Iowa State University Capstones, Theses and

Retrospective Theses and Dissertations

Dissertations

1986

Unsteady flows in a single-stage transonic axial-flow

fan stator row
Michael Dale Hathaway
Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd

Part of the Mechanical Engineering Commons

Recommended Citation
Hathaway, Michael Dale, "Unsteady flows in a single-stage transonic axial-flow fan stator row " (1986). Retrospective Theses and
Dissertations. 8250.
https://lib.dr.iastate.edu/rtd/8250

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University
Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University
Digital Repository. For more information, please contact [email protected].
 INFORMATION TO USERS
While the most advanced technology has been used to
photograph and reproduce this manuscript, the quality of
the reproduction is heavily dependent upon the quality of
the material submitted. For example:
• Manuscript pages may have indistinct print. In such
cases, the best available copy has been filmed.
• Manuscripts may not always be complete. In such
cases, a note will indicate that it is not possible to
obtain missing pages.
• Copyrighted material may have been removed from
the manuscript. In such cases, a note will indicate the
deletion.
Oversize materials (e.g., maps, drawings, and charts) are
photographed by sectioning the original, beginning at the
upper left-hand comer and continuing from left to right in
equal sections with small overlaps. Each oversize page is
also filmed as one exposure and is available, for an
additional charge, as a standard 35mm slide or as a 17"x 23"
black and white photographic print.
Most photographs reproduce acceptably on positive
microfilm or microfiche but lack the clarity on xerographic
copies made from the microfilm. For an additional charge,
35mm slides of 6"x 9" black and white photographic prints
are available for any photographs or illustrations that
cannot be reproduced satisfactorily by xerography.
 8703709

Hathaway, Michael Dale

UNSTEADY FLOWS IN A SINGLE-STAGE TRANSONIC AXIAL-FLOW FAN

STATOR ROW

Iowa State University PH.D. 1986

University
Microfilms
Int©rn£lti0n3.1 300N. Zeeb Road, Ann Arbor, Ml 48106
 Unsteady flows i n a single-stage

transonic axial-flow fan stator row

Michael Dale Hathaway

A Dissertation Submitted to the

Graduate Faculty in Partial Fulfillment of the

Requirements for the Degree of

DOCTOR OF PHILOSOPHY

Major: Mechanical Engineering

Approved;

Signature was redacted for privacy.

In Charge of Major Work

Signature was redacted for privacy.

Fèr the Major Department

Signature was redacted for privacy.

For the Graduate College

Iowa State University

Ames, Iowa

1986
 n

TABLE OF CONTENTS

Page

DEDICATION x

SYMBOLS AND NOTATION x1

ABSTRACT x1v

I. INTRODUCTION 1

II. RELATED RESEARCH 4

A. Experimental Research 4

B. Modeling Efforts 11

III. RESEARCH FACILITY 16

A. Compressor Test Facility 16

B. Research Compressor 18

C. Laser Anemometer System 21

1. Optical configuration 25

2. Flow seeding 27

3. Laser optics cart and traversing mechanism 30

4. Laser anemometer signal processor 33

5. Compressor rotor shaft-angle encoder 34

6. System computer 35

IV. EXPERIMENTAL PROCEDURE AND DATA REDUCTION 36

A. Experimental Procedure 38

1. Survey setup 39

2. Preliminary LFA setup 45

3. Data acquisition 49
 111

B. Data Reduction 52

1. Velocity components 54

2. Summary of average-passage model 61

3. Deterministic-velocity correlations 64

4. Turbulent-velocity correlations 67

V. PRESENTATION AND DISCUSSION OF DATA 68

A. Uncertainty Analysis 69

1. Precision uncertainties 72

2. Bias uncertainties 73

B. Description of the Measured Flow Field 75

1. Stator-relatlve time-averaged description of

measured flow field 76

2. Rotor-relative time-averaged description of

measured flow field 81

3. Time resolved description of measured flow field 93

C. Unsteady-Velocity Correlations 108

1. Deterministic-velocity correlations 109

2. Turbulent-velocity correlations 129

3. Comparison between deterministic- and turbulent-

velocity correlations 136

VI. SUMMARY AND CONCLUSIONS 140

VII. RECOMMENDATIONS FOR FURTHER RESEARCH 143

VIII. REFERENCES 145

IX. ACKNOWLEDGMENTS 157

X. APPENDIX A: NASA DESIGN CODE RESULTS 158

 1V

XI. APPENDIX B: STATOR- AND ROTOR-RELATIVE TIME AVERAGES 190

XII. APPENDIX C: ESTIMATION OF BOUNDS ON TURBULENCE COMPONENTS 193

XIII. APPENDIX D: WAKE ENHANCEMENT PROCEDURE 198

 V

LIST OF FIGURES

Page

Figure 3.1 Schematic diagram of NASA Lewis single-stage

compressor test facility 17

Figure 3.2 Meridional-plane view of research compressor flow

path 19

Figure 3.3 Representative rotor blade sections at three

spanwise locations 20

Figure 3.4 Representative stator blade sections at three

spanwise locations 22

Figure 3.5 Block diagram of laser anemometer data

acquisition system 24

Figure 3.6 Layout of laser anemometer system optical

components 26

Figure 3.7 View of optical access through rotor window 28

Figure 3.8 View of optical access through stator window 29

Figure 3.9 Schematic diagram of optics cart assembly 31

Figure 4.1 Research compressor stage 100 percent speed line

operating characteristic 37

Figure 4.2 Meridional-plane view of research compressor flow

path showing the 10 and 50 percent stream surfaces
along which the laser anemometer data were
acquired 40

Figure 4.3 Blade-to-blade view of the 10 and 50 percent stream

surfaces showing the actual laser anemometer 41
measurement survey grids

Figure 4.4 Illustration showing the relationship between the

actual velocity measurements and the various
averages used in the data reduction process 53

Figure 4.5 Research compressor coordinate system and laser

beam geometry orientation 56

Figure 4.6 Periodicity of rotor flow field 59

 vl

Figure 5.1 Contour maps of the stator-relatlve time-averaged

absolute velocity field along the 10 percent stream
surface 77

Figure 5.2 Contour maps of the stator-relatlve time-averaged

absolute velocity field along the 50 percent stream
surface 78

Figure 5.3 Measured extent of the stator suction surface low

velocity region along the 10 and 50 percent stream
surfaces 80

Figure 5.4 Contour maps of the rotor-relative time-averaged

turbulence field showing the rotor wakes and the
dissipation of turbulence 82

Figure 5.5 Rotor-relative time-averaged absolute velocities

along the 10 percent stream surface 84

Figure 5.6 Rotor-relative time-averaged absolute velocities

along the 50 percent stream surface 86

Figure 5.7 Illustration showing laser anemometer

measurements of the vortex street character of
the rotor wake 88

Figure 5.8 Measured components of the rotor-relative

time-averaged turbulence along the 10 percent stream
surface 89

Figure 5.9 Measured components of the rotor-relative

time-averaged turbulence along the 50 percent stream
surface 91

Figure 5.10 Contour maps depicting the time-resolved transport

and dissipation of rotor wakes through the stator
row along the 10 percent stream surface 94

Figure 5.11 Contour maps depicting the time-resolved

transport and dissipation of rotor wakes through
the stator row along the 50 percent stream surface 95

Figure 5.12 Time-resolved character of rotor wake along the

10 percent stream surface 100

Figure 5.13 Time-resolved character of rotor wake along the

50 percent stream surface 104
 vil

Figure 5.14 Contour maps of the total-deterministic-veloclty

correlations along the 10 percent stream surface 110

Figure 5.15 Contour maps of the total-determlnlstlc-veloclty

correlations along the 50 percent stream surface 111

Figure 5.16 Stator blade-to-blade distributions of the

total-determ1n1st1c-veloc1ty correlations along
the 10 percent stream surface 112

Figure 5.17 Stator blade-to-blade distributions of the

total-determlnlstlc-veloclty correlations along
the 50 percent stream surface 115

Figure 5.18 Axial distributions of the axisymmetric

total-deterministic- and Incident-velocity
correlations along the 10 percent stream surface 119

Figure 5.19 Axial distributions of the axisymmetric

total-deterministic- and Incident-velocity
correlations along the 50 percent stream surface 120

Figure 5.20 Contour maps of the scattered-velocity

correlations along the 10 percent stream surface 123

Figure 5.21 Contour maps of the scattered-velocity

correlations along the 50 percent stream surface 124

Figure 5.22 Contour maps of the cross correlations between

the Incident and scattered unsteady velocities
along the 10 percent stream surface 126

Figure 5.23 Contour maps of the cross correlations between

the Incident and scattered unsteady velocities
along the 50 percent stream surface 127

Figure 5.24 Stator blade-to-blade distributions of the

turbulent-velocity correlations along the 10 percent
stream surface 130

Figure 5.25 Stator blade-to-blade distributions of the

turbulent-velocity correlations along the 50 percent
stream surface 132

Figure 5.26 Comparison between the axisymmetric values of the

total-deterministic- and turbulent-velocity
correlations along the 10 percent stream surface 137
 v111

Figure 5.27 Comparison between the axisymmetric values of the

total-deterministic- and turbulent-velocity
correlations along the 50 percent stream surface 138

Figure 10.1 Rotor and stator blade manufacturing coordinate

system used by NASA design code 189

Figure 11.1 Blade-to-blade view of the rotor and stator at a

particular Instant In time, t , showing the
relationship between the rotor- and
stator-relatlve coordinate systems 191

Figure 12.1 Fringe-plane view of the velocity components

showing the relationship between the
ensemble-average and standard deviations of the
axial and tangential velocity components and the
measured velocity components 194

Figure 13.1 Illustration of wake enhancement procedure 199

 ix

LIST OF TABLES

Page

Table 3.1 Summary of single-stage research compressor design

parameters 23

Table 4.1 Relative coordinates of laser anemometer survey

locations along the 10 percent stream surface (stator
chord = 5.6 cm) 42

Table 4.2 Relative coordinates of laser anemometer survey

locations along the 50 percent stream surface (stator
chord = 5.6 cm) 43

Table 5.1 Estimates of bias uncertainties In velocity

measurements 71

Table 5.2 Comparison between calculated and measured rotor

wake segment nondlmenslonal d r i f t distance 97

Table 10.1 Design code Input parameters 159

Table 10.2 Design code predictions of aerodynamic performance 169

Table 10.3 Design code stage and overall performance

predictions 181

Table 10.4 Rotor blade manufacturing coordinates generated by

NASA design code 182

Table 10.5 Stator blade manufacturing coordinates generated

by NASA design code 186
 X

DEDICATION

This dissertation Is dedicated to my parents,

Mr. Clifford D. Hathaway

and

Mrs. Joanne Hathaway,

In appreciation of their continual love, support,

and guidance, without which this dissertation

would not have been possible, and to

Master Myung Hwan Kim

and

Master Yong Chin Pak

for Instilling in me, through their

traditional teachings of the martial arts of

Tae Kwon Do, Ju Do, and Hap Ki Do, the

s p i r i t , patience, discipline, confidence,

humility, and mental resolve to attain my

highest goals and realize my dreams.

 x1

SYMBOLS AND NOTATION

G Generalized velocity parameter

n Number of measurements or rotor shaft position

Nrp Number of surveyed rotor passages

NR Number of rotor blades

Ns Number of stator blades

Nsp Number of rotor shaft positions per rotor blade passage

N9 Number of circumferential measurement points

PS Pressure surface

R Radial coordinate axis or radial distance (Fig. 4.5), cm

R' Axis of laser beam bisector (F1g. 4.5)

^ Apparent stress tensor (Eq. (4.10)), kg/(ms2)

s Distance measured along streamline, cm

J Estimate of the standard deviation of velocity (from Eq. (4.3)),

m/s

SS Suction surface

t Time measured as a function of rotor shaft position, s

Tp Plenum temperature

Ts • Standard-day temperature, 518.7 °R

V Velocity magnitude, m/s

V/\ Average of stator Inlet and exit velocities, m/s

Vq Standard-day corrected velocity, m/s

Vps Freestream velocity, m/s

Velocity vector (Eq. (4.9)), m/s

Z Axial coordinate axis, or axial distance (Fig. 4.5), cm

 xi1

a Laser beam orientation angle measured from Z axis (Fig. 4.5), deg

B Laser beam orientation angle measured from e axis (Fig. 4.5),

deg

Bj Absolute flow angle, deg

Y Laser beam orientation angle measured from R axis (Fig. 4.5), deg

r Stator blade circulation, m^/s

e Circumferential coordinate axis, or circumferential distance

(Fig. 4.5), deg

0' Circumferential coordinate axis perpendicular to beam bisector

(F1g. 4.5), deg

y True ensemble-average velocity (Eq. (4.2)), m/s

p Density

a True standard deviation of velocity, m/s

<PR Angle between the beam bisector and the radial direction
(Fig. 4.5), deg

<P2 Angle between the fringe normals and the axial direction
(Fig. 4.5), deg

w Rotor angular frequency, s"^

R Rotor angular position (Fig. 11.1), deg

Subscripts

I Stator row Inlet condition

IB Inner bound

E Stator row exit condition

m Measured component

OB Outer bound

R Radial component
 xi11

T Total of axial and tangential components

z Axial component

e Tangential component

1 Measured component In direction of f i r s t beam orientation angle

2 Measured component In direction of second beam orientation angle

Superscripts

' Fluctuating component

Ensemble average

'WA Temporal phase-lock average

//'// Spatial phase-lock average

— Time average

A Aperiodic component

AX Axlsymmetric component

I Component due to Incident field

R Rotor-relative parameter

U Component due to scattered field

S Stator-relative parameter
 x1 V

ABSTRACT

Detailed measurements of the unsteady velocity field within the

stator row of a transonic axial-flow fan were acquired using a laser

anemometer. Measurements were obtained on axisymmetric surfaces located

at 10 and 50 percent span from the shroud, with the fan operating at

maximum efficiency at design speed. The ensemble-average and variance of

the measured velocities are used to Identify rotor-wake-generated

(deterministic) unsteadiness and "turbulence," respectively. Correlations

of both deterministic and turbulent velocity fluctuations provide

Information on the characteristics of unsteady Interactions within the

stator row. These correlations are derived from the Navler-Stokes

equation In a manner similar to deriving the Reynolds stress terms,

whereby various averaging operators are used to average the aperiodic,

deterministic, and turbulent velocity fluctuations which are known to be

present In multistage turbomachlnes. The correlations of deterministic

and turbulent velocity fluctuations throughout the axial fan stator row

are presented. In particular, amplification and attenuation of both types

of unsteadiness are shown to occur within the stator blade passage.

 1

I. INTRODUCTION

Unsteady Interactions are known to affect various aspects of

turbomachlne performance. Including blade loading [ 1 ] , stage efficiency

[ 2 ] , heat transfer [ 3 ] , noise generation [ 4 ] , and energy transfer [ 5 ] . In

fact, the fundamental mechanism for energy transfer In turbomachlnes Is

the unsteady throughflow [6]. However, virtually a l l existing turbomachlne

design systems are based on the assumption that the flow Is steady In

time. Consequently, extensive empirical correlations are usually relied

upon to compensate for an Inadequate understanding of the unsteady flow

effects. Unfortunately, empirical correlations tend to be specific to a

narrow class of turbomachlnes, and are usually only applicable for design

point predictions. Furthermore, empirical correlations frequently Include

effects due to other flow phenomena. In addition to the flow phenomena the

correlations were developed for. Nevertheless, I t Is obvious that some

designers of turbomachlnes have an excellent Intuitive feeling for the

effects of many of the flow phenomena Inherent In turbomachlnery, as Is

evidenced by the high efficiencies obtained In modern turbomachlnes.

Because of the limitations of the empirical correlations used 1n the

design process these high efficiencies are generally limited to design

point conditions and for turbomachlnes which are similar to previously

designed turbomachlnes. Perhaps the key to Improving our ability to

predict the off-des1gn performance of turbomachlnes, and to design

turbomachlnes which are considerably different from our design based

experience may be found through a better understanding of unsteady-flow

interactions. Unfortunately, since current turbomachlne design systems

 2

assume that the flow Is steady, there 1s no provision for incorporating

the effects of unsteady flows into these systems, except through the use

of empirical correlations.

A recent publication by Adamczyk [7] described the development of a

system of equations for simulating the flows in multistage turbomachinery,

which he called the "average passage" form of the Navier-Stokes

equations. This average passage system of equations provides a steady

state description of the flow field in an "average passage" of any

specific blade row. In a multistage turbomachine, the number of stator

blades In successive stages usually differs. Rotor blade numbers also

differ in different stages. Since stator (rotor) blade wake effects are

influential in downstream stator (rotor) rows, different stator (rotor)

blade numbers result in aperiodic flows in downstream stator (rotor) blade

passages. An "average" of the flows in the different passages of a

specific blade row is the "average passage" flow. The average passage

system of equations is ideally suited for use i n a turbomachine design

system. Unlike current turbomachine design systems, however, the average

passage system of equations includes terms which can account for the

effects of the unsteady flows. These terms look, in form, much like

Reynolds stress terms and include major contributors to the generation of

nonaxisymmetric flows in multistage turbomachines. With respect to a

given blade row, the terms identified through Adamczyk's derivation of the

average passage system of equations account for the following

nonaxisymmetric flows:

1. Spatial nonperiodicity of the flow in different blade passages of

a specific blade row due to differing rotor or stator blade

 3a

counts In successive stages (e.g., bladed stator row

followed by an + 1 bladed stator row of a succeeding stage).

2. "Deterministic" unsteadiness, due to the relative motion between

stator and rotor blade rows.

3. "Random" unsteadiness, due to any unsteadiness not correlated

with the fundamental blade passing frequency, for example vortex

shedding, global flow-field variations, and particle mixing.

I t Is Important to note that these terms, as a whole, are what the

empirical correlations used by current design systems have attempted to

approximate. Therefore, even I f the average passage system of equations

are never Incorporated Into a design system, they provide a useful format

for understanding the role of unsteady flows In generating nonaxisymmetric

flows In multistage turbomachlnes.

The magnitudes and distributions of some of these terms have been

determined from detailed laser anemometer measurements of the rotor-wake-

generated unsteady absolute velocity field throughout a single-stage

transonic axial-flow fan. Since the data were acquired from an Isolated

stage the terms due to spatial nonperlodlclty between blade passages are

not present, as they would be In a typical multistage environment.

Therefore, only the terms due to the deterministic and random unsteadiness

w i l l be presented herein. The relative Importance of these terms Is

discussed, and some general suggestions for possible approaches to

modeling these terms are provided. These data are for a loosely coupled

fan ( I . e . , rotor/stator spacing Is 85 percent of rotor chord), and

therefore the strength of the rotor wakes are significantly diminished

from what the stator would see I n a more realistically coupled stage. The
 3b

advantage of the wide spacing between blade rows i s that i t uncouples the

potential flow-field effects from the rotor wake effects. Also presented

are details of the kinematics of the passage of rotor wakes through the

downstream stator row, as well as other relevant features of the stator

flow field.
 4

II. RELATED RESEARCH

The number of research papers published during the last 30 years which

deal with the effects of the unsteady flows generated as a result of the

Interaction between rotating and stationary blade rows Is quite large. It

Is not unusual for these papers to be both experimental and theoretical or

numerical In nature, and often the effects of both the deterministic and

random (turbulent) unsteadiness are considered. As a result, the task of

organizing these papers Into specific categories Is d i f f i c u l t .

Nevertheless, an attempt Is made to organize the research papers dealing

with unsteady blade row Interactions Into two categories: those papers

which are primarily experimental In nature, or make their greatest

contribution from measurements and those papers which are primarily either

theoretical or numerical In nature ( I . e . , deal with modeling of the

unsteady flows). This section Is not Intended to provide a complete

compilation of a l l the unsteady flow research In turbomachlnery over the

last 30 years. I t 1s designed to provide the reader with a representative

sampling of past research related to the present research program.

A. Experimental Research

One of the earliest experimental research papers dealing with

unsteady blade-row Interactions was published In 1954 by Kofskey and Allen

[8] who observed, through smoke visualization of the flow 1n a low-speed

turbine, that thickening of the upstream stator suction surface boundary

layer occurred at certain positions of the rotor blading. In 1966, Smith

[9] described the process of wake chopping and attenuation, which

Indicated that a wake segment is modified by the chopping process, which

 5

disperses I t , and by uneven energy distribution. In 1970, Kerrebrock and

Mikolajczak [10] demonstrated that wake generated unsteadiness caused by

intra-stator transport of rotor wakes would have a significant effect on

the blade-to-blade stator exit temperature profile. Since the work of

Kerrebrock and Mikolajczak, there has been an increasing number of papers

published i n the open literature which deal with blade row Interaction

effects.

The effect of blade row Interactions on stage efficiency was

demonstrated i n 1974 by Yurinskiy and Shestachenko [2] who, using a

rotating bar wake generator ahead of a turbine cascade, found increased

(larger than steady flow values) total-pressure losses occurred for

unsteady flow than when testing in steady flow. Hodson [11,12] measured a

50 percent higher profile loss for a blade section i n a turbine rotor than

for that same section tested in a steady-state cascade. He proposed that

the larger loss was due to a higher growth rate of the rotor suction

surface boundary layer as a result of the interactions of the wakes of the

upstream nozzle row. Similar results were found by Tanaka [13]; he

measured a 1.1 to 1.25 times Increase 1n the total-pressure loss

coefficient of a lightly loaded cascade in a fluctuating flow over that in

a steady flow. Tanaka also observed that the periodic fluctuations tended

to suppress the growth of turbulence within the boundary layers, but that

this effect reached saturation above 8 to 10 percent turbulence intensity.

Okiishi et a l . [14] proposed that, due to blade-row interaction effects,

an appreciable portion of stator row loss can occur i n the blade-to-blade

free-stream region between the edges of the blade surface boundary

layers. Dong et al. [15] measured a 15 percent loss i n the free-stream

 6

total pressure through a stator row. Hansen and Okilshi [16] indicate

that the axial locations of the measurement stations used to assess rotor

row and stator row loss values are important i f a portion of the rotor

wake mixing-out loss occurs within the stator "row," as defined by stator

upstream and downstream measurement stations.

The effects of blade row interactions on blade surface boundary

layers were studied by Walker [17]. He observed that the stator suction

surface boundary layer of a low-speed single-stage axial-flow compressor

underwent periodically unsteady transition under the influence of the

upstream rotor wakes. Evans [18] confirmed the observations of Walker,

and in addition found a much larger rate of boundary-layer growth on a

stator blade disturbed by rotor wakes compared to a cascade blade i n

steady flow. Pfiel and Herbst [19]; Pfiel et a l . [20] used a rotating bar

wake generator to observe the effects of wakes on the boundary layer of a

f l a t plate. They found that the wakes impinging on a f l a t plate forced

the onset of transition to occur at shorter distances than for the

undisturbed plate, and that the wakes also affected the location of the

end of transition. Hansen and Okiishi [16] observed from surface hot-film

measurements of the stator blade boundary layer of a low-speed axial-flow

compressor that the rotor wake segments became wider, much as turbulent

spots do, as they traveled downstream.

The effects of blade-row interactions on blade loading have been

studied by Fleeter et a l . [21] in a large-scale low-speed single-stage

axial-flow research compressor. Their data correlated quite well with an

aerodynamic cascade traverse-gust analysis. They found that the unsteady

pressure differential magnitude decreases in the chordwise direction, and

 7

attains a value very near zero at the blade trailing edge. Their

observation Is significant In that 1t reflects the validity of the Kutta

condition for unsteady flows. Gallus et a l . [1] measured the fluctuating

pressures on the midspan profile surfaces of both rotor and stator blades

for several points of operation. They found that the downstream cascade

Induces stronger potential flow interactions i n the upstream cascade, and

that stronger throttling leads to Increased wake sizes and consequently

stronger fluctuation forces. The wake interactions were observed to be

the main source of fluctuating forces. Franke and Henderson [22]

investigated the influence of solidity, incidence flow angle, and

rotor-stator spacing on blade row interactions. They found a major

Influence of solidity at large values of incidence, where i t is suspected

that stall occurs. Some other investigation of the effects of blade row

interactions on blade loading are listed i n References 23 to 26.

Blade row interactions have also been demonstrated to be a source of

acoustic excitation in turbomachines. In 1977, Schmidt and Okiishi [27]

found that the level of discrete frequency noise at the inlet of a

multistage compressor could be varied appreciably by relative positioning

of the stationary blade rows. Gallus et a l . [4] found that the level of

discrete frequency noise could be reduced considerably by only slight

increases in the axial gap, within the range of small axial distances

between blade rows. They also found that the wake shape Influences the

intensity of the various harmonics.

The effects of blade-row interactions on energy transfer has been

investigated by Zierke and Okiishi [5] in a low-speed axial-flow research

compressor. They found that rotor wakes which have interacted with stator
 8

or Inlet guide vane wakes have lower total pressures than rotor wakes

which have not Interacted with stator or Inlet guide vane wakes. They

also found that periodic unsteadiness of total pressure Is considerable

downstream of a blade row that Involves energy addition and loss, and Is

minimal downstream of a blade row that Involves loss only. In 1985, Ng

and Epstein [28] confirmed, 1n a high-speed fan stage, the previous

findings of Kerrebrock and Mikolajczak [10], by showing that the transport

of rotor wakes through a downstream stator row resulted In a gapwise

temperature nonunlformlty at the stator exit of as much as 15 to 20 °C.

The effects of blade-row Interactions on heat transfer have been

Investigated by Dr1ng et a l . [29] In a large-scale low-speed axial-flow

turbine. They found that the unsteady wake passing affected the surface

Stanton number. Ooorly and Oldfleld [3] used a rotating bar wake

generator to Investigate the effects of wake passing on a downstream rotor

cascade. They observed that each wake produced a turbulent boundary-layer

patch In the blade suction surface boundary layer, which Is swept along

the blade surface and Is responsible for dramatic transient Increases In

heat transfer.

The general characteristics of wake-generated unsteadiness have been

reported by many Investigators. Joslyn et a l . [30] measured the

unsteadiness and three-dimensionality of the flow In a large-scale

axial-flow turbine. They measured large Increases In the pitch angle near

the suction side of the rotor wake as the vane wakes periodically Impacted

on the rotor suction surface. Matsuuchi and Adachi [31] measured the

three-dimensional unsteady flow Inside a rotor blade passage of an

axial-flow fan. They observed that the upstream stator wakes became
 9

deeper (and sometimes thicker) as they moved closer to the rotor blade

pressure surface, that the stator wake decay characteristics within the

rotor passage are very complex, and that the wake does not necessarily

decay monotonlcally. The character of the blade wakes which are the

primary source of blade-row Interactions have been reported by many

Investigators [32-41].

Wake transport characteristics have also been studied by Dunker [42]

who used a laser transit anemometer to map the unsteady flow field through

a high-speed single-stage axial-flow compressor stator row. Dunker used

contour maps of the turbulent kinetic energy to show the transport of

rotor wakes through the stator row. Binder et a l . [43] performed a

similar experiment In a high-speed axial-flow turbine stage In which they

observed an Increase 1n the turbulence In the stator wake as a result of

I t s being chopped by the rotor blades. Subsequently, Binder [44]; Binder

et a l . [45] suggested that this Increased turbulence In the chopped stator

wake was a result of stator secondary vortex cutting by the downstream

rotor blades, which caused a breakdown In the secondary vortex and a

subsequent increase In turbulence, Hodson [46] measured the wake

generated unsteadiness In the rotor passages of an axial-flow turbine. He

showed that the level of periodic unsteadiness on the suction surface was

2.5 times that on the pressure surface, whereas the mean level of

free-stream turbulence was only slightly higher. Hodson also observed

that the energy of the turbulent fluctuations was greater than the energy

of the periodic fluctuations on the pressure surface, and opposite on the

suction surface. Hodson concluded that this was because the wakes behaved
 10

as negative jets, with high loss fluid being convected towards the suction

surface.

