of uniform-
ity of configuration and a lack of definition as to the proper method of com-
puting the losses. Finally, the drive system was assumed to be located in one
or more parallel, annular ducts.
3
Computation Formulas
-
’i

The equations used in this performance analysis were synthesized from

various sources. Some were used in their original (source) form and others
were modified to make them more convenient for use in this analysis. The equa-
tions used are presented below.

Flow-state parameters- The basic flow-state parameters were determined

from input information about the reference control station and the test sec-
tion. These parameters were derived from standard relationships for
compressible flow.

(ref. 17, p. 8)

aT = JVRTT (ref. 17, p. 51)

0.76
PT = USID(%) (ref. 18, p . 19)

aT
a, = (ref. 18, p. 4 )

PT
(ref. 18, p. 4 )

{ }*
2(Y-1)
(appendix A)
A* = (+
2 [1+ y + ]l) : M

Local conditions- The local flow conditions were determined for each end
of each section.

5
 1. Mach number: The local Mach number was found from a Newton's-method
solution of the relationship

(appendix A)
J
2. Reynolds number: The Reynolds number based on the characteristic
length R, usually the local hydraulic diameter, was determined from

0.76
(appendix A)

3. Friction coefficient: A Newton's-method solution was used to deter-

mine the friction coefficient for smooth walls from the expression

[logl,(XRN2) - 0.8]-2 - A = 0 (appendix A)

Section pressure Zosses- The l o s s in total pressure caused by each sec-

tion was calculated in a form non-dimensionalized by local dynamic pressure:
K = ApT/q. (In this study the smallest-area end of each section was used as
the local reference position.) The individual losses were based on the nature
of the section, local flow conditions, and input geometry and parameter infor-
mation. The most appropriate l o s s forms for typical wind tunnel sections are
catalogued on the following pages. The nonstandard formulas, those which are
not directly attributable to the literature, are developed in appendix A. The
precise equations, which were developed from various curve-fitting and inter-
polation techniques based on the plots presented in certain figures, are given
in appendix B.

1. Constant-area ducts: For closed, constant-area sections the pressure

loss due t o friction is given by
K = - XL (ref. 7, p. 5 3 )
Dh

2 . Open-throat duct: The losses from an open-throat test section may be

(<r
found from the expression

K = 0.0845 -- 0.0053 (ref. 7, p. 150)

Dh

3 . Contractions: In contracting sections, where the major part of the

losses is due to friction, the local loss may be approximated as

AL (ref. 6 , p. 5 2 8 )
K = 0.32 -
Dh

6
 4 . Corners with no net area change ("constant area"): A duct can change
direction with or without the aid of flow guide vanes. For a constant-area
turn employing turning vanes for efficiency, with a 'hormal" number of vanes
(ref. 7, p. 241), and with chord-to-gap ratios between 2-to-1 and 4-to-1, the
losses resulting from friction and rotation caused by the vanes are

= KTV
3 2[ + ( loglo REF
loglo m )
2.58

3 (appendix A)

The "Reynolds number used for the turning vane l o s s should be based on vane
chord. The turning vane loss parameter K T ~ is plotted as a function of turn-
ing angle in figure 3(a), with the assumption that KTV = 0.15 is a reasonable
value for a 90' corner. Corners without turning vanes are less efficient and
the loss function may be approximated by a sixth-order polynomial as shown in
figure 3(b) :

K = - 6.021515~10-~
4.313761~10-~ $' - 2.755078~10-~
$ + 1.693778~10-~ (p3

+2.323170~10-~$4 - 3.775568~10-~
$5 + 1.796817~10-~~
$6

(appendix B)

This function assumes a l o s s value of about K = 1.8 for a 90' turn. The
foregoing losses are those associated with the turning of the flow only. The
losses for a corner system (with or without vanes), with the walls of the duct
to be considered as well, requires an additional term for the frictional loss
of the constant-area duct based on the centerline length.

5. Corners (diffusing): Corners with diffusion may well employ longer

vanes in order to improve the efficiency of the diffusion process. For this
reason they were treated as vaned diffusers with the addition of the rota-
tional loss term of the turning vane function:

(appendix A)

where u(20 - 21.5') is the unit step function and the turning vane loss
parameter is defined as for a constant-area corner. This loss function
includes the effects of friction.

6. Diffusers: Diffusion produces both expansion and friction losses in

the duct given by

= CKEXp +
A
(8 sin 0 AR - 1
AR + l)](AR-; !)l (appendix A)

7
where t h e expansion p a r a m e t e r v a l u e s , KEXP, are p l o t t e d a g a i n s t e q u i v a l e n t cone
a n g l e i n f i g u r e 4 and t h e t e c h n i q u e used f o r e s t i m a t i n g t h e KEXP v a l u e s is
d e s c r i b e d i n appendix A.

( I t should b e n o t e d t h a t t h e r e are more s o p h i s t i c a t e d t e c h n i q u e s f o r

e s t i m a t i n g d i f f u s e r performance t h a n t h e one p r e s e n t e d h e r e . However, t h e y
r e q u i r e boundary-layer c a l c u l a t i o n s ; f o r example, see r e f e r e n c e 19. E x p e r i e n c e
w i t h b o t h t h e s i m p l e t e c h n i q u e d e s c r i b e d h e r e i n and more complex t e c h n i q u e s
i n d i c a t e s t h a t t h e two produce comparable r e s u l t s . G e n e r a l l y , l i t t l e i s gained
by t h e s i g n i f i c a n t a d d i t i o n a l e f f o r t r e q u i r e d t o u s e t h e more complex
approaches.)

7. E x i t : The t o t a l p r e s s u r e l o s s a t t h e e x i t of a n o n - r e t u r n wind t u n n e l ,
o r o f any e x p e l l e d flow, i s due t o t h e l o s s of t h e k i n e t i c energy of t h e e x i t -
i n g flow. T h i s i s g i v e n by
Y

(appendix A)

8. Fan (power) s e c t i o n : Fan d r i v e s e c t i o n s are commonly made up of con-

t r a c t i o n s , c o n s t a n t - a r e a a n n u l a r d u c t s , and d i f f u s e r s . A n a l y s i s s h o u l d b e
handled by d i v i d i n g t h e f a n s e c t i o n i n t o t h e s e t h r e e component p a r t s .

9 . Flow s t r a i g h t e n e r s - honeycomb ( t h i n w a l l s ) : The l o s s through t h i n

f l o w - s t r a i g h t e n e r o r honeycomb systems may b e e x p r e s s e d a s

where t h e h y d r a u l i c d i a m e t e r i s t h a t of t h e honeycomb c e l l . The f r i c t i o n

c o e f f i c i e n t i s determined from a Reyriolds number based on t h e s u r f a c e roughness
of t h e honeycomb:

for RN 5 2 7 5 ,

and f o r RN > 2 7 5 ,

10. Flow s t r a i g h t e n e r s - a i r f o i l members ( t h i c k w a l l s ) : Flow through

a d j a c e n t a i r f o i l s w i l l f i r s t c o n t r a c t and t h e n d i f f u s e . It w a s assumed t h a t
t h e p o i n t of minimum d i s t a n c e between p a r a l l e l members would be a t 30 p e r c e n t
o f t h e s t r a i g h t e n e r l e n g t h back from t h e l e a d i n g edge. The forward 30 p e r c e n t
w a s t r e a t e d as a c o n t r a c t i o n and t h e a f t 7 0 p e r c e n t a s a vaned d i f f u s e r . Thus,

(appendix A)
8
where t h e h y d r a u l i c d i a m e t e r is t h a t of each c e l l of t h e f l o w s t r a i g h t e n e r ,
t h e f r i c t i o n c o e f f i c i e n t is determined from a Reynolds number based on t h a t
h y d r a u l i c d i a m e t e r , t h e area r a t i o and e q u i v a l e n t cone a n g l e a r e based on t h e
e x i t and minimum f l o w areas, and u(28 - 21.5') is t h e u n i t s t e p f u n c t i o n .

11. I n t e r n a l f l o w o b s t r u c t i o n - d r a g i t e m : The l o s s due t o t h e d r a g of

i n t e r n a l s t r u c t u r e such a s s t r u t s o r models h a s t h e form

(appendix A)

1 2 . P e r f o r a t e d p l a t e : P e r f o r a t e d p l a t e w i t h sharp-edged o r i f i c e s , used
a s p r o t e c t i o n s c r e e n o r as s c r e e n around the i n l e t of a n o n - r e t u r n t u n n e l ,
produces l o s s e s g i v e n by

13. Mesh s c r e e n : The l o s s e s produced by a mesh s c r e e n may b e e x p r e s s e d

as

K = K R N K ~ S H (1 - AFLOW +
7) - ("AFLOW
1)2 ( r e f . 7 , p. 308)

where t h e Reynolds number i n f l u e n c e f a c t o r , KRN, i s p l o t t e d a g a i n s t Reynolds

number (based on mesh d i a m e t e r ) i n f i g u r e 5, and t h e mesh c o n s t a n t , KMEsH, i s
1 . 3 f o r a v e r a g e c i r c u l a r m e t a l w i r e , 1 . 0 f o r new metal w i r e , and 2 . 1 f o r s i l k
thread.

1 4 . Sudden expansion: For a sudden expansion w i t h d u c t i n g downstream

( t o a l l o w r e a t t a c h m e n t of t h e f l o w and maximize t h e p r e s s u r e r e c o v e r y ) t h e
loss is

( r e f . 7 , p. 128)

15. Vaned d i f f u s e r s : The p r e s s u r e l o s s of a vaned d i f f u s e r , one i n

which s p l i t t e r vanes are used t o improve the performance of a s h o r t d i f f u s e r
by d e c r e a s i n g t h e e f f e c t i v e e q u i v a l e n t cone a n g l e of e a c h chamber, may b e
determined from

K = ( 0 . 3 + [0.006(2e -
21.50) 1}(AR& '> 2

(appendix A)

where u(28 - 21.5') i s t h e u n i t s t e p function. (See f i g . 6 . )

16. Fixed, known l o s s : For a f i x e d l o s s i t e m where t h e p r e s s u r e l o s s

v a l u e i s known, t h a t v a l u e may b e used d i r e c t l y by d e f i n i t i o n :

9
 17. Multiple ducts: In a system of multiple ducts, where the local flow
passes through two or more separate, identical passages at the same time, the
losses have the same value as those for the same type of single duct. Some of
the pertinent parameters, such as hydraulic diameter and equivalent cone angle,
should be based on the geometry of one of the individual ducts. The loss for
the system of ducts may then be determined from the l o s s for a single duct:

18. Loss value transferred to reference location: Each local l o s s

parameter is calculated based on local conditions at the smallest-area end of
each section and may then be referenced to the test section conditions by the
formula

(appendix A)

13. Overall and summary performance: The energy ratio of the wind
tunnel under consideration is given by
1 (ref. 4 , p. 69)
ER =
N

i=1

The pressure difference across the wind tunnel walls, determining the minimum
required structural strength for each section, is given by

The power required to be input into the flow in order to drive the flow
through the wind tunnel at a specified test section speed is expressed as

(appendix A)

The actual drive power required is dependent on the efficiency of the

fan/motor system:

10
 COMPUTER PROGRAM DESCRIPTION

The computer program was written in FORTRAN IV language. It consists of

a main program which calls five subroutines and/or six library routines, as
required. Two of the subroutines are optional and may be abbreviated and
simulated in order to save execution time and/or memory storage space.

Method of Solution

The general technique used is outlined in the computer program functional

flow chart in figure 7. The program was developed in six functional units: a
main program and five specialized subroutines. The main program retains
general control over the computational flow and calls the subprograms as
required.

In the main portion (designated PERFQRM), at first entry into the program,
various section-shape geometry relationships and certain semi-empirical dif-
fuser, turning vane, and honeycomb l o s s functions are defined. The case title
card is read and checked for validity by specified code. The tunnel master
data control card is then read, checked for validity, and checked for content
of pertinent data by the data-checking subroutine. If any errors are found in
either of these two preliminary cards, error messages will be printed.
Although detected errors will not abort the computer run (unless a card of
improper format is encountered where not expected), the case under current
consideration will not be computed - only the checking of input errors will
then be performed on each section card. Prior to reading the section cards,
the units of measure (International System or U.S. Customary System) to be
used for the particulai- case are read. These units of measure are used as the
basis for the development of the appropriate flow parameters and test section
conditions.

The section cards are read and operated on one at a time. They are
checked for validity and input errors by the data-checking subroutine (called
DATACK) and the input information, if sufficiently complete, is then used in
the computation of the section upstream- and downstream-end geometries.
Adjustments to these geometry calculations are made for any multiple-ducted
sections. For diffusing sections where the ex7ansion loss parameter was not
input by the user, that parameter is generated from predefined functions.
Branching of the computational flow then transfers control to the appropriate
block of instructions for the remainder of the calculations which are peculiar
to the particular section under consideration.

After all section cards have been read in and operated on, each in turn,
a case termination card is encountered. The termination card specifies the
optional summary operations to be performed. The encounter of this card
11
signals the end of a case and triggers the final calculations. The codes
contained on this card determine the printing of velocity-optimizing and cir-
cuit summary information, the plottitg of the summary information, and the
return to the beginning for another case.

The data-checking subroutine evaluates the master and section input cards
for completeness of data (based on the requirements for the type of section).
Then, if any error was detected during computation of a case or if the
appropriate termination code was specified by the user, the complete set of
input data is tabulated. Messages about errors, omissions, or superfluous
information are included.

The subroutine SPEED computes the local Mach number based on local cross-
sectional area and determines the local flow velocity.

The subroutine FRICTN calculates the local Reynolds number, usually based
on the local duct hydraulic diameter, and the local friction coefficient for
smooth pipes.

The subroutine dUTPUT accepts the calculated section parameters along

with the section type codes describing the types of information to be output
and prints the section information according to the appropriate format.

The subroutine PLdTIT plots the summary information (cummulative total

pressure losses and/or wall pressure differential) versus circuit centerline
length if requested. The plot is scaled for centimeter or inch plot paper,
determined by whether International or U.S. Customary units are used for
computation.

The program is terminated after the last operations on a case for which a
no-return instruction on the termination card was given by the user.

Computing Equipment Required

Hardware and machine components- Although this program was written for
use on an IBM 360/67 with TSS Monitor, batch mode, an attempt was made to keep
it compatible with any system that uses FQRTRAN IV. No magnetic tapes were
used. In this version, input is made by cards and the data to be plotted are
stored on a disc file for plotting at a later time in an off-line mode. How-
ever, it is possible to use a typewriter-type terminal for conversational or
real-time computation, typing the data by card-image format, and plotting
immediately after the computation has been completed for a case.

The total core required for compilation on the IBM 360/67 was approxi-
mately 82 000 (decimal) bytes. If necessary this figure can be reduced by
eliminating two subroutines, DATACK and PLQTIT. The sizes of the main program
and of each subroutine were approximately as follows:

12
 PERF@M 38 800 bytes
DATACK 25 700
SPEED 800
FRICTN 900
OUTPUT 12 900
PL0TIT 2 900

The program was executed on an IBM 360/67, writing plot data on a disc,
logical unit 10. Later the data file was accessed from a 14.8 character-per-
second binary-coded-decimal terminal and plotted on a Zeta plotter with
0.005-in. step increments. The plot page size was programmed not to exceed
25 by 38 cm (or 10 by 15 in.).

S o f t w e e - This program was written for use on any computer with suffici-
ent core and with a standard FgRTRAN IV compiler.

The Zeta plotter routines, with minor exceptions, are compatible with the
Calcomp routines. The subroutines AXIS, FACTgR, LINE, PLflT, SCALE and SYMBQL
are alike in both Calcomp and Zeta plotting.

CALL AXIS - draws the axis line and annotates the divisions at every two
centimeters or each inch (depending on the units of measure specified).

CALL FACTflR - enables the user to produce normal size drawings with plotters
which have either 0.01- or 0.005-in. increment size. The variable FACT
must be set to 1.0 for 0.01-in. increments and to 2.0 for 0.005-in.
increment plotters.

CALL LINE - plots centered squares connected by straight lines through the
coordinate pairs of data values.

CALL PLflT - is used to establish
movements. Before plotting
desired along the X-axis.
Y-axis. The plot-page size
which are equated to 25 and
a new point of origin for the pen and paper
commences, the pen must be positioned where
The program will position it along the
is defined by the values of YLEN and XLEN
38 cm or 10 and 15 in., as required.

CALL PLgTF - is an alternate plotting initialization routine which is avail-

able in the Zeta but not Calcomp plot package. It is used in place of
PLOTS whenever deferred plotting is desired. The first argument in the
call statement indicates the speed of the terminal with which the plotter
is interfaced. The second argument is the logical-device number of the
plot file.

CALL SCALE - examines the data and determines the proper scaling for the given
dimensions of the plotter paper, 25 by 38 cm or 10 by 15 in.

CALL SMflDE - is available only in the Zeta plot package. It permits the user
to choose from extensive capabilities which affect several of the plotter
routines. In this program the options have been set equal to the usage
found in the Calcomp routines, and therefore, if Zeta plotter routines
are not available, the call to SM@DE should be eliminated.

I 13
CALL SYMB@L - prints the input case title at the top left of each plot page as
it appears in columns 2 through 80 of the title card. For reference pur-
poses, it also draws a small plus sign at the origin of the plot.

The library routines used are standard F@RTEUN routines:

ABS - Absolute value

AL@G10 - Common logarithm, base ten

ATAN - Arctangent (result in radians)

EXP - Exponential of the natural number e

IFIX - Convert from real number to integer

SIN - Trigonometric sine (argument in radians)

SQRT - Square root

Programming Techniques

It was intended that this program be usable on as many different computer

systems as possible. Therefore, in order to make them applicable to some
machines, certain statements were forced into particular forms which would be
less efficient on other systems (e.g., Hollerith instead of literals in format
statements).

C@MM@N and DATA statements were used as much as possible to simplify the
definition of parameter values. In the main program, arithmetic statement
functions were used for three purposes: (1) for the definition of section
hydraulic diameter, area, and equivalent cone angle geometry functions; ( 2 ) for
the conversion function from local to reference-section pressure losses; and
( 3 ) for the definition of the least-squares-polynomial-curve-fit functions.
The last group of functions includes: (1) the corner turning-loss parameters
as functions of turning angle (see fig. 3 ) ; ( 2 ) the diffuser expansion loss
parameters for the different cross-section shapes as functions of equivalent
cone angle (see fig. 4 ) ; and ( 3 ) the mesh screen Reynolds number sensitivity
factor as a function of mesh-diameter-based Reynolds number (see fig. 5 ) .

Certain functions not easily solvable in closed form were solved itera-
tively (some by Newton's method) to 0.01 percent accuracy. These functions
include test section Mach number, local section Mach number, and local section
friction coefficient.

Numeric codes were used for specifying such things as section type, sec-
tion end-shape types, and system of computational units; for decisions on
requirements for inputs to each section type; and for case-termination pro-
cedures and outputs desired. The various important input codes are listed in
tables 1 and 4 . All sections of the multiple-ducted type were assigned high
code numbers for simplicity in selecting them for special handling. The

14
various s e c t i o n types were grouped in code decades for reasonable association
of section code and component function. Where possible, the second digit of
the code (if that second digit is not zero) reflects the basic characteristic
of the section: constant area, contracting, or diffusing.

The information input fields on the master data and section cards were
arranged in three basic groups: (1) qualitatrve information (type and shape);
(2) quantitative geometry information (number of ducts, cross-sectional end
dimensions, and length); and (3) loss-related parameters. The case termination
card employs the same format as the section cards so that it may be encoun-
tered at random intervals without causing a program crash. For the tabulation
of the input data (for error-location and record-keeping purposes), object-
time formatting was used to compile the combination input and annotation data
set for a convenient output.

Much of the output of the program was set up on a demand (i.e., optional)
basis. A section-by-section performance analysis is automatically provided.
A brief summary of the variation of selected parameters through the circuit,
and plots of those parameters, may be selected if desired. An annotated list-
ing of the inputs may be requested or, if errors are detected, the listing is
internally forced in order to provide a simple means of error-detection and
correction and/or simplified record-keeping of case data.

Source Code

A source code listing of the performance estimation program is provided

in appendix C along with the associated notation definitions. The source code
includes the use of comment cards throughout the program for identification of
the operations carried out by each set of instructions.

Operating Instructions

The basic source program deck arrangement is shown in figure 8 .

Input- Sample coding forms for the four types of input cards required are
presented in figure 9. The special symbols required in the first columns of
the title and master data cards are included.

1. Title card: For the title labeling card, with the exception of the
first column which must contain an asterisk (*I, the entire card may be used
as desired. This title was programmed to appear at the top of each page of
the case to which the title refers, including the plots. Only one title card
per case may be used.

2. Master card: The tunnel master control data card provides sufficient
information for defining conditions in the test section (which is the refer-
ence section for all calculations) and conditions of the surrounding external
atmosphere. Table 2 details the inputs included on the master data card. The
first column must contain a minus sign in order to identify the card as a
valid master card. The remainder of the inputs should be positive, with

15
columns 2 through 6 containing five fields of integers only (no decimal points).
Columns 7 through 10 were not used on this card and should be left blank.
Columns 11 through 50 should contain floating-point numbers. These columns
were divided into eight parameters of five columns each, including decimal
point.

3 . Section cards: The individual section information cards were based

on the same format as the master card, except that the section cards require
no special identifying code. Table 3 details the inputs contained on the sec-
tion cards. The first six columns, containing four data fields, require inte-
ger inputs. The remaining 74 columns were divided into two real number fields
of two columns each (with the assumed default decimal points to the right of
the second columns), and 14 real number fields of five columns each (with the
assumed default decimal point between the third and fourth columns of each
field).
Although the input parameter requirements vary from section to section,
certain requirements are basic to all sections. These items include: (1) the
section type code, (2) the section end shape codes, and ( 3 ) the section dimen-
sions (end height(s) and width(s) and/or diameter(s) and usually length). A
detailed list of the additional, specialized requirements for each section is
presented in table 4 .

Although not mandatory in order to obtain a correct total power estima-

tion, it is advisable to input the section data cards in the actual section
order so that the summary calculations and plots have relevance to the,actual
circuit.

4 . Termination card: The case termination card, which signals the end
of the section inputs for a particular case, is identified by the constraint
of blanks in card columns 3 and 4 . The numbers contained on this card are
used strictly as task codes; table 5 shows the details of these codes. In the
event of a request for plotted information, the code determines the type(s) of
information to be presented. For all other tasks the codes dictate a simple
yes/no decision.

As many cases as desired may be input in a single job submission. The

same system of units need not be used in all cases. Any parameters may be
changed as desired from case-to-case since there are no forced carry-overs
(except the specific heat ratio, y, which is fixed at the time of program
compilation).

Output- Based on the foregoing input information the results may be

calculated and tabulated in five different types of information groups.

1. Section performance analysis: The section performance tabulation

fully describes the performance-related parameters of the wind tunnel circuit.
Atmospheric and test section flow reference conditions are stated at the top
of the first page. The various parameters are tabulated for each section in
the order of computation with the upstream end information on the first and
the downstream information on the second of the two lines for each section.
The section sequence number and type (a translation of the code) and the end
16
shapes are given first. The geometry and local velocity information are
presented next, followed by the section length and calculated total pressure
loss values.

2 . Overall performance: If no input errors are encountered during the

analysis of a case, overall performance values are presented at the end of the
section performance tabulation. This includes the total circuit length, the
total pressure losses and energy ratio for the circuit, and the total
operating power required.

3 . Summary characteristics: If requested on the termination card and

barring any errors, a summary of the circuit characteristics is tabulated on a
separate page. This tabulation includes section sequence numbers, Mach num-
bers, cumulative pressure losses, and local wall pressures, all as functions
of distance through the circuit.

4 . Plots: Under proper condition codes, the cumulative pressure losses

and/or the wall pressure differentials will be plotted as functions of dis-
tance through the circuit (centerline length). The straight lines that appear
on the plots connecting the points are for reference only and do not represent
the actual distribution in a component.

5 . Input data tabulation: Finally, if an input error was encountered

during the analysis of the circuit, or if such information was requested by
the user, the input cards are tabulated with annotations regarding missing or
superfluous inputs. A careful look at this section should allow the user to
discover why a given set o f input data did not produce the expected type of
results.

All of the foregoing types of output are shown for the test case.
Computer system restrictions- Certain restrictions and/or assumptions had
to be imposed on the computer system and its methods and abilities in order to
perform the performance analysis within reasonable time, effort, and money
constraints.

1. Hardware: This analysis was programmed for a moderate-sized system

with common components. No special hardware is required with the exception of
a plotter if the plotting option is used. The output printing device is
assumed to have available a minimum capacity of 120 characters per line, but
the number of lines per page may be set by means of the LINEMX parameter in
the main program. (Barring any special requirements, 45 lines for an 8.5-in.
page or 60 lines for an 11-in. page are recommended.)

2. Software: Certain software restrictions were imposed simply as a

starting point to the problem solution. The input card formats were fixed as
shown in figure 9. The specific heat ratio (y) and the number of lines per
output page were fixed for each compilation of the source deck, although
changes can be made by altering the values of G and/or LINEMX, respectively,
near the beginning of the main program.

17
 For reasons of possible memory limitations on smaller systems, the number
of wind tunnel components in each circuit case was limited to 30 sections.
This limit may be changed by assigning a new value to LMTSEC in the main pro-
gram and by re-dimensioning the following variables as denoted by "XX"' . in
the main program (PERF@RM), DELP(XX+2), SEKO(XX) , SEL(XX), SMACH(XX) ,
SSUMEL(XX+2) and SSUMKO(XX+2); in the data-checking subroutine (DATACK),
ENDATA(XX,20), NCHECK(XX,20) and NDATA(XX,4); and in the plotting subroutine
(PLOTIT), DELP(XX+2), SSUMEL(XX+2) and SSUMKO(XX+2). If memory limitations
are a severe problem and/or if computer-controlled plotting facilities are not
available to the user, the data-checking and/or plotting subroutines may be
11 removed" by inserting dummy, one-card subroutines with the same arguments

which would have no effect on the calculations. This would decrease the util-
ity and power of the program, but would retain the basic performance estima-
tion capabilities without crippling them altogether.

The plotting routines were written according to the requirements for a

plotter with 0.005-in. increments.

Optional inputs- Certain of the parameter inputs are designated as

optional and have built-in assumed default values in the event that the user
knows no better values than the ones provided in the sources referenced herein.
These optional parameters are shown in tables 2 through 4 .

On the master card (see table 2), the units of measure should be speci-
fied and an error message will be given if they are specified erroneously
(other than as type 1 or 2). However, the units code will default to 1 (the
International System) and case execution will continue. The test section and
atmospheric total pressures will default to one atmosphere if not specified.

On the section cards (see table 3), the number of items in the duct will
default to unity if not specified. The expansion loss parameter for diffusers
defaults to a value based on figure 4. (It is computed by determining the
shape of each end, the extent to which the diffuser is two-dimensional in
nature (i.e., changing cross-sectional size in height or width only), and the
equivalent cone angle, and then interpolating between the curves of figure 4.
See appendix A . ) The mesh screen loss constant defaults to 1.3, the value for
an average-condition metal mesh screen (ref. 6, p. 527), and the reference
Reynolds number for turning vanes defaults to 0.5 million (ref. 6, p. 527).
The surface roughness for honeycombs defaults to the appropriate equivalent of
0.00001 my the value for new, commercially smooth, non-steel pipe (ref. 7,
p. 62). The factor for the additional influence of a blockage on downstream
sections ( A E ) defaults to zero.

Diagnostic messages- There are a limited number of error diagnostic

messages which were built in to handle many, but not all, of the potential
user errors. The causes and appropriate corrections of these errors should be
evident in each message.

1. Title card: If a card is in the position of a title card and does

not begin with an asterisk as required, the following message will appear:

I 18
 ('...(invalid
L L L L Om
rnTrnT title)...') IS IXCGRRECT GR IWROPER AS IT EXISTS.
THE FIRST CARD COLUMN MUST CONTAIN AN ASTERISK (*) TO BE IDENTIFIED
AS A VALID TITLE CARD.

2. Master card: An invalid master card is denoted by:

MASTER CONTROL DATA ('...(card image)...') IS INCORRECT OR IMPROPER

AS IT EXISTS. THE FIRST TWO CARD COLUMNS MUST CONTAIN A NEGATIVE
NUMBER (-1 TO -9) TO BE IDENTIFIED AS A VALID MASTER CARD. THIS
CASE WILL BE SKIPPED.

A general omission from the master card of required information produces:

CRITICAL OMISSION(S) IN TUNNEL MASTER CONTROL DATA PREVENT EXECUTION

OF THIS CASE. ANY SUCCEEDING CASES WILL NOT BE AFFECTED.

Two master cards, back-to-back, for a given case are identified by:

MORE THAN ONE MASTER CONTROL CARD EXISTS FOR THIS CASE OR INPUT CARDS
ARE OUT OF ORDER. CHECK DECK SET-UP. THE LAST MASTER CARD ENCOUNTERED
WILL BE ASSUMED AS THE CORRECT MASTER CARD FOR THE SECTION O S WHICH
FOLLOW.

Encountering a master card where not expected (generally indicating missing

case termination and title cards) causes this message:

MASTER CONTROL CARD HAS BEEN ENCOUNTERED BEFORE CASE TERMINATION AND
TITLE CARDS. CHECK DECK SET-UP. ERROR-MESSAGE TITLE WILL BE GENER-
ATED AND SUMMARY OUTPUT, NO-PLOT, INPUT DATA TABULATION, AND NEXT-CASE
RETURN TERMINATION PARAMETERS WILL BE ASSUMED.

If an invalid test section upstream end shape geometry is specified, one which
the program cannot handle, an error results:

**ERROR -- INVALID TEST SECTION UPSTREAM END SHAPE CODE WAS SPECIFIED
AS (code used) (SHOULD BE 1, 2 OR 3). THIS CASE CANNOT BE EXECUTED.

If an invalid units code is specified the message is:

THE UNITS OF MEASURE CODE IS IMPROPERLY SPECIFIED AS (code used),

(SHOULD BE 1 OR 2). CHECK MASTER CARD (COLUMN 4 ) . SEE THE DATA
TABULATION AT THE END OF THIS CASE. THE INTERNATIONAL SYSTEM OF
UNITS WILL BE ASSUMED FOR THIS CASE.

If the termination code requests power-matching but the input power value is
such that the calculation would be meaningless, a diagnostic of the following
form is printed:

**ALTHOUGH VELOCITY-OPTIMIZING WAS REQUESTED BY TERMINATION CODE, THE

INPUT POWER VALUE IS ILLEGAL (LESS THAN OR EQUAL TO ZERO). THEREFORE,
NO VELOCITY-OPTIMIZING IS POSSIBLE. RECHECK INPUT VALUE ON MASTER
DATA CARD.
19
 3 . Section card: A general omission of required data from a section
card will cause this message:

**ERROR --CRITICAL OMISSION(S) I N SECTION INPUT DATA. SEE DATA

TABULATION AT END OF OUTPUT FOR T H I S CASE.**

If an invalid section shape code is specified it is not possible for the

program to properly compute section end geometries; as a result an error
occurs:

**ERROR --INVALID SECTION SHAPE CODE WAS SPECIFIED AS (input code)

(SHOULD BE 1, 2 OR 3 ) . T H I S SECTION WILL BE SKIPPED.

An error which arises during computation and causes a non-positive total

pressure loss for a given section prevents completion of the case analysis and
gives rise to an error message:

**ERROR --SOME INCORRECT COMBINATION OF INPUTS OR UNANTICIPATED

SITUATION HAS CAUSED AN INVALID (NON-POSITIVE) TOTAL LOSS LEVEL.
RECHECK SECTION (section number) INPUT DATA.

If the maximum allowable number of circuit sections written into the program
is exceeded by placing too many section cards together in one case, or without
termination, title, and master cards between cases, this diagnostic will
appear:

MAXIMUM LIMIT ON THE NUMBER OF SECTIONS (...(maximum allowable number

of section) ...)
HAS BEEN REACHED. EITHER A CASE TERMINATION CARD HAS
BEEN OMITTED (ALONG WITH TITLE AND MASTER CARDS TO BEGIN A NEW CASE)
OR THIS CASE I S TOO LONG FOR THE PROGRAMMED ALLOWABLE NUMBER OF
SECTIONS. THE CASE HAS BEEN TERMINATED AT THIS POINT.

In this instance, the inputs from the group of sections for which the limit
was exceeded will be tabulated and the remaining section inputs will be
evaluated and tabulated. I f the user fails to cause the test section blockage
amounts specified on the master control card to coincide with that of the test
section card, erroneous analysis may result since inconsistent flow areas
would be calculated. The section card value will be used (since the discrep-
ancy may be desired) and this notice is given:

**NOTE -- TEST SECTION BLOCKAGE FROM SECTION CARD INPUT (...(section

input value) ...
PERCENT) DOES NOT EQUAL THAT OF THE MASTER CARD INPUT
(...(master input value)... PERCENT). CHECK DATA DECK. SECTION CARD
VALUE WILL BE ASSUMED AS CORRECT AND EXECUTION WILL CONTINUE.

An invalid section type code will cause a section to be skipped and a message
to be printed:

*ERROR --
INPUT SECTION TYPE CODE (CARD COLUMNS 3 AND 4 ) CALLS
INVALID SECTION TYPE. DATA CARD IGNORED.**

20
 Any iiipilt errors were deemed justifiable came for judgment as an incmplete
case. As a result, reliable overall and summary information cannot be calcu-
lated. To assist the user in locating the error(s), the input values will be
forced to be tabulated and the following explanation appears:

***DUE TO ERROR(S) IN INPUT CARD(S), VALID SUMMARY INFORMATION IS NOT

AVAILABLE. REFER TO THE TABULATION OF INPUT DATA ON THE FOLLOWING
PAGES. CORRECT THE ERROR(S) AND RESUBMIT THIS CASE. SUBSEQUENT CASES
WILL NOT BE AFFECTED.

4 . Possible errors lacking diagnostics: Certain potential problem areas

remain unprotected by diagnostic and error-recovery systems.

No special provision was made for two test sections in the same circuit
case. As long as the blockage values for both test sections match the one
from the master card, no message will be printed. In any event, the execution
will not be terminated. The test section shapes and dimensions from the
master card are not checked against those of the test section card. Although
a mismatch of these values could cause a mass-flow error, including and
enforcing such a check could inhibit any meaningful tandem-test section cases.
These problems could be avoided, however, by naming only one working section
as a test section and referring to the other by general type.

Also, there was no provision for checking the specified tunnel type
against the types of sections actually used (e.g., checking a non-return, or
open-test-section tunnel for exit or open-throat test section input cards).
This check is not critical and was left to the user.

One error-check was not included due to the program complications which
would have resulted. If a case termination card is omitted at the end of a
case and a computer-system control card or a title card is encountered, the
error will be disastrous due to mismatched format types. Execution and
calculations will be immediately aborted by the computer.

Test Case

The NASA-Ames Research Center 40- by 80-Foot Wind Tunnel was used as an
example of a typical wind tunnel. This tunnel, illustrated in figure 10, is
of the single-return, closed-test-section, continuous-running type. It has a
flat-oval test section 12.2 m ( 4 0 ft) high by 24.4 m (80 ft) wide and is
powered by six 12.2-m (40 ft) diameter, six-bladed fans. It has an eight-to-
one overall contraction ratio and uses multiple-circular-arc type turning
vanes in each of the four 90' corners.

A complete list of the test case inputs and computed information outputs
are presented in figure 11. The machine computing time for this test case
,I (without plots) was about 7 sec on an IBM 360/67.

Although this test case was not an exhaustive exercise of all possible
tunnel components, it does include most of the basic section types: diffusing
test section, single-duct contraction and diffuser, constant-area single duct,
21
c o n s t a n t - a r e a c o r n e r w i t h t u r n i n g vanes, and m u l t i p l e - d u c t f a n s e c t i o n s
( c o n t r a c t i o n , c o n s t a n t - a r e a annulus, and d i f f u s e r ) . Examples of o t h e r t y p e s
of components a r e shown i n t h e sample c a s e s which f o l l o w .

DISCUSSION AND APPLICATIONS

Wind t u n n e l energy r a t i o , r e q u i r e d power, and o p e r a t i n g v e l o c i t y a r e

i n t e r d e p e n d e n t . The energy r a t i o i s a f f e c t e d by v e l o c i t y through t h e e f f e c t
of v e l o c i t y on t h e Reynolds number. The r e q u i r e d d r i v e power, i n f l u e n c e d
d i r e c t l y by o p e r a t i n g v e l o c i t y and i n v e r s e l y by energy r a t i o , i s a l s o con-
t r o l l e d by t h e f a n system e f f i c i e n c y which i s o f t e n o n l y a n e s t i m a t e d q u a n t i t y .
Any e s t i m a t e of o p e r a t i n g v e l o c i t y f o r a given power l e v e l i s , t h e n , dependent
on t h e b a s i c e f f i c i e n c y of t h e c i r c u i t (energy r a t i o ) and d r i v e system e f f i -
c i e n c y , assuming t h e b e s t power estimate a v a i l a b l e t o b e t h a t d e l i v e r e d t o t h e
f a n s . This interdependency means t h a t a n e r r o r i n t h e p r e d i c t i o n of energy
r a t i o (and/or i n t h e e s t i m a t i o n of f a n e f f i c i e n c y ) w i l l c a u s e corresponding
e r r o r s i n power and v e l o c i t y estimates.