In a multistage environment, the blade-row Interactions are even more

complex, and can result 1n unusual effects not observed 1n the single-stage

environment. Wagner et a l . [47] found that the periodic unsteadiness In

an Imbedded rotor or stator row was appreciable, depending on the extent

of the wake Interactions Involved at the particular location considered.

They were able to construct, from hot-wire data, sequential wake transport

Interaction plots throughout the compressor at various rotor/stator

relative positions, which Illustrated some of the possible wake

Interactions that can exist In multistage turbomachlnery. In 1985, Tweedt

et a l . [48] demonstrated that complex Interactions within multistage

turbomachlnery can result In unusual total-pressure distribution patterns

depending on flow rates. The unusual total-pressure distributions

resulted 1n a higher frequency Interaction In the downstream blade row

than would be expected from the number of blades present. Williams [49]

observed that the turbulence Introduced by rotor wakes persists In terms

of Identifiable velocity fluctuations for several stages before the rotor

contributions are mixed out Into unidentifiable random fluctuations. From

a Fourier analysis of his results, Williams identified discrete frequency

levels generated from upstream blade rows as much as four stages

downstream. He also observed that the strength of the periodic wave forms

was a strong function of location within the passage, with the periodic

high frequency wave forms being strongest i n the stator wakes.

 11

B. Modeling Efforts

One of the earliest papers dealing with the modeling of blade-row

Interactions was published In 1953 by Kemp and Sears [50,51] who studied

the aerodynamic Interference between rotor and stater blade rows for

Incompressible Invlscld flows by regarding each blade row as an Infinite

two-dimensional cascade. They obtained expressions for the unsteady

components of l i f t and moment of the blades of each row, and they also

calculated the effects of the stator wakes on the unsteady l i f t of rotor

blades. They found l i f t fluctuation amplitudes of about 10 percent of the

steady l i f t . The process of wake chopping by a th1n a i r f o i l was

theoretically examined In 1958 by Meyer [52] who considered the wake as a

negative jet carried with the free-stream. In 1965, Lefcort [53] extended

the analysis of Meyer to Include the case of a blade of finite thickness.

Lefcort considered that the unsteady forces In turbomachlnes were due to

four main effects: a circulation effect, a blade thickness effect, a wake

effect, and a wake distortion effect. The above mentioned blade-row

Interaction models were a l l developed using the concepts of circulation

theory presented by von Karman and Sears [54].

After these early developments, the focus of Interest 1n unsteady

a i r f o i l theory became compressibility effects, and, later, cascade

effects. The problem of a l i f t i n g a i r f o i l passing through a gust pattern

was examined by Horlock [55]. Using a heuristic approach, Horlock

partially accounted for the second-order effects of small mean-flow

Incidence on the fluctuating 11ft. A similar approach was used by Naumann

and Yeh [56] to account for small a i r f o i l camber. These treatments.

 12

however, were Incomplete 1n that they only accounted for the modified

boundary condition at the a i r f o i l surface while neglecting the coupling

between the unsteady flow and the potential flow around the a i r f o i l .

A complete theory that accounts for the dependence of the unsteady

flow on the mean potential flow of the a i r f o i l was developed by Goldstein

and Atassi [57]. They used their theory to analyze the interaction

between a periodic two-dimensional gust with an a i r f o i l i n uniform flow

which showed that the unsteady velocity field associated with the oncoming

gust was significantly distorted in both amplitude and phase by the steady

potential field about the a i r f o i l . Adamczyk [58] considered the effects

of f i n i t e blade thickness, camber, and incidence, and used his theory to

analyze the interaction between a one-dimensional gust and a cascade of

compressor blades. A combined analytical and numerical approach was used

to solve the field equations which govern the rotational and irrotational

velocity fields. The results of Adamczyk's analysis showed a strong

influence of mean loading on the unsteady force generated by the passage

of the one-dimensional gust through the cascade of compressor blades.

One of the earliest attempts to develop a numerical method for

solving the inviscid, compressible, two-dimensional, unsteady flow on a

blade-to-blade stream surface through an axial compressor stage was in

1977 by Erdos and Alzner [59]. A comparison of their numerical results

with experiment indicated that there was s t i l l much to be learned about

the nature of blade-row interactions before a numerical solution scheme

for a complete stage would be capable of predicting the actual flows. In

1982, Krammer [50] developed a time marching scheme for computing the

unsteady blade forces in turbomachines, which he based on potential flow

 13

theory and simulated viscous wakes. Blade surfaces were modeled by vortex

distributions and viscous wakes were simulated by contra-rotating vortex

rows. Krammer was able to generate sequential pictures of the wake

chopping and transport through the downstream blade row. However, his

analysis only considered the case of an equal number of stator and rotor

blades, and quantitative agreement with measurements was only f a i r .

Hodson [61] solved the two-dimensional Invlscid equations of motion using

a f i n i t e volume scheme based upon the work of Denton [62] In order to

calculate the unsteady flow generated by the Interaction of upstream wakes

with a moving blade row. The wakes were simulated by a velocity deficit

which was periodically moved across the Inlet boundary of the computational

domain at the relative speed of the wake generating blade row. Wake

dissipation was modeled by an a r t i f i c i a l viscosity, and contour plots of

the calculated entropy levels were used to Identify the wake regions.

Hodson's analysis confirmed the existence of such phenomena as wake

chopping, migration, and shearing. He concluded that many of the phenomena

associated with rotor-stator wake Interactions are dominated by Invlscid

rather than viscous effects.

In 1985, another attempt to numerically simulate the Interactions

between rotor and stator rows 1n a single stage turbomachlne with equal

numbers of rotor and stator blades was performed by Ral [63]. He compared

numerical results with experimental data and concluded that. In the case

of the time-averaged surface pressure distributions, there was good

agreement. The comparisons between pressure amplitudes and phase

relationships, however, were not very favorable. Joslyn et a l . [64]

described an Invlscid method for modeling turbomachlnery wake transport,

 14

which was based In part on the linear small-disturbance theory of Smith

[65]. Their calculations compared favorably with smoke visualization

experiments which showed the transport of chopped smoke-trace segments

through a downstream rotor row. Unfortunately, they were unable to

measure the actual wake fluid and Instead attempted to simulate a wake

using a smoke trace. Their experimental results must, therefore, be

questioned as to their validity In modeling real wake transport phenomena,

such as wake centrlfugatlon.

Although most of the above mentioned analyses provide a great deal of

Insight Into the characteristics of blade-row Interactions, they are

generally Impractical for use In a turbomachlnery design system. Usually

the analyses are restricted to a single stage turbomachlne with equal

numbers of rotor and stator blades, and even then they require significant

computer resources to solve the complete time-dependent flow field. For a

turbomachlne design engineer to use these models to predict the effects of

unsteady flows throughout a typical multistage turbomachlne would be a

formidable task.

Perhaps the earliest effort to Include the effects of blade-row

Interactions In a turbomachlnery design system was attempted In 1954 by

Smith [65] who considered the effects of secondary flows produced by

upstream blade rows. In 1981 Adklns and Smith [66] Included the effects

of spanwise mixing. Including wake centrlfugatlon, which might also be

considered an attempt at Including blade row Interactions. A more

fundamental approach to modeling blade row Interactions was presented by

Mitchell [67]. Mitchell developed a three-dimensional nonaxisymmetric

theory to analyze the Interaction effects due to wakes between two blade
 15

rows 1n an axial-flow turbomachlne. Mitchell's theory was based on

I1fting-line and nonaxisymmetric actuator disk principles, and assumed

that the wake was a small three-dimensional perturbation from the free

vortex condition. In 1979, Sehra [68] described a correlation method for

Introducing nonaxisymmetric blade row effects. Including deterministic and

turbulent unsteadiness, Into the axisymmetric flow field computation for a

highly loaded Isolated blade row. Sehra's approach centered about the

modeling of three Important nonaxisymmetric phenomena (apparent stress,

apparent entropy, and mean rothalpy) which control the time-mean momentum

and energy transfer. Adamczyk [7] took a more rigorous mathematical

approach to Including the effects of nonaxisymmetric flows than had been

previously attempted. Starting with the Navler-Stokes equation, Adamczyk

considered a decomposition of the flow field Into a steady-state blade-to-

blade periodic component, a blade-to-blade aperiodic component, a periodic

unsteady component, and a random unsteady component. Then, In a manner

similar to developing the Reynolds averaged Navler-Stokes equation,

Adamczyk performed a series of averages on the Navler-Stokes equations to

resolve, what he called, the average passage equation system. Adamczyk's

average passage equation system was developed to simulate the flows In

multistage turbomachlnery. His average passage equation system reduces to

the equation system developed by Sehra [68], and to the axisymmetric

equations of motion, given the proper assumptions used to develop each. A

model for closing the Invlscid form of Adamczyk's averaged passage

equation system was published by Adamczyk et a l , [69], and the average

passage equation system has been used to solve the Invlscid flow through a

counter-rotating propeller, see Celestlna et al. [70].

 16

III. RESEARCH FACILITY

The single-stage axial-flow compressor test f a c i l i t y of the National

Aeronautics and Space Administration (NASA) Lewis Research Center was used

to accomplish the research described In this dissertation. The test

f a c i l i t y , research compressor, and laser anemometer system are described

In this section.

A. Compressor Test Facility

A schematic diagram of the single-stage ax1al-flow compressor test

f a c i l i t y at NASA Lewis Is shown In Figure 3.1. The drive system for the

compressor consists of a 3000 hp electric motor with a variable-frequency

power supply. Motor speed Is controllable from 400 to 3600 rpm. The

motor Is coupled to a 5.52 gear ratio speed Increaser gear box that In

turn drives the rotor. The f a c i l i t y was sized for a maximum flow rate of

100 lb/sec with atmospheric a1r as the working fluid. Air Is drawn Into .

the f a c i l i t y from an Inlet located on the roof of the building. The air

next passes through a flow measuring station consisting of a thin-plate

orifice, through Inlet butterfly valves, and Into a plenum chamber. The

air Is Is then accelerated to the compressor test section via a nozzle,

through the test compressor, and Into a collector through a sleeve throttle

valve. The air subsequently passes through a spray water cooler and Is

exhausted back Into the atmosphere or through an altitude exhaust system.

The airflow may be controlled through either the upstream butterfly valves

or the downstream collector valve. For the present Investigation the air
ATMOSPHERE ORIFICE
INLET

p? INLET THROTTLE VALVES^

COLLECTOR/THROTTLE
VALVE-V

STRAIGHTENER SCREEN7
rTEST STAGE ,
GEAR B0X-\ \
\ \ !
DRIVE MOTOR-7\ \ ' PLENUM-» /
I i /

.ALTITUDE
EXHAUST

ATMOSPHERE
EXHAUST

CD-1091FR-U

Figure 3.1 Schematic diagram of NASA Lewis single-stage

compressor test f a c i l i t y
 18

was exhausted back Into the atmosphere and the collector valve was used to

control the airflow.

B. Research Compressor

The f i r s t stage of a NASA Lewis designed two-stage axial-flow fan was

used as the research compressor to accomplish the research described In

this dissertation. A complete description of the aerodynamic design of the

f u l l two-stage fan Is provided In References 71 and 72. Therefore, only a

brief description of the relevant design parameters of the f i r s t stage

geometry, used as the research compressor In the present Investigation,

w i l l be provided herein. A sketch of the flow path of the research

compressor Is presented I n Figure 3.2, and the coordinates of the flow path

are provided i n Appendix A.

The research compressor was designed with wide axial spacing between

blade rows (85 percent rotor axial chord at midspan) i n order to reduce

blade-row-interaction noise. The first-stage rotor was designed as a low-

aspect-ratio (1.56) damperless fan with a design pressure ratio of 1.63 at

a mass flow of 33.25 kg/sec. The design rotational speed Is 16 043 rpm,

which yields a design t i p speed of 429 m/sec. The inlet relative Mach

number at the rotor t i p i s 1.38. The first-stage rotor has 22 blades of

multiple-circular-arc design. The rotor solidity varies from 3.114 at the

hub to 1.290 at the t i p . The rotor inlet and exit t i p diameters are. 51.1

and 49.5 cm respectively, and the rotor inlet and exit hub/tip radius

ratios are 0.375 and 0.478 respectively. The running rotor t i p clearance

i s 0.5 mm. An example of the first-stage rotor blade sections at three

spanwise locations i s shown i n Figure 3.3.

 19

2.39 13.21

-20 -10 0 10 20 30
AXIAL DISTANCE, 1. CM

Figure 3.2 Meridional-plane view of research compressor flow

path
 20

^MID SPAN

BLADE
MOTION

2.5 0 2.5
AXIAL DISTANCE FROM ROTOR STACKING AXIS, CM

Figure 3.3 Representative rotor blade sections at three

spanwise locations
 21

The first-stage stator was designed to discharge fluid axlally. The

stator consists of 34 double-circular-arc designed blades which vary In

solidity from 2.485 at the hub to 1.271 at the t i p . The stator t i p

diameter 1s constant at 48.7 cm and the Inlet and exit hub/tip radius

ratios are 0.500 and 0.530 respectively. An example of the stator blade

sections at three spanwise locations Is shown In Figure 3.4.

A summary of the relevant design parameters of the single-stage

research compressor Is provided 1n Table 3.1, and a complete listing of

the Input parameters to the aerodynamic design code [73] which was used to

design the first-stage geometry 1s provided 1n Appendix A. Also Included

In Appendix A are the output tables from the design code which Include

velocity diagram Information on the axisymmetric design streamlines at

each axial computational station and complete blade manufacturing

coordinates at each of the design blade section radii for both the

first-stage rotor and stator.

C. Laser Anemometer System

A laser fringe-type anemometer (LFA) system developed at NASA Lewis

for use In high-speed turbomachlnery was used as the principal measuring

Instrument to accomplish the research described In this dissertation. A

detailed description of the basic anemometer system, as originally

designed. Is reported In References 74 and 75. Because of minor changes

In some of the optical components and system layout which have occurred

since the system was originally designed, a brief description of the

current laser anemometer system is provided herein. A block diagram of

the major components of the LFA system 1s provided In Figure 3.5.

 22

MID SPAN
(R - 18.606 CM)—^
HUB (R = 12.891 CM)

/-TIP

TIP (R = 21.384 CM)

FLOW ^

5.0 -2.5 0 2.5 5.0

AXIAL DISTANCE FROM STATOR STACKING AXIS, CM

Figure 3.4 Representative stator blade sections at three

spanwise locations
 23

Table 3.1. Summary of single-stage research compressor design

parameters

Rotor speed 16042.8 rpm

Corrected mass flow rate 33.25 kg/s

Stage pressure ratio 1.59

Inlet pressure and standard day conditions

temperature

Number of blades

Rotor 22

Stator 34

Blade section profile

Rotor multiple circular arc

Stator double circular arc

Blade stacking axis radial line through center of gravity of

location blade sections

Rotor Stator
hub tip hub tip

Blade aerodynamic chord 9.264 cm 9.522 cm 5.728 cm 5.768 cm

Blade element solidity 3.144 1.290 2.485 1.271

Maximum thickness to 0.085 0.029 0.080 0.060

aerodynamic chord ratio

Leading and trailing 0.005 0.002 0.005 0.013

edge radius to aero­
dynamic chord ratio

Inlet exit Inlet exit

Hub radius 9.583 cm 11.829 cm 12.189 cm 12.931 cm

Hub/t1p radius ratio 0.375 0.478 0.500 0.530
 24

/-CIRCUMFERENTIAL
/ PATH
ROTOR
ROTATION OPTICS
OPTICS MOTOR
DRIVES
17^

JACKING
MOTOR
PHOTO-
MULTIPLIER
SIGNAL
^^ONCE-PER-REVOLUTION PULSE MICROCOMPUTER

ELECTRONIC LASER
SHAFT ANGLE ANEMOMETER
ENCODER PROCESSOR
(TRANSIT TIME)

A
DISK f
MINICOMPUTER
STORAGE
N
1f
CENTRAL PRINTER
COMPUTING CRT
FACILITY TERMINAL

Figure 3.5 Block diagram of laser anemometer data

acquisition system
 25

1- Optical configuration

The basic optical components of the laser anemometer system are shown

i n Figure 3.6. I t is a single-channel, dual-beam, fringe type anemometer

with on-axis backscatter collection optics. The 514.5 nm (green) line of

an argon-ion laser operating at a power level of 1.5 W i s used to generate

the laser beam for the LFA system. After the beam leaves the laser a

series of mirrors turns the laser beam through 180° for compactness. The

beam then passes through a collimator which i s used to position the beam

waist at the probe volume, and to adjust the diameter of the beam waist to

the proper diameter. Because the LFA processing electronics requires at

least 10 fringes, and the fringe spacing was 10.4 ym, a waist diameter of

125 ym was selected. The beam then passes through a beam splitter which

splits the beam Into two parallel beams of equal intensity. Two more

mirrors then direct the beam through a 200-mm-focal-length focusing lens

(50 mm diameter) which causes the beams to converge and ultimately cross

at the lens focal point. At the intersection where the two beams cross

(called the probe volume) a pattern of bright and dark fringes are formed

due to the constructive and destructive interference of the wave fronts of

the two beams. The fringe pattern is perpendicular to the path of the

bisector of the two beams which is perpendicular to the research

compressor centerline. The probe volume length i n the direction of the

bisector of the two beams 1s 2 mm.

Part of the light which i s scattered from seed particles which cross

the bright fringes of the probe volume is then collected through the

200-mm-focal-length focusing lens and directed to a lOO-pm-diameter

pinhole in front of the photomultiplier tube (PMT). The focusing lens in

 26

^ MIRROR

COLLIMATOR
PHOTOMULTIPLIER MIRROR
TUBE
ROTATABLE
BEAM SPLITTER

100-PM PIN HOLE-'

ARGON-ION
160-MM FOCUSING LENS LASER
ORANGE-PASS FILTER-

MOTORIZED
GONIOMETER, VERTICAL
MOUNT 1\ MIRROR

GONIOMETER
MIRROR MIRROR FOR
COLLECTED LIGHT
200-MM FOCUSING AND
COLLECTING LENS
PROBE VOLUME

FRINGES

LASER BEAMS

Figure 3.6 Layout of laser anemometer system optical

components
 27

front of the photomu1t1pl1er tube 1s 50 mm 1n diameter, with a focal

length of 160 mm. The function of the orange-pass f i l t e r In front of the

160-mm-focal-length focusing lens 1s explained In the next section.

The final optical element In the laser beam paths prior to entering

the research compressor flow path, not shown In Figure 3.6, Is a

3.2-mm-th1ck glass window In the compressor shroud. Two different laser

windows were used; a "rotor window," see Figure 3.7, which extended from

about 1.5 rotor chords upstream of the rotor leading edge to just past the

stator leading edge, and a "stator window," see Figure 3.8, which extended

from just past the rotor trailing edge to about one-half stator chord

downstream of the stator trailing edge. The rotor window covered a

constant circumferential width of 20°. The stator window was cut to follow

the stator blade suction and pressure surface contours so that an entire

stator passage could be viewed. Both windows were contoured to follow the

shroud Inner flow path. The windows were made of chemically treated window

glass and were pressure tested to 300 psig prior to their Installation In

the r1g.

2. Flow seeding

The seed material used was Rhodamlne 6g In solution with ethylene

glycol and benzyl alcohol [76]. This seed material has the characteristic

of fluorescing orange light when Illuminated by the green light of the

argon-1on laser. The function, therefore, of the orange-pass f i l t e r In

front of the 160-mm-focal-length focusing lens Is to allow only the

flourescent light emitted by the seed particles to reach the PMT. This

reduces detection of unwanted scattered light and therefore allows

measurements closer to solid surfaces.

 28

ROTOR STATOR

(j)

QC 10

-10 0 10 20
AXIAL DISTANCE, 1, CM

Figure 3.7 View of optical access through rotor window

 29

ROTOR STATOR

i é

u 20 — FLOW

oc 10

-20 -10 0 10 20
AXIAL DISTANCE, 1» CM

Figure 3.8 View of optical access through stator window

 30

A commercially available atomizer was used to atomize the seed

solution and generate liquid seed particles, nominally 1.0 to 1.4 ym In

diameter [77], which were Introduced Into the research compressor flow

field through a 6 mm tube located 35 cm upstream of the rotor. The seeder

tube was mounted In a probe actuator which was mounted on top of a

circumferential positioning pad. In this configuration I t was possible to

remotely position the seeder probe radially and circumferentially within

the research compressor flow field to achieve optimum seeding of the

streamtube which passed through the laser probe volume.

3. Laser optics cart and traversing mechanism

A sketch of the optics cart assembly Is shown I n Figure 3.9. All of

the optics were located on a f l a t rigid metal table and secured In place

using commercially available magnetic Instrument bases. The beam splitter

was mounted In a motorized rotator assembly which allowed the angle of

orientation of the fringes to be rotated about the optical path of the

Input beams, thus enabling different components of the flow velocity to be

measured. The focusing lens and final turning mirror of the transmitting

optics are mounted on a motorized goniometer cradle so that they move

together as a unit. The motorized goniometer allows the direction of the

input beams to be directed off-radial i n order to be able to access areas

which would normally block the beam paths i f they were transmitted

radially into the compressor flow path.

A metal framework between the optics table and a large horizontal x-y

traversing table formed a channel in which the laser sat. The laser was

secured to the bottom of the channel and the entire assembly was mounted
 OPTICS
TABLE

LASER

RESEARCH
COMPRESSOR LASER

\ HABLES - ooo' SUPPLY

ASSEMBLY ^TRAVERSING

Figure 3.9 Schematic diagram of optics cart assembly

I
!
 32

on top of a wheeled metal cart. A motorized jacking mechanism at the

front of the optics cart, In conjunction with the x-y traversing tables

and the motorized goniometer, enabled positioning of the probe volume at a

particular axial, radial, and circumferential position within the

compressor flow field, to within ±0.05 mm axlally, ±0.05 mm radially, and

±0.03° circumferentlally. The angular orientation of the laser beams in

the R-0 plane could be controlled to within ±0.01* using the motorized

goniometer, and the beam splitter enabled control of the angular

orientation of the fringes i n the Z-0 plane to within ±0.03° (refer to

Fig. 3.6).

The optics cart was originally designed for surveying a rotor-only

flow field which usually does not require circumferential traversing of

the probe volume, since the rotor sweeps by the probe volume. However,

the research described in this dissertation required surveying a stator

flow field making circumferential traversing of the flow field necessary.

The jacking screw mechanism, therefore, was adopted as a temporary

modification to the optics cart to provide the capability for

circumferentlally traversing the probe volume through the stator flow

field. A remotely controlled high-speed stepping motor was used to drive

the jacking screw. Positioning of the jacking screw, however, was

controlled by the operator and typically required about 30 seconds to

position.

High-speed stepping motors were also used to drive the beam splitter,

goniometer, and x-y traversing table. An optical encoder attached to each

motor shaft provided position feedback to each motor drive controller.

 33

The motor drives were of the accelerate-decelerate type (with regard to

the output pulse train) and typically positioned the optics i n about

1 second. Each motor had a "home" position to which i t could be moved for

remote position calibration. An 8-bit-word microcomputer served as the

Interface between the motor drives and the data acquisition control

minicomputer.

4. Laser anemometer signal processor

A commercially available counter-type laser anemometer signal

processor with a 500-MHz clock was used to process the signal bursts from

the photomultiplier tube. The processor measures the transit time of a

seed particle across eight fringes. To Insure measurement validity and to

help discriminate against noise, the processor performs several tests on

each seed particle passing through the probe volume. One test i s a 5/8

comparison test where the transit times for crossing five and eight

fringes is compared. The ratio of these times must be sufficiently close

to the value of 5/8 to be accepted. This test i s used to reject

measurements from seed particles which are accelerating as they pass

through the laser probe volume. Another test performed by the processor

Is a sequence check such that during each signal burst the high-pass

filtered signal from the photomultiplier tube must alternatively rise

above a threshold level and f a l l below the zero level at each fringe in

order to be accepted. This test is used to reject extraneous signal noise

and to reject signal bursts occurring from multiple particles or very

large particles which pass through the laser probe volume. The processor

also provides an additional test which attempts to discriminate signal

 34

bursts occurring from multiple particles or very large particles within

the probe volume by requiring that the high-pass filtered signal from the

photomultlpller tube does not rise above an overload level,

5. Compressor rotor shaft-angle encoder

The LFA system Is free to acquire velocity measurements whenever a

seed particle crosses the LFA probe volume, which results In the random

acquisition of many velocity measurements during every rotor revolution.

Therefore, a commercially available compressor rotor shaft-angle encoder

was used to simplify the task of tagging each velocity measurement with

the proper angular position of the rotor. The shaft-angle encoder

provides a continuous measure of the rotor shaft position relative to a

once-per-rev signal obtained from the rotor disk. When a velocity

measurement occurs, the seed particle fringe-crossing frequency and

corresponding rotor shaft position are simultaneously transmitted to the

LFA system minicomputer which records them as a data pair.

The encoded angular position of the rotor Is produced by a counter

that 1s clocked by a frequency synthesizer. The frequency Is adjusted

near the beginning of each rotor revolution so that the number of counts

for each revolution (selected by the operator) remains approximately

constant. An optical sensor generates the once-per-rev pulse as a result

of a passing target on the face of the rotor disk near the hub flow path.

For the results cited In this dissertation the encoder counts per

revolution (selected as 4400) was divided by four, which for the 22-bladed

research rotor yielded 50 angular positions (shaft positions) per blade

passage.
 35

I t 1s Important to note that the velocity measurements do not really

occur at a discrete rotor shaft position, but rather occur anywhere within

an Interval between two adjacent rotor shaft positions. With 50 rotor

shaft positions per blade passage, the Interval length between shaft

positions varies between 0.7 mm at the hub and 1.4 mm at the t i p of the

rotor blade. Therefore, In this dissertation the term shaft position Is

used with the understanding that measurements attributed to a rotor shaft

position actually occur 1n an Interval about that position. In order to

prevent velocity data from being assigned to the wrong rotor shaft

position, the data acquisition program discarded data acquired during any

rotor revolution In which the encoder count for the succeeding revolution

varied by more than ±20 counts from the desired count of 4400. Thus, the

actual average error In the rotor shaft position

was less that ±0.0045 of a rotor shaft position.

6. System computer

A l6-b1t-word minicomputer with 32 k (k = 1024) words of core memory

and hardware floating-point mult1ply-d1v1de capability was used to provide

automatic control of the LFA data acquisition process. The minicomputer

used dual removable cartridge-type magnetic disk storage, with each disk

having a capacity of 1.25 million 16-b1t words. Two cathode-ray-tube type

display terminals were used for monitoring the LFA data acquisition

process. The minicomputer automatically controlled positioning of the

laser probe volume and was Interfaced to the laser anemometer signal

processor and compressor rotor shaft angle encoder from which I t acquired

the fringe crossing frequency and rotor angular position, respectively.

 36

IV. EXPERIMENTAL PROCEDURE AND DATA REDUCTION

The objective of the experimental procedure was to obtain spatially

and "temporally" ( I . e . , rotor shaft position, not real time) detailed

measurements of the rotor-wake-generated unsteady absolute-velocity field

at selected sites within the research compressor stator row. With the

compressor operating at design speed and maximum efficiency, a laser

fringe-type anemometer (LFA) system was used to acquire ensemble-averaged

measurements of the absolute velocities along two design axisymmetric

stream surfaces through the stator blade row. A design axisymmetric

stream surface refers to a surface of revolution on which LFA measurements

were acquired, and does not represent an actual stream surface along which

fluid particles travel. For each design axisymmetric stream surface, the

absolute velocity measurements were acquired, spatially, from

blade-to-blade across the stator pitch and from upstream through

downstream of the stator blade row, and temporally, for a number of rotor

shaft positions evenly distributed across several rotor pitches.

The 100 percent speed line operating characteristic of the research

compressor Is shown In Figure 4.1, which Identifies the maximum efficiency

point at which LFA survey data were acquired. From observations, the

rotor speed was maintainable to within ±0.3 percent of design speed, and

the corrected mass flow rate was maintainable to within ±0.14 kg/sec. The

corrected mass flow rate was calculated from a calibrated orifice plate

located 1n the Inlet ducting upstream of the plenum chamber. The overall

pressure ratio was determined from a 5-element rake downstream of the

stator and from the upstream plenum pressure.