These e r r o r s r e s u l t i n g from an e r r o n e o u s p r e d i c t i o n of t h e c i r c u i t energy

r a t i o c a n b e found from t h e r e l a t i o n s h i p governing r e q u i r e d power, t e s t sec-
t i o n v e l o c i t y , and energy r a t i o , assuming given motor electrical and f a n
e f f i c i e n c i e s . For a f i x e d t e s t s e c t i o n v e l o c i t y ,

a P - 1~
- 1
~ ~ ~ ~
PREQUIRED 1 --
AER
ER
and, f o r c o n s t a n t power, a n e r r o r i n energy r a t i o y i e l d s t h e performance
penalty

-AVO
=
VO +-E) 1/3

The expected t r u e power and v e l o c i t y levels can t h u s b e o b t a i n e d from t h e

performance estimate:

f o r a given s e t of test s e c t i o n c o n d i t i o n s , and from

1/ 3
"E st i m a t e

f o r a given l e v e l of r e q u i r e d power.

22
 This adjustment of the estimated performance values is pointless for a
known, existing wind tunnel, but necessary for new, or proposed facilities.
Before the adjustment can be made the probable error in the energy ratio esti-
mate must be determined. It is desirable, therefore, to consider several
existing facilities of different circuit types in order to gain a degree of
confidence in the performance estimation routine.

Results

The input parameters and output performance values for the several sample
cases, other than the test case shown in figure 11, are compiled in appendix D.
The estimated energy ratios for the seven sample wind tunnels are presented in
table 6 . The corresponding sketches for all these sample tunnel circuits are
shown in figures 10 and 12.

The actual energy ratios for the first three wind tunnels presented in
table 6 were estimated from the best available information on fan and electri-
cal efficiencies from known input power levels. The actual performance of the
other four facilities was taken from measured data.

The test case and first sample case was the circuit of the NASA-Ames
Research Center 40- by 80-Foot Wind Tunnel as described previously in the test
case discussion. The predicted energy ratio for this rather conventional
tunnel was only 1 percent higher than the actual value when new.

The performance of the NASA-Ames 7- by 10-Foot Tunnel was predicted at a

slightly optimistic level. However, this tunnel is one with several known
problems which complicate the prediction process. With the air exchanger oper-
ating, the primary diffuser is known to have some local flow separation,
having been designed at a 6 O equivalent cone angle, an angle too great for its
cross-sectional shape and length (see fig. 1). Also, the drive fan is stalled
from the centerbody out to about 45 percent of the fan radius, causing some
back-flow along the nacelle centerbody. (The impact of the stall on the fan
efficiency has only been estimated; it was assumed that the fan efficiency
would suffer by about an additional 10 percent.) In spite of these things,
the predicted energy ratio was only about 3 percent too high relative to the
original value of approximately 7.85, both values taken in the zero-air-
exchange configuration. This agreement may indicate that much of the above-
mentioned off-design performance is triggered by the air exchanger operation
and is not as significant with the air exchanger closed. Although insuffici-
ent data are available to resolve this question, the fact remains that the
prediction accuracy, for the stated conditions, was good.

The Lockheed-Georgia Low-Speed Wind Tunnel employs a tandem test section

design. For this analysis, the larger, V/STOL test section was used as the
only reference station. Because of the two area restrictions to cross sec-
tions smaller than that of the reference area (those of the smaller test sec-
tion and of the fan), the tunnel efficiency would be expected to be low.
(This in no way reflects on the tunnel's usefulness as a research tool or on
its design or capabilities. The ''low efficiency" value results only from the
point of reference used in the calculations.) In other than these features
23
the facility is basically of conventional design. The computerized
performance prediction was in error by less than 2 percent from the true value
of about 1.10.

The Indian Institute of Science 14- by 9-Foot Tunnel at Bangalore stands

out among non-return wind tunnels as a facility with an unusually high energy
ratio. Although the determination of circuit dimensions for the program input
was somewhat hampered by the limitations of small drawings, the estimated
energy ratio was within 1 percent of the true facility value of 6.85. It is
interesting to note that the fan performance data of reference 23 would indi-
cate a fan design efficiency of about 69 percent. The power requirement calcu-
lations based on energy ratio and test section maximum velocity, however, show
that the fan efficiency must be higher than was expected; in fact, greater
than 90 percent.

The Hawker Siddeley 15-Foot V/STOL Tunnel at Hatfield, England was con-
structed under economy constraints and is a compact, cost-effective facility.
The estimated performance was about 1.6 energy ratios higher (i.e., more
optimistic) than the actual value of 2 . 3 8 . This is an error of about 67 per-
cent. The primary performance difference was probably caused by the fan sys-
tem. The losses of the ducting in this area are difficult to predict because
the area changes are not gradual and are even difficult to define.

The University of Washington 8- by 12-Foot Double-Return Tunnel has a

surprisingly high measured energy ratio of 8 . 3 . This would indicate a very
carefully designed circuit powered by carefully designed fans. The perform-
ance estimate produced by the computer program is lower than the actual energy
ratio by about 13 percent, showing that the achieved performance level is
higher than would normally be expected.

The NASA-Langley Research Center 30- by 60-Foot Open-Throat Tunnel is

unusual in configuration, having a double-return system with the twin fans
located less than two fan-diameters downstream of the test section. The loca-
tion of the data point for this tunnel on the diffuser design curves of fig-
ure 1 would not indicate that any diffuser-related problems should be expected
forward of the fans. The diffuser between the fans and the first corner, how-
ever, does have a rather large equivalent cone angle (more than 8 ' ) . If the
fans cause or contribute to diffuser flow problems (see the Cautionary Design
Guidelines for Diffusers) and if those problems lead to corner flow ineffici-
ency in a region critical to overall performance, then the circuit energy
ratio may be well below the normal estimated level. Although it is not clear
whether this is the case in the NASA-Langley 30-by60-Foot Tunne1,the perform-
ance estimate was about 27 percent higher than the actual value of about 3.71.

Evaluation

To summarize what may be learned from the sample cases:

1. The Ames 40- by 80-Foot and 7- by 10-Foot Tunnels and the Lockheed-
Georgia Low-Speed Tunnel, although at opposite ends of the energy ratio

24
spectrum, are all basically standard, single return, closed-test-section
facilities; the computer program estimates of actual performance were good.

2. The Indian Institute of Science Bangalore tunnel, being of the non-

return variety, is a different and less common type of facility; the computer
program closely estimated its actual performance.

3 . The University of Washington double-return tunnel is a third major

circuit type; the program produced a reasonably accurate prediction of its
performance.

4. The Hawker Siddeley V/STOL and Langley 30- by 60-Foot Tunnels are
examples of facilities which may have flow problems due to too-rapid area
changes and, as a result, lower than optimum performance levels for their
respective circuit types. For these tunnels, because of their flow quality
and not because of their circuit types, the program provided a poor estimate
of actual performance.

Based on these results one thing is immediately clear: the performance

of a wind tunnel of conventional, conservative design can be evaluated accu-
rately. On the other hand, the performance of a tunnel whose design generates
or contributes to flow problems (separation or grossly non-uniform) will be
overestimated by the loss equations and computer program.

Flow peculiarities and off-optimum designs, even though seemingly only

slight, can cause operational performance to fall significantly below the pre-
dicted levels. Such problems can be expensive whether considered in terms of
modifying the facility or in such terms as reduced testing capability and
increased power costs. Judicious, iterative use of the estimation techniques
presented in this report, simplified by computerized automation, can lead to
the improvement of an existing facility through guidance of design changes or
to the optimization of a proposed new wind tunnel design.

Ames Research Center

National Aeronautics and Space Administration
Moffett Field, California 94035, January 8, 1976

25
 APPENDIX A
~

1 NON-STANDARD FUNCTIONAL FORMS

Due t o the nature of this analysis, certain of the local flow-state, sec-
tion loss, and summary parameter formulas were used in a form more convenient
than that usually found in the literature. The relationships which were
altered or derived are outlined on the following pages.

I
I Local Flow-State Parameters

The calculation of several local parameters was based on the local Mach
number, determined from the relationship between the local area and the area
for choked flow:

I S o l v i n g f o r the area for choked flow, knowing the test section area and Mach
number ,
v+l
+ 1 2(Y-1)

2 [1+(9
M
O
2
)]

Mach nwnber- Another form of the same area relationship,

3%

can be rewritten to produce a polynomial equation in Mach number which may be

solved b y Newton's method if the areas are known:
y-l

M)
+- 9] 2
Y - 1
=

26
 ReynoZds number- The local Reynolds number was calculated based on other,
known, local conditions and from basic principles:

m = P-
u
(conservation of mass)

(ref. 18, p. 19)

(ref. 18, p . 4 )

Friction coefficient- The Reynolds number-friction coefficient function

used was

-
1 = 2 loglo(RNfi) - 0.8 (ref. 6 , p. 70)
fi

A Newton's method solution was performed on a rewritten form of the equation:

[log10(ARN2) - 0.8]-2 - X = 0

Section Pressure Losses

The losses for some types of sections were derived in forms not found in
the literature. For others, a curve-fit of data points or a simplification of
analysis was performed.

Corners (constant area)- The frictional and rotational losses through

turning vanes are additive: K = K~ICTION+ KROTATION. Also,

Assuming that the frictional loss value has a form similar to that for a flat
plate, then at 9 0 " :

1 -- 0.455B (ref.6, p.527)

KFRICTION = 7K T V ~ ~
(1ogl0 RN) Z 58

27
 Thus, t h e c o n s t a n t B i s dependent on t h e t u r n i n g vane l o s s c o n s t a n t and t h e
r e f e r e n c e Reynolds number a t which t h a t c o n s t a n t w a s determined:

1 2.58
7 KTVgO(log10 RNREF)
B =
0.455

I Therefore,

KFRICTION =
1
7 KTVgO ( l o g l o RNREF
loglo RN

S i n c e t h e r o t a t i o n a l term i s assumed independent of Reynolds number,

KROTATION = (2/3)KTvgo. The a d d i t i o n a l c o m p l i c a t i o n of l o s s parameter varia-
t i o n w i t h t u r n i n g a n g l e i s p r e s e n t e d i n f i g u r e 3 f o r a l o s s parameter a t 90'
e q u a l t o 0.15. It w a s assumed t h a t t h e r e l a t i o n s h i p between t h e a c t u a l and
reference loss constants is linear:

where f ( $ ) i s t h e f u n c t i o n a l r e l a t i o n s h i p p l o t t e d i n f i g u r e 3. The complete

t u r n i n g vane l o s s f u n c t i o n t h e n becomes

Corners (diffusing)- D i f f u s i n g c o r n e r s were t r e a t e d as vaned d i f f u s e r s

w i t h t h e a d d i t i o n of r o t a t i o n a l l o s s e s dependent on t h e t u r n i n g a n g l e . The
expansion and f r i c t i o n a l l o s s e s used were t h o s e f o r a vaned d i f f u s e r :

I The r o t a t i o n a l l o s s i s as f o r a c o n s t a n t - a r e a c o r n e r where

- 2
KROTATION - 7 KTv =

The d i f f u s i n g c o r n e r l o s s f u n c t i o n i s t h e n

where u(28 - 21.5') is t h e u n i t s t e p function.

28
 ,R*fft?aers- The diffuser lmses are due to both friction and expansion.
The friction term may be derived theoretically from

Making the simplifying assumptions that the density and the friction coeffi-
cient are approximately constant and applying conservation of mass,

which, for a conical diffuser, becomes

KFRICTION -
2
,

and transforming variables from surface to centerline distances,

dx

Completing this integration the friction loss becomes

The influence of the expansion term is given by

K = KEXPANSION + KFRICTION
Thus, it may be rewritten:

K =
KEXPANSION + KFRICTION (1 -

29
The expansion loss parameter curves shown in figure 4 were determined using
the approximation

K - KFRICTION (CONICAL)
=
'>'

and figure 5(a) of reference 9, which shows complete diffuser losses plotted
as functions of equivalent cone angle and independent of area ratio for circu-
lar, square and rectangular, and two-dimensional diffusers. (This implies an
assumption that the expansion part of the losses is dependent only on cross-
sectional shape, the extent to which the diffusion takes place in only one
direction, and the equivalent cone angle.) Thus, the complete loss for
diffusers is given as

using KED from figure 4 .

Exit- The kinetic energy loss at an exit of a non-return wind tunnel was
derived from basic compressibility relationships and with the assumptions that
the exit flow static pressure is equal to the atmospheric pressure and that
the exit velocity is uniform.
(rewritten
fromref.17,p.53)

Rewriting, the local total pressure is

V
Also, since ApT = pT - PTATM
-- PT - p, the total loss parameter is

since

= I2 pv2 1
= -2 YPM2 (ref.17,~.55)

Simplifying, the exit loss becomes

F Z m straighteners: a i r f o i l members (thick)- Thick flow straightener

losses were assumed to be made up of two parts: contraction and subsequent
diffusion:

The contraction was estimated as being about 30 percent of the length of the
straighteners:

0.32h (0.30L)
KCONTRACTION =
Dh

The diffusion portion was based on the aft 70 percent of the length and on the
exit and minimum areas for the computation of the area ratio and equivalent
cone angle. As for a vaned diffuser,

Hence the l o s s for thick flow straighteners becomes

31
 Internal flow obstruction: drag item- The loss due to internal structure
may be derived from the relationships governing power losses:

PINPUT~~G
- K~~~~ P 02 ~ 0 ~ 0 3
2pF

and PDWG = DVE = (1/2)pv3scDE, where E is the factor accounting for addi-
tional effects on downstream sections. Since PIN PUT^^^
= PDMG, the loss
becomes

and therefore

Since in general the flow density at the fans is unknown at the time a given
section l o s s is calculated, and since for incompressible flow the density
ratio is unity, the ratio of the densities at the fan and the local station
was assumed as unity for the analysis. (If the user prefers not to make such
an assumption, an approximation of the ratio may be made by way of a change in
the downstream influence factor E.) The loss due to a flow obstruction is

Vaned d i f f u s e r s - The expansion and friction losses for vaned diffusers

were combined into one parameter which is reasonably independent of area ratio
and is presented in figure 6 . The loss curves shown were approximated by a
two-segment, straight-line curve fit so that, for vaned diffusers

where u(28 - 21.5') is the unit step function.

Loss value transferred t o reference location- The change of reference for

l o s s values is defined as

32
Using the law of conservation of mass, this may be rewritten in terms of areas
and Mach numbers:

1
pv2 AoV
a
Qo
=-
=-
1 p V 2 AVO
2 0 0

and

Wall pressure differential- The pressure across a section of wall was

determined from the exterior atmospheric pressure, internal static pressure,
and cumulative pressure losses through the circuit. Since the wall pressure
differential for a given section is Apwi = p~~~~ - pi and

PT:

and, using the test section as the reference location,

The wall pressure differential may be written as

33
 Input power required- The power i n p u t t o t h e flow r e q u i r e d f o r o p e r a t i o n
of a wind t u n n e l c i r c u i t having predetermined l o s s e s w a s c a l c u l a t e d from t h e
p r e s s u r e rise r e q u i r e d a t t h e f a n s , w i t h t h e s i m p l i f y i n g assumption t h a t t h e
s t a t i c and t o t a l p r e s s u r e rises a c r o s s t h e f a n are e q u a l .

FpoAovo
PINPUT = APT~AFVF PFPoAoVo

Considering c o n s e r v a t i o n of mass,

Also,
N

Thus,

34
 APPENDIX B

NUMERICAL FUNCTION-APPROXIMATIONS

The formulas t h a t f o l l o w r e s u l t e d from c u r v e - f i t t i n g a n d / o r i n t e r p o l a t i o n

t e c h n i q u e s a p p l i e d t o c e r t a i n f u n c t i o n s a r i s i n g from t h e l o s s a n a l y s i s .

Corners

The c o r n e r loss p a r a m e t e r s f o r c o r n e r s w i t h and w i t h o u t t u r n i n g vanes are

shown i n f i g u r e 3. For a c o r n e r w i t h vanes, u s i n g l e a s t - s q u a r e s polynomial
c u r v e - f i t t i n g t e c h n i q u e s , t h e t u r n i n g vane l o s s f u n c t i o n of f i g u r e 3 ( a )
becomes, f o r 0" I 8 I 30":

KTV = 1.395066X10-2 + 5 . 6 7 2 6 4 9 ~ 1 0 - ~(8

+ 7 . 0 8 1 5 9 1 ~ 1 0 ' ~g2 + 1 . 3 9 4 6 8 5 ~ 1 0 - ~g3

-4. 885101X10-8 O4 (B1)

and f o r 30" < 8 I 90":

KTV = -1.605670X10-1 4- 1.446753X10-2

- 2 . 5 7 0 7 4 8 ~ 1 0 - ~d2 + 2 . 0 6 6 2 0 7 ~ 1 0 - ~g3

-6.335764~10-' d4 032)

For a c o r n e r w i t h o u t t u r n i n g v a n e s t h e l o c a l l o s s f u n c t i o n of f i g u r e 3 ( b ) w a s
found u s i n g a l e a s t - s q u a r e s polynomial t e c h n i q u e and is g i v e n by

K = 4 . 3 1 3 7 6 1 ~ 1 0 - ~- 6 . 0 2 1 5 1 5 ~ 1 0 - ~8

+ 1 . 6 9 3 7 7 8 ~ 1 0 - ~a2 - 2.755078~10+ 83
+ 2 . 3 2 3 1 7 0 ~ 1 0 - ~f14 - 3 . 7 7 5 5 6 8 ~ 1 0 - ~g5

+ 1 . 7 9 6 8 1 7 ~ 1 0 - ~ ' f16 (B3)

For a l l t h e above e q u a t i o n s , 8 i s t h e flow t u r n i n g a n g l e i n d e g r e e s .

Diffusers

The d e t e r m i n a t i o n of t h e d i f f u s e r l o s s p a r a m e t e r i s a complex o p e r a t i o n .
It depends on t h e c r o s s - s e c t i o n a l shape and e q u i v a l e n t cone a n g l e of t h e sec-
t i o n . For a c o n i c a l d i f f u s e r t h e expansion f u n c t i o n s a r e , f o r 3' L 2 0 I10':

35
 KEXPCircular
= 1.70925x10-1 - 5. 84932X10'2 (20)

+ 8 . 1 4 9 3 6 ~ 1 0 ' ~(20)2 + 1 . 3 4 7 7 7 ~ 1 0 - ~( 2 0 ) 3

- 5 . 6 7 2 5 8 ~ 1 0 - ~(20) - 4 . 1 5 8 7 9 ~ 1 0 - ~(20)

+ 2 . 1 0 2 1 9 ~ 1 0 - ~(20) 034)

for 0" < 20 < 3":

K E n c ircu 1ar
-
= 1 . 0 3 3 3 9 5 ~ 1 0 - ~ 1 . 1 9 4 6 5 ~ 1 0 - ~( 2 0 ) (B5 1

and for 28 > 10":

KEXPCir cular
= -9.66135~10-~ + 2 . 3 3 6 1 3 5 ~ 1 0 - ~(20) (B6)

For a square cross-section diffuser the expressions are, for 3" I 2 0 I 1 0 " :

= 1 . 2 2 1 5 6 ~ 1 0 ' ~- 2 . 2 9 4 8 0 ~ 1 0 - ~(20)
KE*Square
-
+ 5 . 5 0 7 0 4 ~ 1 0 - ~( 2 € 1 ) ~ 4 . 0 8 6 4 4 ~ 1 0 - ~( 2 0 )

- 3 . 8 4 0 5 6 ~ 1 0 - ~(20) + 8 . 7 4 9 6 9 ~ 1 0 - ~( 2 0 )
- 3 . 6 5 2 1 7 ~ 1 0 - ~(20) (B7 1

for 0" < 20 < 3":

KExp square
= 9.62274~10'~ - 2 . 0 7 5 8 2 ~ 1 0 ' ~ (20) (B8)

and for 28 > 10":

KEXPSquare
= -1.321685~10'~ + 2 . 93315X10-2 (20) 09)

For a two-dimensional diffuser with a square upstream-end cross section the

expansion l o s s functions are, for 3" 5 2 0 I 9":

= 3.23334~10-1 - 5 . 82939X10'2 (20)

KEXP 2DRectangular
- 4 . 9 7 1 5 1 ~ 1 0 ' ~ (20) + 1 . 9 9 0 9 3 ~ 1 0 ' ~(20)

- 1 . 9 8 6 3 0 ~ 1 0 - ~( 2 0 ) 4 + 2 . 0 6 8 5 7 ~ 1 0 ' ~ (2015

+3.81387~10+ (20) 15 (BIOI

KEXP = 5.72853 - 1.21832 (20)

2DRec tangular
+ 7 . 0 8 4 8 3 ~ 1 0 ' ~ (20)

36
 = - 5.333333~10'~ (29)
1.0~10-~
KEXP2DRec tangular
and for 28 > 10':

Since the expansion function for a two-dimensional diffuser with circular

sides was not given (and is not defined), it was assumed for computational
purposes that this value would be the same fraction of that for a two-
dimensional rectaFlgular diffuser as the l o s s of a conical is of that for a
three-dimensional square diffuser:

-
KEXP
2% ircular KEXP2DRectangular

For cross-section shapes somewhere between rectangular and circular, such as

flat oval (flat ceiling and floor with semi-circular sidewalls), or for dif-
fusers which have one end rectangular and one end either circular or flat oval,
a loss value between that for circular and rectangular may be more appropriate;
thus ,
KEXP + KEXI?
-- 2DRectangular '"Circular
K EXP2DAverage 2

and

- KEaSquare + KEmCircular
KEXP DAverage 2

The extent to which a diffuser is planar in nature was computed from the
ratio of the changes in size of the two characteristic dimensions from end to
end :

6, = smaller of
h2 - hl or w2 - w1
w2 - w1 h2 - hl

or if the ratio is negative,

6, = 0

Then, based on the geometries of each end, the basic loss constant, KEDBasicy
and the addi-
may be selected Erom K E X P ~ KEXP ~ ~ ~ or~ KEmSquare
~ ~3DAverage ~ ,
tional l o s s fact. may be selected from the corresponding
KEXPAdditional '

37
KEXP
2% i r c u ar
~
3 KEXP
2DAverage
Or
2DRect angu l a r
. Finally, the applicable

d i f f u s e r expansion loss c o e f f i c i e n t i s g i v e n by

Mesh S c r e e n

The mesh s c r e e n Reynolds number s e n s i t i v i t y f a c t o r p l o t t e d i n f i g u r e 5

c a n b e expressed i n f u n c t i o n a l form a s , f o r 0 I RJ < 400:
78.5(1 - 7s~;)
RN
Km =
100
+ 1.01

and f o r RN 1 4 0 0 :

Vaned D i f f u s e r s

The vaned d i f f u s e r loss c o e f f i c i e n t f u n c t i o n s p l o t t e d i n f i g u r e 6 were

approximated by two l i n e segments; f o r 28 < 21.5':

Kv = 0.3

and for 21.5' L 28 I_ 90':

K, = 0 . 3 + [0.006(28 - 21.5')]

Thus, over t h e e n t i r e r a n g e of e q u i v a l e n t cone a n g l e s of i n t e r e s t ,

K, = 0.3+ [0.006(28 - 21.5')~(28 - 21.5'11

where u(28 - 21.5') is the u n i t s t e p function.

38
 !l??E?!DIX c

COMPUTER PROGRAM FdRTRAN CODES

The following pages contain the F8RTRAN codes developed to implement the
wind tunnel performance analysis techniques presented in this report.

The Notation section explains the variable names used in the program.
(Note that in the notation sections, as throughout this report, all letter 0 ' s
occurring in FflRTRAN names are shown with slashes, as 8 ; all number zeros are
shown unslashed.) This notation section is similar to that for engineering
symbols presented in the main body of the report, but this section was changed
in two respects. First, it was rearranged alphabetically by F8RTRAN variable
name. Second, it was expanded to include many variable names which were not
used elsewhere and which have significance only in the context of the computer
program. The "titles" shown in parentheses in the first column of this nota-
tion section are column heading titles which appear on the program output
pages.

Immediately following the Notation are the listings of the six actual
FORTRAN program codes: the main program (PERF8RM) and the five subroutines
(DATACK, SPEED, FRICTN, OUTPUT, AND PL8TIT). Each program routine page is
titled and numbered for clarity. The last seven columns of each line on each
page contain a two-letter program routine name abbreviation and a line
sequence number (in ten-count increments). Thus, the user can know at a
glance to which routine (and where within that routine) any given line or
instruction belongs. Each instruction line in the program is uniquely
identified.

39
 NOTATION (FflRTRAN)

F~~RTRAN
name Engineering
and/or (title) symbol Description
A A cross-sectional area of local section,
m2 (ft2)

A I1 cross-sectional area of individual duct

at upstream end, m2 (ft2)

AI2 cross-sectional area of individual duct

at downstream end, m2 (ft2)

AL AFLOW cross-sectional area of local flow, m2

Ut2)

AMACH M local Mach number

AMACHl Mach number at section upstream end

(MACH1)

AMACH2 Mach number at section downstream end

(MACH2)

AR AR ratio of cross-sectional areas at

(AR,CR) upstream and downstream ends of section

ASL speed of sound in moving flow at local

section, m/sec (ftlsec)

ASTAR cross-sectional area for sonic flow at

specified flow conditions, m2 (ft2)

AS0 speed of sound in moving flow at upstream

end of test section, m/sec (ft/sec)

AT speed of sound in still gas, computed at

total (stagnation) conditions, m/sec
(ft/sec)

AVGPWR average power consumed by each drive fan

at specified conditions: PWR(?JP/ENFAN,
w (hP)
A0 cross-sectional flow area of test section
at upstream end, m2 (ft2)

A1 cross-sectional flow area of section at

(AREA1) upstream end, m2 (ft2)

40
F@RT”
R ~ UA Yn1 name Engineering
and/or (title) symbol Description
Alt8AO ratio of local section upstream area to
(Al/AO) test section area, m2 (ft2)

A2 A2 cross-sectional flow area of section at

(AREA2) downstream end, m2 (ft2)

A20AO ratio of local section downstream area to

(A2/AO) test section area, m2 (ft2)

BLKAGE blockage to flow in local section (at

upstream end for all applicable sec-
tions except fan contraction, for which
it is at downstream end), fraction of
local area

(BLKGE) blockage to flow in local section (at

upstream end for all applicable sec-
tions except fan contraction, for which
it is at downstream end), percent of
local area

CD CD drag coefficient of flow obstruction,

CH#RD cV chord of turning vanes, m (ft)

D D diameter of circular duct, m (ft)

DATA data array of master, section, and

termination card floating-point inputs

DELP APWi local pressure difference across wind

tunnel wall, N/m2 (lb/ft2)

(D EPS) AE increment of flow-obstruction downstream

influence factor greater than unity:
E - 1, (greater than or equal to zero)

DFAN drive fan diameter, m (ft)

DH hydraulic diameter:
Dh
4 x (cross-sectional area)
perimeter ¶ m (ft)

DHL hydraulic diameter of single cell in flow

straightener, m (ft)

41
F ~ R T R A Nname Engineering
and/or (title) symbol Description
DHUB diameter of drive fan hub and/or spinner,
m (ft)

DHO hydraulic diameter of test section, m

(ft)

DH1 hydraulic diameter of upstream end of

local section, m (ft)

DH2 hydraulic diameter of downstream end of

local section, m (ft)

DMESH diameter of mesh element in woven-mesh

screen, m (ft)

D1 diameter of upstream end of circular

section, m (ft)

D2 diameter of downstream end of circular

section, m (ft)

K local total pressure loss of section:

EKADD additional diffuser expansion l o s s factor

KEXPAdditional due to more diffusion in one plane than
in another (i.e., partially two-
dimensional diffusion)

EKBASE basic diffuser expansion loss factor for

purely three-dimensional diffusion

EKC expansion l o s s value for conical

diffusers

EKCNTR local total pressure loss from contract-

ing portion of thick-airfoil flow
straighteners

EKCSAV estimated expansion loss coefficient for

KEXP 3DAverage three-dimensional, combination circular
and square cross-section diffuser

EKD KD IFFUs ION local total pressure loss from diffusing

portion of multi-loss-type sections

EKEXP net expansion l o s s coefficient for

1 diffusers

42
F~RTRAN Engineering
and/or (title) symbols Description
EKMESH KMESH mesh screen-type loss constant
(KMESH)

EKS expansion loss value for three-

KE*Square
dimensional expansion in square cross-
section diffusers

EKSTRT local total pressure loss coefficient due

to strut drag in fan section

EKTE local total pressure loss parameter for

corners without turning vanes

EKTE90 vaneless-corner loss parameter for given

(KT 90) corner at a 90" turn

EKTV turning vane loss coefficient

EKTV90 turning vane loss parameter for given

(KT 90) vanes at a 90" turn

EKV local total pressure l o s s coefficient for

vaned diffusers

KO section total pressure loss referred to

test section conditions: -
q0
EKi local cotal pressure loss coefficient due
to diffusion and vanes in a diffusing
corner

EK2 local total pressure loss coefficient due

to rotational flow in a diffusiing
corner

EK2DC estimated expansion loss coefficient for

KEXP2~C
ircular hypothetical, two-dimensional diffusion
with circular sides:

KEXPcircular
KED
2DRectangular

EK2DCS estimated expansion loss coefficient for

KEXP2DAverage two-dimensional diffuser with cross-
section shape of some square/circular
hybrid

43
NRTRAN
name Engineering
and/or (title) symbo1 Description
EK2DR KEXP expansion loss coefficient for two-
2DRectangular dimensional rectangular cross-section
diffusers

EL L centerline length of section, m (ft)

(L)
ELC length of contracting portion of thick-
airfoil flow straighteners, m (ft)

ELD length of diffusing portion of thick-

airfoil flow straighteners, m (ft)

EMDH length-to-hydraulic-diameter ratio of

(L/DH) flow straightener cell

EMDATA data array containing master-card

floating-point inputs

EMF Mach number at the fan section

EMU lJ flow viscosity, N sec/m2 (lb sec/ft2)

EMUSTD pstd standard-day value of viscosity,

N sec/m2 (lb sec/ft2)

EMUT lJT reference viscosity at a known tempera-

ture, computed for a still gas (stagna-
tion conditions), N sec/m2 (lb sec/ft2)

EMWRIT master card output array containing data

and/or annotation(s)

EM0 MO
Mach number at upstream end of test
section

ENDATA data array containing section-card

floating-point input

ENDUCT number of ducts in multiple-duct sections

ENFAN number of fans in fan drive section

ENITEM number of drag or blockage items in each

local duct

ENU V kinematic viscosity of gas, m2/sec

( f t /set)

44
FgRTRAN name Engineering
and/or (title) system Description
ENWRIT section-card output array containing data
and/or annotation(s)

EPS E: flow-obstruction downstream influence

factor (greater than or equal to unity)

ER ER energy ratio: ratio of energy of flow at

test section to the output energy of
the fans

ETAFAN fan aerodynamic efficiency, percent

(ETA)

ETWRIT case termination-card output array con-

taining termination request de-codings

FAC function defining the area of sections

with circular cross sections

FACT scaling factor for plot size

FAF0 function defining the area of sections

with flat-oval cross sections (flat
floor and ceiling, semi-circular walls)

FAR function defining the area of sections

with rectangular cross sections

FDHC function defining the hydraulic diameter

of sections with circular cross
sect ions

FDHF0 function defining the hydraulic diameter

of sections with flat-oval cross
sections

FDHR function defining the hydraulic diameter

of sections with rectangular cross
sections

FEKC function defining the diffuser expansion

loss for three-dimensional, circular
cross-section diffusers

FEKCH function defining the diffuser expansion

loss for three-dimensional, circular
cross-section diffusers at high diffu-
sion angles (TH2 > 10')

A-5944 45
F@RTRANname Engineering
and/or (title) system Description
FEKCS function defining the diffuser expansion
loss for three-dimensional, circular
cross-section diffusers at small diffu-
sion angles (TH2 < 3
'
)

FEKS function defining the diffuser expansion

l o s s for three-dimensional, square
cross-section diffusers

FEKSH function defining the diffuser expansion

loss for three-dimensional, cross-
section diffusers at high diffusion
angles (TH2 > 10")

FEKS S function defining the diffuser expansion

loss for three-dimensional, square
cross-section diffusers at small diffu-
sion angles (TH2 < 3
'
)

FEKO function defining the change-of-reference

station for total pressure losses from
local section to test section

FEK2DL function defining "two-dimensional"

(rectangular) diffuser expansion loss
for low diffuser angle range (TH2 < 9")

FEK2DU function defining "two-dimensional"

(rectangular) diffuser expansion l o s s
for high diffuser angle range ( T H 2 2 9 ' )

FKTE function defining corner turning loss

parameter EKTE for corners without
turning vanes (based on a value of
EKTE = 1.80 at P H I = 90")

FKTVl f (0) function defining turning vane loss param-

eter EKTV (based on a value of
EKTV = 0.15 at P H I = 90") for lower
turning angle range ( P H I I30")

KTV2 f (0) function defining turning vane loss

parameter EKTV (based on a value of
EKTV = 0.15 at P H I = 90") for upper
turning angle range (30" < P H I I 90")

FTH function converting diffuser equivalent

cone angle, TH2, in degrees to half-
angle, THY in radians

46
nA
ryJRTRAN naiiie E-l15LllccL
rr:
ing
-A,.-..

and/or (title) symbol Description

FTH2 function defining diffuser equivalent
cone angle, TH2

G Y specific heat ratio of gas

H1 hl height at the upstream end of a non-

circular section

H2 h2 height at the downstream end of a non-

circular section

IFLAG parameter indicating the sequence number

assigned to the fan section

IPL0T decision parameter for selecting which

(if any) plots are to be plotted

IPRINT decision parameter for requesting or

omitting output of summary character-
istics page

ISEC section type-description code

ISEQ input section sequence number

ISHAPl section upstream-end cross-sectional

shape code

ISHAF' 2 section downstream-end cross-sectional

shape code

ITITLA assumed case-title array in the event the

title card is omitted

ITITLE input case-title array

1TU"L wind tunnel circuit-type code

ITYPE code for type of output format required

for printing section information

IU units-of-measure type code

LINEMX maximum number of output lines per page

LMTSEC limit for maximum number of sections in

any given case

47
F~RTRANname Engineering
and/or (title) system Description
MCHECK master-card input-requirement checking
code array

MDATA master-card integer input data array

MF#RMT master-card output format array

MWRITE master-card output array containing data

and /or annotation (s)

N N section assigned sequence number for

order of occurrence in circuit

NCHECK section-card input-requirement checking

code array

NDATA section-card integer input data array

NF#RMT section-card output format array

NN section type number for printing proper

section title

NWRITE section-card output array containing data

and /or annotation( s)

P input tunnel total (stagnation) pressure,

standard atmospheres

PA input atmospheric (barometric) pressure,

standard atmospheres

atmos heric (barometric) pressure,

’TATM Y
N/m (lb/ft2)

PHI corner flow turning angle, deg

PI ratio of the area of a circle to the

square of its radius

PRSTY porosity of certain non-solid flow

obstructions: AL/A

PT PT tunnel total (stagnation) pressure, N/m2

(lb/ft2)

PWRI decision parameter for requesting or

omitting the matching of power consump-
tion with given input value

48
F$RT?AY name Engineering
and/or ( t i t l e ) system Description
PWRIP power r e q u i r e d t o be i n p u t t o f l o w i n
order t o d r i v e wind t u n n e l a t s p e c i f i e d
speed, W (hp)

PWRMCH t o t a l power v a l u e f o r which t h e maximum

test s e c t i o n v e l o c i t y is t o be deter-
mined ( i f r e q u e s t e d ) , W (hp)

PwRdP t o t a l f a n motor o u t p u t power r e q u i r e d t o

d r i v e wind t u n n e l a t s p e c i f i e d speed,
w (hP)
QO t e s t s e c t i o n upstream-end dynamic
povo2
pressure:
2
, N/m2 (lb/ft2)

R gas constant, m2/sec2 OK (ft2/sec2 OR)

P l o c a l s t a t i c d e n s i t y , N sec2/m4
(lb sec2/ft4)

RHdSF PF r
s t a t i c d e n s i t a t t h e f a n s , N sec2/m4
( l b sec2/ft )

RH~SO s t a t i c d e n s i t y a t u stream end of test

PO
t
s e c t i o n , N sec2/m ( l b s e c 2 / f t 4 )

RHdT d e n s i t y computed f o r t o t a l ( s t a n a t i o n )
PT 8
c o n d i t i o n s , N sec2/m4 ( l b sec / f t 4 )

RN RN Reynolds number : pVR

u
RNREF r e f e r e n c e Reynolds number a t which turn-
R N ~ ~ ~
i n g vane 90°-loss c o n s t a n t , EKTV90, was
determined

RNV Reynolds number f o r t u r n i n g vanes based

PVCV
on vane chord: - u
RUFNES s u r f a c e roughness i n honeycomb c e l l s ,
m (ft)
(RUFNES) s u r f a c e roughness i n honeycomb c e l l s ,
10-6 m (10-6 f t )

SEKO section t o t a l pressure l o s s array (refer-

enced t o t e s t s e c t i o n c o n d i t i o n s ) used
i n summary c a l c u l a t i o n s

49
F~RTRANname Engineering
and/or (title) symbol Description
SEL section centerline length array used in
summary calculations, m (ft)

SERRgR section input error occurrence code

SLAMDA friction coefficient for smooth pipes

SLMDAE calculated friction coefficient in test

section at the requested power-matching
condi.tion

SLMDAl friction coefficient at section upstream

end

SLMDA2 friction coefficient at section down-

stream end

SLR diffuser side length ratio: ratio of

change in height to change in width
from upstream to downstream end, or its
inverse, whichever is less than or
equal to unity

SMACH section downstream-end Mach number array

used in summary calculations

ratio of flow-obstruction drag area to

local flow area

SSUMEL summation array of total centerline

length from start of circuit to end of
local section

SSUMKO summation array of total pressure losses

from start of circuit to end of local
section

SUMEKO N summation of all section total pressure

c
i= 1
KO;I losses referenced to test section
conditions

summation of all section centerline

SUMEL
5
i= 1
Li lengths (total circuit flow length),
m (ft)
T tunnel total (stagnation) temperature,
*C (OF)

TH 0 diffuser half-angle, rad

50
F(?!?.T?AA
r?aEe Engineering
and/or (title) symbol Description
TH2 28 diffuser equivalent cone angle, deg
(2 THETA)

TLIST case-fatal error occurrence code

TLISTI decision parameter for requesting or

omitting tabulation of input data

TRETRN decision parameter for requesting return

for additional case or final
termination

TSBLKG test section blockage used for computa-

tion of basic test section conditions,
percent of test section cross-sectional
area

TT TT tunnel total (stagnation) temperature,

OK (OR)

V V local flow velocity, m/sec (ft/sec)

voc calculated test section velocity at

adjusted power level, m/sec (ft/sec)

VOK test section flow velocity at input

conditions, knots

v1 section upstream-end flow velocity, m/sec

(ft/sec)

v2 section downstream-end flow velocity,

m/sec (ft/sec)

w1 W1 width of upstream end of non-circular

section, m (ft)

w2 w2 width of downstream end of non-circular

section, m (ft)

51
 52

I
 Y
-3
s
4
s
3
Y
3
Y
4
4
4
6
4
6
4
OM (r
4
6
1
1
1
7
1
I
1
7
_I
1
_c
a

53
 -BE -1.08 0
PP 1090

54
55
 300
sro
320
510
340
$50
ab0
370
380
39 0
400
1u10
'400
1430
'440
1450
1460
!e70
!y00
1490
1500

56
57
58
59
 10
20-
so
40
so
m-
170
80
90
.no-
lo
.20 -
30

60
61
63
64
65
66
67
 *
GO T O Paaa ..__I __ -. . .. . . _.,.

c...