 37

1.7 MAXIMUM
EFFICIENCY
OPERATING
POINT
1.6

1.5

1.4
30 32 34 36
STANDARD DAY CORRECTED MASS FLOW,
KG/SEC

Figure 4.1 Research compressor stage 100 percent speed line

operating characteristic
 38

Extensive software written for the LFA-system minicomputer were used

to support the data acquisition procedures. This software Included

programs to generate data files describing the research compressor flow

path and the complete three-dimensional stator and rotor blade geometry,

programs to allow the experimenter to prescribe complete survey

Information along selected design axisymmetric stream surfaces prior to

beginning data acquisition, programs to allow remote positioning of the

optics by the experimenter, programs to provide complete LFA-system

computer control of the data acquisition procedures, and many other

programs, some of which are briefly described In the succeeding sections.

Although extensive data reduction software was also available on the

LFA-system minicomputer. Its processing times were quite long and Its

memory was considerably limited as were printing and plotting

capabilities. Therefore, the NASA Lewis central computing facilities were

used for a l l reduction, tabulation, and plotting of the data presented

herein. The data disks generated from the LFA-system minicomputer were

copied to data tapes which could be read by the mainframe computer used to

reduce the data.

A. Experimental Procedure

The experimental procedure consisted of setting up a survey f i l e to

prescribe the coordinates of points to be surveyed, following extensive

setup procedures to Insure accurate positioning of the probe volume, and

the actual data acquisition procedures used to acquire the data. The

survey setup, preliminary setup, and data acquisition procedures are

described below.
 39

1. Survey setup

A computer program was used to set up the coordinates of the survey

points at which laser anemometer data would be acquired prior to actually

beginning data acquisition. The "survey setup program" generates a disk

f i l e consisting of the coordinates of the LFA survey points, specified

along a selected design axisymmetric stream surface, and their required

beam orientations. This survey f i l e 1s then read by the data acquisition

program which automatically sequences through the specified survey points

and results In collection of the desired data.

Figure 4.2 Is a meridional view of the research compressor flow path

which shows the 10 and 50 percent stream surfaces along which the LA data

were acquired. The dots are positioned at approximate axial locations of

measurements. The complete LFA survey grids for the 10 and 50 percent

stream surface surveys are shown In Figure 4.3 and the relative

coordinates of the survey points are given 1n Table 4.1 and Table 4.2,

respectively. The solid circles (dots) In Figure 4.3 Indicate survey

points where data were actually acquired, whereas the open circles

Indicate where data acquisition was attempted but data could not be

acquired, or was of poor quality. Problems with window blockage,

Insufficient entralnment of seed particles across flow boundaries, seed

particle accumulation on solid surfaces, separated flow regions, and very

high flow unsteadiness contributed to an Inability to acquire data at some

of the survey points.

Each stator flow-field survey was divided into two separate surveys;

an interblade-row survey, and an intrablade-row survey. The interblade-

row survey consisted of nominally eight axial survey stations, six in

 30 |—
STREAM-
SURFACE,
PERCENT
7 10
LU
U
z FLOW / \ 50
<
H-
w / ROTOR \ STATOR

<
s 10 o
<

-%0 -10 0 10 20 30
AXIAL DISTANCE, CM

Figure 4.2 Mer1d1onal-plane view of research compressor flow

path showing the 10 and 50 percent stream surfaces
along which the laser anemometer data were
acquired
 41

INTERBLADE-ROM SURVEY INTRABLAOE-ROW SURVEY

7.5
lOOr

50 • 5.0

2.5-
a • ACTUAL SURVEY
• ATTEMPTED SURVEY

i TROTOR TRAILING EDGE

I -50^
A J
5.0
,
7.5 10.0 12.5 15.0 17.5 20.0
CM

-100 -50 0 50 100 150

AXIAL DISTANCE, PERCENT STATOR CHORD
(A) 10% STREAM SURFACE.

INTERBLAOE-ROH SURVEY INTRABLADE-ROW SURVEY

5 150 r CD E F H JK 0R
7.5r I I I I III I I I I 1 I
GI

100 •

5.0
50 •

2.5

^ -50 L
rRoroR TRAILING EDGE

J I
5.0 7.5 10.0 12.5 15.0 17.5 20.0
CM

-L I I
-100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD
(B) 50X STREAM SURFACE.

Figure 4.3 Blade-to-blade view of the 10 and 50 percent stream

surfaces showing the actual laser anemometer
measurement survey grids
 42

Table 4.1. Relative coordinates of laser anemometer survey locations

along the 10 percent stream surface (stator chord = 5.6 cm)

AXIAL STATOR INTERBLAOE- CIRCUMFERENTIAL

SURVEY CHORD ROM GAP COORDINATE OF
STATION PERCENT PERCENT SURVEY POINT

A -104 0 5 to 95 percent
B -82 20 of stator lead­
Interblade-row C -62 40 ing edge gap In
survey D -41 60 10 percent gap
E -28 Increments
F -20 80
G -10
H 0 100
I 10
J 20
K 35 5 to 95 percent
Intrablade-row L 50 of local stator
survey M 65 gap In 10 percent
N 85 gap Increments
0 100
P no
Q 130
R 154
 43

Table 4.2. Relative coordinates of laser anemometer survey locations

along the 50 percent stream surface (stator chord = 5.6 cm)

AXIAL STATOR INTERBLADE- CIRCUMFERENTIAL

SURVEY CHORD ROW GAP COORDINATE OF
STATION PERCENT PERCENT SURVEY POINT

A -70 20 5 to 95 percent
B -52 40 of stator lead­
Interblade-row C -34 40 ing edge gap In
survey D -30 10 percent gap
E -18 80 Increments
F -10
G -8
H -5
I 0 100
J 5
K 10 5 to 95 percent
Intrablade-row L 20 of local stator
survey M 35 gap In 10 percent
N 50 gap Increments
0 65
P 85
Q 100
R 110
S 130
T 150
 44

the Interblade-row gap between the rotor trailing edge and stator leading

edge, and two at 0 and 10 percent of stator axial chord. The Interblade-

row gap Is defined as the axial distance between the rotor trailing edge

and stator leading edge at the radii corresponding to the design

ax1symmetric stream surface. At each of these axial survey stations, ten

equally spaced circumferential survey points were specified across a

stator circumferential gap. The stator circumferential gap varies with

axial position and Is defined as the circumferential distance across the

stator flow field between the pressure and suction surfaces. Ahead of the

stator leading edge, the circumferential survey points were specified from

5 to 95 percent of the stator leading edge circumferential gap In 10

percent gap Increments. Therefore, upstream of the stator leading edge,

the ten circumferential survey points were at the same absolute

circumferential locations at each axial survey station. At the stator

leading edge, and at a l l survey stations downstream of the stator leading

edge, the circumferential survey points were specified from 5 to

95 percent of the local stator circumferential gap In 10 percent gap

Increments.

The Intrablade-row survey consisted of nominally 15 axial survey

stations between -10 to 154 percent of stator axial chord. At each of

these axial stations there were ten circumferential survey points

specified, equally spaced from 5 to 95 percent of local stator

circumferential gap 1n 10 percent gap Increments. Outside of the stator

blade row the percent gap was referenced to the extension of the stator

blade mean camber line.

 45

In order to verify matching between the data acquired for each

survey, the 0 and 10 percent stator axial chord stations were surveyed

twice, once for the 1nterblade-row survey and again for the Intrablade-row

survey. The Interblade-row surveys at 0 and 10 percent stator axial

chord, however, could not be completed near the stator suction surface as

the rotor window frame substantially blocked the collection of scattered

light from seed particles. Therefore, only the Interblade-row results

upstream of -10 percent stator axial chord, 1n conjunction with the

1ntrablade-row survey results, were used In constructing a l l data figures

presented herein (see Fig. 4.3 and Tables 4.1 and 4.2).

2. Preliminary LFA setup

The LFA setup procedures, performed each day prior to beginning data

acquisition, consisted of orienting the LFA system relative to the

research compressor coordinate system, and orienting the LFA probe volume

and rotor blades relative to the surveyed stator blade passage. The LFA

system Itself, with optics optimally aligned, was used In the following

manner to accomplish the above mentioned LFA setup procedures. With the

orange-pass f i l t e r removed from the optics path, thereby sensitizing the

LA system to light scattered from metallic surfaces, the LFA probe volume

was moved relative to known "sharp-edged" reference points until the

output current from the PMT was maximized. At this point, from visual

observation, the LFA probe volume was aligned to the known reference point.

To orient the LFA system to the research compressor coordinate

system, the LFA system was positioned. In the above manner, such that the

LFA probe volume was located at the upstream edge of the stator foot-ring,
 46

see Figure 3.2, the position of which was known relative to the research

compressor coordinate system. The LFA system motor drive offsets were

then updated, according to the difference In the measured and expected

positions of the stator foot-ring reference point, such that the motor

drive positioning program would reflect the probe volume position In the

research compressor coordinate system.

To orient the LFA probe volume to the surveyed stator blade passage,

the LFA system was positioned such that the probe volume was located at

the leading and trailing edges of the lower surveyed stator blade at the

radii which corresponded to the design stream surface along which the

stator flow field was surveyed (refer to Figures 3.7 and 3.8). The laser

system was then used, as described above, to determine when the LFA probe

volume was positioned at the lower surveyed stator blade edges. A visual

Inspection was made for verification. By knowing the measured locations

of the lower surveyed stator blade edges, at the radii corresponding to

the design axisymmetric stream surface along which the stator flow field

was surveyed, the position of the probe volume with respect- to the

surveyed stator blade passage could be determined with good precision.

The final LFA setup procedure was to orient the rotor blades with

respect to the surveyed stator blade passage. This was accomplished with

the research compressor operating at design speed and at the desired

survey flow conditions. Prior to determining the orientation of the rotor

blades relative to the surveyed stator blade passage, the above determined

stationary reference coordinates were remeasured, based on their measured

nonrotating positions, to Insure that they did not change after the

research compressor was allowed to stabilize at the survey operating

^...
 47

conditions. The LFA system was again used as a measuring tool, this time

to determine the position of the rotor blades relative to the surveyed

stator blade passage. The LFA probe volume was positioned at the radius

which corresponded to the design stream surface along which the stator

flow field was surveyed, and at the circumferential position which

corresponded to the lower stator blade leading edge, as measured above.

The probe volume was then moved axlally away from the stator leading edge,

towards the trailing edge plane of the rotor blade row, until the PMT

output current was maximized, which In this case corresponded to the probe

volume being positioned at the trailing edge plane of the rotor blade

row. With the orange-pass f i l t e r removed from the optics path and the

laser power and PMT high-voltage power supply properly adjusted, the data

acquisition program was run to acquire and save on disk approximately

15 000 measurements of the rotor blade trailing edge as I t passed through

the laser probe volume. As each measurement was acquired I t was

simultaneously tagged with the corresponding angular position of the rotor

wheel (I.e., rotor shaft position) at which the measurement occurred. For

this experiment there were 850 equally spaced rotor shaft positions

distributed across 17 rotor blade passages, which resulted In 50 rotor

shaft positions per rotor blade passage. Therefore, assuming that the LFA

probe volume was optimally aligned to the rotor blade row trailing edge

plane, there would be a clustering of measurements about every 50 shaft

positions which correspond to the locations of the rotor blade edges. For

a l l other shaft positions between rotor blade edges, the number of

measurements would be zero. The shaft positions corresponding to the

rotor blade edges could, therefore, be Identified as the shaft positions

 48

which have the greatest number of measurements and which are spaced apart

about every 50 shaft positions. These measured rotor trailing edge shaft

positions were found to be consistently repeatable. By referencing these

measured rotor trailing edge shaft positions to the blade passage in which

they occurred ( i . e . , shaft positions from 1 to 50) an average rotor

trailing edge shaft position was determined. This average rotor blade

shaft position was used in subsequent surveys of the stator flow field to

trigger the beginning of data acquisition, during every revolution of the

rotor, to occur at the circumferential coincidence of the rotor trailing

edge with the surveyed stator leading edge, and to offset each shaft

position, as i t was saved on disk, such that the coincident shaft position

was identified as shaft position one, subsequent shaft positions were

incremented from there. Therefore, for a l l subsequent experiments and

displayed or saved results, shaft position one corresponds to

circumferential coincidence of the rotor trailing edge with the surveyed

stator leading edge.

Finally, the stator survey f i l e was updated to reflect the measured

locations of the surveyed stator blade edges and to reflect the current

measured rotor trailing edge shaft position which would be used to trigger

the beginning of data acquisition during each rotor revolution. This

Insured that the coordinates of all survey points were accurately

positioned relative to the surveyed stator passage and that, for a l l

survey points, the rotor blade trailing edges were accurately in sync with

the lower surveyed stator blade leading edge.

 49

3. Data acquisition

The LFA system minicomputer was used to control the data acquisition

procedures via an automated data acquisition program [75], which Is set up

to sequentially acquire data at each of the survey points specified In the

survey f i l e . At each of these survey points, the LFA system was free to

acquire velocity measurements whenever a seed particle crossed the LFA

probe volume, which resulted In the random acquisition of many velocity

measurements during every rotor revolution. The seed particle

fringe-crossing frequency, which Is proportional to the particle velocity,

was measured by the laser anemometer signal processor. A continuous

measure of the rotor shaft position, relative to a once-per-rev signal

obtained from the rotor disk, was provided by the compressor rotor

shaft-angle encoder. When a velocity measurement occurred, the seed

particle fringe-crossing frequency and corresponding rotor shaft position

were simultaneously transmitted to the LFA system minicomputer which

recorded them as a data pair. The actual recorded rotor shaft positions

were referenced to the lower surveyed stator blade leading edge, as

explained In the previous section.

At each survey point, nominally 30 000 measurements (I.e., data

pairs) were acquired at each of two fringe angles, oriented at ±20° of the

t1me-averaged absolute flow angle In order to reduce flow angle biasing.

The time-averaged absolute flow angle was previously determined from a

series of short LFA experiments and was found to vary blade-to-blade by

less than about ±3°. Therefore, the circumferentially averaged values of

the time-averaged absolute flow angle distributions were used to determine

the fringe angle orientation at each axial survey station. The data at
 50

each survey point were recorded at one of 850 different rotor shaft

positions, which were evenly distributed as 50 positions per blade passage

across 17 of the 22 rotor blade passages.

After a l l the requested data for a particular survey point had been

acquired, the following information was stored on the LFA system disk

storage medium: the number of measurements, the sum of the seed particle

fringe-crossing frequencies, and the sum of the frequencies squared, at

each of the 850 rotor shaft positions. In addition, "background

Information" was also recorded on disk which Included; the coordinates of

the survey point position, the number of rotor shaft positions and rotor

blades surveyed, the total number of measurements requested, the plenum

temperature, and the research compressor operating conditions ( i . e . ,

pressure ratio, weight flow, and rotational speed). The LFA data were

eventually transferred from the data disks to a data tape which could be

read by the mainframe computer used to reduce the data.

As previously mentioned in the section describing the LFA setup

procedures, the data for each stream surface survey were acquired in two

parts, an interblade-row survey and an intrablade-row survey. The

interblade-row surveys for each stream surface were accomplished in the

f a l l season ( i . e . , October), about five months prior to beginning the

intrablade-row surveys which were accomplished during the later part of

the winter season ( i . e . , March). The reason for the large time difference

between acquiring the interblade-row and intrablade-row surveys was due to

facility scheduling and hardware availability problems. Fabrication of

the stator window was not complete at the time the stator surveys were

scheduled. Therefore, in order to keep on schedule, i t was decided to go

...
 51

ahead and accomplish the 1nterblade-row surveys with the rotor window In

place. The rotor window, however, did not extend much past the stator

leading edge (see Figure 3.7), and therefore the 1ntrablade-row surveys

had to be postponed until the stator window was available. Once the

stator window was available, the Intrablade-row surveys were Initiated.

An Interblade-row survey required about 6 hours to complete, and an

1ntrablade-row survey required about 12 hours to complete. Approximately

4 hours were required to complete the preliminary setup procedures. On

the average, each survey point required about 3 min to complete, which

Included the time required to position the optics, write the data to disk,

and print the survey log. In general, the data for each Interblade-row

and Intrablade-row survey were acquired with the rig In continuous

operation. However, brief shutdown periods were occasionally required In

order to clean the laser window or to recheck reference points. Although

a substantial span of time elapsed between completion of the Interblade-

and 1ntrablade-row survey measurments, there appears to be good continuity

between all results. In any case, the conclusions obtained" from these

survey results are specific to measurements acquired completely within

either the Interblade- or 1ntrablade-row surveys, and therefore are not

affected by any possible mismatch between these surveys.

The PMT supply voltage and laser power were set at nominally 1600 V

and 1.5 W, respectively, which resulted 1n a typical data rate of about

1000 measurements per second. However, the data rate varied from less

than one measurement per second to as much as 2500 measurements per

second, depending on various factors, such as; Insufficient seed particle

entralnment across flow boundaries which reduces seed particle density 1n

 52

the laser probe volume, proximity to solid surfaces and window fouling

which degrades signal to noise ratio, and low velocity regions which

reduce the number of seed particles crossing the probe volume per unit of

time. In order to minimize the overall time to acquire data for a

complete survey, data acquisition was suspended I f the data rate could not

be maintained above about 50 measurements per second. However, In cases

where the data was considered Important enough to acquire even though the

data rate was below 50 measurements per second, the number of measurements

requested was reduced to as few as 10 000 In order to obtain some measure

of the velocity and yet s t i l l minimize the overall data acquisition time.

B. Data Reduction

The f i r s t level of data reduction consisted of determining the

velocity components which characterize the fundamental nature of the

unsteady-deterministic and random features of the flow field. These

velocity components were then used to calculate a l l other results

presented herein. Including the rotor- and stator-relatlve time-average

velocities, and the velocity correlations associated with Adamczyk's

average passage model. In order to facilitate Interpretation of the

results, a brief summary of Adamczyk's average passage model Is presented

herein. For a l l calculations, density fluctuations were neglected as

there was no means of measuring the time-resolved density. The symbols

used In the equations are defined In the Symbols and Notation section.

The relationships between the actual laser anemometer measurements and the

various averages used In the data reduction procedures are provided In

Figure 4.4 which will be referred to In the ensuing sections.

 53

DISTRIBUTION OF DISTRIBUTION OF
LASER ANEMOMETER LASER ANENOMTTER
VELOCITY MEASUREMENTS VELOCITY MEASUREMENTS
AT ROTOR SHAFT POSITION AT ROTOR SHAFT POSITION
(R,Z,6 CONSTANT) (R.Z.B CONSTANT)

A
H (-"A

Jr-
VELOCITY. V VELOCITY.

ENSEMBLE-AVERAGED VELOCITY DISTRIBUTION MEASURED

ACROSS 6 OF THE 1/ MEASURED BLADE PASSAGES
(R.Z.E CONSTANT)

I DENOTES MAGNITUDE OF « 2 VV^'

275 505 335 365 395 125 456 186 516
ROTOR SHAFT POSITION
_/

TEMPORAL PHASE-LOCK
AVERAGED
VELOCITY DISTRIBUTION
(R.Z.9 CONSTANT)

, , 1

IS>
PS" j ss T
1
RDENOTES
MAGNITUDE
OF
TIME-AVERAGED
VELOCITY DISTRIBUTION
(R.Z CONSTANT)
10 20 30 10 50
ROTOR SHAFT POSITION

20 10 60 80 100/
PERCENT ROTOR PITCH

0 20 10 60 80 100
PS SS
PERCENT STATOR GAP

Figure 4.4 Illustration showing the relationship between the

actual velocity measurements and the various
averages used In the data reduction process
 54

Velocity components

The measured seed particle fringe crossing frequencies are converted

to velocity by multiplying by the fringe spacing. The velocities are then

converted to standard day conditions using the equation

(4.1)

where is the corrected velocity, T^ is the plenum temperature, and

Tg Is the standard-day temperature. In a l l subsequent equations V will

always refer to the standard-day corrected velocity and, therefore, the

subscript c will be eliminated henceforth.

As previously mentioned, the laser anemometer system acquires

velocity measurments over many successive rotor revolutions which results

in a distribution of velocity measurements at every surveyed rotor shaft

position. Examples of two such velocity distributions acquired at

different rotor shaft positions are provided in the upper two plots of

Figure 4.4. Each measured velocity distribution contains information

about both the unsteady-deterministic ( i . e . , rotor wake) flow-field

features and the "random" unsteadiness. Because the flows in

turbomachinery are nonstationary, "ensemble averaging" [78] is required in

order to decouple the unsteady-deterministic flow-field features from the

"random" fluctuations. For the purpose of this thesis, when ensemble

averaging, an experiment Is defined as a l l measurements acquired during a

given rotor revolution. As measurements are acquired during each rotor

revolution, the LFA data acquisition system automatically sums the measured
 55

particle fringe crossing frequency and Its square at every rotor shaft

position. Therefore, the ensemble average velocity and corresponding

variance are readily determined at each rotor shaft position as

n w

1=1

where the superscript denotes the ensemble average, Is the

Instantaneous velocity measured at a particular rotor shaft position

during a given rotor revolution, and n Is the number of measurements

acquired at that particular rotor shaft position. Since there were

typically 30 000 measurements acquired for each measured component, and

assuming these measurements were equally distributed across a l l 850

measured rotor shaft positions, there would be an average of n = 35

measurements per rotor shaft position. A typical distribution of

ensemble-averaged velocities across 6 of the 17 surveyed rotor blade

passages Is shown In the middle plot of Figure 4.4 ( I . e . , third plot from

the top).

The ensemble-averaged velocity components In the research compressor

coordinate system can be calculated, at every rotor shaft position of each

survey point, using the ensemble-averaged results determined from runs

made at multiple beam orientations. The research compressor coordinate

system and LFA beam geometry orientation are shown In Figure 4.5. The

measured velocity component V lies along line AA, which Is In the

 56

YR LASER
/\BEAMS
BEAM
BISECTOR

!-/

STATOR

FLOW

\
\
A
Figure 4.5 Research compressor coordinate system and laser
beam geometry orientation
 57

plane of the beams and perpendicular to the bisector of the crossing

beams. The beam bisector can be deflected In the off-radial direction by

the goniometer mirror (see Fig. 3.6). The beam bisector 1s restricted to

the (R,0) plane, and the deflection angle 1s denoted by <p . The
R
rotatable beam splitter Is used to rotate the direction of the fringe

normals about the R'-ax1s (which 1s aligned with the beam bisector). The

angle between the fringe normals and the Z-ax1s 1s denoted by and

1s measured In the (Z.e') plane. The z, e, r velocity components

are related to the measured velocity components through the following

three equations.

V cos a + V cos B + \/„ cos Y = V (4.4)

z m 0 m R m m

where subscript m Is assigned the values 1, 2 , 3, corresponding to each

different beam orientation, and

cos a = cos <p^

cos (3 = cos <Po sin <p

K Z
cos Y = sin <p_ s1n (p
K Z
For the present Investigation, there were only two runs made at

different (p angles. Therefore, assuming (p = 0, the following

two equations can be used to calculate the axial and tangential components.

V cos o + V cos (3 = V (4.5)

z me mm

In practice * = 0 for most survey points. However, near the stator

K
surface, the (p angles for the two runs are equal, but are set to
K
some nonzero value In the range -6° to +12° (pressure to suction surface,

respectively) In order to eliminate beam blockage due to the window

frames. Thus, the calculated tangential component of velocity actually

 58

lies along the 0' direction when ^ 0. Although this

Introduces an additional error source In the measurements, I t was

considered a reasonable sacrifice In order to obtain some Indication of

the nature of the flow field close to the stator surface, which would

otherwise not be possible. Since the cosine of 12 degrees Is 0.9781, the

difference between the calculated velocity In the e' direction and the

velocity In the e direction (2.2 percent) was considered acceptable.

The ensemble-averaged axial and tangential velocity components are

calculated from the ensemble averages of the measured velocity components

using the ensemble averaged form of Equation (4.5).

(4.6)

The total absolute ensemble-averaged velocity, therefore. Is

(4.7)

Earlier Investigations [79] have demonstrated the periodicity of the

flow field between the rotor passages of this research compressor.

Figure 4.6 shows, at several different survey locations, ensemble-averaged

axial-velocity profiles measured across each of 17 rotor blade passages,

plotted to the same scale. These measurements, acquired during the scope

of this research program, support the periodicity of the flow field

between the rotor blade passages of this research compressor. Therefore,

as standard practice, all ensemble-averaged velocity components presented

herein are temporally phase-lock averaged as follows.

(4.8)
 59

200 r

52% STATOR CIKWD

75% STATOR LEADING CDGC CAP

(A)

-35% STATOR CHORD

UST STATOR L6ADIN0 CDCC CAP

(B)

-18% STATOR CHORD

5% STATOR LEADING EDGE GAP

(C)
120
20 (JO 60 80
PERCENT ROTOR PITCH

Figure 4.6 Periodicity of rotor flow

 /xrw
where G represents any flow-field parameter (e.g., ^ or V ) , the

subscript j=l to 50 denotes a particular shaft position measured

relative to the suction surface of the 1'th rotor passage, the

superscript ««« denotes the temporal phase-lock average, and N^p is

the total number of rotor passages surveyed ( I . e . , 17 passages). Since

the ensemble-averaged axial and tangential velocities were determined from

the measured data at 850 rotor shaft positions, equally distributed across

17 of the 22 rotor blade passages, the successive application of Equation

(4.8) results in a description of the flow field in a representative rotor

passage at 50 equally spaced shaft positions across the rotor passage

(e.g., second plot from the bottom of Fig. 4.4). I f a l l 30 000

measurements were equally distributed across these 50 rotor shaft

positions there would be an average of 600 measurements per shaft

position. Note that all velocity components are temporally phase-lock

averaged after the ensemble-averaged velocities and their corresponding

variances are determined. This was done In an attempt to preclude any

aperiodic flow-field variations, such as due to rotor passage geometry

differences, from occurring In the temporally phase-lock averaged

results. The temporally phase-lock averaged velocity and variance are

assumed to characterize the fundamental nature of the typical

unsteady-deterministic and random features, respectively, of the flow

field, and therefore are used to determine a l l other reduced flow-field

quantities, including the velocity correlations.

The lower plot of Figure 4.4 shows a typical distribution of

time-averaged velocity measurements across a stator circumferential gap at

one axial survey location. The temporal-phase-lock-averaged velocity

 61

measurements at each gap-wise location are used In determining the

corresponding time-averaged velocities In the absolute ( i . e . ,

stator-relatlve) frame of reference (see Appendix B). The difference

between the temporal-phase-lock-averaged results and the time-averaged

results provides an estimate of the deterministic velocity fluctuations.

The circumferential average of the gap-wise distribution of time-averaged

velocities at each axial survey location provides an estimate of the

corresponding axisymmetric velocity.

2. Summary of average-passage model

In the derivation of the average-passage system of equations,

Adamczyk [7] assumed that the absolute velocity field, with respect to a

given blade row, could be decomposed In the following manner. First,

Adamczyk assumed that the absolute velocity field could be decomposed as,

1n vector notation,

4/ = 4/+

where Is the ensemble-averaged (unsteady-deterministic) component,

and Is the random fluctuating component. Next, Adamczyk assumed that

the ensemble-averaged velocity field could be decomposed as

/VA

J\/\ /vol
where <v 1s the t1me-averaged component, and y is the deterministic

fluctuating component. Finally, Adamczyk assumed that the t1me-averaged

velocity field could be decomposed as

I
•v/v /v/v /v/v
<p =i> >>-<u
 62

/////
«/\A
where y Is the passage-to-passage averaged (periodic) component, and

^ is the passage-to-passage aperiodic component. The passage-to-passage

averaged component describes a steady-state flow field which Is the same

In each blade-to-blade passage of a specific blade row. The passage-to-

passage aperiodic component arises from differing rotor or stator blade

counts 1n successive stages. In a single-stage turbomachlne the aperiodic

component, V, would be zero, provided passage-to-passage flow variability

Is not Introduced by passage-to-passage geometry differences (e.g., due to

manufacturing tolerances of Installation). Combining a l l of these

separate velocity decompositions together yields,

///// I
F\F\ R>J^
= ^ f (4.9)

I t should be noted that a l l averages In the above equation are density

weighted averages, according to Adamczyk's derivation. Adamczyk then

substituted the above velocity decomposition Into the Navler-Stokes

equations and. In a manner analogous to Reynolds averaging the

Navler-Stokes equations, performed the following averages.