._ , I . _." - ___. .. .

68
69
 9230
424 0
9250
9260
9270
9280
9190
93oo
9310

70
71
72
73
74
1 1820
1830
i e49
1 1850
1 166@
1 1e70
1 1886
1 1890
1 1900
1 1910
1 1920
1 1930
1 1940
1 1950
1,1960
1, 1 9 7 0
1,1980
1,1990
1,2000

75
C...

76
240
770
?BQ.
790
609
810
420
830
8U.
850
169
e70
B80
890
900-
1910
r9a.
1930
lPYQ
!950
!Qbo
!970
!9b0
!990
Ian0

77
78
79
I
80
81
 Me -2- oK 1 7 4 0
- - -
Ob0 O K 1770
000 -OK 1?00
000 DK 1790

82
a3
85
 10
zn
30
4.0-
so
40-
TO
Be
90
QL

86
87
 bK 4860
OK 4870

88
DK SO70
DK 5080

89
 DATACK VIPE tL

90
91
92
93
 QATACK -

94
 95
:I ,
96
97
 CN 00

-- - -

98
99
100
. - QUTPUT PAGE J

101
102
103
 Pl 490
BT 500

104
105
106
 INPUT AND OUTPUT FOR SAMPLE CASES

Six wind tunnels were used, in addition to the test case (fig. ll), asI

sample cases to establish the reliability and accuracy of the computer program
analysis technique for the various types of duct components and wind tunnel
circuits. Each case included here is titled with the appropriate wind tunnel
name and its pages are numbered. The performance analyses are presented on
the first two to three pages of each case. The summary characteristics tabu-
lations and the plotted information were omitted. The annotated tabulations
of the input data were included for reference.

The results of the performance analyses are summarized in table 6 . They

are discussed and critiqued in the Results and Evaluation sections of this
report.

107
 aJ
+I U
U
I
I
0
0
0
.
+I
I
I

+I
I
+I
+I
+I
+I 0 .
I -44
N V

i
+
I
I
I
I

+
I

+I
.
t
+I
I
9- 9
I *
*.
Y VILA
n +
\
2:
I-I-

?n + $1 J.
I
v
I
I +Im, I-

3 !
3
4
I
I
It +I 00
n
&
I
I
s;
1Q
9
K I
Y
E ~

0
*D
5
c
4
+I
I
a70
99
A
.. m
n
+I
t I-]
vu
I-
1 : u u
doc
~+

I
,
3
~1I
z
a,
w

-
l l
c
I

I
/ II U
ll,

If
L
!Y
0
vi
d
I+

108
 U t
dc:
"U
v a
*0
%I

d
I-
C
' I V
51
-1
II p.9
01
VI
t- I LL
u' I I
UII I D
VI, i
z
Q
U

U
d

109
m
I&
rl
d
a

!
4 I
4
4
4
!
0
I
4 I
1
t
c b
z I
3
c:
4

,
4

!
1
I
* !
!
J

I
I

-
L
2
3
I-
C
2
c
3
I-
C
C
LL

-
C

>
1

n:
I
F
U
U
I-
Y
L
I
c
3:
4
U
U
Ll
a
U
U
I
4
I
v
4
2

110
 C
n v
I U
.( >

-
c

4
c
5
.c

U
2 U
z c U
-
C
I
c
7
u
>
c
2 C
C 1
c 4
c L

-i
4

4
2
c

U
-
2
c Y U
4

4
1
4

a
-
C
I
U
C
3
c
C 0 .
I
C
c t U
z2 -
Y
U

C
E
!A C
c U
4 (I
I-
lu
2
2
4
c
c
C
a
I
P
:
C
2
2
L

c
4
2
a
LA
c
U
c:
W
c .c C
I
U 2

I
I

111

' \
 +I
I
9 1
l-+
+I
I
4 1
(c+

+I
I
9 1
Q +
+
I
1
4 1

-
Q *
+I
I
z
c
a
-
2
I-
n
9 1
rr\+ 5 0 0
+I
I
d l
m +
+
I
I
9 1
V +
+
I
I

7;;1
6 1
m+
+
.
D Q

. .
9
I
d l
;3 D 0
4
0,
In

m + m r) m 9
+I r
. . . .
d F d
m r) V m
I d 4 .
I( 9
Q I
N + N U N In

I
- 4 1
+I d

m
. .
rl
!
n
l-
d
V
m .
V
4.
-4

Ni i
rl ). 9 Q.
0

. .
9 0 al N N

14
2
3
0
4
.
r
>
e
(L,
0
m
Ln
I-

m
m
I-
Ln

;j d
.t
4
r
2
J
m
4
l-
m
;o
Z/i P
e
v N V

=I
I+ +

I-+
+

9 + N v v

N kl T
U

0
N
0
4
I 0
4

!
I
N
*i al

I
112
 **
4 +
I
I
U
(s
a
$4 n
n + W
c
I a
I
;c4 f
.
f
a
+ ? w
I I-
I c
a G I a (z
a 9+ D P
C
:
d
+I P
Y
z I I-
I(
- 1 W
c 9 + d
z n
I + ?: 2 I

.
U I .
) 0
a I ~

c V
a
I
:I 5 0 W
+ m
1
- 4 1
I
0
. -*
c)

0
4
u

* VI+
+I
e

*I
e+
+I 4
N

4
4
*+
I
1

+I
m
.
-4

0
A
A c ILI
tI1 a I z
z C 9 1 X
2 m+ 3
3 2 e
e
Q
z
c
.(
C

-
r
I-
::
+
I .
3
?
D
9
0
0
.
-4
9
0
0
.
rl
4
c.
I
1 0
c +I
.
J U
I-
C
Q
U
I
1

2
91
N +
U
.-I

0' 3
.
-7
-4
0
a
U
I

. .
0
2 C +I 9 .r
.
OI 4
c 4 N
*'9 c
L 4 1
I VI
N

0
*
m
U N + 4 -4 rl
V

.
I
I
r- +I 3 9
0
9
W
t-
I -l 0 (z
a
u 9 1 v 0 0 UJ
I + 3 -4 -4 c
I-
+I z
E 4
*
J U w
u
I
u
I
rlI
.-+
D
3
v
-l
.
-4

o\
U
4
v
J)
U
I
V
P
rr 4
a
u +
oL+
u1
u1
VI G
u! a
PL
VI
ul
+ v)
Y
L:
I-+ r
a
a I
I a
Q v)
VI o+ 4 N N U
4
2
z

rl N N

a 0 N
r I m

* > 02 d
-*
.
I -4 N
*
it

113
 In 9
N m

.
0 N
0 0
0
0
9
0

N 9
01 9

.
N nJ
0 0
0
0
90
0

9
.
r-
01
0
-4

ma nu
UU *Ul
?0 90 ?
00
?
NN
Y?
Inn Inry
-44 dIn

r-
In
0
u
.
0
:
9

00
UJ
0
m n
.
00
4 0
m d
..
Wr-
4:
9 N
Ur-
N

..
Inin
4-4
O Y
r-O'
inr-
4'
0 .

i I

I
n
..
ZI 2
1m
4.4
1U
0
r-I d
*I
mO'
-41

cI c
I+ !

iiI+ id 111
LVV
l I-
t
C) u
a i
+I
0
i: I

3
i* 3
iI z
I 2 I,
L A

I '
I
i
%
.
I I- I-
: I z U
I Q U
I I- 0
w l-
2 2
1
C 0
L, V

++ rl N

114
 I *+
" N QI
XXUJ I
0
rc
n ml
i
ai
C!
.'
4 41

I
c *I
uv
w

wl

>
I
I 1
C I

2
a
i
c j '
V
W
4

115
 \

2
I-
e
c
9-
5
L

n
U
3
2-
r
c
U
I c

I U
c
L
U
4
U
<
I 0
I C
I
>
c
c
b
L

U
>

i
:
4

L
I

i X
a
5
,

i
I
,
I
I

!
*i

116
 *I I
.I

w
c3
4
n

9
..
a

UJ
3
2
c1
I-

..
2 v
0 c
u

U
J
.i

it 9
C
U

I
g
U
C
2 .(
z 5
0 P
3 c a m
t < c
U
'U
- a
a
7.
U
v) a<
I- c
-I >
3
d c
v) 0
W < c
I- *< z
0
1
U
C
L

I
-I
u
c
C

c-
V
rr
x
0
m Iy
\ a I- I-
w 2 v)

*
I
a 4
A
Iy
C
5
A
-
4

z I-
a
w 0
2
c
z
3
e C
z
3

-
O
Z
3
O
2
a
c n a
I

111
n c(

W t
n a
v) z
I4
I r
z e
0 w
A I-
a Lu
I
(3 v)
0:
c a
3 V .(
u U
0
I
O c
Y I
rll
r
U
13
I
I

+i

117
 +
.I 1
91
f-+
+I
I
.+I
f-+

+I 2
c
I I-
91 a
9 + C
+I
I
4 1
9 +
+
I
I
-
z
I-
a
-
2
c
0
9 1 I C
in+ 0
+I C

I C
- 4 1
m +
+
I

*+
91

+ r
.
L - 1
.$+
I
I
.i
c

+ 4 +
t- I
a I
0 01
m +
+ 2
+I r

.
0 I
n in
4 J C
w N
34 >
4 c
n 4
I
1 0: e a <
< w m+ I-
0:
+ *U m

.
f
! u I m 3 I
v)
3
b:
";
Ly I 4
! 0
m * U I

I
I
I
2
0
w
I-
0
.
I-
o
4
d

m
.
c

n
I
I

"li
V d l
d 4 U
W
! v1 7

-.
0 IC
P m 4.
U
m 3
14
2 d I-

+ 3
I-
r)
In
0
n
I
I o, r
d I in
d + -4 U j.

+
m +

I-I

4 +
Iu r.U
v 11

\1 Y

L. (ri 0 V
I
Ry
N U
N
"?

(\I m U LT* h

118
 91 I +
I
UJ
I
0
a $4
n +
I
I
Zl
. +I
I
-
z
t-

4
$1 o
J
W
3
+
I
f I
c
z 21
0
u
..
a
41
+
I
I
? z
t-
P
-
c
n
-
t
t
n
3 3 0
+

* 1:
I
I
0
0
0
.
JI
+I 0
4
0
4
I -4 4
*
*+
9 1
In In
+I * *11 rn

-
z
0 *+
I
d l
N
l-

.+
*
U
.
r-
v
P
4
.
e
U
4
+ +1
111 < I
v) C 9 1
I- 2
m+
v)
Ly
c
A
0
C
e
c
"
a
4
m+
+I
I
1
m
3
-4
-4
o\
m
4
-4
'3
r-
tn
-4
. 0
r-
-4
.
In
c
v)
c
If
+ 0 0
\
I r- IC
> 2 I
r 2
21 In
d
In
4
+I
-I
UJ
2
z
2
C
L
c
c
U
I
- 4 1
nl+
f-
nl
b)

r-
.
Ln
9

N
d
I-
d
I
In
In
*
In
r-
rl
.
n
z
I
3 9
4+
+
I
I
1
3D
In

-4
rl
W
d
4
. .
OI m
m
d
-4
0
r-
ln
4
0
UJ
Lu
P
v)
- 4 1
+I
I
0
r-
In
.
.
4 + d
I
3:
9
2 m+
+ ln In
a
I
0
a
2 r-+
+ .
-4 d
.
W
I
0
W 9 + 4 -4
w r\l cy
I
*
V
3
A 4 -4 -4 ry

j N
w
g~
0
U
ry
m

a
QUI + N
d
VI W
4 -4

119
d
*N
r(

rl
w
0
4
9
0
a
4
U
N
d
0
0
.
0
rl

4t 9

tu
d
0
A
-I
..
Q U
a39
NN
0 0
CI)
z
a
m d
e .
m 9 O r
*a
c
4

.
-I
tu 9
z
z N
3 d
I-
n
5
x
c
+
0
.
e*
0

O,
d

Of-
U O
..
ml drl

I
5 4 m
Inm
W e .
V
z
UJ
L
V
VI
LL
0
ty
I-

:i
2
t
U
I-
v1
2 In
U (F

-
z
a
P
z 2" I +I
N

n. 0

i 2
u.

*
+I
I

i
1
I
I
I
I
1
w1
I
;1
t-A
w

I-'
z
++

120
 NI m 9
9
0
cu
0
u) 0 0
0 0
2 0
e 3
0

I
9
N
9
*
I-
I- v)
I
v)
e 0 0 cz I
* 0
e
0 u)
t
0
e 0 0 u)
P I
UJ
3
z 8 aD
m*
.
I
0 4
t 0
II
z
0
ue
e
e Nr- m m
uiU
0 0
mm r
0 0 t
0 . * e (1
0 0 0 0 z
W

* e4
I-Q
..
- 4 4
mm
m e
mm
.+.-I
d

W
z
c(
A
cz
W
u c
(*
0
0
z
0 w
e V
.J 0
3
z
@ -.I
a
4 c
m 3
c
0,
c
4 9
.J
UJ

3
I-
o
z
U
I:
I-
O
0
K
OI
*m
5
d
L I
dNv)*
a n.Wa I
I
m-l
m0 m.
94
4
y1

Yu
..
CI
u 99
Nnl
v)

U
mul
0
w J J +
I- P I Ot
3
t-
a V
V
I-
3 , i d 'L1
u.a
v)
z
w

z
4
I
0
2
c.

0
a 3
w

z
+
+ m
rl

121
 0 1
+
N I U
A'
41
U
¶

I
+

*I
I
I
bnl
;'I
- 1
(1
U
9
- 1
I
i 4
I
,
I +I 1
I I 1
0
4
1
1
!
!
( A +i I
t

cn
! Ly

!
I

,
I
I
+
i
71 W
z
"II 0
z
+I

+I
I

9
i 1 0
2
I I

"i I
I
I
+

I
I
n
-zv)
- 2
n a
AClS
IUIY-I
-zu
iL-U
a
aco
cza
-0e
PUG
 I
.I +
Wl I
91
*+
+I
I
*d+l
+
I
I
91
9 +
+
I
I
4 1
.a+
+ ? z z
I-
a
- -
I-
a
c
5
c
2
0

I-
a a n
0 0 0 0 0

N
4

4
W
N
.
-4

* **
In
N
4

b)
N
4

.3.
(0

VI v\ 0 In
0

.d . .
0 0 0
0 0 0 0
0 0 0 0 . .
a 0 0 3
t-
3
m a
0
4
I- 2

*.
a 0 m aD Q

.
t
M n d VI
5 m Q
a I-
n .3.
5 0, n4 I-
.*
0
Q
0 U *
I-
I
Q
u
I
N N N In 4 * 0
2 9 9 9 P .
.
v) 0 .3.
a
2
Iu Q, Q, 0, 9 (r In N
0 9 39 ? .3. a
0: I-
**
c(

0
z 9 9 m n N

.
Ly
0
I-
I 9 4 9 In
a I- W a m Q
u
s ?
I- 0
u) rg
o v) U 9 m
2
2 W m m m 0 m U
a m m m
d U
* 1
U .?
a 0
U
m l-
4
N N N N U d U
9 Q 9 v

. .
9 -I 0
0,
0
9
0
(n

9
0,
0
9
P
U
9
.c
d
,
IC
cu
m
N
P
N
.
Y\

N N N m rn

cu N ry v m m

*
In
.?
ul
U
In
3 m
v

N ?
. VI D rn

123
 >
I-
n
r
U'

0
9
m
9

0
0
4
0
.
0
-z
I-
n
0

sd i =a
0 s
N U
m m
9 9
$ 5
d r l

4
I-
<
r

2
r 4 3)
n
L
I- 0
4
2
d
O
4 -I
L
n
n m
L
U
U
c
A
U
J\
. N
m
9
.
c
).
L
L
c
9 W
W m
m
d
9 rl

m
N
m

9
m

i
i'2

124
 4'
hl
I
r- 0 N 0 dl
m
w *9
4
4- N *
4 N
Ia: 9 0
9
M

9
0 *
0

0
0 0
9
0
9
0
OD
N 0 N
m 0, 0
*0
-4
*
.
9 -4

. . .
9 N

.
d 0
9 0 J
0 0 0 0
O
N 0 0 0 0

* **. *
*.
F4
m 9
r-
N IC hl 9
-4

n 9 9 mln
00 mln m m
.*
4 m 9 O N
Lu 00 -4-4
mm m m mQ
* e 4.4 44
I
00 * * e *
40
e .
U 0 0 0 0 00
t- 00

. * ..
a w
I
c
a
V
z
a
x
rz
hlN
0
NN
. r-l-

**
mu\
IC-

v)v\
*e
rcrc
m a
U*
. * ..
aril
V I 4
.thl
-I 0
W U
z
z e
w
3 a m
c 4
*
-I
0 w W
z z
c(
z
I: 3
c +
c
2
0
l
0
2
I
I
*
4
v) LI I
+ *
\ S ry
w
z +I **
c
O
0
0
I
c
u
* I
I
v m
.*
00 .. . *
88 00 8 0 8 0
0 . ..
ou
04

I
d
tu
v)
I
z
3
v)
NN

m m
4 4 4 4 4 4
4N

c +I 00 0 0
c v) 1 P3 90: ? ? 30:
00 -0
U
4
rr
V
a
CI
a
z
Lu
v)
0
W
0

N
m
4
w
In '+
I
I
rcrc
mm
e*
0 0
NN
0 0
N N
0 0
N N
W

C
c
h

* 4
2
w
4
Ly
n
V

z
rz
. W
+
I
I
I
W N
23
m v .. * .
I-+ I-l-
nln m m
r-r-
lnln
. *
4

I m m U* U* e*. u
0
w
3 v)
w +
v) t;
a
a
w
a I I +I m m

*** . *..
W 2 a d d *rc ICrc r-rc
Y 0 Inn
v)
* * lnln m m
S
a
I
z
::
c
I
I
I
4.4
4.4 * ** *.
+I cc t-I- t-t tc c
i U V
ww
acz
UU
ULU
acz
U U
w w
LLK
U V
wul
acz
a
c
L
w t a
t t
V V
a
w
a
+ 4
a
3
I
t-
Y,
I3
d
I
> 1 z
t- I LL U
w
0 a
a u w
z
E! i w
m
a
4
c
a
a
e I c t- V t- a
V
u
v)
1 z
Ly
w
2
a
t
VI
u,
v)

c
2
a
t
W
&
VI
3
a Z v) z LL
V D K
v) W 0 I
u I- V 0
a c 4
z u1 9 l- (F

125
 3 m In
Q r n N W
0 o r ; I
n
f-
0
9
0
N
d
0
0
.
O
o
0
0
. .
u
0
0
m
0
0
N
0
e

+ d 9 9
(v
U u * P 9
m N
m
0
N

0
. m
9
0
o
3
0
v)
Z
4
LL

In
3
m
o
3
O
m
?
N
o
o
o
m
l
m
-
m
. 1
:I
lr'
9
LL
0

II "
I
II
N
9 s:
oc
a m
*-r
e r
rlr
"
N N
0 0
GzI
3
0 0
9 3 3' m e
0 ddl dc 0 0 o< 00
Lu

? < ? Y'
m 9 9'
4 d
N

0 0
d 0
e
0
e
0 0
N e A
N 0, Q
l-
o
I-
In N
m m
e
d d

s
N
. 22
e .
"
2g
e .
"
m
9
.
in 00
N N
0
99
.

0
l-
U
I
9)m
+I-
e
mu!
*
m a
I-In

In0
0 .
N
.
In

In
U U U *9 5

P m
(Dl-
e .
NcO
N O
O m
s:
..
99
m m
22
..
UI I-+
a UU
a w w
f &rz I l n I
v)

v)
u
ru 2
a.
*
I-
a
>
z
U
4
LI c
I- a
U &AI
u L
W# U
a

0
. U
9
4
I
I/

126
 m
Lu
0
U
a

0
..
u1
a
5
I-
2
0
Y

it

0
2
ru
w
Y
c
4
E
c
U
a
(Y
z
J
c
n
Z
L(
%
-1
0
c
m
\
>
c
O
0
Y
I
In
r(

c
Y
2
u
a
r(
U

w
-1
UJ
g
L.
u)

a
::
(Y

U
I

127
* *I
UI
za

e
e
a
e
0
Lu
3
2
CI 0
I-
z I
I
0
0
0 I a
ua 4

0 c In *
e I
I m. 0
z
I
c 5 a
c 0
I 0
I
I
+
t
I
3
d

N
N
z
v)

a
0

. ru V
0
A
i In
0
0
(L
C +I 4

.
W N
u
4 r- c)
U c I e
c I-
a I In z
4
r P + c1

C
-1 +I > Yz
I-
4 b 0 I-
a I-El
-1
; ; i x
Lu
4
d
m
UI

3c t s +I N
I-
In
0
e
a I
! a

!
W
I- +I U Lu
3
8
u I
v)
a +I N
r- 2
3 I In
I
I (3
a
a
0
I
I
-1
W +I U w
a
c z rr
z 3
v)
\ I 3
N u
> I- I+
u
0
I-
o z
0 0
U
I
+ N e
c
VI 4
4
fS
t
U
4
+ c K
U
(L c
U
a U
H V
a w a
c
>
Lu
-1
W
P U
0 I
U
v)

a?
*I
UJ

q
L

128
 '+I
I
E1
+I
I
21
+I
I
9 1
e+
+I
-41
9 +
+
I
-?
91
In
3
+I
4i
ln+
+I
I
9 1
U +
+I
I
-(I
I U +
I
+I
I
I 9 1
m+
+1 0, n N N r-
.
4 P r- t
h
d l
I
ry
U ln ul
In
4 t
I
m+ 4 4 * 8
U
e
In
+I m
d P
n nJ N
I
IC w
, 91 4
U
<
In
?
N+ d 4 * .r
+I $ U (0 to
s 0,
!
!
(
b I
d l
N+
4
I
m
w
e
a N
In
8
D
I
0, 0 N N 4 1
+ N 0 N N

.*
m N
b I c r- I- b
C I
C
21 m
m * *In
3
I
8
In
I
UI
8
1
b
I
U +I m m N N
I
r( 3 + n(
I- r-

.
a VI ul In
c
U 21 4
4
4
d U * $
a
< +
.
a

2
0,+

U +
++
U
C

U N N N 4 I

3 nI ry N N 4
.
a

f
O 0 0 3
N rl r( t
4l
N y\ PD P

129
 *I- *
4t
9

w
a
S
Lu
0
.
u1
W I-
4 0 a
0
9 Et
In
01 L
Lu
t-
e
e

0
.
e
0
L
0
P
u
Lu 0 t-
3 4 Lu
z 0
* -1
I
0
a
t S
z
.
0
0
u
u
W
e m
e a

4 1
In+
+I
I E
C
-*
V

* +
.
9
d
I m
*9 1+ m
0
+ 9
CP
-1

.
I Lu
a I
0 I
A 9
w
n *+
4 1
0
U
I-
a
+ m

.
U X
c I m
a I 9 !-
r .o I
N
a
m+
I- A
a + Q I N
u
2 I1 15 I
, 94- z
z
Lu 4 3
z m + m I d
I-
2
3
t
+ 9
m a
z
I e
I 0 CI
0 91 rl I
2 N +
CI A
3 + *N 0
I-
I In
-1 VI
a 4
I
1 * \
c N + rl >
VI
+I al > 4

. .
\ t
w .* D b)
0
I 0 0 r- 0
e
I-
O
0
LL
2L
+
m U aD
i
4
I
In
4
I I-
U
2:
I-
LL
4
L
V
m
+
+ nl
. a
L
V
L
CI
a
5 >
a * ru
P +
+ P c A
u
Ill P
A 3
W I
a hl v)
a Q + 4 -I
I 4
m d
u
%
x
i
a
3
4 d 4 a
x
5

4 u a
0) W r.

* N n In
-
*
it
4 rl 4

130
 lm
+ In
91
PD
W
N
Wl
OI
%=I
l-
t-1
k m aD

.
U U 0 N
i 0 d
0 W 0 d 0
4
0
0
0
0
e
0
8
0
e
0
e
0 0 0 0

m 9 (I)
n
c
I *m m
m
N 6 co
4
e
I 3
0
0
0
t
0 0
r-
0
OD
z:
VI
-4
0
e 8 8 e
0 0 8 e
0 0 0 0

c
I
I
I
b
N
0
9
e
N
In
e
-4
d
m
9
d
e
0
m
N
8
In
0
m
8
9
.
rl
l-

Nrl 0 0 O N Ne
In* LnIn InN
99 9 m
m m mrl
N d -Id d m
S? -49-4 ..Id 40
00 ;; 8
0 0
. e .
00
8 .
00 00
- 8

? $ 7: 74 ?? mm mQ
* e
r-* r-I- +-I e m
* 8
dd 4 4 4 U
4- *e
d rld rl
Urn mm mcu

VI
*.
U
?
r-
9
0:
4
rl
m
U
e
U

U 0
rl In d
b 0
9
8 e
N 4
e 7
d

Fec8* 8 8
9-4
m e
drl
00
Of-
dN
.. ge 2. $ S ZE
NN
e .
cuw
8 .
rum
Ind 0 0 ou Inn Inn VIVI
4:
mPD
Inn
e .
m m
e . o".? rlrl
e .
dr-
8 .
m W W N N U
VI N NN $ $ %:

+
m
4 .ln
* d r-r- I-m OIn nu) V I 0
o"? SY In-
e .
* O
8 .
0 0
8 8
OIn
e *
Y Wm mm mm cum m m

4.4 mm m a 9 9
e*
. e
mm
0 .
m e
. e
(Fm 9 9 9
m
e . e . 8
m N Nlv Nm m m mm m
+I
cc cc el- el- I-+ tu
vu
i UUJ
aa
UI
U IU
=a
L' UU
WUI
aa
uu
UJU
aa
UV
W U
aQ:
ua
UV
P V
t- (z a
Lu V (z
w 2
+I d
Z
a v)
3
u.
a
u
Lu
I
n
3
I V a u. u.
a U Ghu
l I v) Lu M
-I u.
-1

i
I
(
a 0 U 0
L 4 I
z t I t
I I- 4- Lz V
I V z u
r(
V
I 4 3 2
U VJ
I a c 3
0 0 a
c v) U I-
z z P I-
U d 4 d
0 0 CI 3
V V a 3 3
P E E
++ IC m *
' I

131
 o
:
$
I
+

1
m
o
0
0
90 90
U
u
0
0
0,
N
0
0
0
0
.
9
0
d
0
0
0
*
m
0
0
Od
m
o
0
0

o9
r
I
N
0
70
-
-
N
0
O
0
90 $1
0

I *o , sm U
o,
m
m
9
m
N
O
O
m
N
d "If
2 0
0
90 09
0 0
0
0
. d
0
0
N
0
d
O
0
d
4

d ( d
m 0
9
d
*,
9
I!
i
NN

. mm
dl Q I f
01< 0 c
O ( 0 ' 09 c3
O ( 0 ' 0 0 0 do E
dNV)
g' ..
mu.
o,m m
m
rl<
4 ,
m'
m
or
N
. 33 9
mI-
Nd
I-
d
>>\ Nm
L

0
m
m
m
d
N
. 2.
m
rl
d l

U
w 5
0

.%
*,
* h I-
o,
2 .
is
'i. ? d
d
e rl
4
i
m or
1-N
mm
.. or
N
m
m
ln

mm
.* e'"(
4 9
!
9
. 8..%
R
9 99

lnln lnln 4-0 lnm mm mm

l-@
9 7 40 :4 $ 4 $m 4o 5m 4m
a4 a s
l-l-
m m lnm
u1 Lu 0 rll-
m N NN m m 0 m lnln

aa

*..
00

* ..
lnln
00
lnln ..
00
lnln
**

8%
lnm
.
vv cc
UU
aa W W
VU
as
a
u,
z
IY
I- u
3
cz
c A
-1
V )
a
1
s) c r - I C I
I-
U
4
Y0 I ' 0
3 Q
t-
2
3
L
9
4

132
2
C
e
I-

I-
U
2
C
c
0
U
3
C

C
U
c
U
c
c
0
U
4

a
C
Y
I
>
c c3
c v)
L \
z
-
C
IL

z-
m
.
In
(Ir

f 3
I-
4
L
I
I
UI
cz
3
v)
v)
UJ
cz
a
u
I
I
4
f
a

133
 w
A
m
I

X
+I C
C
U
>
c I
5
U
*+
Q I
c
a
u + v
a
J
I
0 I
w
0 *+ 4
d l
c
5
a
I
I-
tI CC U
a 9 1 c
m+ c U
U
+I I-
P 2
1
9 c
d l
i! m+ c C
C n
+ I- d

: d
c
U 9
I
1
I-
C
L

2
0 N+ c c
3
< +I > U 2

.:
A
c 0 c x
cx I 0 2
C a <
;;I
C
z
0
V
a
+I P (r
a
C
-
! I U t
L U

.
c
w
c
v)
a
S
9
4 +
1

+
I
n
C
a
!!
-
2
U
>

C i-
L 2 4 1 C
Lu 4 + 0 U L
z 0 I
2
I
..
b I!
5 3
c c
9 + (r

c
C
II U
C
I
I
In+ r
t I
c
c
<
‘
r
a
*+ C
U
c I
I

i
L
(I
m+ c
< 0

i z

4 +
+ U
i
I

134
 +
i

i
+
I

i
+
I
I
I
-
z
c
n
2
0

c
a
0 0
+
I
I
I
+
+I z
0
z -
z
I-
z
c
i I- I-
n n n a
0 D 0 0
+I
I
1
.
0
0
0
0,
0
0
0
.
0,
+I
I
+ 4
I $
-
z
C
c
I
I

+1
.
d
P

3
3
P
0
.
--
a
3

b:
I
1
4
I-
+
.
co
4 I
VI
4
4
6
9
* 9
0,
c
a
C +I
9
m
e
l-
v\ N
?
m
:
t + rn

. .
N N N
I 0 9 9 9
a
2 I 4 OI DL
Q:
c N m m m
C
U
c
a
I-
+I
I
I
0
N
0
9
.
d
m
9
d
.. 9
P
V
N
*
(0

?
m
.
9
0
b

9
C
z
..
N rrl m

.
9 Q
2
a
t Q, +
In
0, 4 4
I 0:
Q m
V
N
?
m
0
m
+I 4 0 N rrl N
I
I
$
e
VI
m
m
N
0
9
p.
m
d
m
. 9
P
e
m
+

++ . .
N N
e
N

+ N N N N d

+ N rrl N N N

++ 0 0 d
r-
rl J
m
N 4- I-

+ N cn 9 OD 6

135
 +
I
I
$4
+
4+ -
2

9
I t
n
91 5 0
+I
I
d l
a+
+ ? L z
0

c I-
n n
a
ul i 0 i 0
+I a
a
0
0
e e
dI
ul+
>
L
0
0,

+I 1
t
>.
t

s! :
n
I:
W
D 4
tI 5 t
d n

*+ J
-41
e
*
d

0
e
t

;
In
W
Q
.
+I
I
6 1
m+
+ D 9 9 D 9

:t
I c
t
m
*.
4
?* <
P

t
?
4
+ 4
J
r
r f
4+
N
I E 0: P
n
e 0:
In
N In
9 3 0 d
I 2 l- N P 0,
a r I n 9
e
9
l- *. ?
d
Nul
+ 5 -4

9
0
8
a
0
P
4
9
0,
I P 0,
rid
10 X
24 :
t
I 5
4
*.
4
:
t
$
4
+
L
d
I
I
\t
%.
d
s 3
dl In n In

a + e ?
c
0 e+ d
2 Q
0

+ e e s e

P IC+ N N N N N

9 + 4 -4 N N N

4 d N N N

d d d
N 4
P 0 Q l- l-

In 9 (D
2 N
r) rl d 4

136
c
I m
0
9
0
LL 0 0
c
a 3 0
a 0 0

N m
ap *
m 0
0 0
0 0
e e
0 0

rl r-
d In
c e
m $
N

4In m 4
mvl m*
0-4 Orl

-
U
e m
0 0
e m
00

2 UI
5 V
z
3’;
O N
4:
mw
c 4 e 4In N U
L L
E OC
t3 0 a
U v)
OC \ 0 m

-
t
C
UJ
a
A
iLI
z
0
.
In
m
e
m
m
OD
e
In
4

li
.c
2
z
3
In
m
9
t rl m r-
> I 0 9
a P e e
z .+ d
I
d
(r
L
I
z m 0 * O
mrl
Q
U
c
2t CFO
m e e
- 4
e
2 V
U w
c v)
I 90 G 9
I e
c: v) ?? ?Y
w m*
(I
4
U
I-
In0
m e -*
111
lL
i
Lu
- 4 N4

Q a
0
>
U
-1
c1
z
4
z
a
- WOI
?
mm
?
m 4
* O
3:
mo,
N-4
-1 a
I c
4 w
v: a
d I lnU NO
z w 0 4 r-9
-I e * * e
m 4 * m e
3 N 4
8
to 00
V
W d d A
UU

y1
.A

4
.
v)

I-
U
:c I L a
a
r-
L
0
U
m

137
 In
"
In o,
d 0,
d
0
-42
m

.
9 U 0 U
2 0 0 0 0
0
9 90 9
0
0
e
0
0 0

A
m
N
N
m
ul
N
N
4
r-
I
0 d 0 0 I- v) v)
s
"
0
e
In
3 2e
0
0
9
o,
0
e
w
t-
3
0
-1
0 0 0 0 0 w d
r 0
LL
0
0
9
-4
e
0
9
0
0
4

9
.
In

rl
m
In
9
0
cv
0:
9
4
2
Q
9
*
v)

*
4 N -1
tu
t-
U a
9 4 U-4 *m mn m- r
InU In* mm m* mn I I
Y
O C
e , ?o cc ?0 90 O C
?C
o c
t-
(3
2
0
e
O C
w
o n
.
N-4
?? 4 - d . 4
-1
i
oIa
4-
* I
Q r
d r
drl
dd
4 -
d.
9
d C
4 -
( UJ
2
I
e v)
3
-1
lY f W
UJ
t- z
oI
9
9
d In
In
9
Q
7
Z
w
V
I7
a
C
L
3
I
4
d N N N
A w
4 a U
t-
a 3 &

0
f-
In
In
0
ln
0
t-
s w
%
0
e 9 9 c a
4 d 4 4 U
c
U

.
L
In4
r-a s: QO
*r- Ze $ 0 -
* U
c
a
Yl-
e
N*
e *
uu *, e4
9
U
c
mn mu uo oc 0 4
me < ur
N n
or
9
O U
e *
m N
*.'
wr
e

dl-
U.r
K :
4..
90
9r-
O <
r-r
N U
O n
r-r
>
c
c

t
m, U. uo O < 0.4
y
~r N O O < or
e e 9
9.
e 1
5: l n b
d. In9
4 4
4.
-43-4
d r a
c
>
4
a
Nr N Y InIn In& I n U
oc oc m e m< o,C
z< 9

3:
9 .
d d
Nnl
4.
nlr
m
4.
rye
e

c c t-t t-t- t-) cc

V C V ( V U V ( V C
UJU w I UJW LUI UJU
a C P < arz at l Y C

rz rz rr: a (z
z w z
5 c V
c)

3
LL 2
LL 2
u
v)
2
V
LL
"
u 3 U
0 L
-1 A Y, d
-I t- d c A
Q U 9 V Q
3
I
N
3
n
I:
I
N
2 3
d
I

:
t- t-
0 0
0
S E I I
N m
d
4 d " 5, v\
4

138
m
I
UJI
LLI
0
cl
n

,
4
4
4
C
% +1
z
c
0
NI
I t
V
U
2 9 1 >
C ' I
c I
4
+
,
4

+I
4t I
I
m t
V
7 1
d l
&A
>
d l

I
II
C
-U. c
c
-
<
2
z-
I-
a
<
c
-
7
U
C V
2 < C
0,
-
z
c
3
c
L
C
I-
C t V
C U
U ii >
I
c c
4
C
>
c
<
I c
C k
(r
< U
7
a c C
U 7
c
2
c
I
L
a
4
U
U
a
U
> n
-c
U Y-
c
C