1. An ensemble average to decouple the unsteady-deterministic

flow-field features from the random unsteadiness.

2. A time average to decouple the steady state flow-field features

from the deterministic unsteadiness.

3. A phase-lock average to decouple the passage-to-passage periodic

flow-field features from the passage-to-passage aperiodic

flow-field features.

..
 63

As a result, the average-passage system of equations was developed.

From the derivation of the average-passage system of equations, the

following second-order tensor was Identified.

/\jr\j r\j 'VfV'o/' •O'fWV

^1J = + ,V,Vj f pV^Vj (4.10)

where subscripts 1 and J are assigned the values 1, 2, and 3, which

correspond to the axial, tangential, and radial coordinates,

respectively. The variable p Is the fluid density, St.. Is the total

TITTUNN IJ
apparent stress, V. Is the stress due to passage-to-passage
LLUJJJJJ. W

aperlodlclty, pV,V Is the stress due to deterministic unsteadiness,

tftttttit 1 J
/VA/X/N
and pV^Vj 1s the stress due to random unsteadiness. This total

apparent-stress tensor 1s generic to the average-passage system of

equations, and along with body forces and energy terms, Its evaluation

constitutes the closure problem for this equation system.

Since the data presented herein were acquired In only one stator

passage of a single-stage compressor, there Is no measure of stator

passage-to-passage aperlodlclty. Therefore, the aperiodic stress term of

t tittt tu
Equation (4.10), pV^Vj = 0. In addition, although the measured flow field

Is compressible, there was no means of measuring the time-resolved

density, therefore. Equation (4.10) cannot be Identically calculated from

the measured data. However, 1f the density fluctuations are negligible

compared to the velocity fluctuations, we can rewrite Equation (4.10) as

— / v r > y v v

^1J = T l ^ j ' VjVj ^ (4.11)

 64

The following sections describe the procedures and equations used to

resolve the correlations of the deterministic and random velocity

/v' r\j I A/w/v
fluctuations ( i . e . , V^Vj and VjVj, respectively).

3, Deterministic-velocity correlations

A possible approach for further decomposing the deterministic-velocity

correlations, , which might be useful for developing a model of

the deterministic-velocity correlations was proposed by Adamczyk et al.

[69]. Drawing from an analogy to gust theory [80], Adamczyk further

decomposed the absolute deterministic velocity component as

4%R,e,z,t) ='pAX(R,z) +[f^(R,e,z)

+ [iyR(R,6 - wt,z) -(pAXfR.z)] +f^(R,e - cot.z) (4.12)

where, with respect to the current research compressor,4^^'* is the

axlsymmetric velocity,4^ is the absolute velocity field as observed in

a frame of reference fixed to the rotor blade row ( I . e . , describes the

Incident gust), 1s the corresponding velocity field observed in a

frame of reference fixed to the stator blade row ( I . e . , the interacted

velocity field),4^^ is the component of velocity which Is unsteady in

either frame of reference ( i . e . , response of the Interacted field to the

Incident gust),4^^ 1s the component of velocity which is unsteady in

the rotor frame of reference (due to spatial aperiodiclty of the stator

flow field, and rotor geometry Induced flow-field variations), and

, ^ (n - 1) 2*
(N N w
SR R)

where n denotes a particular rotor shaft position, is the number

of rotor blades, N^p is the number of rotor shaft positions per rotor
 65

blade passage (N^p = 50 for this rotor), and w Is the angular speed of

the rotor. For the test compressor used In this research program

Is assumed negligible, and therefore was eliminated by temporally

phase-lock averaging, as explained previously.

Fixed to the stator frame of reference the velocity will

appear steady 1n time, while the components and will appear

unsteady In time. Therefore, i f we define the velocity component as

=[4yK (R,e - wt,z) -4 /A *(R ,z)] (4.13)

then the deterministic-velocity correlations which appear in the

average-passage equation system associated with the stator blade row can

be obtained by forming the time average, relative to the stator blade row,
I U 2
of (<(/ + < ( / ) , assuming density fluctuations are negligible. The

result, in tensor notation, is

RV'/WL T T T II II T II II
V^Vj = vjvj + vjvj + vjvj + V^Vj (4.14)
TDVC IDVC ^T'lSVC SDVC

where the subscripts 1,j take on the values 1, 2, and 3, corresponding

to the axial, tangential, and radial velocity components, respectively.

The first term on the right of the equal sign are the correlations of the

deterministic-velocity fluctuations associated with the rotor-relative

time-averaged rotor wake ( i . e . , incident gust), and may be thought of as

representing the deterministic unsteadiness which might be present in the

absence of rotor/stator Interactions. The remaining terms on the right of

the equal sign represent the stators response to the incident rotor-wake
 66

gust. The meaning of the acronyms below each term of Equation (4.14) are

explained In a later section.

The temporally phase-lock averaged velocity field, determined using

Equation (4.8), contains a l l Information about the total deterministic

unsteadiness, which Is a function of r , z, e, t (where t Is a

function of rotor shaft position, not real time). The Incident velocity

field components, determined using Equation (ll.l)(see Appendix B), are

Independent of circumferential position, e, and provide the necessary

Information about the Incident rotor-wake gust. Therefore, the

total-deterministic- and Incident-velocity correlations are calculated by

f i r s t determining their respective fluctuating velocity components In the

research compressor coordinate system. Then, at each spatial survey

point, the square of the fluctuating velocity components at every rotor

shaft position are determined. Finally, these results are time averaged

In the stator frame of reference (see Appendix B) to yield the

corresponding total-deterministic- or Incident-velocity correlations at

every spatial survey point. Thus, the total-determlnlstlc-veloclty

correlations are determined as

(4.15)

and the Incident-velocity correlations are determined as

(4.16)

where superscripts / N / W %, , , refer to ensemble average.

temporal phase-lock average, rotor-relative time average, and

stator-relative time average, respectively. The corresponding

 67

ax1symmetric values of the total deterministic-velocity correlations are

determined from arithmetically averaging the results determined from

Equation (4.14) for a l l circumferential survey points at each axial survey

station.

4. Turbulent-velocity correlations

In Equation (4.10), the term representing the turbulent-velocity

correlation consists of three components (see also Eq. (4.11)). These

three components are essentially the stator-relatlve time averages of the

axial and tangential components of the Reynolds normal stresses and the

Reynolds shear stress. Since a single component laser anemometer system

was used during this research program to acquire the data, the Reynolds

shear stress component could not be determined. Furthermore, at each

survey point, velocity measurements were made at only two beam

orientations, which were acquired Independently of each other. Therefore,

I t was not possible to calculate the standard deviation of the velocities

for other than the measured velocity components. However, a procedure to

estimate the upper and lower bounds of the standard deviation of the

velocity components 1n the axial and tangential directions 1s explained 1n

Appendix C. Unfortunately, these bounds are extremely large and therefore

they provide very l i t t l e useful Information. Therefore, the standard

deviations of the actual measured velocity components are generally used

to provide Information about the nature of the turbulent-velocity

correlations.
 68

V. PRESENTATION AND DISCUSSION OF DATA

The results of the LFA measurements acquired along the 10 and

50 percent axisymmetric stream surfaces are presented and discussed In

this section. The results are presented In graphical form only, Including

contour maps which are frequently relied upon In order to help assimilate

the large amounts of data Involved. All measured data points are graphed

with a symbol marking the actual data point and statistical error bars

which represent the 95 percent confidence Intervals. Contour maps are

constructed to present the results across two blade passages In order to

afford better visualization of the flow patterns. For contour plots of

the time-resolved data which are not periodic from passage to passage, the

flow field 1n any other passage can be determined from the results of the

measured stator passage through their phase relationship with the rotor.

In constructing the two-passage contour plots the contour plotting program

has no knowledge of the presence of the middle stator blade of the contour

plot, and therefore only the upper and lower stator blades should be used

to Infer conclusions about the results near the stator blade surfaces.

The actual LFA measurements are stored on magnetic tape which Is available

for distribution.

The unsteady flows which will be discussed herein are classified Into

two categories; (1) deterministic unsteadiness, and (2) turbulence.

Deterministic unsteadiness refers to the rotor-wake-generated unsteadiness

caused by the periodic chopping and subsequent transport of rotor wakes

through the stator blade row. Turbulence, In the context of this

discussion, refers to any unsteady flow fluctuations which are not

 69

correlated with the blade passing frequency, for example; global flow-

field fluctuations, vortex shedding from blades, shock Induced flow

oscillations, and purely random flow fluctuations.

In typical turbomachines, the unsteadiness caused by the relative

motion between blade rows is normally due to both potential flow-field

interactions and viscous interactions due to wake chopping. However, as

previously mentioned, the data presented herein were acquired within a

compressor designed for minimal acoustical interactions, which was

accomplished by increasing the spacing between the blade rows. As a

result, the interactions between the potential flow fields of each blade

row were minimized. Therefore, only viscous interactions due to

deterministic ( i . e . , rotor-wake-generated) unsteadiness and turbulence are

present in the results contained herein.

As previously mentioned (Section IV.B.4), only bounds on the minimum

and maximum expected turbulence levels could be provided for turbulence

components, except in the direction of the measured fringe angle

orientations. To provide some indication of the turbulence variation in

the flow field, the lower bound of turbulence will be used to provide

qualitative indications of the axial, tangential, and total components of

turbulence. However, when a quantitative measure of turbulence is

required the actual measured turbulence estimates in the directions of the

fringe angle orientations will be presented.

A. Uncertainty Analysis

The uncertainty analysis presented herein makes use of the terms

precision and bias as defined by Kline [81], wherein precision refers to

 70

random components of uncertainty or repeatability and bias refers to fixed

components of uncertainty or systematic errors. The uncertainties In

positioning the laser probe volume and In determining the rotor shaft

position are estimates of the total (precision plus bias) uncertainties,

and have been previously quoted at a 95 percent confidence level 1n

Sections III.C.3 and IV. A. 3, respectively. The error In any single LFA

measurement Is a function of many different parameters Including flow

turbulence and random noise In the photomultlpller tube signal. As a

result, I t Is difficult to make estimates of the total uncertainty In an

Individual LFA measurement since this noise Is generated by background

radiation which varies with each measurement. However, estimates of the

precision uncertainty Intervals of the ensemble-averaged velocities were

calculated, based on the number of Individual LFA measurements acquired at

each rotor shaft position. The estimated precision uncertainties of the

ensemble-averaged velocities were then used to estimate the precision

uncertainties of all other parameters which were calculated from the

ensemble-averaged velocities using the method of Kline [81 ]• for

calculating the propagation of uncertainties In calculations.

All graphed quantities Include error bars to Identify the estimated

precision uncertainty Intervals of a l l measured or calculated parameters.

A summary of some of the other error sources associated with LFA

measurements are presented below, and the estimates of the uncertainties

In the measured velocities resulting from each error source are provided

In Table 5.1. All uncertainties are quoted for a 95 percent confidence

level.
 71

Table 5.1 Estimates of bias uncertainty In velocity measurements

Statistical bias 1.2 percent within rotor wakes

0.3 percent outside rotor wakes

Angle bias 3 percent w1th1n rotor wakes

1 percent outside rotor wakes

Negative velocity bias could not be estimated

due to flow reversals

Bias due to probe volume 0.2 percent

position uncertainty
 72

!• Precision uncertainties

A statistical measure of the uncertainty In the precision of V as

an estimate of the true mean velocity, ji , 1s given by

where v' and J are the ensemble-average and standard deviation

(Eqs. (4.2) and (4.3)), respectively, and t^yg the value of the

student-t distribution at the (1 - a) 100 percent confidence level with

(n - 1) degrees of freedom. For a 95 percent confidence level (a = 0.05)

and n > 30, t^yg = 2.0.

Similarly, a statistical measure of the uncertainty In the precision

of J as an estimate of the true variance, a , Is given by

(n - n f . 2 (n - l)j2
2 2
^a/2 *l-*/2

where Is the variance of a random sample of size n (Eq. (4.3)),

2 2
and x^/2 '^l-a/2 values of a chl-square distribution

at the (1 - o) 100 percent confidence level with (n - 1) degrees of

freedom [82].
c2
I t should be noted that J resulted from the sum of flow

fluctuation effects such as turbulence, velocity variations due to rotor

speed d r i f t , and velocity gradients In the tangential direction across the

measurement shaft position Interval, plus the previously mentioned random

noise In the photomultlpller tube signal. The maximum velocity gradient

In the tangential direction occurs across the rotor wake and Is on the

order of 3 percent per rotor shaft position at the rotor trailing edge and

decreases to less than 0.1 percent per rotor shaft position at the stator
 73

trailing edge. Observations indicate that the rotor speed d r i f t during a

run Is on the order of 0.3 percent.

The axial and tangential velocity components are related to the

measured velocity components and the fringe orientation angles through

Equation (4.5). The 95 percent confidence level of the precision

uncertainty of the fringe orientation angle, <p^, is 0.03°, and is

Included along with the precision uncertainties of the measured components

in determining, through the propagation of uncertainties analysis, the

precision uncertainties of the axial, and tangential components.

2. Bias uncertainties

The uncertainties associated with LFA measurements, including

methodologies for quantifying them, have been fairly comprehensively

documented [83-85]. Only those sources of bias which Introduce the

greatest uncertainty, and for which estimates of the uncertainties can be

determined will be described below. Two of the most often cited sources

of LFA bias uncertainty are statistical and angle biasing. Statistical

biasing [86] arises due to the variation of velocity magnitude with time.

For a uniformly seeded flow, more particles cross the probe volume per

unit time when the velocity Is higher than when the velocity is lower than

the mean. Therefore, an arithmetic average of measurements made over a

given period of time yields a calculated mean velocity which is higher

than the true mean. The greatest velocity variations occur in the rotor

wake In which the velocity variations are as much as 15 percent of the

mean, which would result in a 1.2 percent correction for velocity bias.

However, these large excursions from the mean velocity are a result of the
 74

vortex nature of the rotor wakes [87]. Because of the vortex nature of the

wake there are two equally probable velocity states which can occur during

any given measurement (see Fig. 4.4), and the probability of measuring a

velocity equal to the mean of the distribution can be virtually n i l . All

other velocity distributions outside the rotor wake are single peak

distributions. The greatest standard deviation from the mean of a single

peak distribution, or from either peak of the double peaked velocity

distributions Is about 5 percent, which would result In a 0.3 percent

correction for velocity bias. Because the correction for velocity bias Is

typically small, and since the nature of the velocity distribution cannot

be ascertained, as only the mean and standard deviation of the velocities

are usually recorded, the data have not been corrected for statistical

bias.

Angle biasing [88] occurs because the flow direction fluctuates with

time. More measurements per unit time occur when the flow direction Is

parallel to the fringe normal direction, e^, than when the flow

direction fluctuates away from the fringe normal direction. The bias In

the arithmetic average of velocity measurements acquired over a given time

period Is proportional to the angle between the fringe normals and the

mean flow direction. For the data presented herein, the angle between the

fringe normals and the mean flow direction Is 20°, which yields an angle

bias uncertainty of about 1 percent. However, based on LFA measurements

the mean flow angle changes by as much as 30° across the rotor wakes, and

therefore there are regions In which the angle between the fringe normals

and the mean flow direction are on the order of 35°, which could result In

a 3 percent error In the measured velocity.

 75

Another source of velocity bias [83] occurs from the Inability of the

current LFA system to distinguish between seed particles traveling 1n the

positive or negative direction. Therefore, In regions containing flow

reversals, such as 1n the separated boundary layers, the mean velocity

will be biased high. The use of Bragg shifting Is one means of

eliminating this source of velocity bias.

The uncertainty In positioning the probe volume within the test

section Is also a source of velocity bias [83]. This source of velocity

bias can be calculated as the product of the estimated spatial flow

gradients and the positional uncertainty 1n the direction of the

gradient. The largest spatial gradients w1th1n the measurement grid are

on the order of 1.5 percent per millimeter, and occur around the stator

leading edge, near the stator suction surface, and across the stator

wake. The maximum positional uncertainty of the probe volume Is on the

order of 0.1 mm, which would result 1n less than a 0.2 percent bias

uncertainty In the measured velocity.

B. Description of the Measured Flow Field

The relative motion between the rotor and stator blade rows

complicates a description of the measured flow field. However, I t Is

possible to provide a reasonably clear picture of the unsteady

Interactions resulting from both the deterministic unsteadiness and

turbulence by presenting several descriptions of the measured flow field

with respect to different viewpoints. The f i r s t viewpoint Is a

"t1me"-averaged description of the measured flow field as would be seen by

an observer moving with the stator blade row. This Is, of course, a view
 76

from the measurement frame of reference. The second viewpoint Is a

"t1me"-averaged description of the measured flow field as would be seen by

an observer moving with the rotor blade row. This view provides a clear

picture of the rotor wakes which eventually Interact with the stator blade

row. The final viewpoint 1s again from the measurement frame of

reference, but now the "time" resolved description of the measured flow

field Is presented. This view shows the relative motion between the rotor

and stator rows, and the actual chopping and subsequent transport of the

rotor wakes as they pass through the stator blade row.

In the context of this discussion, "time" refers to the relative

position of the rotor and stator rows, not real time. In the stator frame

of reference, time 1s Indicated by rotor shaft position. In the rotor

frame of reference, time 1s Indicated by circumferential position. Of

course, to correctly represent units of time the rotor wheel speed must be

considered. However, for Interpretation of the results I t Is not

necessary to display the correct units for time.

1. Stator-relative time-averaged description of measured flow field

Figures 5.1 and 5.2 show the components of the stator-relatlve

time-averaged absolute velocity field through the stator row for the 10

and 50 percent stream surfaces, respectively, which Illustrates the extent

of the stator potential flow field. I t Is apparent that the stator

potential flow field extends only about 30 percent of stator chord

upstream from the stator leading edge plane, whereas the rotor trailing

edge Is almost 100 percent of stator chord upstream of the stator leading

edge. Therefore, I t 1s obvious that potential Interactions between the

 77

SJSJÎEY 10% STREW SURFACE

STATIONS A B C DEFGHIJKLN NOP Q R
8.5r I

ZOO CURVE VALUE, CURVE

M/S LABEL
160 (A)
150 170 (B)
180 (C)
100 190 (D)
200 (E)
210 (F)
50 220 (6)
230 <H>
240 (I)
0 250 (J)
260 (K)
ROTOR TRAILING EOCE 270 (L)
-50

(A) ABSOLUTE TOTAL VELOCITY. VY.

8.5 R I I I I II II II I I I I I I I
i 200
C \
1 8.0 \ CURVE VALUE. CURVE
M/s LABEL
ft ISO
145 (A)
s 7.5 155 (B)
i™ 5
165
175
<C)
(D)
5.0 185 (E)
k 195
205
(F)
(G)
^ 0 215 (H)
2.5 225 (I)
i
I -50
rROTOR trailing EDGE
0
(B) ABSOLUTE AXIAL VELOCITY, V,.
8.5 - I I I I II I I I I I I I I I I I I CURVE VALUE. CURVE
M/s LABEL
200 (A)
-180
3.0 -165 (B)
150 -150 (c)
-135 (D)
-120 (E)
7.5
100 -105 (F)
-90 (G)
-75 (H)
50 5.0
-60 (I)

0
( -45
-30
(J)
(K)
2,5 -15 (L)
0 (M)
,'ROTOR TRAILING EDGE 15 (N)
-50
30 (0)
5.0 7.5 10.0 12,5 15.0 17.5 20.0
CM
I I I _l I I
-100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD
PI»
( c ) ABSOLUTE TANGENTIAL VELOCITY. Vg.

Figure 5.1 Contour maps of the stator-relatlve time-averaged

absolute velocity field along the 10 percent stream
surface
 78

50% STREAM SURFACE

AXIAL G.
H ,1
SURVEY \ /
STATIONS A B CD EFlfjKL n N 0 P OR S T
I I II I I I I I I II I I
200 r- 8.0 CURVE VALUE. CURVE
M/S LABEL

150 160 (A)

7.5 170 (B)
180 (C)
100 190 (D)
5.0 200 (E)
50 210 (F)
220 (G)
230 (H)
0 2.5 210 (I)
250 (J)
-50
ROTOR TRAILING EDGE 260 (K>
oL_f^ 1 1 1 270 (L)

(A) ABSOLUTE TOTAL VELOCITY, V|.

S I I II I INilI I I I I I I I I I
&200 8.0
S CURVE VALUE, CURVE
giw M/S LABEL
7.5 145 (A)
§100 155 (B)
165 <C)
55.0 175 (D)
W 50
185 <E)
195 (F)
2.5 205 (G)
215 (H)
,"ROTOR TRAILING EDGE 225 (I)
J I I ' '

(B) ABSOLUTE AXIAL VELOCITY, %. CURVE VALUE, CURVE

200 r- 8.0
I I i l l IIII I I I I I I M I I M/S LABEL
-180 (A)
150 ofc» -165 (B)
7.5 -150 (C)
-135 (D)
100 (E)
-120
-105 (F)
5.0
50 -90 (G)
-75 (H)
-60 (I)
0
2.5 -45 (J)
-30 (K)
-50 ROTOR TRAILING EDGE -15 (L)
0 (M)
5.0 7.5 10.0 12.5 15.0 17.5 20.0 15 (N)
CM
I I I I I I
-100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD

(c) ABSOLUTE TANGENTIAL VELOCITY, V,

»
8-

Klgure 5.2 Contour maps of the stator-relatlve t1me-averag.u

absolute velocity field along the 50 percent stream
surface
 stator and rotor are very weak (acoustic interactions, however, may s t i l l

be present), and that the only significant blade row interactions are due

to rotor wake generated unsteadiness and turbulence. I t is also apparent

from Figures 5.1 and 5.2 that a significant low velocity region exists

along the stator suction surface which eventually grows to become part of

the stator wake as i t leaves the stator blade. From the extent of this

low velocity region near the stator suction surface, i t is probable that

the boundary layer has separated. Figure 5.3 shows the locus of maximum

velocities measured along the suction surface for both the 10 and

50 percent axisymmetric stream surfaces. These maximum velocities were

determined at various chordwise locations by observing the pitchwise

location at which the magnitude of the total time-averaged absolute

velocity was a maximum, which occurred just prior to the laser probe

volume entering the low velocity region near the stator suction surface.

I t is probable that reverse flows are present within this low velocity

region, but this could not be ascertained with the present LFA system. A

similar effort to determine the stator pressure surface boundary layer was

attempted, but was unsuccessful due to the spatial resolution of the LFA

probe volume compared to the thickness of the pressure surface boundary

layer. Other observations of the stator-relative time-averaged flow field

include the approximate location of the stagnation point and the resulting

high velocity gradients around the stator leading edge.

The stator-relative time-averaged description of the stator flow

field as depicted in Figures 5.1 and 5.2 is essentially the kind of

picture a turbomachine designer has in mind when trying to design a blade

row to match a given set of velocity triangles. I f any attempt is made to

 80

Q
7.5 I
100
5
y Ul
œ
50 y 5.0
</) —-o
a
|s: 5!
en os

i " /- ROTOR
/ TRAILING
-50 / EDGE .

5.0 7.5 10.0 12.5 15.0 17.5 20.0

AXIAL DISTANCE, CM
-100

-100 -50 0 50 100 150

AXIAL DISTANCE, PERCENT STATOR CHORD
(A) 10% STREAM SURFACE.
G-I RI
150r A CD E F\H/JK L M N 0 P Q R
7.5 I II I I lui I I I I I I II

a 100
5
5.0
si 50
S
<o —o o
Q o fj

< </) 2.5

C9
-50
i /-ROTOR
/ TRAILING
S /•' EDGE

•lool— 5.0 7.5 10.0 12.5 15.0 17.5 20,0

AXIAL DISTANCE, CM

-100 -50 0 50 100 150

AXIAL DISTANCE, PERCENT STATOR CHORD
(B) 50% STREAM SURFACE.

Figure 5.3 Measured extent of the stator suction surface low

velocity region along the 10 and 50 percent stream
surfaces
 81

account for the unsteady flow effects 1t 1s usually through "turbulence

modeling." Turbulence modeling, as used In this paragraph, refers to the

usual turbulence associated with external flows In which turbulence Is the

only unsteadiness Involved. This kind of turbulence modeling Is

appropriate for a Reynolds averaged model of the flow field. However, In

turbomachlnes the flow field Is nonstatlonary, and therefore the Reynolds

averaged equations of motion are time dependent. Therefore, the use of

turbulence modeling of the type referred to above would require solution

of the time-dependent Reynolds averaged equations of motion throughout an

entire multistage turbomachlne, which Is currently Impractical.

2. Rotor-relative time-averaged description of measured flow field

To provide a clearer picture of the unsteadiness a stator row sees as

a result of an upstream rotor row we can recast the measured LFA data In

the rotor frame of reference (see Appendix B). Contour maps of the

rotor-relative time-averaged turbulence along the 10 and 50 percent stream

surfaces are shown In Figure 5.4. The shaded regions of Figure 5.4

Indicate the rotor wake fluid (see Appendix A). Figure 5.4, therefore,

represents a time-averaged picture of the rotor wakes as they would be

seen by an observer riding with the rotor blades. In this case, the

stator blades appear smeared out, since they are moving relative to the

rotor. Referring to Figure 5.4(B), the rotor wake appears to broaden more

rapidly within the stator passage until eventually the adjacent rotor

wakes begin to mix together by about 120 percent of stator chord. The

broadening of the rotor wake within the stator passage Indicates that the

stator row promotes mixing. The turning of fluid by the stator row can
 82

A B C D E F G H I J K L H N O P Q R
I I I I I M I I I I I I I I I I I
200 r

150 •

100 •

50 -

S -50^

200r
z A
(A) 10% STREAM SURFACE.

B CD EF HJK L
I I li I lllin I I I 1
M N 0 P
I M
OR S
t
Ï
1

150-

100-

50 R

-50-

-100^
_1_ I
-150 -100 -50 0 50 100 150
AXIAL DISTANCE* PERCENT STATOR CHORD
(6) 50% STREAM SURFACE.

Figure 5.4 Contour maps of the rotor-relative time-averaged

turbulence field showing the rotor wakes and the
dissipation of turbulence
 83

also be observed by the upturning of the rotor wakes as they move through

the stator blade row region.

Figures 5.5 and 5.6 show further details of the character of the

rotor-relative t1me-averaged rotor wake as I t moves through the stator

blade row region along the 10 and 50 percent axisymmetric stream surfaces,

respectively. Along the 50 percent stream surface, see Figure 5.6(A), the

rotor-relative time-averaged axial velocity component within the rotor

wake looks much like the typical gausslan model of rotor wakes which has

been proposed from previous Investigations of wakes In low-speed

compressors [32-41]. The tangential velocity component of the

time-averaged absolute total velocity, however, does not at a l l f i t a

gausslan model. At the 10 percent stream surface, see Figure 5.5, the

character of the rotor wake 1s much different than at the 50 percent

stream surface, and the gausslan model Is completely Inadequate to

describe this rotor wake. I t Is Important to note that this particular

rotor wake Is comprised of shed vortlcles from the rotor blade trailing

edge [87] (see Fig. 5.7), and thus the wake has a different character than

that which would occur from a simple mixing of two shear layers.

Therefore, I t Is apparent that a gausslan model cannot, 1n a l l Instances,

adequately model the characteristics of the time-averaged rotor wake

associated with a high-speed compressor.