2
4
-
I
0
U
*
4 v)
v
a
2

139
.? +I
I
&I 9 1
a I-+
a
a +I
I
d l
e+
I
+I
a I
a 9 1
a 9 +
a
w +
a I
2 I
I 4 1

-
I- 9 +
2:
a
L1
a
+I 2
c
.
z
c
o.
a 0
a $$ 0 0
+I
I
4 1
* In+
+I 0,
9
I
1 2
e+ N
+I m
In I
4
0 4
I
1 4 I

I .?+ m I
c
U a a +
z 2 I-
a
1
3
3I- m
a
0
9,i
c z
13
2
I
3
sa
0
0
LI
I-
a
I
+ a*
I
I
d l
m+
a
c
4
?
m
cy
9
01
0
N
e
a
m
.
I?
N
9U
4
I
,
I
a

. .
I-
C c U + 4
4
N
r-
d
N
cy
0 I
C 3 v) I
w I c
U
I
C
4
c(
Q
4: 0
m
rl
0
3
rt
d I
.c z
>
a
0
ul
c
a
I-
0
c.l
t
U
W :$
+I
I
L
c

n
0
r-
9
r-
d
N
011
d

VI
a
6
4-
'm+. d
.
In
VI

rn
I
,
I
I
dI?
.
O v)
2 +I 0 m 0 h OD I
z
a
U N N VI N
e
I
,
a I
n m
e I
rl
@ rl
U
I- S! n rl d d $4 I

.
2
U + U
0 s d
a
rl
N
I
1
c I
I rl 9 I
x z: *
L
m
e
Q
I 0

.
c 4 I
U
U
V m +
+ N
U
U

>
-
c
r-+
+ N
*
N
1
I

2
-I Q + b m F 4 <v 4
4
V
4
2
e m F m 4 1

C
c
Y\ C
d
cy
(r d \

U N 4 VI OD .
140
 v) I I +
n a I it
UJ U I
a 0
D $1 0
u)
U I + I-
I- a I 4
u1 U I z
a I
$4 a
+I
I
?-
L
-
z
I-
a
a
UI
I-
a
9
91 3 0 0
+I a
w
1 t
u
:I J
+I
I
Z
- -
z
l-
a
a
x
0
u
L
$1 3 0 u
cn
+I
:1
I .
3
3
3
P
0
0
0
0
. . a
0

it
+
I
91
U +
+
I
-
4
a
I-
9
0
4 1 0 0
U + d 4 -I
'11
a + 2
2
!- I
4 I 3
C 9 1 I-
m+

.
2 O
C c 3 C 9 2
c t 3 C I- I
c I I
n 4 9
c :4 2 c 4 I-

:
a O
c
V

2 1
9 1
nl+
+I n
P
4
Y
. U
0
c
n
VI
(I,

4
h)
. 9
2
C
c
c
c
4
I
d l
cy+
+ n
d

0
d
. ;1U
N
P
9
4
. >
m
Am
9 E

I
91
4 +
+I 0
Y

2
0
C

..
4
0
0
9
4
. K
W
k
z
u
+I n U VI u
I
3,
0
-4
9
0: I
u
r(l d
d + N * N K
a
w
01+
+ ul
W
a

r-+
+
N
. r N
. >
W
-I
0
z
4
-I
I
a+ N r N 4
ul
4
z
cnl
Z I VI+ N r N
I4

. I VI w
b v I-

.
it

2 VI
rl **

141
 REFERENCES

1. McDonald, Alan T.; and Fox, Robert W.: Incompressible Flow in Conical
Diffusers. Tech. Rep. No. 1, Army Research Office (Durham), Project
No. 4332, 1964. (Available from Armed Services Technical Information
Agency, U.S. Department of Defense.)

2. Reneau, L. R.; Johnston, J. P.; and Kl.ine, S. J.: Performance and Design
of Straight, Two-Dimensional Diffusers. Trans. ASME, Journal of Basic
Engineering, Vol. 89, March 1967, pp. 141-150.

3. Rouse, Hunter; and Hassan, M. M.: Cavitation-Free Inlets and Contrac-

tions. Mechanical Engineering, Vol. 71, March 1949, pp. 213-216.

4. Pope, Alan; and Harper, John J.: Low-Speed Wind Tunnel Testing. John
Wiley & Sons, Inc., N. Y., 1966.

5. Pankhurst, R. C.; and Holder, D. W.: Wind-Tunnel Technique. Sir Isaac

Pitman & Sons Ltd., 1952.

6. Wattendorf, Frank L.: Factors Influencing the Energy Ratio of Return

Flow Wind Tunnels. Fifth International Congress for Applied Mechanics,
Cambridge, 1938, pp. 526-530.

7. Idel’chik, I. E.: Handbook of Hydraulic Resistance. AEC-TR-6630, The

Israel Program for Scientific Translations Ltd., 1966. (Available from
Clearinghouse for Federal Scientific and Technical Information, U.S.
Department of Commerce.)

8. KrEber, G.; (translated by Dwight M. Miner): Guide Vanes for Deflecting

Fluid Currents With Small Loss of Energy. NACA TM-722, 1933. (Transl.
into Engligh of “Schaufelgitter zur Umlenkung von Fliissigkeitsstromungen
mit geringem Energi everl.ust TngPnieur--Arkiv,Vol - 3, 1932,
) ”

pp. 516-541.)

9. Henry, John R.; Wood, Charles C.; and Wilbur, Stafford W.: Summary of
Subsonic-Diffuser Data. NACA RM L56F05, 1956.

10. Moore, Carl A., Jr.; and Kline, Stephen J.: Some Effects of Vanes and of
Turbulence in Two-Dimensional Wide-Angle Subsonic Diffusers. NACA
TN 4080, 1958.
11. Cochran, D. L . ; and Kline, S. .J.: I J s e of Short Flat Vanes for Producing
Efficient Wide-Angle Two-Dimensional Sub-sonic Diffusers. NASA TN 4309,
1958.

12. Wallis, R. A. : Axial Flow Fans. Academic Press , N. Y. , 1961.

13. Chmielewski, G . E.: Boundary-Layer Considerations in the Design of Aero-

dynamic Contractions. J. o f Aircraft, Vol. 11, No. 8, Aug. 1974,
pp. 435-438.
142
14. Eckert, William T.; Mort, Kenneth W.; and Piazza, J. E.: Wind-
Sensitivity Studies of a Non-Return Wind Tunnel With a 216- by 432-mm
(8.5- by 17.0-in.) Test Section - Phase I. NASA TM X-62,171, 1972.

15. Eckert, William T.; Mort, Kenneth W.; and Piazza, J. E.: Wind-
Sensitivity Studies of a Non-Return Wind Tunnel With a 216- by 432-mm
(8.5- by 17.0-in.) Test Section - Phase 11. NASA TM X-62,307, 1973.

16. Mort, K. W.; Eckert, W. T.; and Kelly, M. W.: The Steady-State Flow
Quality of an Open Return Wind Tunnel Model. Canadian Aeronautics and
Space Journal, Vol. 18, No. 9, Nov. 1972, pp. 285-289. (Also NASA
TM X-62,170, 1972.)

17. Liepmann, H. W.; and Roshko, A.: Elements of Gasdynamics. John Wiley
& Sons, Inc., N. Y., 1957.

18. Staff of Ames Research Center: Equations, Tables, and Charts for
Compressible Flow. NACA Report 1135, 1953.

19. Sovran, Gino; and Klomp, Edward D.: Experimentally Determined Optimum
Geometries for Rectilinear Diffusers with Rectangular, Conical or
Annular Cross-Section. Fluid Mechanics of Internal Flow, Gin0 Sovran,
ed., Elsevier Publishing Co., Amsterdam, 1967, pp. 270-319.

20. Annon: Low-Speed Wind Tunnel User Manual. Lockheed-Georgia Company,

ER-11,000, 1970.

21. Krishnaswamy, T. N.; Ramachandra, S. M.; and Krishnamoorthy, V.: Design

and Characteristics of the 14' x 9' Open Circuit Wind Tunnel. Proc.
of the 11th Seminar on Aeronautical Sciences, National Aeronautics Lab.,
Bangalore, 1961, pp. 417-434.

22. Krishnaswamy, T. N.: Selection of the Electric Drive for the 14' x 9'
Wind Tunnel. Journal of the Aeronautical Society of India, Vol. 7,
No. 2, May 1955, pp. 19-28.

23. Krishnaswamy, T. N.; and Ramachandra, S. M.: Fan System of the 14' x 9'
Open Circuit Wind Tunnel of the Indian Institute of Science. Journal
of the Aeronautical Society of India, Vol. 18, No. 2, May 1966,
pp. 47-61.

24. Kirk, J. A.: Experience With a V/STOL Tunnel. Journal of the Royal
Aeronautical Society, Vol. 71, Sept. 1967, pp. 606-622.

25. DeFrance, Smith J.: The N.A.C.A. Full-scale Wind Tunnel. NACA
Report 459, 1933.

143
 TABLE 1.- NUMERIC INPUT CODE DEFINITIONS

Code
Code t y p e D e s c r i p t i o n of code meaning
value

Tunnel t y p e 1 Closed t e s t s e c t i o n , s i n g l e - r e t u r n t u n n e l
2 Closed test s e c t i o n , d o u b l e - r e t u r n t u n n e l
3 Closed t e s t s e c t i o n , non-return t u n n e l
4 Open-throat, s i n g l e - r e t u r n t u n n e l
5 Open-throat, d o u b l e - r e t u r n t u n n e l
6 Open-throat, n o n - r e t u r n t u n n e l

+
Units of measure

S e c t i o n shape
1
2
1
I n t e r n a t i o n a l System of U n i t s ( S I )
U.S. Customary U n i t s
Circular cross section
2 Rectangular c r o s s s e c t i o n
3 F l a t o v a l c r o s s s e c t i o n ( c e i l i n g and f l o o r
p a r a l l e l with semicircular sidewalls)
Section -type (See t a b l e 4)
P l o t type 10.0 No p l o t s
1.0 C u m u l a t i v e p r e s s u r e l o s s e s vs C i r c u i t
length
2.0 Wall p h e s s u r e d i f f e r e n t i a l v s c i r c u i t
length
>?.O Cummulative p r e s s u r e l o s s e s and w a l l
pressure d i f f e r e n t i a l vs c i r c u i t length
(on separate p l o t s )

144
-
m
u
-d
c
3

n
rl alwa
al a al al r l d al
a
a za n
23
a n w u
m n h al
s u
a 8
0
U 7.1
al
1 0
vd
c c
0
d
aJ
k
30
a rl m 3 4 al U u 7
n
m
G w
k al
s 7.1
d
k 0
o u
k
m
2
M
a
C
M
m
m
al 5
al

v E 0 k al W 1.1 a a k
a u a v) u u a
al k m 0 h .I4 M m m
+I k n a a fi u C W W 0
U u r l rl 7 d C d .d
a m 3 W 0 0 N rl rl
a
k
rl Mu d 01.1 a
.rl al
k
V n O
$2 C lu
P a
W rlu
aJa
6u
1.1 0
JJ
0 .da
m rl a U
a u c ?rl U U U

n al
rl
al d
gal k U
0 0
d
U n
7
c0 u4
a
0 c
I 0
z7 us
7.1a V d a 8h dal 7.1 u a
2 u m rw al
U *
aal u u U u k U a
u
:- m
mal u u
0 m a
7
U
alk
d 0 3 u krl
u a m a ala
0 m m m a G
al
al
m u o
a
3m 3';f
a: a
m a l
b
u o
rl al
u
W m v
9)
al
7.1 U
a
u.4
i-
ma
al
? umal k
a m u u
al X
w
- r9 X El H H

rl
Ual m k k k k &I
aa
ah
aalaJal
MMMbO
al
M
c u alalalal al rl rl rl rl rl r l d
H u u u u u a a a a lu a a
c
C C O C
d d d d 2 2 2
- H H H H H

%*
U
n
0
h
0
8 a
n
rl
4rl-a rl a rl a rl
rl
W a
rl
W
01 cdcdual a al lu al u al u
k C C r l k k k rl
rl
7
k
7.1
7
oos7.1
7.17.1cd7
E
l
7.1
k
d
7
g
7.1
7.1
3 d
rl
3
a
d
3
7
a
w w u u w w u w u w U W w w
a a a
d e
al aaalal
?one 0 d 0 d 0
al
n d al
n

w
c3
M
i-i i4
w
m
z E+wU
2
G cn rl
d
n
n
E
4
3 m 2 X rl
3 s3 0
3
H
PI
H
H PI

m o m
r l N N
4 elmum \o I I I
rl
d r f N

145
 - ~ ~

u u u u u N U
m w u - l w w w k - u w
U I l l I I I O U N w
d I l l I I I k k k k k W E l k
G I l l I I I 0 0 0 0 0 E l l N o
3 \urn M U
6 E E E E EW E E
- O w

o ( d
TI aJ - k
0) u k G G O
a C d k C d O 0
""a,
b)
n a ) 0
3
k 0 l . d
3 u
9 4 -
u s

n
m
W

d
U
d
a
b4
V
m
al
t
a

U
M N " N
6E
H U
aJ0a
M M M
aaJa
u u u
aJ
M
a- rl rl 74 rl rl rl
E G G
H W H
U
c
H
3
d
(d

d
(d

2
(d

2
(d

2
(d

2
n n
0
rl
0
rl a
..
aa a a
a W W aJ aJaJ a al al
aJ U u n nn a, n n
k l-i rl
.d 9 3
5 a a
w Iy Iy
hi hi
d n FI

rl N
n n
m
rl N
5 J X

ul
Nt
.
I
*
rl

146
-
u
m E w k 1
U d I I a, h
d w(da I I ru) w > O N rl
C 00al bo I I - 1 I 0 0 rl
3 o w al 00 k O 0 (d a,
M d c d a rlrl O d N N r l C P
k
a,
.
)
U
k
% k
(do
g
d fi0 d
C

al u C
.
I u u 0 (d
o u ( d a
uc)
7
k
0
k k
2
U
0
LI
ala a l a
.) m
7 u u u P rn
ad c u
n
(XI aJ0
a (d 8
U ( d ( d
8 0
rl
7
k
0
W M O m k k (d
C ( d r l ( d ( d d w m
a a ru c
d
0
U
3al
o s
0
m m W
a
a,
0
.r( .
a ro m m
d
k
d
P
U
C
m 0
0d d
0
0
U
5-rl
0
m
l d d
al
d
$ s
al d
3
k C
n (do
d
fii
a,
U
d a,
n

w a
g:: l
d
k
U
v1
w
a,
h
U
w k al 0
U kvlmal
d m
(dP
al 5al
E o w m a,
% U
M *rl
C
0
0 0 wow- e (d
O d k U
3 u t a ta
I a,
v)

ual
s a
T

ah
c u
$ k
H
a 0
k
$a k
a,
- a,
a U
7
(d nn
V I 0
n
0 U
7
n
2
.rl
..
6.
u dQ) 0 a
C 0 . c 0
aJ a O d a rla -rl u
w w al
8
k
a,
n u u
drf
n
-a,
un
rl
rl
(d
a,
7
4 7 ' a
1
CT
u
0 w w
1 9
(da U
U
(do
w 0
g
.rl a
al
d cn
@ a ,
n t a
a,
cn
@a,
ncn
U
a
a,
U

- 0
m
(d
C
.rl

n
a,
U
(d
E
a,
m U U
a- .rl
dal a a,
a d
d U
c v)
.rl (d
&Id
U - U
1 -
a
- I L I
h
C
(d
n a
VI u a,
W rl a
ac 5 a
$67
ud
(d
ru
a,
-r(

%
0 n
- 0

147
 TABLE 4.- ADDITIONAL, SECTION-DEPENDENT INPUT REQUIREMENTS

Section
- d d it i o n a l
Card
input 2quir emen t ? a
YPe column ( s )
Type d e s c r i p t i o n title(s)
3de
-
S i n g l e ducts:
Test s e c t i o n , c l o s e d , c o n s t a n t 31 ---
a r e a , empty
T e s t section, closed, constant 32 SIAL Required 36-40
a r e a w i t h model BLKGE Optional 46-50
CD Required 61-65
D EPS Optional 76-80
T e s t section, closed, 33 KEXP Default 56-60
d i f f u s i n g , empty
T e s t section, closed 04 SIAL Required 36-40
d i f f u s i n g , w i t h model BLKGE Optional 46-50
KEXP Default 56-60
CD Required 61-65
D EPS Optional 76-80
T e s t s e c t i o n , open-throat, 05 ---
empty
T e s t s e c t i o n , open-throat, 06 sIAL Required 36-40
w i t h model BLKGE Optional 46-50
CD Required 61-65
D EPS Optional 76-80
Constant-area d u c t 10 ---
Contract i o n 20 ---
Corner, c o n s t a n t - a r e a , turning 30 CHORD Required 41-45
vanes o n l y PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Corner, c o n s t a n t - a r e a , w i t h 32 CHORD Required 41-45
t u r n i n g vanes and w a l l s PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Corner, c o n s t a n t - a r e a , w i t h 33 PHI Required 51-55
walls and w i t h o u t t u r n i n g KT 90 Default 56-60
vanes
Corner, d i f f u s i n g , w i t h 34 CHORD Required 41-45
t u r n i n g vanes and w a l l s PHI Required 51-55
KT 90 Default 56-60
RNRE F Default 66-70
Diffuser 40 KEXP Default 56-60
E x i t k i n e t i c energy from flow 45 ---
dump
Sudden expansion 46 ---
-

148
 TABLE 4 - AZDITIClNAL, SECTION-DEPENDENT INPUT REQUIREMENTS - Contftlued .
Section
dditional
Card
input Requirement? a
Type description Type column ( s )
title (s)
code
~

Flow straighteners, thin 51 L/DH Required 36-40

honeycomb PRSTY Required 46-50
RUFNESS Defau1t 66-70
Flow straighteners, thick 52 L/DH Default 36-40
airfoils PRSTY Required 46-50
Perforated plate with sharp- 53 PRSTY Required 46-50
edged orifices
Woven mesh screen 54 DMESH Required 41-45
PRSTY Required 46-50
U4ESH Default 56-60
Internal structure (drag 56 ITEMS Default 9-10
item(s)) at upstream end of S/AL Required 36-40
section BLKGE Optional 46-50
CD Required 61-65
D EPS Optional 75-80
Fixed, known local loss item 57 K Required 61-65
at upstream end of section
Multiple ducts:
Constant-area ducts 61 DUCTS Required 7-8
Contractions 62 DUCTS Required 7-8
Corners, constant-area; 70 DUCTS Required 7-8
turning vanes only CHORD Required 41-45
PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Corners, constant-area, with 71 DUCTS Required 7-8
turning vanes and only one CHORD Required 41-45
side-wall each PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Corners, constant-area, with 72 DUCTS Required 7-8
turning vanes and walls CHORD Required 41-45
PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Corners, constant-area, with 73 DUCTS Required 7-8
walls and without turning PHI Required 51-55
vanes KT 90 Default 56-60
Corners, diffusing, with 74 DUCTS Required 7-8
turning vanes and only one. CHORD Required 41-45
side-wall each PHI Required 51-.55
KT 90 Default 56-60
RNREF Default 66-70

149
 TABLE 4.- ADDITIONAL, SECTION-DEPENDENT INPUT REQUIREMENTS - Concluded.

Section
Additional
Card
input
Type column(s)
Type description title(s)
code
~~

Corners, diffusing, with 75 DUCTS Required 7-8

turning vanes and walls CHORD Required 41-45
PHI Required 51-55
KT 90 Default 56-60
RNREF Default 66-70
Diffusers 84 DUCTS Required 7-8
KEXP Default 56-60
Vaned diffuser 85 ---
Sudden expansion from multiple 86 DUCTS Required 7-8
ducts to single duct
Sudden expansion from multiple 87 DUCTS Required 7-8
ducts to multiple ducts
Fan, constant-area annular 91 DUCTS Default 7-8
duct(s) with motor-support ITEMS Default 9-10
strut( s ) s /AL Required 36-40
DHUB Required 41-45
BLKGE Optional 46-50
CD Required 61-65
ETA Default 71-75
D EPS Optional 75-80
Fan contraction(s) to annular 92 DUCTS Default 7-8
duct(s) with motor-support ITEMS Default 9-10
strut (s) DHUB Required 41-45
BLKGE Optional 46-50
Fan diffuser(s) from annular 94 DUCTS Default 7-8
duct(s), each with tapering, DHUB Required 41-45
cone-shaped centerbody BJXGE Optional 46-50
KEXP Default 56-60
Internal structure (drag 96 DUCTS Required 7-8
item(s)) at upstream end of ITEMS Default 9-10
each duct S/AL Required 36-40
BLKGE Optional 46-50
CD Required 61-65
D EPS Optional 75-80
Fixed, known local loss item 97 DUCTS Required 7-8
at upstream end of each duct K Required 61-65
'"Default" indicates the input is optional and has a default value if omitted
(see table 3 ) .
"Optional" indicates the input may be selected and included as desired.
"Required" indicates the input must be non-zero and included for all sections
of the specified type or the section will be skipped and the case not
completed due to input error.