The rotor-relative time-averaged picture of the measured turbulence

In the directions corresponding to the fringe angle orientations (I.e.,

±20° from the measured time-averaged absolute flow angle) are shown In

Figures 5.8 and 5.9 along the 10 and 50 percent stream surfaces,

respectively. From Figures 5.8 and 5.9, 1t Is also apparent that

 10% STREAM SURFACE
250
-LOM STATOR CHORD -11% STATOR CHORD 20% STATOR CHORD

230

AXIAL SURVEY STATION,

210 SEE FIGURE 1.3(A).

190

170 "•d
150
J .1 I I I I
^ 250
-82% STATOR CHORD 20% STATOR CHORD 65% STATOR CHORD
B
^ 230

0 210
S
J 190

W 170 PS ss
1 150 -J 1 I I I -! L I I I

S
2501- -62% STATOR CHORD 0% STATOR CHORD 100% STATOR CHORD

230 -

210

190
-•|I
I"- ....
170

150 J
20 40 60 80 100 20 10 60 80 20 10 60 80 100
PERCENT ROTOR PITCH

Figure 5.5 Rotor-relative time-averaged absolute velocities

along the 10 percent stream surface
 10% STREAM SURFACE
-60 - -60 -35
-LOTX STATOR CHORD -41% STATOR CHORD 20% STATOR CHORD
-85 -85 -60
AXIAL SURVEY STATION.
SEE FIGURE 4.3(A).
-110 -no -85

-135 -135 -no

-150 ' -160 -135

-185 ' -185 I . 1-

-160 .A-

-60 -60 50
Ifi^ -82% STATOR CHORD -20% STATOR CHORD 65% STATOR CHORD
-85 -85 25

i -110 -no 0

-135 . • -135 -25

-160
ss -160 00
-50

-185 ' -185 _L. 1 i- -75 I

I -60
-62% STATOR CHORD
-35
0% STATOR CHORD
50
100% STATOR CHORD

-85 -60 25

-110 -85 0

-135 ... -110 -25

ill'
-160 . -135 -50

-185 1 . I I ! ' -160 _J 1 _- - 1 1 J -75 _JL_

10 20 30 40 50 I 10 20 30 40 50 10 10 30 40 50
PERCENT ROTOR PITCH

(B) ABSOLUTE TANGENTIAL VELOCITY. V,

e-

Figure 5.5 Continued

I
: !
 50% STREAM SURFACE
210,-
-70% STATOR CHORD -18% STATOR CHORD 35% STATOR CHORD
190

170
r-"
150

130 AXIAL SURVEY STATION.

SEE FIGURE '1.3(B).

no
210
-52% STATOR CHORD OX STATOR CHORD 65% STATOR CHORD
190

170

150

130
CO
110 a>

210 -35% STATOR CHORD 20% STATOR CHORD 100% STATOR CHORD
190

170

150

130

110 _L. J
40 GO 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT ROTOR PITCH

(A) ABSOLUTE AXIAL VELOCITY. V^.

Figure 5.6 Rotor-relative time-averaged absolute velocities

along the 50 percent stream surface
 50% STREAM SURFACE
-90 -90 20
-70% STATOR CHORD -18% STATOR CHORD 35% STATOR CHORD
110 -110 0

130 AXIAL SURVEY STATION. -130 -20

SEE FIGURE 4.3(B).
-150 -150 -10

-170 -170 -50

-190 -J -190 -80

-90 -90 20
-52% STATOR CHORD 0% STATOR CHORD 65% STATOR CHORD

-110 -110 0

-130 -130 -20

-150 .. -150 • -10

'Ik..
-170 -170 -60

-190 -J -190 J -80

-90 r- -90 20% STATOR CHORD 20

-35% STATOR CHORD 100% STATOR CHORD

-110 -110 0

-130 -130 -20

-150 -150 -10

-170 -170 -60

-190, _L _L -J -190 J -80

10 MO 60 80 100 20 10 60 80 100 20 10 60
PERCENT ROTOR PITCH

(B) ABSOLUTE TANGENTIAL VELOCITY. V@.

Figure 5 . 6 Continued
 88

LASER ANEMOMETER
VELOCITY MEASUREMENTS
111111111M M IRI N 1! !! MI M 111!II'F!' :11 I: I: I ' I

A AND B DENOTE PARTICULAR ROTOR

SHAFT POSITIONS (R/Z,0 CONSTANT)
SEE FIGURE 4.6

VELOCITY

Figure 5.7 Illustration showing laser anemometer

measurements of the vortex street character of
the rotor wake
 10% STREJM SURFACE
25
-104% STATOR CHORD -11% STATOR CHORD 20% STATOR CHORD
20

15

10

5 AXIAL SURVEY STATION.

SEE FIGURE 1.3(A).
0 I I I I I i I I I I I I I I I I

25
-82% STATOR CHORD -20% STATOR CHORD 65% STATOR CHORD
20

15

10

5
00
0 I I I I I I 1 1 1 I I J I I I I kO

25
-62% STATOR CHORD 0% STATOR CHORD 100% STATOR CHORD
20

15

10

J I 1 I I L I I I I I
20 10 60 80 100 0 20 10 60 80 100 0 20 10 60 90 100
PERCENT ROTOR PITCH
(A) FIRST MEASURED COMPONENT.

Figure 5.8 Measured components of the rotor-relative

time-averaged turbulence along the 10 percent stream
surface
 10% STREAM SURFACE
25 r -LOTZ STATOR CHORD -11% STATOR CHORD 20% STATOR CHORD
20 -

15 -

10 -
AXIAL SURVEY STATION.
SEE FIGURE 1.3(A).

5-

J 1 ! I I J 1 I I 1 J
0 -

25r -82% STATOR CHORD -20% STATOR CHORD 65% STATOR CHORD
20 -

15 -

10 -

5-

OL J I I I

25 -
-62% STATOR CHORD 0% STATOR CHORD 100% STATOR CHORD
20 -

15 -

10 -

5 -

I _L _L _L J I L
20 40 60 80 100 0 20 10 60 80 100 20 10 60
PERCENT ROTOR PITCH
(B) SECOND MEASURED COMPONENT.

Figure 5 . 8 - Continued
 50% STREAM SURFACE
30 -
« -70% STATOR CHORD -18% STATOR CHORD 35% STATOR CHORD

25 -

20 -
AXIAL SURVEY STATION.
SEE FIGURE 4.3(8).
15 --

10 -

5 — J I _l L
30r -52% STATOR CHORD 0% STATOR CHORD 65% STATOR CHORD

25 -

20 -

15 -

10 -

5L- I I I 1 1 J 1 I I I I. I I I I

30R -35% STATOR CHORD 20% STATOR CHORD 85% STATOR CHORD

25 -

20 -

15 -

10
5 — I I I I I I I I I I J I I I
20 10 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT ROTOR PITCH
(A) FIRST MEASURED COHPONENT.

Figure 5.9 Measured components of the rotor-relative

time-averaged turbulence along the 50 percent stream
surface
 50% STREAM SURFACE
JOf - -70% STATOR CHORD -18% STATOR CHORD 35% STATOR CHORD

25 -

20 - AXIAL SURVEY STATION,

SEE FIGURE 4.3(B).
15 -

^ 10-

0 - J I I J 1 1 I I I 1 I I I I

-52% STATOR CHORT»

â 30r 0% STATOR CHORD 65% STATOR CHORD

i
É
P
g

à
OL

30r -35% STATOR CHORD

J 1 1 I I J I I L
(O
IN>

f 20% STATOR CHORD 85% STATOR CHORD

s
G 25-
œ
20

15

10
_L J_
20 40 60 80 100 20 40 60 80 100 20 40 60 80 100
PERCENT ROTOR PITCH
(B) SECOND MEASURED COMPONENT.

Figure 5.9 - Concluded

 93

turbulence Increases 1n the rotor wake. I t 1s obvious, therefore, that

the stator Is subject to both periodic fluctuations In absolute velocity

and turbulence as the rotor wakes pass through the downstream stator row.

Due to the vortex street character of the rotor wake, the magnitudes of

turbulence within the rotor wake are higher than might be expected.

3. Time resolved description of the measured flow field

As previously mentioned, the dominant blade-row Interactions present

In the current research fan are due to viscous Interactions caused by the

chopping of rotor wakes by the downstream stator blade row. Figures 5.10

and 5.11 show, along the 10 and 50 percent stream surfaces respectively, a

sequence of "snapshots" In time (I.e., rotor/stator orientations or rotor

shaft positions) of the measured flow field as the rotor wakes are chopped

and subsequently transported through the stator blade row. The shaded

regions overlayed on the contour maps of turbulence are provided to

Indicate rotor wake fluid (see Appendix 0). An Indication of the decay of

the rotor wakes can be Inferred from the contours of the lower bound of

turbulence. Similar contour maps depicting the movement of wakes through

adjacent blade rows have been reported by Bunker [42]; Ounker et al. [89]

for a compressor blade row, and Binder et al. [43]; Binder [44]; Binder et

al. [45] for a turbine blade row.

From Figures 5.10 and 5.11 we can see that after the stator blade

chops the rotor wake, the upper and lower stator-passage wake segments

begin to d r i f t apart as they convect through the stator blade row. Thus,

by the time the rotor wake segments reach the stator exit there Is an

appreciable mismatch between the rotor wake segments, which were

 A B C D C F G H I J K L M NO P 0 R A B C D E F G H I J K L M NO P 0 R
> I > I I T 1 II I I I I I I I I 1

/. . ' /.:v •
g 0

20 r- & 20
tiO § 40 TURBULENCE. CURVE
LABEL
60 S 60 * ^FS M/S
80 80
. i i- 2.2 5 (A)
100 S 100 4.4 10 (B)
6.5 15 (C)
SP = 1 SP - 30 8.7 20 (0)
10.9 25 (E)
13.0 50 (F)
15.2 35 (6)

0 0 / J
20 20
40 7 40
60 60
80 80
100 100

/ SP = 10 SP DENOTES ROTOR SHAFT

POSITION (THERE ARE 50
ROTOR SHAFT POSITIONS
PER ROTOR PITCH).
0 % 0
20 20
40 S 40
60 ° 60

80
100
I«
St 100
7
^ SP = 20 SP = 50
-loi) 50 100 150 -100 -50 50 100 150
PERCENT STAIOR CHORD

Figure 5.10 Contour maps depicting the time-resolved transport

and dissipation of rotor wakes through the stator
row along the 10 percent stream surface
 b H1 G HI
A B CD Ef jljjcI n N 0 P QR S 1 A B CD EF IIJK L N N 0 P OR S T
I I M t I 111 t • t i I I 1 I I I I I II T illLi I I I I I II I I

0 u 0
1
Q. 20 Z 20
40 0 no
1
S « 60
60 ? TURBULENCE. CURVE
80 1 80 LABEL
i 100 S 100 * *FS M/S

2.2 5 (A)
SP : JO 4.4 10 (a)
6.5 15 (c)
8.7 20 (0)
10.9 25 (E)
15.0 30 (F)
15.2 35 (C)
a. 20
A' .

vO

SP = 10 SP = qo

SP DENOTES ROTOR SHAFT

POSITION (THERE ARE 50
r- ROTOR SHAFT POSITIONS
PER ROTOR PITCH).

v
SP = 20 SP = 50

50 100 150 -100 -50 50 100 150

PERCENT STATOR CHORD

5.11 Contour maps depicting the t1me-resolved

transport and dissipation of rotor wakes through
the stator row along the 50 percent stream surface
 96

originally part of the same rotor wake. Using linear small-disturbance

theory. Smith [65] showed that for thin airfoils with relatively l i t t l e

turning the d r i f t distance can be related to the airfoil circulation, r ,

and the average of the Inlet and exit velocities, V^, according to the

following equation.

(5.1)

where

The d r i f t distance, AS , 1 S measured along the stator trailing edge

streamline. Table 5.2 shows a comparison between the calculated and

measured d r i f t distances along the 10 and 50 percent stream surfaces,

which have been normalized by the stator axial chord. The measured

nondlmenslonal dri ft distance was determined from Figures 5.10 and 5.11

based on the relative rotor/stator orientation corresponding to SP = 10

(20 percent rotor pitch). Joslyn et al. [64] used an Invlscid analysis to

successfully predict the motion of a narrow smoke stream (I.e., an

1nv1sc1d zero-veloc1ty-def1c1t "wake") through a downstream rotating blade

row by tracing Isochronous lines measured relative to the I n i t i a l upstream

generated smoke stream. For a cusped leading edge a i r f o i l , their approach

reduces to that of Smith [65].

I t Is also evident from Figure 5.10 and 5.11 that the rotor wakes

tend to pile up at the stator exit, and that there Is some spreading of
 97

Table 5.2 Comparison between calculated and measured rotor wake segment
nondlmenslonal d r i f t distance

Percent Stream surface

Parameter Symbol 10 50

Stator axial chord 5.6 cm 5.6 cm

Stator inlet

Radius Rl 23.2 cm 18.5 cm

Tangential velocity 135 m/s 150 m/s

%
I
Total velocity Vl 230 m/s 235 m/s

Stator exit

Radius RE 23.23 cm 18.68 cm

Tangential velocity 0 m/s 0 m/s

Total velocity VE 200 m/s 185 m/s

Drift distance

Measured 41.6 percent C 34,6 percent C

Calculated 44.7 percent C 38.4 percent C

 98

the rotor wake width, although very slight. The piling up of the rotor

wakes at the stator exit 1s a result of the static pressure field which

retards the absolute velocity, and thus slows down the transport of the

rotor wakes through the stator row. Therefore, In light of the above

observations. I t seems apparent that the kinematics of the transport of

rotor wakes through the downstream stator blade row Is largely controlled

by the time-averaged potential flow field. Of course, for this stage, the

rotor wake has already substantially decayed prior to entering the stator

blade row. In a more closely coupled stage most of the rotor wake mixing

would occur while the rotor wake was within the stator row. In addition.

Interactions between the respective potential flow fields would be

present. Therefore, the above picture of the time-resolved flow field

could be quite different In a more realistically coupled turbomachlne. In

which case a simple Invlscid analysis may not be adequate to predict the

rotor wake motion through the stator blade row.

A generally accepted theory, f i r s t proposed by Meyer [52], Is that

the rotor wake behaves like a negative jet. Therefore, when the wake

(negative jet) Is cut by the stator row the width of the wake will tend to

diminish on the suction surface, and Increase on the pressure surface.

Several Investigators have since shown experimental evidence In support of

this theory [10,28,31,46], Most notable was an experimental Investigation

by Kerrebrock and Mikolaczyck [10] which Indicated that the higher energy

fluid of the rotor wake Impinged and collected on the stator pressure

surface, as evidenced by a measured Increase In the stagnation temperature

on the pressure side of the stator wake. However, there Is no apparent

evidence In Figure 5.10 or 5.11 of the rotor wake width either diminishing
 99

on the stator suction surface, or Increasing on the pressure surface.

Most likely, the manner In which the data was collected and presented, and

the spatial resolution of the data have obscured this effect since, for

this loosely coupled stage. I t Is probably quite small. However, I t Is

apparent from Figures 5.10 and 5.11 that the rotor wakes are moving towards

the pressure surface, and away from the suction surface, as expected.

Depending on where the rotor wakes are relative to the stator, the

rotor wakes may be affected differently due to the Influence of the stator

, flow field. Figures 5.12 and 5.13 show the character of the rotor wake,

as measured at different points within the stator passage, along the 10

and 50 percent stream surfaces, respectively. Each picture Is in the

stator frame of reference, and Is measured at a particular radial, axial,

and tangential position. Therefore, the abscissa Is a function of

rotor/stator relatl.ve position ( I . e . , stator-relatlve.tlme measured In

rotor shaft positions). At a constant axial and radial position the rotor

wake velocity deficit Increases and decreases depending on the stator

pitchwise location, therefore. I t Is obvious from Figures 5.12 and 5.13

that the stator flow field has a significant Influence on the rotor wake.

Similar behavior was found for the turbulence variation across the rotor

wake, as shown In Figures 5.12(C), 5.12(D), 5.13(C), and 5.13(D). The

stators Influence on the rotor wake velocity deficit and turbulence will

be discussed further, In the following sections, with respect to the

velocity correlations (see Eq. (4.11)) Identified from the development of

the average passage equation system.

Similar blade-row Interaction effects have been observed by Matsuuchi

and Adachi [31] who measured the flow field inside a rotor passage of an
 10% STREAM SURFACE
(0% STATOR CHORD)
250 5% STATOR SAP 35% STATOR GAP 65% STATOR GAP

220

190 jUlAliili,,
160

130

100

250 P 15% STATOR GAP 15% STATOR GAP 75% STATOR GAP
i
220

190
PS ss
160

130

100 _L I I J J_ J_ J

250 25% STATOR GAP 55% STATOR GAP 85% STATOR GAP
220
*'XTTTX*T*XX*XXXX**T2*XXTRNI
•'•""'""c'zn^tzuziazziztza"''
190

160

130

100
20 40 60 80 100 0 20 40 60 80 100 20 40 60 80 100
PERCENT ROTOR PITCH
(A) AXIAL VELOCITY. V^.

Figure 5.12 Time-resolved character of rotor wake along the

10 percent stream surface
 10% STREAM SURFACE
(0% STATOR CHORD)
-50
5% STATOR GAP 35% STATOR GAP 65% STATOR GAP
-80

-110

-mo

-170

-200 J I I I 1 1 1 I ' L L I I I I I I

-50
15% STATOR GAP 15% STATOR GAP 75% STATOR GAP

-80

-110
liUllIir
-140
PS ss
-170

I I I I I I ! I 1 1 I I I T I
-200

-50 r-
25% STATOR GAP 55% STATOR GAP 85% STATOR GAP

-80

-no
It'" ^ "iiiii .T%T^
-110

-170

I I I ! I J L J I I I I
-200
0 20 10 60 80 100 0 20 10 60 80 100 0 20 10 60 80 100
PERCENT ROTOR PITCH
(B) TANGENTIAL VELOCITY.

Figure 5.12 Continued

 10% STREAM suRFAct
(0% STATOR CHORD)

5% STATOR GAP 35% STATOR GAP 65% STATOR GAP

J 1 i I J ! L

15% STATOR GAP 15% STATOR GAP 75% STATOR GAP

PS SS

! 1 J I I J I I I I J 1 L

25% STATOR GAP 55% STATOR GAP

85% STATOR GAP

_L _L J _L I I J I
10 60 80 100 0 20 10 60 80 100 0 20 10 60
PERCENT ROTOR PITCH

(c) FIRST MEASURED COMPONENT OF TURBULENCE, V V

Figure 5.12 Continued

 101 STREAM SURFACE
(OX STATOR CHORD)

51 STATOR GAP 35% STATOP GAP 65% STATOR GAP

J I I. . ! I .J ! L

15% STATOR GAP 45% STATCR GAP 75% STATOR GAP

PS ss

85% STATOR GAP

25% STATOR GAP 55% STATOR GAP

I I ! L I L I _L
20 40 60 80 100 0 20 40 60 80 100 0 20 40 60
PERCENT ROTOR PITCH

(D) SECOND MEASURED COMPONENT OF TURBULENCE.V V, .

Figure 5.12 Continued

 50% STREAM SURFACE
(5% STATOR CHORD)
250 5% STATOR GAP 45% STATOR GAP 75% STATOR GAP

220
'"«iiini»""
190

160
PS ss
130

100 J I I I I I I I _1_ J I J I I L

250 25% STATOR GAP 55% STATOR GAP 85% STATOR GAP

220 IIII'ITI,,..

190

160 "II,»,,'"""

130

100 J I I I J I 1 L J I I I

35% STATOR GAP 65% STATOR GAP 95% STATOR GAP

250

220
Jl
190

160

130

100 L J I J I I I I ' L L

0 20 10 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

PERCENT ROTOR PITCH
(A) AXIAL VELOCITY, V,.

Figure 5.13 T1tne-resolved character of rotor wake along the

50 percent stream surface
 Ï
TIME-RESOLVED ABSOLUTE TANGENTIAL VELOCITY, " Q. N/S

SOL
 50% STREAM SURFACE
(5% STATOR CHORD)
5X STATOR GAP 45% STATOR GAP 75% STATOR GAP
25

20
PS SS
15

10

0 J ! I I ! ' ! I ! L

25% STATOR GAP 55% STATOR GAP 85% STATOR GAP

25

20

15

10
O
0»
5

0 I I I

25 35% STATOR GAP 65% STATOR GAP 95% STATOR GAP

20

15

10

_! ! L J l_ J L _L
20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT ROTOR PITCH
•WV.'
.2
(c) FIRST MEASURED COMPONENT OF TURBULENCE.V

Figure 5.13 Continued

 50% STREAM SURFACE
(5X STATOR CHORD)
5% STATOR GAP 25, 15% STATOR GAP 25 75% STATOR GAP

20: PS ' SS 20

,si.- 15
!
10 I- 10
I

I I I
T 5

ol- -' 0 _L L

25% STATOR GAP 25 55% STATOR GAP

25 85% STATOR GAP

20 20

15 15

10 10

5 5

o'- J I L
—' 0

35% STATOR GAP 65% STATOR GAP 95% STATOR GAP

25 - 30

20 | - 25

20
"I
15

= !- 10
i
L_ —' 5
40 60 80 100 20 no 60 60 100 20 40 60
PERCENT ROTOR PITCH

(D) SECOND MEASURED COWONENT OF TURBULENCE

Figure 5.13 Continued

 108

axial flow fan using a rotating hot-wire anemometer. They observed that

the wakes from the upstream Inlet guide vanes became deeper and sometimes

wider as the hot-wire was circumferentlally traversed closer to the rotor

pressure surface. They attributed this behavior to the jet like character

of the Inlet guide vane wakes sucking fluid from the suction surface and

Impinging on the pressure surface.

B. Unsteady-Velocity Correlations

The average passage equation system was developed In order to provide

an Improved mathematical model for simulating the three-dimensional flow

field 1n multistage turbomachlnes which would be consistent with current

turbomachlne design systems. As a result, the average passage equation

system depicts a steady-state flow field 1n which, with respect to a given

blade row, the flow 1s spatially periodic from passage to passage. In

addition, the average passage equation system Includes spatial derivatives

of the aperiodic-, unsteady-deterministic-, and random-velocity

correlations which were Identified by Adamczyk as the major contributors

to the generation of nonaxisymmetric flows In multistage turbomachlnes.

In order to solve the average passage system of equations 1t Is necessary

to develop models to predict the behavior of these velocity correlations,

and 1n order that modeling efforts are focused on the most significant

terms, the relative Importance of each of these velocity correlations

should be assessed. The magnitudes and distributions of the unsteady-

deterministic- and random-velocity correlations which were determined from

LFA measurements of the unsteady velocity field w1th1n the stator row of

the research compressor are presented 1n this section. All velocity

 109

correlations presented herein are Independent of time. The total-

deterministic- and turbulent-velocity correlations are functions of R,Z,e,

while the Incident-velocity correlations are functions of R,Z only.

!• Deterministic-velocity correlations

Figures 5.14 and 5.15 show contour maps of the axial, tangential, and

cross components of the total-determlnlstlc-veloclty correlations (TDVC,

see Eq. 4.14) along the 10 and 50 percent axisymmetric stream surfaces,

respectively. As expected, see Figure 5.15(A) for example, the axial

component of the TDVC decreases rapidly downstream of the rotor trailing

edge as a result of the rapid decay of the rotor wake. However, Instead

of continuing to decrease, the axial component of the TDVC appears to

Increase again locally, near the stator leading edge region at about mid

passage. This same behavior also occurs for the tangential and cross

components of the TDVC, except less pronounced, and also along the

10 percent stream surface, see Figures 5.14 and 5.15.

Circumferential (blade-to-blade) line plots of the axial, tangential,

and cross components of the TDVC at various axial survey stations are shown

In Figures 5.16 and 5.17 for the 10 and 50 percent axisymmetric stream

surfaces, respectively. These line plots provide a more quantitative

Indication of the Increase In the TOVC which occurs near the stator

leading edge. The axisymmetric free stream velocity (V^^ = 235 m/s)

upstream of the stator potential field 1s used to assess the relative

Intensity of the total determ1n1st1c-veloc1ty fluctuations. Referring to

2 2
Figure 5.17, the axial component of the TDVC decreases from about 190 m /s
2 2 2
(6 percent Vp^) at -70 percent stator chord to about 80 m /s
 no

AXIAL 10% STREAM SURFACE

SURVEY
STATIONS A B C D E F G H I J K L M N O P
8.5
I I I II I I I I I I I I I I
200

8.0
150
CURVE VALUE. CURVE
(M/S)2 LABEL
100 r
7.5,-
0 (A)
20 (B)
50- 5.0- W (c)
60 (D)
80 (E)
0-
2.5- 100 (F)
120 (G)
-ROTOR TRAILING EDGE
-50'-
O*-
(A) DETERMINISTIC AXIAL VELOCITY CORRELATION,
8.5r

: 200-
8.0- CURVE VALUE. CURVE
s 150- LABEL
( M /S)^
0 (A)
7.5- 20 (B)
100-
10 (c)
60 (0)
50- 5.0- 80 (E)
100 (F)
120 (G)
2.5- 110 (H)
160 (1)
- ROTOR TRAILING EDGE 180 (J)
5 -50- 200 (K>
0 —
(B) DETERMINISTIC TANGENTIAL VELOCITY
njrj
CORRELATION, VgVg,
8.5-

200 p CURVE VALUE. CURVE

LABEL
(M/S)^
!
8.0- -20 (A)
150 j- 0 (B)
20 (c)
10 (D)
100!-L
7.5-
60 (E)
80 (F)
50 5.0- 100 (G)
120 (H)
110 (I)
160 (J)
2.5-
180 (K)
-ROTOR TRAILING EDGE 200 (L)
-50L
OL.t
5.0 7.5 10.0 12.5 15.0 17.5 20.0
CM

-100 -50 0 50 100 ISO

AXIAL DISTANCE, PERCENT STATOR CHORD

( c ) DETERMINISTIC VELOCITY CROSS CORRELATION. V'^V^.

Figure 5.14 Contour maps of the total-determ1n1st1c-veloc1ty

correlations along the 10 percent stream surface
 m

50% STREAM SURFACE

G, H I
SURVEY
STATIONS A CD E F JK L M N 0 P O R S
200r 8,0
CURVE VALUE. CURVE
(M/S)' LABEL
-40 (A)
0 (B>
w (C)
80 (0)
120 (E)
160 (F)
200 (G>
240 (H)
280 (I)
/-ROTOR TRAILING EDCE

(A) DETERMINISTIC AXIAL VELOCITY CORRELATION

II IIIII I I I I I I III I
2 200- 8.Op

:i 150-

-50- ROTOR TRAILING EDGE -ROTOR TRAILING EDGE

Vo L\L ICTO 12.5 15.0 17.5 20.0 5.0 7.5 10.0 12.5 15.0 17.5 20,0
CM
-100 -50 50 100 150 -100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD
(B) DETERMINISTIC TANGENTIAL VELOCITY CORRELATION. (c) DETERMINISTIC VELOCITY CROSS CORRELATION.

Figure 5.15 Contour maps of the total-determlnlstlc-veloclty

correlations along the 50 percent stream surface
 10% STREAM SURFACE
200 -104% STATOR CHORD -10% STATOR CHORD 20% STATOR CHORD
150

100

50

0 AXIAL SURVEY STATION, ^ 19 m iH M II '1 m rt

SEE FIGURE 1.3(A).
I I 1 I 1 I ' I I I J I I I
-50

200 -62% STATOR CHORD 0% STATOR CHORD 50% STATOR CHORD

150

100

50

0
. a o s O Q a s Q i iD a { F L Q A A O Q Q O ' a o a 13 a w is
I I I * I I ! I I J 1 I I I
-50

200 -29% STATOR CHORD 10% STATOR CHORD 85% STATOR CHORD

150 I

100

50

0
a a -o B s a o Sa a s s a Q a
J L _1 I J I I I I I I I I
-50
0 20 < 4 0 E 0 80 100 0 2 0 I | 0 6 0 8 0 100 0 2 0 4 0 6 0 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP

(A) AXIAL COMPONENT, V^.

Figure 5.16 Stator blade-to-blade distributions of the

total-determlnlstlc-veloclty correlations along
the 10 percent stream surface
 10% STREAM SURFACE

-LOW STATOR CHORD -10% STATOR CHORD 20% STATOR CHORD

250

230

210

190
AXIAL SURVEY STATION.
170 SEE FIGURE 4.3(A).
m o m
150
oi a I

-62% STATOR CHORD 0% STATOR CHORD

250 R 50% STATOR CHORD

230

210

190

170

a1a mIm mIm I g 1 I P lli. Bl P I JÏ

aIP B| B =1 a
150 flL

250r -29% STATOR CHORD 10% STATOR CHORD 85% STATOR CHORD

230

210

190 _ E

170
m o
™ B 1 w—m 1 ni—ffl I «—° 1 m I a PI J_ ?lm I I ° wl " It "iBB
150 20 40 GO 80 100 20 40 60 80
0 20 40 60 80 100 0 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP

(B) TANGENTIAL COMPONENT,

Figure 5.16 Continued

 10% STHEAU SURFACE
200p
-104% STATOR CHORD -10% STATOR CHORD 20% STATOR CHORD
150 J

100
AXIAL SURVEY STATION.
50 SEE FIGURE 4.3(A).

0 s. a o a ® o

-50 _1 I _J I
s
200 -62% STATOR CHORD 0% STATOR CHORD 50% STATOR CHORD
f
150

100

50

0- O Q Q a a a a o ^ e g BOA
'Sa
-50 J I

200
-29% STATOR CHORD 10% STATOR CHORD 85% STATOR CHORD
150

100

50

0 •A B Q Q E O o B a B B SE

-50
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT STATOR LEASING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP

(c) CROSS COMPONENT,

Figure 5.16 Continued

 50% STREAM SURFACE
-70% STATOR CHORD -51 STATOR CHORD 201 STATOR CHORD
250

200

150

100

50

0 J j ! L I L I 1J I I
250
-35X STATOR CHORD 01 STATOR CHORD 50% STATOR CHORD
200

150

100 * * * *
50

0 -1 I I J o °ia "1° ° m m I ™
I I
250
-181 STATOR CHORD SX STATOR CHORD 100% STATOR CHORD
200

150 AXIAL SURVEY STATION, *

SEE FIGURE 4.3(8).
100
C $
50

J I I I I g Im a I = l« »ia I
20 W 60 80 100 0 20 10 GO 80 100 0 20 10 60 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR SAP PÎRCEHT LOCAL STATOR GAP

(A) AXIAL COWOKEHT. V^.

Figure 5.17 Stator blade-to-blade distributions of the

total-determ1n1st1c-veloc1ty correlations along
the 50 percent stream surface
 50% STREAM SURFACE
-701 STATOR CHORD -5% STATOR CHORD 20% STATOR CHORD
250
?
200

150 AXIAL SURVEY STATION,

SEE FIGURE 4.3(8).

100

50
IS 0
m
0 o -L J I ra—glJ 5 I I*" IM " I Io I
250
-35% STATOR CHORD 0% STATOR CHORD 50% STATOR CHORD
200

150

100

50

0
a o im m rm a im o i" a I m ra 1° -LS B_L_ B " l " ° l g rnla ° l °
250
-18% STATOR CHORD 5% STATOR CHORD 100% STATOR CHORD
200

150

150

50

a Id B IB T, Im lO m I IN ° I J DLL 2 m I m H, I a ° i a n. i m i
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR SAP

(B) TANGENTIAL COMPONENT. VgVg.

Figure 5.17 Continued

 50% STREAM SURFACE
200 -70% STATOR CHORD -5% STATOR CHORD 20% STATOR CHORD

150 - L
100

m
50 m o
O IS
0 O Œt (5 ®
I , I I 1 J ! I 1 \ \ \ \
-50

200 -35% STATOR CHORD OX STATOR CHORD 50% STATOR CHORD

150

100

50 m m

0
O Q to O O Q O Q g 19
o
-50 1 I, I 1 J L J 1 J L

200 -18% STATOR CHORD 5% STATOR CHORD 100% STATOR CHORD

150 J

100
AXIAL SURVEY STATION.
50 SEE FIGURE 1.3(8).

0 0 0 0 m m B O B D ° s Q
(S O Q
-50 J L _L L J L
0 20 40 GO 80 100 0 20 10 60 80 100 0 20 10 60 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP

(c) CROSS COMPONENT. V^VG.

Figure 5.17 Continued

 118

2
(4 percent Vp^) at -18 percent of stator chord. From -18 percent stator

chord to the stator exit the axial component of the TOVC continues to
2 2 2
decrease to less than 25 m /s (2 percent Vp^) from about 50 percent

stator chord on, except In a region from about -18 to 10 percent of stator

chord where the axial component of the TDVC Increases near mid passage to
2 2 2
a maximum of about 150 m /s (5 percent V^g) at 5 percent stator chord.

Although considerably weaker, similar Increases In the TOVC occur along

the 10 percent stream surface and for the tangential and cross components

of the TOVC.

In Section IV.B.3 an analogy to gust theory was used to further

decompose the TDVC. In the following paragraphs this gust analogy will be

used to help explain the origins of the observed Increase In the TDVC. In

Equation (4.12), the parameter Is the rotor-relative time-averaged

absolute velocity (Appendix B) which provides a description of the flow

field from the perspective of an observer riding on the rotor blades, see

Figures 5.4 to 5.6. The rotor-relative time-averaged velocity Is used as

a model of the undisturbed rotor wake which would occur In the absence of

the stator blade row. The rotor-relative time-averaged velocity Is, of

course, Independent of stator circumferential position, and only varies In

time (rotor/stator relative position) and with axial distance. I f the

axisymmetric absolute velocity Is subtracted from the rotor-relative

time-averaged absolute velocity (Eq. 4.13), the result Is the deterministic

unsteadiness associated with the undisturbed rotor wake, and 1n the analogy

with gust theory represents the Incident gust.

Figures 5.18 and 5.19 show the axial distributions of the axial,

tangential, and cross components of both the TDVC and the deterministic
 119

200"

ISO-

100 - -
, T
10% STREAM SURFACE
• TOTAL
« INCIDENT
50 -

i r
:

-50' ^ :
(A) AXIAL COMPONENT.
250 - 200

200-- - 150

1501- 100

100- 50

50- ' 0 -
!
• • ® •
: ;•' - 2 -50 -
-120 -GO 0 60 120 180 -120 -60 0 GO 120
PERCENT STATOR CHORD

(B) TANGENTIAL COMPONENT. (C) CROSS COMPONENT,

Figure 5.18 Axial distributions of the axisymmetric

total-deterministic- and Incident-velocity
correlations along the 10 percent stream surface
 120

250 r

200 f

150

50% STREAM SURFACE

" TOTAL
100 r O INCIDENT

50 !-
a a•
I

0
(A) AXIAL COMPONENT.
250 R 200 f-

200: 150i-

150 100-

100 50

V.
M"
I § M # M
501 01-
m S
i
»# S 1 a S ï-2_ _FL M -501
-100 -50 0 50 100 150 -100 -50 0 50 100
PERCENT STATOR CHORD

(B) TANGENTIAL COMPONENT. (G) CROSS COMPONENT.

Figure 5.19 Axial distributions of the axlsymmetric

total-deterministic- and Incident-velocity
correlations along the 50 percent stream surface
 121

velocity correlations of the Incident gust (IDVC, see Eq. (4.14)) along the

10 and 50 percent axlsymmetrlc stream surfaces, respectively. The TDVC

and IDVC are about the same magnitude, which indicates that.the effects of

blade row interactions must be small. The greatest variation between the

TDVC and IDVC occurs within the stator row, see Figure 5.19(A) for example,

but by this time the rotor wake has substantially decayed. A model which

would predict the incident rotor wake decay could, therefore, be used to

estimate the TDVC, at least for the wide axial spacing between blade rows

in the present research compressor.

Also shown in Figures 5.18 and 5.19 are abrupt increases in the TDVC

and IDVC which occur at the interface between the interblade- and

intrablade-row surveys (which also coincides with the start of the stator

potential field, see Figs. 5.1 and 5.2). I t 1s unclear whether these

abrupt Increases in the TDVC and IDVC are real or due to a mismatch between

the Interblade- and intrablade-row surveys. A comparison between the

interblade- and intrablade-row measurements of the TDVC and IDVC at 0 and

10 percent stator axial chord was inconclusive as the 1nterblade-row survey

at those axial locations could not be completed. During the Interblade-row

survey, the rotor window frame blocked the collection of data near the

stator suction surface and significantly Increased the data acquisition

times for the remainder of the survey points at 0 and 10 percent stator

axial chord, which affected the quality of the measurements.

Again referring to Equation (4.12), the parameter represents

the deterministic unsteadiness resulting from blade-row interactions. In

other words, 4^^ represents the departure from which results from

the Influence of the stator blade row, and will be subsequently referred
 122

to as the scattered velocity. The total deterministic velocity

unsteadiness Is the sum of the unsteadiness associated with the Incident

and scattered velocities.

Figures 5.20 and 5.21 show contour maps of the axial, tangential, and

cross components of the scattered deterministic velocity correlations

(SDVC, see Eq. (4.14)) along the 10 and 50 percent axisymmetric stream

surfaces, respectively. Referring to Figure 5.21(A), the axial component

of the SDVC 1s predominately zero throughout the stator flow field, except

near the stator surfaces. Immediately downstream of the rotor trailing

edge, and near the stator leading edge at about mid passage ( I . e . , where

the axial component of the TDVC was also high). Similar behavior Is

observed for the tangential and cross components of the SDVC, along the

10 percent stream surface. The contour maps of Figures 5.20 and 5.21

Illustrate the effects of blade row Interactions which are a result of the

stators response to the Incident rotor wake.

Figure 5.21(A) also showed an Increase In the axial component of the

SDVC near the stator leading edge suction surface which was not apparent

In the corresponding contour map of the axial component of the TDVC

(Figure 5.15(A)). In fact, the axial component of the TDVC Is decreasing

towards zero at the stator leading edge suction surface, which Is In

direct contrast to the Increase In the axial component of the SDVC at the

same location. Referring to Equation (4.14), there Is an additional

component of the TDVC to be considered, which 1s associated with

correlations between the the Incident and scattered deterministic velocity

unsteadiness.
 123

AXIAL 10% STREAM SURFACE

SURVEY
STATIONS A B C DEFGHIJ KLM NOP 0 R
g ' ' ' I M I I I I I I I I I I I I

200

8.0
ISO
CURVE VALUE. CURVE
(M/S)^ LABEL
100 7.5 0 (A)
15 (B)
30 (C)
50 - 5.0
y 15 (D)
60 (E>
75 <F)
0
2.5

-50 ^ -ROTOR TRAILING EDGE

(A) AXIAL COMPONENT OF SCATTERED VELOCITY

CORRELATION, V^V^.

8.5 r I I I I M I I I I I I I I I I I I

Z 200

!..
S
8.0

7.5
CURVE VALUE.
(M/S)
0
CURVE
LABEL
(A)
S5 LOOR 15 (B)
5 30 (C)

I 50 - 5.0 15 (D)
60 (E>
75 (F)
90 (G)
2.5

ROTOR TRAILING EDGE

I -50
JÙ , ,
(B) TANGENTIAL COMPONENT OF SCATTERED VELOCITY
CORRELATION, V^VG.
I I I I I I I I I I I I I I I I I I
8.5
200

8.0 CURVE VALUE. CURVE

150
(M/s)' LABEL
0 (A)
7.5 15 (a)
too
30 (c)
45 (D)
50 5,0 60 (E)
75 (F)
90 (G)
0 105 (H)
2.5

ROTOR TRAILING EDGE

-50
JL
5.0 7.5 10.0 12.5 15.0 17.5 20.0
CM
-100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD

( c ) CROSS COMPONENT OF SCATTERED VELOCITY CORRELATION, V^V^.

Figure 5.20 Contour maps of the scattered-velocity

correlations along the 10 percent stream surface
 124

50* STREAM SURFACE

AXIAL G H I
SURVEY
STATIONS A B CD E FLFJKL
F M N 0 P QR S T
200 r 8.0
I I II I ill 11 I I I I I I I I I

150
CURVE V/AUE, CURVE
<M/S)' LABEL
100 0 (A)
15 (B)
50 30 (C>
M5 (0)
60 (E)
0
75 (F)

-50 ROTOR TRAILING EDCE

(A) FLXIAL COMPONENT OF SCATTERED VELOCITY CORRELATION.

V^.
:200r" 8.0 I I II I I III 11 I I I I III I I
i
G 150
CURVE VALUE. CURVE
7.5
(M/S)^ LABEL
0 (A)
100 15 (B)
£ 30 (C)
5 5.0 45 (D)
60 (E)
75 (F)
« 0
O 2.5 90 (6)
105 (H)
S -50 -ROTOR TRAILING EDGE

(B> TANGENTIAL COMPONENT OF SCATTERED VELOCITY CORRELA-

TION ÛÛ
230R *-0

CURVE VALUE. CURVE

150 (M/S)2 LABEL
7.5 -15 (A)
-30 (B)
100
-15 (C)
5.0 0 (0)
50 (E>
15
30 (F)
0 45 (G)
2.5
60 (H)
75 (I)
-50 /-ROTOR TRAILING EDGE
ll '
5.0 7.5 10.0 12.5 15.0 17.5 20.0
CM
-100 -50 0 50 100 150
AXIAL DISTANCE, PERCENT STATOR CHORD
(c) CROSS COMPONENT OF SCATTERED VELOCITY CORRELATION.

Figure 5.21 Contour maps of the scattered-velocity

correlations along the 50 percent stream surface
 125

Figures 5.22 and 5.23 show contour maps of the axial, tangential, and

cross components of the correlations between the Incident and scattered

unsteadiness, (ISVC, see Eq. (4.14)) along the 10 and 50 percent

axisymmetric stream surfaces, respectively. Referring to Figure 5.23, the

axial component of the ISVC Is Increasing towards greater positive values

near the stator leading edge mid passage region, where the TOVC and SOVC

are also Increasing. However, near the stator leading edge suction

surface the ISVC Is decreasing towards greater negative values where the

SDVC Is Increasing to greater positive values and the TDVC Is zero. Thus,

I t Is apparent that the deterministic velocity fluctuations associated

with the Incident gust are out of phase with the scattered velocity

fluctuations resulting from the stators response to the Incident gust.

When the Incident and scattered unsteadiness are out of phase they can

effectively cancel the TDVC.

The physics behind this Island of Increased unsteadiness Is s t i l l

unclear. Perhaps, the straining of the rotor wakes as they enter the

stator potential field results 1n the stretching of the vortlcles within

the rotor wake, which results In a greater wake deficit. This seems

unlikely, however, as there Is no evidence of straining of the rotor wakes

near the region where the TDVC Increases, see Figures 5.10 and 5.11. Of

course, the wake In actuality Is three-dimensional, therefore. I t Is

possible that the rotor wake Is being strained In the radial-tangential or

radial-axial plane.

A second possible scenario which deserves further Investigation, and

which may explain the Island of Increased deterministic unsteadiness, Is

that the fluctuations toward more positive Incident angles which occur
 126

AXIAL
SURVEY
tO% STREAM SUBFACE
STATIONS A B C DEFGHIJKLM NOP 0 R
g 5r ' ' ' I I I M I I I I I I I I I I
200

8.0-
1501-
CURVE VALUE. CURVE
I (M/S)' LABEL
1001- 7.5-
-30 (A)
-15 (B)
0 (C)
sot- 5.0 (D)
15
30 (E)

2.5-

-SOL /-ROTOFL TRAILING EDOE

0—1^
(A) AXIAL COMPONENT OF CORRELATION BETWEEN INCIDENT
AND SCATTERED UNSTEADINESS. VJV^.

S.Sr

200 r CURVE VALUE» CURVE

(H/S)^ LABEL
8.0;- -60 (A)
150- -45 (B)
-50 (C)

100 r
7.5- -15 (D>
0 (E)
15 (F)
^ SO,- 5.1 30 (G)
45 (H)
60 (I)
75 (J)
2.5-
90 <K)
105 (L)
-50L -ROTOR TRAILING EDGE
OU-U
(D) TANGENTIAL COMPONENT OF CORRELATION BETWEEN
INCIDENT AND SCATTERED UNSTEADINESS. VG'V^.

8.5-

200-

8.0-
150- CURVE VALUE« CURVE
(M/S)^
(A)
7 5- -JO
-15 (B)
0 (C)
15 (D)
50-- S.Of 30 (E)
45 (F)
60 (F)
2.5-
• ROTOR TRAILING EDGE
-50-
5.0 7.5 10.5 12.5 15.0 17.5 20.0
CM
-1M -50 0 50 Î0Ô î'so
AXIAL DISTANCE, PERCENT STATOR CHORD
(c) CROSS COMPONENT OF CORRELATION BETWEEN INCIDENT
AND SCATTERED UNSTEADINESS*

Figure 5.22 Contour maps of the cross correlations between

the Incident and scattered unsteady velocities
along the 10 percent stream surface
 127

50% STREAM SURFACE

AXIAL G I
SURVEY \
STATIONS A B CD EF JKL H N 0 P QR S T

200.. 8.0-

CURVE VALUE. CURVE

ISO- (M/S)^ LABEL
7.5-
-60 (A)
100- -15 (B)
-30 (c)
50.. 5.0- -15 (D)
0 (E)
15 (F)
0 -
30 (G)
2.5-
45 (H)
-50- 60 (1)
^ROTOR TRAILING EDGE

(A) AXIAL COMPONENT OF CORRELATION BETWEEN IN­

U
CIDENT AND SCATTERED UNSTEADINESS, ViV'
Z'Z'

200- 8.0-

150 R
7.5-

100;-

5.0-
50I-

0 -
2.5-

-50 - /• ROTOR TRAILING EDGE -ROTOR TRAILING EDGE

S.0 " 7*5 M.Y 12.5 15.0 12.5 20.0 5.0 '
7.5 foTô 12.5 15.0 17.5 20.0
CM
_L_
-100 -50 50 100 150 -100 -50 ~ "50 Too "iso
AXIAL DISTANCE, PERCENT STATOR CHORD

(8) TANGENTIAL COMPONENT OF CORRELATION BETWEEN IN­ ( c ) CROSS COMPONENT OF CORRELATION BETWEEN IN­
CIDENT AND SCATTERED UNSTEADINESS, V^VG. CIDENT AND SCATTERED UNSTEADINESS, V^V^,

Figure 5.23 Contour maps of the cross correlations between

the Incident and scattered unsteady velocities
along the 50 percent stream surface

•S TÙIj V*'-^
 128

during the passage of the rotor wakes causes an Intermittent separation of

the stator suction surface boundary layer or perhaps a shedding of a vortex

from the stator leading edge, which Is then convected towards the adjacent

stator-blade pressure surface as a result of the wake slip velocity.

Previous experimental Investigations [90-94] of Isolated airfoils In

oscillating flows have demonstrated the existence of significant secondary

vortical structures which appear near the airfoil leading edges at higher

angles of Incidence. I would expect, however, that this effect would

result In a "tongue" of Increased deterministic unsteadiness originating

from the stator leading edge and extending outward. In the streamwise

direction, toward the adjacent stator-blade pressure surface. Perhaps

though, another as yet unexplained competing effect suppresses or cancels

any Increases In the deterministic unsteadiness nearer to the stator

leading edge, which results In the measured Island of Increased

deterministic unsteadiness out away from the stator leading edge.

A third scenario which may explain the Island of Increased

deterministic unsteadiness Is related to Binder's et al. [45] hypothesis

to explain the measured Increase In turbulence that he observed near the

leading edge region of a turbine rotor blade, which was interacting with

wakes generated from an upstream stator row. Binder concluded that stator

secondary vortex cutting by the downstream rotor blades results in a

breakdown in the secondary vortex and a subsequent Increase in the

turbulence. I t seems likely that whatever phenomenon is occurring to

cause the Island of increased deterministic unsteadiness observed in the

stator row of the present research compressor is also responsible for the
 129

increased turbulence observed near the leading edge of Binder's turbine

rotor. However, just how a secondary vortex breakdown can result In an

Increase In deterministic unsteadiness, which Implies an Increase In the

rotor wake velocity deficit. Is not yet clear. Although Interactions with

the secondary vortex generated from the upstream blade row may be

responsible for the observed Increases In unsteadiness In both the present

research compressor and Binder's turbine, something more than a breakdown

of the secondary vortex must be occurring to cause an Increase in the

deterministic unsteadiness observed in the present research compressor.

However, Binder's hypothesis certainly deserves further scrutiny.

A final proposed explanation for the increase in the TDVC is that as

the rotor wake impinges on the stator surface a potential or acoustic

perturbation emanates from the stator surface, as a result of the stators

response to the incident rotor wake, and acts to destroy or Induce a

vortical flow near the region where the TDVC is high. For the present

research compressor i t may not be possible to fully explain the physics

behind the Increase in the TDVC, as the Interactions are quite small due

to the wide axial spacing between the rotor and stator rows. Further

planned experiments in a more closely coupled stage may help to better

explain this phenomenon.

2. Turbulent-velocity correlations

Figures 5.24 and 5.25 show circumferential (blade-to-blade)

distributions of the measured components of the turbulent-velocity

correlations at various axial survey stations along the 10 and 50 percent

axisymmetric stream surfaces, respectively. Since the fringe angle

 10% STREAM SURFACE

500 -104% STATOR CHORD -41% STATOR CHORD

20% STATOR CHORD

400 .
J
300
AXIAL SURVEY STATION,
200 SEE FIGURE 4.3(A).

100 .

0. i-J iI I J I I I I J 1 I L
500.
-82% STATOR CHORD -20% STATOR CHORD 65% STATOR CHORD
400 .
n
300 .

200

100

0 J 1 I I I -1 1 1 1J_ J 1 1 I I

500p -62% STATOR CHORD 0% STATOR CHORD 100% STATOR CHORD

400 -

300

200

100 -

J 1 I I I J 1 1 I -J 1 1 I I
20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP
(A) FIRST HEASURED COMPONENT.

Figure 5.24 Stator blade-to-blade distributions of the

turbulent-velocity correlations along the 10 percent
stream surface
 10% STREAM SURFACE
104% STATOR CHORD -11% STATOR CHORD 20% STATOR CHORD

t I
I t
! ' ;
Îi
^ 'f ^ Ii
—-1 J -I J_ _
82% STATOR CHORD -20% STATOR CHORD 65% STATOR CHORD

w
If 1 I i!_
I L J I L_ -J
-62% STATOR CHORD OX STATOR 100% STATOR CHORD

AXIAL SURVEY STATION.

SEE FIGURE 1 . 3 ( A ) .

f t ,l i
0 20 40 60 80 J L. -J J ! I L -L
0 20 itO 60 80 100 0 20 10 60 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP
(B) SECOND HEASURED COMPONENT.

F i g u r e 5 . 2 4 Continued
 50% STREAM SURFACE
500 -70% STATOR CHORD -18% STATOR CHORD 35% STATOR CHORD

400
• M
300
i i
200 0 ;
î> Il II
100 ; I
01 J I— ! —1 ! 1 1 l I I 1_
500 -52% STATOR CHORD 0% STATOR, CHORD 65% STATOR CHORD

400 I

300

200 Il "
100

0 -J J 1 1 I 1 I ! I I !,

500 -35% STATOR CHORD 20% STATOR CHORD 100% STATOR CHORD
AXIAL SURVEY STATION,
400 SEE FIGURE 4.3(B).
L

300 • -

200 I I' ' i ' à

J 11 '1 '
100 __
I L
20
lis
40 60
^^ 80
T
100
L i
0
i
20
h
40
i
60 80 100 0
J20 !40 1
60
1
80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP
(A) FIRST HEASURED COMPONENT.

Figure 5.25 Stator blade-to-blade distributions of the

turbulent-velocity correlations along the 50 percent
stream surface
 50% STREAM SURFACE
-701 STATOR CHORD -18X STATOR CHORD 35% STATOR CHORD

AXIAL SURVEY
STATION, SEE
.FIGURE 1.3(B).

?1 '

I. I J L J I i.,

-521 STATOR CHORD OX STATOR CHORD 65% STATOR CHORD

W
LO

_J L__ J I ! I L J L

-35X STATOR CHORD 20% STATOR CHORD 100% STATOR CHORD

) '

T f Î
I _L_ -J U_
0 20 10 60 80 100 0 20 1)0 60 80 100 0 20 10 BO 80 100
PERCENT STATOR LEADING EDGE GAP PERCENT LOCAL STATOR GAP PERCENT LOCAL STATOR GAP
(B) SECOND MEASURED COMPONENT.

F i g u r e 5 . 2 5 Continued
 134

orientation varies with the direction of the axisymmetric total-absolute

velocity, which varies with axial position, the components of the

turbulent- velocity correlations which are measured In the direction of

the fringe angle orientations may also vary, due to the varying fringe

angle orientations. Therefore, caution should be exercised when

attempting to draw conclusions from apparent axial gradients 1n the

measured turbulent-velocity correlations. However, no such caution Is

required In drawing conclusions from the blade-to-blade gradients of the

measured turbulent-velocity correlations as the fringe angle orientation

along which the turbulent-velocity correlations are measured 1s Invariant

with circumferential (blade-to-blade) position. The fringe angle

orientations of the first and second measured components, and a^,

respectively, are shown In Figure 12.1 of Appendix C.

Referring to Figure 5.25(A), the f i r s t measured component of the

turbulent-velocity correlation Is approximately constant at a level of

2 2 2
about 190 m /s (8 percent Vpg) from -70 to -18 percent stator chord.

The Intensities of the turbulent-velocity correlations are quoted as

percentages of the axisymmetric free-stream absolute velocity upstream of

the stator potential field (V^^ = 235 m/s), as was done for the

deterministic-velocity correlations. At about -18 percent stator chord

the f i r s t measured component of the turbulent-velocity correlation begins

2 2
to Increase, near midpassage, to a maximum of about 350 m /s
2 2 2
(15 percent Vp^) at -5 percent stator chord, and remains above 300 m /s
2
(13 percent Vpg) , near midpassage, until about 20 percent stator chord,

beyond which Its magnitude becomes obscured by the stator suction surface

boundary layer.
 135

Binder et al. [43]; Binder [44]; Binder et al. [45] observed a similar

Increase In turbulence within the stator wake of a single stage turbine as

I t was transported through the downstream rotor row. Binder found that

the turbulence within the stator wake Increased near the rotor leading

edge and attributed the Increased turbulence to cutting of secondary

vortlcles by the rotor which caused the vortlcies to breakdown, thus

generating higher turbulence levels. Although this seems plausible as an

explanation for the turbulence Increases found by Binder, I t does not seem

to suffice as an explanation of the observed Increases In both the

deterministic- and turbulent-velocity fluctuations observed In the present

research compressor. A breakdown of the secondary vortex would Imply

Increased mixing out of the rotor wake velocity deficit, which Is contrary

to the Increases 1n the rotor wake velocity deficit shown In Figures 5.12

and 5.13.

I t Is Interesting to note that there are two blade-to-blade regions

at which the first component of the measured turbulent-velocity correlation

2 2 2
Increases, that both regions peak at about 300 m /s (13 percent V^g) ,

and that both regions appear to move towards the stator suction surface

with Increasing axial distance. In reality the two regions of Increasing

turbulent-velocity correlation are circumferentially constant, and the

changing stator camber causes the stator suction surface to move

circumferentially towards the region of Increased turbulent-velocity

correlation. Eventually, both regions of Increased turbulent-velocity

correlation appear to mix out across the stator passage, except In the

stator wake where the magnitude of the turbulent-velocity correlation

remains high. Similar behavior occurs for both components of the

 136

turbulent-velocity correlation at both 10 and 50 percent span. Figures

5.12(c), 5.12(d), 5.13(c), and 5.13(d) Indicate that the turbulent

velocity fluctuations within the wake remain approximately constant

blade-to-blade, while the turbulent velocity fluctuations within the free

stream Increase and decrease blade to blade. Therefore, the blade-to-

blade Increase In the turbulent-velocity correlation occurs because of an

Increase In the free-stream turbulence, and Is not due to an Increase In

the wake turbulence.