150
 TABLE 5 . - CASE TEWTNATION TASK DESCRIPTIONS

Card Input Input

Task description
column (s) type value

3-4 Blanks Blanks Case termination card identification

6 Integer Summary characteristics page(s) :
0 Non-pr int
#O Print
7-8 Real Plotting of summary information as a function
of distance through circuit:
30.0 N o plots
1. Cumulative pressure loss
2. Wall pressure differential
>2. Cumulative pressure loss and wall pressure
differential
9-10 Real Complete, annotated tabulation of input
values :
0.0 No print unless internally forced by
omission of required inputs
fO.0 "Cho sen" tabulation
11-15 Real Power-matching (optimizing velocity for a
specified power level) :
0.0 No velocity optimization
fO.0 Velocity optimization
16-20 Real Return to beginning for evaluation of another
case :
0.0 No return, program termination
fO.0 Return

151
c
cbp 7
4
h
W

1
3 3

> a

h
m rl
N
;
I a
II >3 24 1
a . m
U 0
4 D.3
3

I I
I I

I II

.
.

0
m u d
U
e m
o m m
O d U s
Mc
3
Y u .-
U P P
4 %
u m w m
0 u u
u m c
c m v 0)
m m d >
M U m
2 ,mgu.4
-3
g
w u
m e m o
s m m m
9 u 3 m
c c c c c
m a m
rl
3 3 -+c
a c
.M .
M .Mom .
M
w P C U U
m mc
vm cu a m V G
o m 0
m u s 4
o m
m u
m u
0 0 0 u u r l o u
3
0 0 d c
u m 3 O d d ru l c
m
u e v c 3 w

c
rl
. crl
c
a a
-1M m >
u o
U P G
3 w m
M ~ U
I U
c o u
.4 4 m e
z m
0 u4
4
m u &

'u
rl w N 0
m e .
.4 rl 'u
M O 01
cc c
o m -
m. P -
U
I - c
P O
m v d
w m u
c w u
* a w
u m w
0
.
.I
 0
0

0
co

0
(D

0
d

0 m
hl
k
01
u
3
k
crl
a
G
bo
co
0 al
v- a
k
al
co
3
w
w
.rl
(D
!a
I
4
al
k
3
d bo
c, rl
0 Ih
0
Y-
0
co
>
4
3
hl

ea ..
2
z
a,
+
0
Z 4
\
\
c

153
 a
a
a
9
rl

rc
0

154
\
+
C
8-
c .E
c-"g
0
d
d
u
-4
C
d
w
a,
T
J

u
I?
8
0
a,
M
I C
0
.rl
u
V
rd
k
u
V
n
a
W

155
 m
a,
U
tecl
a,
w
a,
ec
--- ec
W
a
a,
a
3
rl
a, U
E3 E
V
U
I
e
M hl
.rl
v1 a,
a, ec
a 3
M
E
.d
.rl
Fr
c,
U
(d
ec
u
u
co
n
P
W

-7

oooa A ~3 a

156
 aJ
d
M
C
Id

-I+
i??
C
Fc
3
u
w
m 0
aJ

s :
C
0
-rl
aJ u
-a LI
-I+ C
3 3
M w
M m
C (d
.I+
m
E3 C
Fc
U 3
u
c
u Fc
-84 0
3 w
m v)
C aJ
Fc m
3 m
H 0
d
n
(d d
W (d
u
0
4
I
m
aJ
Fc
3
M
.I+
k

157
 v1
a
!
3
ia
a
a
7
u
7
0

I
v1 m
d
Ll 9)
7 Ll
k 3
M
h
P
v

158
 J,
W
z
n
>Q

160
 at
r4
M
E
(d

a,

U
U

rl
s
rl
7
CT
PI
W
0
E
0
rl
U

27
W
rd

3
U
c
aJ
.rl
U
rl
u-l
u-l
aJ
0
U
v)
v1
0
rl
k
aJ
In
7
W
u-l
art
a
a
aJ
G
3
I
\o
aJ
b
7
M
d
crr

161
 Main Program (PERF0RM)

fead
Title Card

Card Error, Transfer to

Section
-------
Subroutine DATACK Geometry

-------
Skip Analysis, Only
U

nits
Basic
Parameters Measure Parameters v
Test
n
Default Valve
for Diffuser
Expansion L o a

Section Flow

to Section

(a) Main program.

Figure 7.- Basic functional flow chart.

162
a
al
a
1
rl a
c”0 al
5
C
V d
U
I C
0
6 V
(d
Fc I
80
0 h
Fc
a al
Fc
E 7
M
i
n
d
cr
(d
v

163
 Data-CheckingSubroutine (DATACK)

Section A Tabulation of
Card Checking Input Data

-
"
# Card
__ Checkina
- 4
Assignment of Fixed
Determination
Required General
of of
Determination
Required Format Parameters
Inputs w

I Checking
Checking and Setting Requirements and
Error Codes Filling Arrays with
1 Data or Message

Checking and Setting

Error Codes
d

(b) Data-checking subroutine.

Figure 7.- Continued.

164
 (3
Local Speed Subroutine (SPEED)

Start

Local Mach Number

and Velocity

1
Return

Local Reynolds Number and Friction Coefficient Subroutine

(F R ICTN)

P 1

Start

Reynolds
1
Number

1
I
Friction
Coefficient

T Return

( c ) Local speed and Reynolds n u m b e r / f r i c t i o n c o e f f i c i e n t s u b r o u t i n e s .

F i g u r e 7.- Continued.

165
 Section Analysis
Information
output Summary Information
Subroutine Plotting Subroutine
(OUTPUT) (PL0TIT)

D Start

Definition of Plot
Section Titles Titles and Parameters

Section Type
us Units of

SI
Modify Parameters
for SI Plots

Typical
Section

Section
Q Pressure
Information
Losses

Plot Wall
Pressures

*
1
No
Plot Pressure Losses

Return

(d) Section information output and plotting subroutines.

Figure 7.- Concluded.

166
h
(d
W

167
 (d
91
3
0
Q)
91
Ll a
5 Q)
a
a
Ll 4
0
W

s
Q)
d
a

168
h

rn
U
Q
aJ
c
rn
a
U
a
a
U
1
2
d
&.I
0
w
rn
M
C
d
a
a
0)
c
Q
4
a
6
a
VI
I
Q\
aJ
&I
1
ta
r(
h

169
170
 !
i
1 !
! I [
'0 i
I8 I C
i
I
I ! *
VI II
I F
I I
4 i
i
! i
0 0
1
0 0;

i
rl dl
I 0 0:
' 0

i I

i
i
I

I
*I
!i
v1

!i i
I
d
rl
rd

! I

II
iI
I
I

171
 4
+I 0
0
tn 31 rl

. .
0 0
Y I 0 0
r! I
a
n
9 0 0
0 0 a
4
+ 0
QI
In
IF
rl

. .
I

.
tn N
I
I
0
0
2 3
+ 0
0 0
0 3

il
+I
I
I
I
.
N
r-

*
In
VI
,

*
-

-I
I
e
0.
m
0
e

+
+I m0, **
..
0

..
9 * 0
m 4
o m
4
-4 z6" Q
3 4

+I 00
0 .
3 0 0 0 3

-
4
2
3
+I
I
a m
L
40
r-
..
U
?
P
n
J?
(Tm
mcu
..
9 4
00 0
N 4-
I -4
I- t
C
2
c
3
+I
1
I
r-
m
0
. 3
. .
cr
r-
U
Q

co
-
t
L
I
I
+
U 4

1
. .
1
5
a
21
a
t
I
I
+
0
0
0)
. h
m
4
0
0
4

4
a
+I
I
00
00
0 . .
-4
D
-I-
a m
. e
m
..
In v
-7

I N m m v
U
c

.. . .
2
U
c +
I
40
r-m
.. -.
P (Td
r-0
0
tn n
0

I mu\ 4 4 0 cn v
Nu9 P mln .s 1'
4 4 N
N
D 90, (T n

+I
I
I
I
corn
a m
N,U
tnN
.. 3
4

Q
n
.. Y .
cry
4ul

mr-
m m
0

N
4 4
E,
v

t
I
1
I
@o,
m 4
0-
U.4
0 . .
>
P
3
\1
.. .
3 N
e m
3ul
WCJ
9
9
N
4

tI t O
U >
I-+
U V
I- 3
s5
U
I 'U J I LUW 8
all L (La cr 3

UI
-I
L (2 2
I z 0
I I
I
I Ul I-

i
I
2
t-
v.
-
I:
I-

3 cf
u
a
a
t
1
I
U
4 u:
UI z
I U
v) 0
Q ' 3 U
z
g!
r U
rY y. z
0 4
U! u U
L ul (T
e

172
 d
m
m
N
I-
N
*
3
c
N
s
0
d
0
N
0
0
0
N
0
e 9 9 3 9
0 0 0 C 0

N J OL N (F
0 P 9 bt 9
0 r- N

. . .
N N
In cn
OD
3
e
0" d
0
0 d

0 0 0 0 0

co
4

r:
N
9
0
m
m
e .* * .
d

m
m
9
4
*
Cl
U
.t
In
a
PI
a
1
l-i
c)
In m CD
g
.
130

. .
d
m 9 m 1 rn m W
d 0 O C 0 0
e e
0 0 oc 0 0 I

:
$
4

N
. Q
*
N-
O m
w
9
a0
N d

i
N VI? 4 A

0
$
m
9
9
.
r-
e
d
al

3
9
N
m
f-
d
e I
0
w
k

. . . E4.
v m oc
PI
m 9 oc
0
0 a
N .t mc 41 W d
d
a
.
* VI- d -4
k
.Y PI
d I-). IC r-
? e I

s Q
N
mr
N r
m
N
m
N
a
s
d
9
2 d.
N r
.
.
N
I
N

0
"4

N
d
. 9
3
rl
4
mc
0
e
NC
VI1
1
m
In

N
VI
*
OD
m
N
In
. d
JJ
u
0)
(0

l-i
m
401
N
.
m:
mr
*.
0 :
m
?
0
.Y
o\
7-l

*
0
PI
e
e
7
JJ
a
4 +I
i
U
cz
n
I-
u
u
t l
V I
I
UI
I-
cl
u1
I-
V
Lv 3
c
.rl

3i
U U s o! t
z
n
>. I- P
0 v) U v) W
UJ' m w 3 ul
+ 3> z
3
i
(Z
I- 4
2 LI
I u I lr I
z
0' I CJ
a
I-
U
a I-
M

I-I
/I t-
I VI U 3c 3
U U U z
UI;
W!
+
U
0 -
!!
U
a
UJ
L
4
t
a
Lu
4
! z U U L !Y

! 4 c
C
0 C
L
c
u
+I
U c)

z
.i
0;
+I
-
c (r
,-
*
d
IC
r -
9

173
 m

YI
2P

I
I
a
I
B
IL
1
2
I
t
z
3
u4
I
I

i
.d
u
2
0
u-c a
d

c aJ
. *
-
: z
d
d
C
!5
d
t- 0 U
C d E
2 u 0
c
3
a V
N
d I
t
C a
d rl
C
U u rl
I a
C
a
0 aJ
k
> h 1
u M
d d
I u Er
C 0
4 rl
a
U
I-
2
2 n
U c)
L W

I
c
U
4
U
v
U
V
U
I
4
I
v
4
7

174
 +
I
r l
I
11
3 1
i lI E
0
I d
I

>
a
r
8
0
W
-
3
W
+
I
G
d

V
I
I
m
c I 0
c I d
c U
il
c
I m
OL
d
+I d
ec
I- aJ
C: U
3 4 U
E
I
3
IJ:
4
I + 2a
c 1
c
0 c
I
2
0
U
I
-
c

c
I
I
R
a
m
>
X

I
U
e
c
-
U
d
+
1
1
v1
I
a 3
4
I-
2i l1
a I - I r(
(u C J I U
c z : I 0
z c I Q)
u!
U
3 + cn
I n
u a
W
z.
4
w
VI
+
1
w I
Lz I
t
w
I

i
+I
Ly
I
4
I
v:
4
z

175
 a
aJ
1
C

C
0
.. u
0
w I
C
rl
rl

176
177
 m

C
n
c C I
I C 4
I
I

c
-
C
,
Lp
-
4
c

I a
I C
I
c
i
c
2
a
c C (r
4t I C
I
1 : L;
c
”
4
-L- U
L

I
r
(d
u
I a c
, c:
c
(d
a
z
Ic ;f C
a u
5 c rf
2 1
a
- - 4
II
I- t
4 I-
a
P
2
c
U 3 c
C
c L
.r( a
2 a 9)
1
3 4 -4 c > v VI
c 0 I I- Y 0 E
c n 2 .d
C
2
4
c
d
t
z
0
i x
-
a
2
C
r
I
g
d
u
c C u c a 0
3 I C 4-l u
I- c cz
w
I
I *n I- (d
rl I
C n I-
w -
Y 1
C
U
I
L U
L
c 0
-
C
P
(d
d
d

.
I r
u
C
C
c
-J
w
I n
c
C
U a
9)
k
> 2 a v) al 9
c z u M
a ! a
v)
w
rz (d d
I 2 6 t Ir c u
C c I a .k
4
a
4 C
4 s
.I
I- t v
2
C
c
2
a
I 4
I- h
U 4
L 2 w
I W
1 S
L t a
a U
a I-
U
V U
U v
a t c 4
c 4
U I
U U
s
4
I c c
a c I
U
4
2

I
I

178
 +I
I
9 1
r-+
+I
I
d l
r-+
tI ?
!-
9 1 a
9 + P
+I
I
-(I
9 +

9
+I
I
1
z
.
I

c.
a
.z
c
0
In+ 0 0
+I 0
0
I a
d l 0 Q)
In+ 91 3
C
+I c
> .rl
I c u
*+
9 1
.
.
e
I El
V
+1 m
.*
c
0
z I (0 v I
u d l e e
c
c
*+ 4 d (d
U
4 4 +I (d
2
t
U I
a
e 0 9 1 u
4 m+
z 1
. . .
C
0 + a
U
CD 0 >
4
N
UI
>
-4 $
4 c.l
I ? 4

-
c
2
c
a
I
a
:l
n
U N
m
m
a

n
h
.?
v
-l
rl
w
c
n
t
U
v)
UI
0 91
N +
+I 0
(r

C
-4
z*
N
4
0
3
3
e
-4
.
rl
r,
a
4
.
ry
m
m
ry
0

rl

. .
2
C
U E! +I -4 N
r- 2 4 m
4
N
P
U
(d
c t I e e m rl
4 u 4 1 U v\ 9 D 4 e 3
C
C
YI N + c U m r) U P P

. . .2
v)
z +I a
U
m
In
0 3
0, 3
n
(d
u
5 9
I
1 n N
4

m
4
e
rl m
e
a
4 + U VI m m Q)
u
+I u CT
m
0
CP
U 0 Q
9
(d
I
(r
* e
I\
e 0: e
0
z1
+
C
4
0
-?
0
(v
3 0
N
N
+
*
z
4
m + N n
w
v

r-+
I+ 9
e

9 + 0 m h( U N 4

P N .
II Y N N

C 0 cu 3 0 cu
c N 0 -l .* rn

N UI 0 al OI

179
 + c
>
I-
a n
L L
a
t- i Lo Lu
+I 0
0
I e
In
Zl 0,

+
I
I I-
n
$1 0
+I 0 Q
0
E:
x
I rl
d l 0
9 +
+I d
?
I I-
(L
1: 0

:I
a si
+I

+ *
0,
In
9
0
90
Q,

~
a
-0
1
d
U

ug
?
4
5
In
w 9
9 0 I
N Q,
e 0,
e
m
4 U
0 Q Q
3 f- a Q)
N
m
N
3
a
4 U 7
* e 7 r(
U

.Y
0 0
6 'JI m
In
OD $ 0
rl
e ? d 0
u
N m v N
v1
d 4 In In
0 I
OI
4
* Q,
me rn d

f 0 0
e
g
-r4
d
4 4 a
U
rl 4 Q k
0 4 ;f; d 7
VI 7 M
9
0
;i 9
d
0 d
R
Q,
4

cy
rl
. 0,
4

rJ
4
. m
In
e
N
In
m
In
e
cy
ul
U

a
a
U
w OI Q
m m c,

l
e 0
0 0
4 4

n(
. m
. n
u*
W

9
. 9
e

d rJ ly nl

rl 4 N cy

d Q N 0
e e m rl

0
4
N
d s In
d

180
181
 d

i?&
:c-
W

m
E

a
c

;
0

n
P
W
;
I-

182
183
 8EP'Z N

184
 Pi5 R t

4 *e

‘31-

1
Y
g 4 -.I
d
I \

185
 I-
r N i
;i ?

6 d
e
I

186
*U.S. GOVERNMENT PRINfING OFFICE: 1976 - 635-275/120