3. Comparison between deterministic- and turbulent-velocity correlations

Figures 5.26 and 5.27 show a comparison between the axial

distributions of the axial and tangential components of the total- -

deterministic- and turbulent-velocity correlations along the 10 and

50 percent axisymmetric stream surfaces, respectively. The axial and

tangential components of the turbulent-velocity correlations are

represented as the minimum and maximum bounds that the values of the

turbulent-velocity correlations can attain, as the actual values could not

be determined from the measured data, see Appendix C. The axial and

tangential components of the total-deterministic- velocity correlations

are denoted by symbols, and Include uncertainty Intervals which denote the

95 percent confidence level. The main conclusion which can be drawn from

this plot Is that the turbulent-velocity correlations are significantly

greater than the total-determ1n1st1c-veloc1ty correlations. Only near the

rotor trailing edge and 1n some Instances near the stator leading edge are

the total-determ1n1st1c-veloc1ty correlations of the same order of

magnitude as the turbulent-velocity correlations. Therefore, through-flow

..
 137

10' 10X STREAM SURFACE

10- BOUNDS ON TURBULENT VELOCITY CORRELATION-;

/

1I
10'
ES ! I II-'

10'

! ' =' .'..

s LTOTAL DETERMINISTIC VELOCITY CORRELATION

i
WITH UNCERTAINTY INTERVAL

I a l l
_U
a 10"
(A) AXIAL COMPONENT.
IO''F

10^

10'

a" 0

10^
Î £

180
^-120 -60 0 60 120
PERCENT STATOR CHORD
(B) TANGENTIAL COHPOHEHT.

Figure 5.26 Comparison between the axisymmetric values of the

total-deterministic- and turbulent-velocity
correlations along the 10 percent stream surface
 138

10^
50% STREAM SURFACE

10Ï

r
10' SO
O =.
,3"

3 a m

s
1 L
ï _1_
i ,0', 0 - (A) AXIAL COMPONENT.
lO^c

Is
BOUNDS ON TURBULENT VELOCITY CORRÉLATION-^

10^ t r
T

10'

10'

1TOTAL DETERMINISTIC VELOCITY CORRELATION

WITH UNCERTAINTY INTERVAL

J_
10" 100 150
-100 -50 0 SO
PERCENT STATOR CHORD
(B) TANGENTIAL COMPONENT.

Figure 5,27 Comparison between the axisymmetric values of the

total-deterministic- and turbulent-velocity
correlations along the 50 percent stream surface
 139

mixing appears to be largely controlled by the turbulence and not the

deterministic unsteadiness, at least for the presently considered research

fan. As previously mentioned, however, the turbulence as defined herein

Includes any unsteadiness not correlated with the blade passing frequency,

which in the case of this research fan Includes the unsteadiness due to

vortex shedding from the rotor trailing edge. As a result, the large

values of the turbulent-velocity correlations are to a great extent due to

the increased turbulence within the rotor wake resulting from the vortex

shedding. Since this increased turbulence 1s "frozen" with the rotor wake

i t may also be possible to use a gust analogy to help model the turbulent-

velocity correlation, as was suggested for the total-deterministic-velocity

correlation.
 140

VI. SUMMARY AND CONCLUSIONS

The time-resolved unsteady velocity field along two stream surfaces

In the stator row of a single-stage transonic axial-flow fan was mapped

using a laser anemometer. The measured velocities were used to determine

the magnitudes and distributions of the deterministic- and turbulent-

velocity correlations identified from the average-passage equation system.

Since the present research compressor was designed for minimal acoustical

interactions, the rotor/stator axial spacing (85 percent of rotor chord)

is much greater than usual in a typical compressor. As a result, the rotor

wakes are considerably diminished by the time they reach the stator inlet,

reducing the possibility of significant blade row interactions. However,

the advantage of the wide axial spacing between the rotor and stator rows

is that i t allows for investigation of the rotor wake generated

interactions, independently of potential flow interactions. Based on the

measured data, the following conclusions can be made.

1. Rotor wakes retain their identity as they are chopped and

subsequently transported through the stator row.

2. The kinematics of the transport of rotor wakes through the

downstream stator row is largely controlled by the time-averaged potential

flow field.

3. Simple linear disturbance theory is adequate for predicting the

d r i f t distance between wake segments convecting along the stator pressure

and suction surfaces.

4. Due to the diffusing nature of the stator row the wakes tend to

pile up at the stator exit.

 141

5. Spreading of the rotor wake segments appears to be minimal.

6. Blade row Interactions can result 1n the amplification or

attenuation of both the deterministic and turbulent unsteadiness.

7. Regions of amplified unsteadiness may occur In Isolated regions

away from the stator surfaces.

8. I f the scattered unsteadiness Is out of phase with the Incident

unsteadiness, the total-deterministic unsteadiness may be negligible In

areas where the scattered and Incident unsteadiness are significant.

9. An analogy to gust theory appears to provide a useful means for

analyzing the behavior of rotor-wake-generated unsteadiness and should aid

the modeling of the deterministic unsteadiness.

10. As a result of the "frozen" character of the turbulence within

the rotor wakes, I t may be possible to use a similar gust analogy for

helping to model the turbulent unsteadiness.

11. The turbulent-velocity correlations are significantly greater

than the total-determ1n1st1c-veloc1ty correlations, which Indicates that

through-flow mixing Is largely controlled by the turbulence and not the

deterministic unsteadiness (at least for the presently considered research

fan).

Presently, no satisfactory explanation of the presence of the

Increased deterministic and turbulent unsteadiness near the stator

leading-edge plane 1s apparent. In a more closely coupled stage where the

blade row Interactions are expected to be much greater, this region of

Increased unsteadiness could be detrimental to the performance of the

stator (e.g.. Induce separation of the boundary layer) particularly I f the

Increased unsteadiness Is nearer the stator suction surface. Of

 142

course, depending on the state of the boundary layer, I t Is also possible

that this region of Increased unsteadiness could be beneficial to the

performance of the stator. I t Is Important, therefore, that further

Investigation of this region of Increased unsteadiness be continued.

An animated movie of the transport of the rotor wakes though the

downstream stator row was produced from the LFA measurements. A

representative sample of figures Illustrating the rotor wake transport

through the stator row for six of the 50 rotor shaft positions for which

measurements were acquired was shown In Figures 5.10 and 5.11. The power

of movie animation In revealing details of the rotor wake transport which

was not evident from viewing s t i l l figures (e.g.. Figs. 5.10 and 5.11),

was found to be Invaluable. The movie animation revealed a dramatic

distortion of the rotor wake occurring near the stator leading edge as I t

Is chopped by the stator. This rotor wake distortion was not obvious from

examination of the s t i l l figures.

 143

VII. RECOMMENDATIONS FOR FURTHER RESEARCH

The next phase of research will be to determine the magnitudes and

distributions of the unsteady-deterministic- and turbulent-velocity

correlations In a more closely coupled stage, wherein the rotor trailing

edge Interacts strongly with the stator potential field. In a closer

coupled stage the blade row Interaction effects will be stronger and rapid

mixing of the rotor wake will occur within the stator passage. A

comparison between the unsteady-velocity correlation presented In this

dissertation with the unsteady-velocity correlations of a more

realistically coupled stage may provide some Insight Into how blade row

spacing affects performance. Following the mapping of the close coupled

stage will be an experimental program to determine the unsteady-velocity

correlations within a transonic compressor which has an Inlet guide vane

row. The entire rotor and stator flow field will be mapped In order to

resolve the effects of the Inlet guide vane wakes and the downstream

stator row on the rotor flow field, especially the effects on the rotor

shock system. Another planned experimental effort Is the mapping of an

embedded stage of a multistage compressor to assess the Importance of the

aperiodic-velocity correlation, as well as the unsteady-deterministic- and

turbulent-velocity correlations.

Other recommended experimental research programs are, determining the

magnitudes and distributions of these unsteady-velocity correlations In

the vaned diffuser of a centrifugal compressor, and In a large scale

low-speed compressor or turbine where the boundary layers are thick enough

to allow measurements within. The spatial gradients of the unsteady-

 144

velocity correlations appear In the average passage equation system, and

since there are large velocity gradients across boundary layers I t Is

reasonable to expect that the effects of the unsteady-velocity

correlations might also be large within the boundary layers.

In conjunction with the experimental research programs should be an

on-going effort to develop Instrumentation which will allow the

determination of all components of the total apparent-stress tensor

(Eq. (4.10)) of the average passage equation system. This would require

the simultaneous time-resolved measurements of the axial, tangential, and

radial components of the Instantaneous total-absolute velocity, as well as

measurement of the time-resolved density. A laser anemometer or hot wire

capable of simultaneous measurement of at least two components of velocity

Is a necessity for measuring two-dimensional turbulence (see Appendix C).

To measure three-dimensional turbulence a three component laser anemometer

or hot wire Is required.

Also In conjunction with the experimental research programs should be

an effort at modeling the various components of the total apparent-stress

tensor. As models are developed, the average passage equation system

could be used for a parametric study of the effects of the aperiodic-,

unsteady-deterministic- and turbulent-stress components of the total

apparent-stress tensor on the performance of the turbomachlnes. The

results of a comprehensive experimental, modeling, and computational

research program should lead to a better understanding of the role of

unsteady flows on turbomachlnery performance.

 145

VIII. REFERENCES

1. Gal lus, H.E., Lambertz, J,, and Wallmann, Th. "Blade-Row Interaction

1n an Axlal-Flow Subsonic Compressor Stage." ASME Paper No. 79-GT-92,

March 1979.

2. Yurinskiy, V.T., and Shestachenko, I . Ya. "Losses In an Impulse

Turbine Cascade in an Unsteady Flow." Fluid Mechanics Soviet Research

3, No. 1 (January-February, 1974): 22-27.

3. Doorly, O.J., and Oldfield, M.L.G. "Simulation of Wake Passing in a

Stationary Turbine Rotor Cascade." AIAA Journal of Propulsion 1,

No. 4 (July-August, 1985): 316-318.

4. Gallus, H.E., Grollius, H., and Lambertz, J. "The Influence of Blade

Number Ratio and Blade Spacing on Axial-Flow Compressor Stator Blade

Dynamic Load and Stage Sound Pressure Level." ASME Paper No.

81-GT-165, March 1981.

5. Zierke, W.C., and Okiishi, T.H. "Measurement and Analysis of

Total-Pressure Unsteadiness Data from an Axial-Flow Compressor."

Transactions of the ASME, Journal of Engineering for Power 92A

(1982): 479-488.

6. Dean, R.C., Jr. "On the Necessity of Unsteady Flow in Fluid

Machines." Transactions of the ASME, Journal of Basic Engineering

(March, 1959): 24-28.

7. Adamczyk, J.J. "Model Equation for Simulating Flows in Multistage

Turbomachinery." ASME Paper No. 85-GT-226 and NASA TM-86869, November

1984.
 146

8. Kofskey, M.G., and Allen, H.W. "Smoke Study of Nozzle Secondary Flows

1n a Low-Speed Turbine." NACA TN-3260. 1954.

9. Smith, L.H., Jr. "Wake Dispersion In Turbomachlnes." Transactions of

the ASME, Journal of Basic Engineering 880 (1966): 688-690.

10. Kerrebrock, J.L., and Hikolajczak, A.A. "Intra-Stator Transport of

Rotor Wakes and I t ' s Effect on Compressor Performance." Transactions

of the ASME, Journal of Engineering for Power 92A (October, 1970):

359-368.

11. Hodson, H.P. "Boundary Layer and Loss Measurements on the Rotor of an

Ax1al-Flow Turbine." ASME Paper No. 83-GT-4, March 1983.

12. Hodson, H.P. "The Development Unsteady Boundary Layers on the Rotor

of an Axial-Flow Turbine." Viscous Effects In Turbomachlnes,

AGARO-CP-351, June 1983.

13. Tanaka, S. "Influence of a Periodical Fluctuation on a Profile Loss

of a Cascade (Part I , Determination of the Total Pressure Loss

Coefficient)." Bulletin of JSME 27. No. 226 (April 1984): 660-674.

14. Okllshi, T.H., Hathaway, M.D., and Hansen, J.L. "A Note on Blade Wake

Interaction Influence on Compressor Stator Row Aerodynamic

Performance." Transactions of the ASME, Journal of Engineering for

Gas Turbines and Power 107, No. 2 (April, 1985): 549-551.

15. Dong, Y., GalUmore, S.J., and Hodson, H.P. "Three-D1mens1onal Flows

and Loss Reduction 1n Axial Compressors." ASME Paper No. 86-GT-193,

June 1986.

16. Hansen, J.L., and Okllshi, T.H. "Axial-Flow Compressor Rotor-Stator

Interaction." ISU-ERI-Ames-86135, TCRL-31, December 1985.

 147

17. Walker, G.J. "The Unsteady Nature of Boundary Layer Transition on an

Axial-Flow Compressor Blade." ASME Paper No. 74-GT-135, March 1974.

18. Evans, R.L. "Boundary-Layer Development on an Ax1al-Flow Compressor

Stator Blade." Transactions of the ASME, Journal of Engineering for

Power 100 (April, 1978): 287-293.

19. Pf1el, H., and Herbst, R. "Transition Procedure of Instatlonary

Boundary Layers." ASME Paper NO. 79-GT-128, March 1979.

20. Pflel, H., Herbst, R., and Schroder, T. "Investigation of the

Laminar-Turbulent Transition of Boundary Layers Disturbed by Wakes."

Transactions of the ASME, Journal of Engineering for Power 105

(January, 1983): 130-137.

21. Fleeter, S., Jay, R.L., and Bennett, W.A. "Rotor Wake Generated

Unsteady Aerodynamic Response of a Compressor Stator." Transactions

of the ASME, Journal of Engineering for Power 100 (October, 1978):

664-675.

22. Franke, G.F., and Henderson, R.E. "Unsteady Stator Response to

Upstream Rotor Wakes." AIAA Journal of Aircraft 17 (July, 1980):

500-507.

23. Fleeter, S., Bennett, W.A., and Jay, R.L. "Time-Variant Aerodynamic

Measurements In a Research Compressor." Proceedings of ;yTe Dynamic

Flow Conference. 1978.

24. Adachi, T., and Murakami, Y. "Three-Dimensional Velocity Distribution

Between Stator Blades and Unsteady Force on a Blade Due to Passing

Wakes." Bulletin of the JSME 22. No. 170 (August, 1979): 1074-1082.
 148

25. Murakami, Y., Hirose, T., Adachi, T., and Kato, H. "Unsteady Force on

a Cambered Blade Passing Through an Oblique Wake." Bulletin of the

JSME 24, No. 195 (September, 1981): 1543-1549.

26. Capece, V.R., and Fleeter, S. "First Stage Stator Vane Unsteady

Aerodynamic Response In a Multi-Stage Compressor." Paper No.

AIAA-84-1209, presented at the AIAA/SAE/ASME 20th Joint Propulsion

Conference, June 11-13, 1984.

27. Schmidt, O.P., and Okllshi, T.H. "Multistage Axial-Flow

Turbomachlnery Wake Production, Transport, and Interaction." AIAA

Journal 15 (1977): 1138-1145.

28. Ng, W.F., and Epstein, A.M. "A Quasl-Three-Dlmenslonal Model for

Intra-Stator Transport of Rotor Wakes." International Symposium on

Three-Olmenslonal Flow Phenomena In Fluid Machinery, ASME Winter

Annual Meeting, November 1985.

29. Dring, R.P., Joslyn, H.D., Hardin, L.W., and Wagner, J.H. "Turbine

Rotor-Stator Interaction." ASME Paper No. 82-GT-3, 1982.

30. Joslyn, H.D., Dring, R.P., and Sharma, O.P. "Unsteady

Three-Dlmenslonal Turbine Aerodynamics." Transactions of the ASME,

Journal of Engineering for Power 105 (April, 1983): 322-331.

31. Matsuuchi, K., and Adachi, T. "Measurements of the Three-Dimensional

Unsteady Flow Inside a Rotor Blade Passage of an Ax1al-Flow Fan."

Tokyo International Gas Turbine Conference Proceedings 2 (October,

1983): 523-530.

32. Raj, R., and Lakshmlnarayana, B. "Characteristics of the Wake Behind

a Cascade of Airfoils." Journal of Fluid Mechanics 61, Part 4

(1973): 707-730.
 149

33. Raj, R., and Lakshmlnarayana, B. "Three Dimensional Characteristics

of Turbulent Wakes Behind Rotors of Axial Flow Turbomachlnery."

Transactions of the ASME, Journal of Engineering for Power 98A (April,

1976): 218-228.

34. Hirsch, Ch., and Kool, P. "Measurement of the Three-D1mens1onal Flow

Field Behind an Axial Compressor Stage." Transactions of the ASME,

Journal of Engineering for Power 99A, No. 2 (April, 1977): 168-179.

35. Kool, P., DeRuyck, J., and Hirsch, Ch. "The Three-Dlmensloal Flow and

Blade Wake In an Axial Plane Downstream of an Axial Compressor

Rotor." ASME Paper No. 78-GT-66, April 1978.

36. Reynolds, B., Lakshmlnarayana, B., and Ravlndranath, A.

"Characteristics of the Near Wake of a Compressor of a Fan Rotor

Blade." AIAA Journal 17. No. 9 (September, 1979): 959-967.

37. Ravlndranath, A., and Lakshmlnarayana, B. "Mean Velocity and Decay

Characteristics of the Near- and Far-Wake of a Compressor Rotor of

Moderate Loading." Transactions of the ASME, Journal of Engineering

for Power 102, No. 3 (July, 1980): 535-548.

38. Ravlndranath, A., and Lakshmlnarayana, B. "Structure and Decay

Characteristics of Turbulence 1n the Near- and Far-Wake of a

Moderately Loaded Compressor Rotor-Blade." ASME Paper No. 80-GT-95,

March 1980.

39. Ravlndranath, A., and Lakshmlnarayana, B. "Rotor Wake Mixing Effects

Downstream of a Compressor Rotor." ASME Paper No. 81-ST-98, March

1981.
 150

40. Or1ng, R.P., Joslyn, H.O., and Hardin, L.W. "An Investigation of

Axial Compressor Rotor Aerodynamics." Transactions of the ASME,

Journal of Engineering for Power 104. No. 1 (January, 1982): 84-96.

41. Inoue, M., and Kuroumaru, M. "Three-01mens1onal Structure and Decay

of Vortices Behind an Axial Flow Rotating Blade Row." Transactions of

the ASME, Journal of Engineering for Gas Turbines and Power 106 (July,

1984): 561-569.

42. Ounker, R.J. "Flow Measurements In the Stator Row of a Single-Stage

Transonic Axial-Flow Compressor with Controlled Diffusion Stator

Blades." Viscous Effects In Turbomachlnes, AGARD-CP-351, June 1983.

43. Binder, A., Forster, W., Kruse, H., and Rogge, H. "An Experimental

Investigation Into the Effect of Wakes on the Unsteady Turbine Rotor

Flow." Transactions of the ASME, Journal of Engineering for Gas

Turbines and Power 107 (1985): 458-466.

44. Binder, A. "Turbulence Production due to Secondary Vortex Cutting 1n

a Turbine Rotor." ASME Paper No. 85-GT-29, 1985.

45. Binder, A., Forster, W., Mach, K., and Rogge, H. "Unsteady Flow

Interaction Caused by Stator Secondary Vortices 1n a Turbine Rotor."

ASME Paper No. 86-GT-302, June 1986.

46. Hodson, H.P. "Measurements of the Wake-Generated Unsteadiness In the

Rotor Passages of Axial-Flow Turbines." ASME Paper No. 84-GT-189,

June 1984.

47. Wagner, J.H., Okllshi, T.H., and Holbrook, G.J. "Periodically

Unsteady Flow In an Imbedded Stage of a Multistage, Axial-Flow

Turbomachlne." ASME Paper No. 78-GT-6, April 1978.

 151

48. Tweedt, O.L., Hathaway, M.O., and Ok11sh1, T.H. "Multistage

Compressor Stator/Rotor Interaction." AIAA, Journal of Propulsion and

Power 1, No. 6 (November-December, 1985): 449-455.

49. Williams, M.C. "Laser Veloclmetry Study of Stator/Rotor Interactions

In a Multi-Stage Gas Turbine Compressor." Presented at the AGARD PEP

6th Symposium on Advanced Instrumentation for Aero Engine Components,

May 1986.

50. Kemp, N.H., and Sears, W.R. "Aerodynamic Interference Between Moving

Blade Rows." Journal of the Aeronautical Sciences 20 (1953): 585-598.

51. Kemp, N.H., and Sears, W.R. "The Unsteady Forces due to Viscous Wakes

In Turbomachlnes." Journal of. the Aeronautical Sciences 22 (1955):

478-483.

52. Meyer, R.X. "The Effects of Wakes on the Transient Pressure and
Velocity Distributions 1n Turbomachlnes." Transactions of the ASME 80

(1958): 1544-1552.

53. Lefcort, M.D. "An Investigation Into Unsteady Blade Forces In

Turbomachlnes." Transactions of the ASME, Journal of Engineering for

Power 87A (1965): 345-354.

54. Karmen, T. von, and Sears, W.R. "Airfoil Theory for Non-Uniform

Motion." Journal of the Aeronautical Sciences 5, No. 10 (1938):

379-390.

55. Horlock, J.H, "Fluctuating Lift Forces on Airfoils Moving Through

Traverse and Chordwise Gusts," Transactions of the ASME, Journal of

Basic Engineering 90 (1968): 494-500.

 152

56. Naumann, H., and Yeh, H. "Lift and Pressure Fluctuations of a

Cambered Airfoil Under Gusts and Application to Turbomachlnery." ASME

Paper No. 72-GT-30, 1972.

57. Goldstein, M.E., and Atassi, H. "A Complete Second-Order Theory for

the Unsteady Flow About an Airfoil Due to a Periodic Gust. Journal of

Fluid Mechanics 74, Part 4 (1979): 741-765.

58. Adamczyk, J.J. "The Passage of a Distorted Velocity Field Through a

Cascade of Airfoils." Unsteady Phenomena In Turbomachlnery,

AGARD-CP-177, September 1975.

59. Erdos, J.I., and Alzner, E. "Numerical Solution of Periodic Transonic

Flow Through a Fan Stage." AIAA Journal 15. No. 11 (November, 1977):

1559-1568.

60. Krammer, P. "Computation of Unsteady Blade Forces In Turbomachlnes by

Means of Potential Flow Theory and by Simulating Viscous Wakes." ASME

Paper No. 82-GT-198, 1982.

61. Hodson, H.P. "An Invlscid Blade-to-Blade Prediction of a

Wake-Generated Unsteady Flow." ASME Paper No. 84-GT-43, June 1984.

62. Denton, J.O. "An Improved Time Marching Method for Turbomachlnery

Flow Calculation." ASME Paper No. 82-GT-239, 1982.

63. Ral, M.M. "Navler-Stokes Simulation of Rotor-Stator Interaction Using

Patched and Overlaid Grids." Presented at the AIAA 7th Computational

Fluid Dynamics Conference, AIAA-CP-854 (July, 1985): 282-298.

64. Joslyn, H.D., Caspar, J.R., and Dr1ng, R.P. "Invlscid Modeling of

Turbomachlnery Wake Transport." Presented at the AIAA/SAE/ASME/ASEE

21st Joint Propulsion Conference, AIAA-85-1132, June 1985.

 153

65. Smith, L.H., Jr. "Secondary Flow In Axial-Flow Turbomachlnery."

Transactions of the ASME 77, No. 7 (Oct. 1955): 1065-1076.

66. Adklns, G.G., Jr., and Smith, L.H., Jr. "Spanwise Mixing In

Axial-Flow Turbomachlnes." ASME Paper No. 81-GT-57, March 1981.

67. Mitchell, N.A. "Non-Ax1symmetr1c Blade Row Interaction In Axial

Turbomachlnes." ASME Paper No. 80-GT-133, March 1980.

68. Sehra, A.K. "The Effect of Blade-to-Blade Flow Variations on the Mean

Flow-Field of a Transonic Compressor." Presented at the AIAA 12th

Fluid and Plasma Dynamics Conference, AIAA-79-1515, July 1979.

69. Adamczyk, J.J., Mulac, R.A., and Celestlna, M.L. "A Model for Closing

the Invlscid Form of the Average-Passage Equation System." ASME Paper

No. 86-GT-227 and NASA TM-87199, June 1986.

70. Celestlna, M.L., Mulac, J.J., and Adamczyk, J.J. "A Numerical

Simulation of the Invlscid Flow Through a Counter-Rotating

Propeller." NASA TM-87200, 1986.

71. Cunnan, W.S., Stevans, W., and Urasek, D.C. "Design and Performance

of a 427-Meter-Per-Second-Tlp-Speed Two-Stage Fan Having a 2.40

Pressure Ratio." NASA TP-1314, October 1978.

72. Urasek, D.C., Gorrell, W.T., and Cunnan, W.S. "Performance of

Two-Stage Fan Having Low-Aspect-Ratio, First-Stage Rotor Blading."

NASA TP-1493, August 1979.

73. Crouse, J.E., and Gorrell, W.T. "Computer Program for Aerodynamic and

Blading Design of Multistage Axial-Flow Compressor." NASA TP-1946,

1981.
 154

74. Powell, J.A., Strazlsar, A.J., and Seasholtz, R.G. "High-Speed Laser

Anemometer System for Intrarotor Flow Mapping In Turbomachlnery."

NASA TP-1663, February 1982.

75. Powell, J.A., Strazlsar, A.J., and Seasholtz, R.G. "Efficient Laser

Anemometer for Intra-Rotor Flow Mapping In Turbomachlnery."

Transactions of the ASME, Journal of Engineering for Power 103 (April,

1981): 424-429.

76. Stevenson, W.H., dos Santos, R., and Mettler, S.C. "Fringe Mode

Fluoresence Veloclmetry." Applications of Non-Intrusive

Instrumentation In Fluid Flow Research, AGARD-CP-193, 1976.

77. Strazlsar, A.J., and Powell, J.A. "Laser Anemometer Measurements In a

Transonic Axial Flow Compressor Rotor." Transactions of the ASME,

Journal of Engineering for Power 103, No. 2 (April, 1981): 430-437.

78. Bendat, J.S., and Plersol, A.G. Random Data: Analysis and

Measurement Procedures. New York: W1ley-Intersclence, 1971.

79. Strazlsar, A.J. "Investigation of Flow Phenomena 1n a Transonic Fan

Rotor Using Laser Anemometry." Transactions of the ASME, Journal of

Engineering for Gas Turbines and Power 107 (April, 1085): 427-435.

80. Goldstlen, M.E. Aeroacoustlcs. New York: McGraw-Hill, Inc., 1976.

81. Kline, S. J. "The Purpose of Uncertainty Analysis." Transactions of

the ASME, Journal of Fluids Engineering 107, No. 2 (June, 1985):

153-160.

82. Walpole, R.E., and Myers, R.H. Probability and Statistics for

Engineers and Scientists. 2nd edition. New York: MacMlllan

Publishing Co., Inc., 1978.

 155

83. Patrick, W.P., and Paterson, R.W. "Error Analysis for Benchmark Fluid
Dynamic Experiments, Part I: Error Analysis Methodology and the

Quantification of Laser Velocimeter Error Sources." United

Technologies Research Center, Report No. R85-151772, June 1985.

84. Orloff, K.L., and Snyder, P.K. "Laser Doppler Anemometer Measurements

Using Nonorthogonal Velocity Components: Error Estimates." Applied

Optics 21, No. 2 (January, 1982): 339-344.

85. Snyder, P.K., Orloff, K.L., and Reinath, M.S. "Reduction of

Flow-Measurement Uncertainties in Laser Velodmeters with

Nonorthogonal Channels." Presented at the AIAA 21st Aerospace

Sciences Meeting, AIAA-83-0051, January 1983.

86. Barnett, D.O., and Bentley, H.T. "Statistical Bias of Individual

Realization Laser Velodmeters." In Second International Workshop on

Laser Velocimetry, Vol. 1, pp. 428-444. H.O. Thompson and

W.H. Stevenson, eds. Purdue University, 1974.

87. Hathaway, M.D., Gertz, J., Epstein, A., and Strazisar, A.J. "Rotor

Wake Characteristics of a Transonic Axial Flow Fan." AIAA Journal 24,

No. 11 (November 1986): 1802-1810.

88. Seasholtz, R.G, "Laser Doppler Velocimeter System for Turbine Stator

Cascade Studies and Analysis of Statistical Bias Errors." NASA TN

D-5427, 1977.

89. Ounker, R., Rechter, H., Starken, H., and Weyer, H. "Redesign and

Performance Analysis of a Transonic Axial Compressor Stator and

Equivalent Plane Cascades with Subsonic Controlled Diffusion

Blades." ASME Paper No, 83-GT-208, 1983.

 156

90. Miller, J.A., Pucci, P.P. "Heat Transfer to an Airfoil In Oscillating

Flow." Transactions of the ASME, Journal of Engineering for Power 93A

(October, 1971): 461-468.

91. McAllster, K.W., Carr, L.W. "Water Tunnel Visualizations of Dynamic

Stall." Transactions of the ASME, Journal of Fluids Engineering 101,

No. 3 (September, 1979): 376-380.

92. Ericsson, L.W., and Reding, J.P. "Unsteady Flow Concepts for Dynamic

Stall Analysis," AIAA Journal of Aircraft 21, No. 8 (August, 1984):

601-606.

93. Walker, J.M., Helln, H.E., and Strickland, J.H. "An Experimental

Investigation of an Airfoil Undergoing Large-Amplitude Pitching

Motion." AIAA Journal 23. No. 8 (August, 1985): 1141-1142.

94. Francis, M.S., and Keesee, J.E. "Airfoil Dynamic Stall Performance

with Large-Amplitude Motions." AIAA Journal 23. No. 11 (November,

1985): 1653-1659.
 157

IX. ACKNOWLEDGMENTS

The author would like to express his appreciation to Professor

Theodore H. Ok11sh1 and Dr. Anthony J. Strazlsar for their continual

advice, encouragement, and patience throughout the course of this research

program. In addition, the author would like to express his appreciation

to Dr. John J. Adamczyk for his advice and patience throughout many hours

of helpful discussions of his development of the averaged passage equation

system, and his Ideas on modeling flows In multistage turbomachlnes. The

author Is also grateful to Mr. Kenneth L. Suder for his assistance In

acquiring the data, and for many thoughtful questions which helped to

expand the author's knowledge and understanding of many aspects of this

research program. Furthermore, the author Is grateful to the other members

of his Doctor of Philosophy degree graduate program of study committee for

their assistance, and for their patience and understanding throughout this

research program. The author also extends his thanks to the members of

Dennis M. Sender's Graphics and Exhibits Office, and Margaret C. Appleby's

Editorial Branch for their help In preparing this dissertation, especially

to Mary Eltel for being so conscientious In preparing the figures. The

author Is also grateful to his supervisor, Mr. Gilbert J. Weden, and to

the director of the U.S. Army Propulsion Directorate, Mr. John Acurlo, for

providing a work environment which encourages employees to pursue advanced

educational programs. Finally, the author wishes to acknowledge the

National Aeronautics and Space Administrations Lewis Research Center for

supporting this research program (NASA Grant NAG 3-356), and the U.S. Army

Propulsion Directorate for their support of this dissertation.

 158

X. APPENDIX A: NASA DESIGN CODE RESULTS

The output from the NASA design code [73] used to design the present

research compressor Is presented 1n the following tables. Table 10.1

lists all Input parameters and user defined correlations, In tabular form,

required for the design code analysis. Table 10.2 lists the aerodynamic

output (e.g., velocity triangle Information, blade element performance)

for eleven streamlines at each axial computational station. The NASA

design code gives the streamline radial positions as a percentage of the

flow path span measured from the hub end wall. Table 10.3 lists the stage

and overall mass-averaged aerodynamic performance parameters. Tables 10.4

and 10.5 list the manufacturing coordinates at 15 spanwise locations for

the rotor blade and 11 spanwise locations for the stator blade.

Figure 10.1 shows the coordinate system and nomenclature used for the

manufacturing coordinates of the rotor and stator blade sections.

Table 10.1 Design code input parameters

*** INPUT DATA FOR COMPRESSOR DESIGN PROGAM »**

SCALE FACTOR IS 1.0000

THE COMPRESSOn ROTATIONAL SPEED IS 16 012.8 RPtl. THE INLET FLOW RATE IS 7 J. 30 0 CLB/StC).

THE DESIRED COMPRESSOR PRESSURE RATIO IS 1.590 . THE MOLECULAR WEIGHT IS 28.97 .

CALCULATIONS WILL BE PERFORMED ON 11 STREAMLINES. THE COMPRESSOR HAS 2 BLADE ROWS.

CALCULATIONS WILL BE MADE AT THE BLADE EDGES AND AT 16 ANNULAR STATIONS.

THE SPECIFIC HEAT POLYNOMIAL IS IN THE FOLLOWING FORM

CP = 0.23747E 00 + 0 .21962E-04*T + -0.87791E-07*T**2 + 0 .13991E-09»iT**3 + -0.76056E-13*T**«i + 0 .150<i JE-X6«T**5

INPUT DISTRIBUTIONS BY STREAMLINE OR STREAMTUBE

INLET TOTAL INLET TOTAL INLET WHIRL STREAMTUBE STREAMTUBE

STREAMLIKE TEMPERATURE PRESSURE VELOCITY NO. FLOW FRACTION
NO. (DEO*. R.) (PSIA) (FT/SEC)

1 518.700 .125 000 1 0.1000

2 518.700 .670 000 2 0.2000
3 518.700 .700 000 3 0.3000
4 518.700 .700 000 4 0.4000
5 518.700 .700 000 5 0.5000
6 518v700 .700 000 6 0.6000
7 518.700 .700 000 7 0.7000
8 51C.700 .700 0.000 8 0.8000
9 518.700 .700 0.000 9 0.9000
10 518.700 .700 0.000 10 1.0000
11 518.700 .660 0.000
Table 10.1 Continued

INPUT DATA POINTS FOR TIP AND HUB CONTOURS

TIP AXIAL TIP HUB AXIAL HUB

COORDINATE RADIUS COORDINATE RADIUS
(INCHES) (INCHES) (INCHES) (INCHES)

-14.000 10.100 -14.000 3.750

-6.000 10.100 -12.790 3.750
-1.110 10.100 -12.800 3. 750
-1.100 10.100 -11.300 3.725
-0.200 10.080 -9.900 3.675
0.527 10.000 -7.800 3.600
1.100 9 .870 -5.700 3.525
2.700 9.650 -4.800 3.500
4 .000 9.600 -4.790 3.500
4.010 9.600 -2.000 .580
7.300 9.600 -1.250 .690
7.310 9.600 0 .000 .000
8.400 9.600 1.000 320
9.150 9.600 2.500 .630
9 .900 9.600 3.700 4.750
10.600 9.600 5.300 4.940
10.610 9.600 6.700 5.120
14.000 9.600 6.800 5.124
15 .450 9.600 7.600 5.140
16.459 9.600 8.700 5.140 O
17.000 9.600 9.500 5.140
18.000 9.600 9.900 5.140
18.683 9.623 10.700 5.140
20.000 , 10.149 11.300 5.140
22.000 11.605 13.590 5.140
13.600 5.140
15.700 5.140
15.709 5.140
17 .000 5.140
18.000 5.140
18.683 5. 14U
20.000 5.140
22.000 5.140

WARNING ONLY. AT INPUT POINT, 6, THE TIP CONTOUR DATA IS NOT VERY SMOOTH.

WARNING ONLY, AT INPUT POINT, 23, THE TIP CONTOUR DATA IS NOT VERY SMOOTH.
Table 10.1 Continued

THE INPUT PROFILE LOSS TABLES - OMEGA(BAR)«COS(BETA)/(2.0 *SIGnA)

** PROFILE LOSS TABLE NO. 1 **

PCT. PASS. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM. D-FACTOR LOSS1 PARAM. D-FACTOR LOSS PARAM.

0.00 0. 3000 0 . 0139 0.4000 0.0166 0.5000 0.0203 0.6000 0.0260 0.7000 0. 0 338
10 . 00 0.3000 0.0112 0.4000 0.0130 0.5000 0.0160 0.6000 0. 0202 0.7000 0.0263
20.00 0 . 3000 0. 0100 0.4000 0. 0113 0.5000 0.0132 0.6000 0.0163 0.7000 0.0210
iO .00 0.3000 0. DOBO 0. 4000 0. 0089 0 .5000 0.0103 0.6000 0.0130 0 .7000 0 . OUO
40 . 00 0 . 3000 0. 0080 0 . 400 0 0.0089 0. 5000 0.0103 0 . 6000 0.0130 0. 7000 0 . 0161.
DO. 00 0. 3000 0 . 0080 0 ..4000 0.0089 0. 5000 0.0103 0.6000 0. 0130 0. 7000 0.0165
60.00 0 ..3000 0 . 0080 0..4000 0.0089 0. 5000 0.0103 0..6000 0.0130 0 . 7000 0.0165
70. 00 0 . 3000 0. 0080 0..4000 0.0089 0..5000 0.0103 0..6000 0.0130 0..7000 0.0165
80.,00 0..3000 0.0090 0..4000 0.0103 0..5000 0.0122 0..6000 0. 0153 0.,7000 0.0200
90 ..00 0 .3000
. 0 . 0092 0..4000 0. 0110 0..5000 0. 0140 0..6000 0. 0182 0 ..7000 0. 0243
100..00 0..3000 0. 0104 0..4000 0. 0127 0..5000 0. 0168 0..6000 0. 0221 0..7000 0. 0296

** PROFILE LOSS TABLE NO. 2 **

PCT. PASS. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM. D-FACTOR LOSS PARAM

Q .00 0 .3000 0..0309 0.4000 0..0336 0 .5000 0..0373 0 .6000 0..0430 0 .7000 0 .,0508
10 .00 0 .3000 0..0272 0 .4000 0..0290 0 .5000 0..0320 0 .6000 0..0362 0.7000 0 . 0423
20 .00 0.3000 0 .0250 0 .4000 0 .0263 0 .5000 0..0282 0 .6000 0..0313 0 .7000 0.,0360
30 .00 0.3000 0 .0230 0 .4000 0 .0239 0 .5000 0..0253 0 .6000 0..0280 0 .7000 0.. 0310
40 .00 0 .3000 0 .0211 0 .4000 0 .0220 0 .5000 0 .0234
. 0 .6000 0 .0261 0 .7000 0 .0296
,
50 .00 0.3000 0 .0212 0 .4000 0 .0222 0 .5000 0..0236 0.6000 0 .0264 0 .7000 0 .. 0299
60 .00 0.3000 0 .0214 0 .4000 0 .0226 0 .5000 0 .0241 0 .6000 0 .0269 0.7000 0 .0306
,
70 .00 0.3000 0 .0218 0 .4000 0 .0231 0 .5000 0 .0248 0 .6000 0 .0278 0 .7000 0 .0317
80 .00 0 .3000 0.0233 0 .4000 0 .0248 0 .5000 0 .0270 0 .6000 0.0303 0 .7000 0 .0347
90.00 0 . 3000 0.0272 0 .4000 0 .0290 0.5000 0 .0320 0.6000 0 .0362 0 .7000 0 .0423
100.00 0.3000 0 .0294 0 .4000 0 .0317 0 .5000 0.0358 0 .6000 0 .0411 0.7000 0 . 0486
Table 10.1 Continued

*** PRINTOUT OF INPUT STATION DATA ***

** INPUT SET NO. 1 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
CINCHES) (INCHES)

-11.0000 -11.0000 0.0000 0.0000 0.0000

** INPUT SET NO. 2 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
IxNCHES) (INCHES)

-9 . 5200 -9.5200 0.0000 0.0000 0.0000

** INPUT SET NO. 3 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

-a.9400 -6.5600 0.0000 0.0000 0.0000

** INPUT SET NO. 4 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

-6.8100 -4.4200 0.0000 0.0020 0.0000

** INPUT SET NO. 5 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

-3.5000 -3.5000 0.0030 0.0030 0.0000

Table 10.1 Continued

*** PRINTOUT OF INPUT STATION DATA ***

. '"
** INPUT SET NO. 6 IS AN ANNULAR STATION **
TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(IIICHES) (INCHES)
-2.2000 -2.2000 0.0050 0.0050 0.0000

** INPUT SET NO. 7 IS AN EXTRA ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)
-1.8240 -1.8240 0.0059 0.0059 0.0000

** INPUT SET NO. 8 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION O>
(INCHES) (INCHES) W

-0.9400 -0.9400 0.0080 0.0080 0.0000

Table 10.1 Continued
"** PRINTOUT OF INPUT STATION DATA **»

** INPUT SET NO. 9 IS ROTOR NO. 1 **

K FOR THIS BLADE ROW THE INPUT OPTIONS ARE COORD. MERIDL PUNCH *
TIP C.G. AXIAL LOCATION HUB C.G. AXIAL LOCATION INLET TIP BLOCKAGE INLET HUB BLOCKAGE INLET MASS BLEED
(INCHES) (INCHES)
0.9410 0.9410 0.0100 0.0100 0.0000
LOSS SET USED BLADE TILT ANGLE OUTLET TIP BLOCKAGE OUTLET HUB BLOCKAGE OUTLET MASS BLEED
(DEGREES)
1 0.0000 0.0150 0.0130 0.0000
TIP D FACTOR LIMIT HUB FLOW ANGLE LIMIT TIP SOLIDITY NUMBER OF BLADES CUM ENERGY ADD FRACT
(DEGREES)
0.4600 -20.0000 I.3000 22 1.0000

* POLYNOMIAL COEFS. FOR RADIAL PROFILES OF A BLADE AERO. PARAMETER AND BASIC BLADE ELEMENT GEOMETRY PARAMETERS *
COEF. ROTOR OUTLET PRESSURE L.E. RADIUS/CHORD T.E. RADIUS/CHORD MAX. THICKNESS/CHORD CHORD/TIP CHORD
CONSTANT 0.0018 0.0018 0.0290
LINEAR 0.0000 0.0000 0.0000 0.0000 0.0000
QUADRATIC 0.0000 0.0090 0.0090 0.1680 0.0000
CUBIC 0.0000 -0.0060 -0.0060 -0.1128 0.0000
QUARTIC 0.0000
o*
QUINTIC 0.0000

* INPUT BLADE ELEMENT DEFINITION OPTIONS *

INCIDENCE DEVIATION TURNING RATE TRANSITION MAX THICKNESS CHOKE BLADE MATERIAL DENSITY
ANGLE ANGLE RATIO POINT POINT MARGIN LB/(IN)**3
TABLE (S.S.SEF.) TABLE TABLE TABLE TABLE (L.E.REF.) NONE 0.00000
* TABLE OF BLADE SECTION DESIGN VARIABLES INPUT *
(VARIABLES CONTROLLED BY OTHER OPTIONS WILL APPEAR AS ZEROS IN THE TABLE.)
SUCTION SURFACE
STREAMLINE INCIDENCE ANGLE DEVIATION ANGLE INLET/OUTLET TURNING TRANSITION/CHORD MAX. THICKNESS
NUMBER (DEGREES) (DEGREES) RATE RATIO LOCATION LOCATION/CHORD
1 0.4500 8.0000 0.0750 0.7000 0.6400
2 0.5100 6.8000 0.1800 0.6474 0.6300
3 0.4000 4.8000 0.4300 0.6042 0.6200
4 0.3700 4.5000 0.6600 0.5627 0.6100
5 0.3500 4.6000 0.7900 0.5193 0.6000
6 0.2600 5.7000 0.8300 0.4705 0.5800
7 0.2000 6.6300 0.8600 0.4180 0.5600
8 0.1700 7.5200 0.9600 0.3592 0.5400
9 0.0000 8.6400 0.9800 0.2862 0.5000
10 0.0000 10.3900 1.0000 0.2243 0.5000
11 0.0000 12.5200 1.0000 0.1629 0.5000
Table 10.1 Continued

**» PRINTOUT OF INPUT STATION DATA ***

** INPUT SET NO. 12 IS A GUIDE VANE OR STATOR **

* FOR THIS BLADE ROW THE INPUT OPTIONS ARE COORD, MERIDL PUNCH *
TIP C.6. AXIAL LOCATION HUB C.G. AXIAL LOCATION INLET TIP BLOCKAGE INLET HUB BLOCKAGE INLET MASS BLEED
(INCHES) (INCHES)
5.2000 5.2000 0.0170 0.0170 0.0000
LOSS SET USED BLADE TILT ANGLE OUTLET TIP BLOCKAGE OUTLET HUB BLOCKAGE OUTLET MASS BLEED
(DEGREES)
2 0.0000 0.0200 0.0200 0.0000
HUB D FACTOR LIMIT INLET HUB MACH LIMIT TIP SOLIDITY NUMBER OF BLADES
0.7000 1.0000 1.2800 34

* POLYNOMIAL COEFS. FOR RADIAL PROFILES OF A BLADE AERO. PARAMETER AND BASIC BLADE ELEMENT GEOMETRY PARAMETERS »
COEF. STATOR OUTLET V(0) L.E. RADIUS/CHORD T.E. RADIUS/CHORD MAX. THICKNESS/CHORD CHORD/TIP CHORD
INV.SQ. 0.00
INVERSE 0.00
CONSTANT 0.00 0.0130 0.0130 0.0800
LINEAR 0.00 -0.0080 -0.0080 -0.0200 0.0000
QUADRATIC 0.00 0.0000 0.0000 0.0000 0.0000
CUBIC 0.0000 0.0000 0.0000 0.0000

* INPUT BLADE ELEMENT DEFINITION OPTIONS *

INCIDENCE DEVIATION TURNING RATE TRANSITION MAX. THICKNESS CHOKE
ANGLE ANGLE RATIO POINT POINT MARGIN
TABLE (S.S.REF.) TABLE TABLE S.S. SHOCK TABLE (L.E.REF.) NONE
* TABLE OF BLADE SECTION DESIGN VARIABLES INPUT «
(VARIABLES CONTROLLED BY OTHER OPTIONS WILL APPEAR AS ZEROS IN THE TABLE.)
SUCTION SURFACE
STREAMLINE INCIDENCE ANGLE DEVIATION ANGLE INLET/OUTLET TURNING TRANSITION/CHORD MAX. THICKNESS
NUMBER (DEGREES) (DEGREES) RATE RATIO LOCATION LOCATION/CHORD
1 -3.0000 16.2000 1.0000 0.0000 0.5000
2 -3.0000 12.3000 1.0000 0.0000 0.5000
3 -3.0000 10.5000 1.0000 0.0000 0.5000
4 -3.0000 9.7000 1.0000 0.0000 0.5000
5 -3.0000 9.1000 1.0000 0.0000 0.5000
6 -3.0000 8.8000 1.0000 0.0000 0.5000
7 -3.0000 8.6000 1.0000 0.0000 0.5000
8 -3.0000 8.8000 1.0000 0.0000 0.5000
9 -3.0000 9.0000 1.0000 0.0000 0.5000
10 -3.0000 10.3000 1.0000 0.0000 0.5000
11 -3.0000 14.2000 1.0000 0.0000 0.5000
Table 10.1 Continued

*** PRINTOUT OF INPUT STATION DATA ***

** INPUT SET NO. 10 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

3.0000 3.0000 0.0150 0.0150 0.0000

** INPUT SET NO. 11 IS AN EXTRA ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

3.6850 3.4850 0.0150 0.0150 0.0000

Table 10.1 Continued

*** PRINTOUT OF INPUT STATION DATA ***

** INPUT SET NO. 13 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

7.3400 7.3400 0.0200 0.0200 0.0000

** INPUT SET HO. 14 IS AN EXTRA ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

7.4460 7.4460 0.0200 0.0200 0.0000

** INPUT SET NO. 15 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

11.0100 11.0100 0.0200 0.0200 0.0000

** INPUT SET NO. 16 IS AN EXTRA ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

11.3500 11 .3500 0.0200 0.0200 0.0000

** INPUT SET NO. 17 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)

14.4400 14.4400 0.0200 0.0200 0.0000

Table 10.1 Continued

*** PRINTOUT OF INPUT STATION DATA ***

** INPUT SET NO. 18 IS AN ANNULAR STATION **

TIP AXIAL LOCATION HUB AXIAL LOCATION TIP BLOCKAGE FACTOR HUB BLOCKAGE FACTOR MASS BLEED FRACTION
(INCHES) (INCHES)
16.0000 16.0000 0.0200 0.0200 0.0000

I IFT ZdFT.JH) AR I Z(I.J) AR

1 1 -11.0000 0.0000 1 -11.0000 0.0000
2 2 -9.5200 4.3508 2 -9.5200 4.3508
3 3 -8.0203 4.3628 3 -8.0203 4.3628
4 4 -5.8920 3.0886 4 -5.8920 3.0886
5 5 -3.5000 2.7375 5 -3.5000 2.7375
6 6 -2.2000 4.9657 6 -2.2000 4.9657
7 8 -0.9400 4.9348 7 -1.8240 17.0221
8 10 2.2106 1.5723 8 -0.9400 7.0337
9 11 3.0000 6.0503 9 -0.3071 9.5886
10 13 4.3107 3.5121 10 2.2106 1.9675
11 14 6.2673 2.2070 11 3.0000 6.0503
12 15 7.3400 3.9783 12 3.4850 9.7171
13 17 11.0100 1.1621 13 4.3107 5.5750
14 19 14.4400 1.2434 14 6.2673 2.2070
15 20 16.0000 2.7339 15 7.3400 3.9783
16 7.4460 40.2481
17 11.0100 1.1967
18 11.3500 12.5439
19 14.4400 1.3802
20 16.0000 2.7339

FACTl = 2.5392 FACT2 = 6.1569

Table 10.2 Design code predictions of aerodynamic performance

*» VALUES OF PARAMETERS ON sTREAMLIHES AT STATIOH, 1, WHICH IS AN AHHULUS **

STREAMLINE A/ lAL AXIAL PIERD. TANG. ADS ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
KO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CUP.V. PRESS. TEMP. PRESS. TEMP.
CIH. ) (IH. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEC) (DEG) CI./IN.) CPSIA) CDEG.R.) CPSIA) CDEG.R.)
TIP 10. 100 -11.000
1 10. 100 -11. COO 506.80 506.80 0.00 506.80 0.1635 0. 00 0.00 -0 000 11 125 518.70 12.193 197.33
•>. 018 -11. 000 563.66 563.66 0. 00 563.66 0 5181 0.00 -0.05 -0. 001 11.670 518. 70 12.215 192.22
J 9. UO -11. UQO 57'..17 571.17 0 00 571. 17 0 5283 0 00 -0.10 -0 001 11 700 518.70 12.153 191.22
4 8.668 -11 000 572.82 572.82 0 00 572.82 0 5270 0.00 -0.17 -0.001 11.700 518.70 12.165 191.35
5 8. 147 -11.opc 573.64 573.65 0.00 573.65 0. 5278 0.00 -0.26 -0.001 11.700 518.70 12.158 491.27
6 7.590 -11. 000 573.87 573.88 0.00 573.88 0.5280 0. 00 -0.38 -0.001 11.700 518.70 12.156 191.25
7 6 989 -11. 000 571.29 571.32 0.00 571. 32 0 5261 0 00 -0.52 -0 001 11.700 518.70 12.152 191.21
8 6. 331 -11.000 571.62 571.67 0. 00 571.67 0. 5288 0. 00 -0.72 -0. 001 11.700 518.70 12.119 191.17
9 5 597 -11 oca 575.05 575.11 0 00 575.11 0 5292 0 00 -0.99 -0 001 14 700 518.70 12.115 491.13
10 1 752 -11.000 571.87 575.01 0. 00 575.01 0. 5292 0. 00 -1.39 -0. 001 11.700 518.70 12.116 191.11
11 Î 715 -11. 000 571.25 571.62 0 00 571.62 0 5258 0 00 -2.08 -0 001 11 660 518.70 12.111 191.16
HUB 3.715 -11. 000

«* VALUES OF PARAMETERS ON STREAMLINES AT STATION, 2. WHICH IS AH ANHULUS **

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC , STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CUP.V. PRESS. TEMP. PRESS. TEMP.
(IK. ) (IN.) (FT/SEC) (FT/SEC] (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IN.) CPSIA) CDEG .R.) CPSIA) CDEG.R.)
TIP 10 100 -9 520
1 10 100 -9 520 501.23 501.23 0 00 501 23 0.1610 0.00 -0.00 -0.000 11.125 518.70 12.208 497.50
2 9 626 -9 520 561.31 561.31 0 .00 561 31 0.5159 0.00 -0.06 -0.000 11.670 518.70 12.234 192.41
3 9 157 -9 520 571.72 571.72 0 .00 571 72 0.5259 0.00 -0.11 -0.000 11.700 518.70 12.171 191.46
8.663 -9 .520 570.11 570.11 0.00 57 0 11 0.5211 0.00 -0.17 -0.000 11.700 518.70 12.187 191.61
5 8.139 -9 .520 570.61 570.61 0 .00 570 61 0.5219 0.00 -0.25 -0.000 11.700 518.70 12.183 191.55
6 7 .579 -9 .520 570.52 570.53 0.00 570 .53 0.5218 0.00 -0.36 -0.000 14.700 518.70 12.183 191.57
7 6 971 -9.520 570.57 570.59 0.00 570 59 0.5218 0.00 -0.50 -0.000 11 .700 518.70 12.183 191.56
S 6 .312 -9 .520 570.19 570.53 0.00 570 .53 0.5218 0.00 -0.69 0.000 11 .700 518.70 12.183 191.57
9 5.571 -9 .520 570.19 570.57 0.00 570 .57 0 .5218 0.00 -0.96 0.000 11.700 518.70 12.183 191.56
10 1 .715 -9 .520 569.86 570.03 0 .00 570.03 0.5213 0.00 -1.39 0.000 11.700 518.70 12.188 191.62
11 3 .661 -9.520 566.12 566.52 0.00 566 .52 0.5209 0.00 -2.16 -0.001 11.660 518.70 12.183 191.95
HUB 3 .661 -9 .520
Table 10.2 Continued

*» VALUES OF PARAMETERS OH STREAMLINES AT STATION, 3, WHICH IS AH AHNUiUS **

STREAMLINE AXIAL AXIAL HERD. TANG. ABS ABS, ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
KO. RADIUS COORD. VEL. VEL, VEL. VEL. • MACH HO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN. ) (IN. ) (FT/SECJ (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 10..100 -6.94 0
1 ID . 100 940 503. 17 503. 17 0.00 503. 17 0 . 4600 0. 00 0. 00 0..000 14. 125 518.70 12. 216 497.59
2 9. -8..767 560.16 560. 16 0.00 560.16 0 ..5147 0. 00 -0. 05 0..000 14..670 518. 70 12. 24 3 4 92.50
i 9.. 11/S -fi. 596 5/0. 28 570.28 0.00 570.28 0 . 5245 0..00 -0. 09 0..000 14 . 700 518. 70 12. 186 491.59
«1 8 .660
. -5 .416
, 568. 31 568. 31 0.00 568. 31 0 .5226
. 0 ..00 -0..14 0..000 14..700 513. 70 12..202 491.78
5 5..134 -Ô..225 568. 34 568. 34 0. 00 568. 34 0 .5227
, 0 ..00 -0..20 0..001 14..700 518,.70 12..201 491.78
6 7 ..571 -a,.020 567, 53 567 .54 0.00 567. 54 0 .5219 0 .00 -0,.27 0..001 14 .700 518..70 12 .208 491.85
7 6 .. 961 -7 .798 566 .60 566.61 0. 00 566.61 0 .5210 0,.00 -0,.36 0..002 14..700 518.,70 12..216 491.94
S 6 .291 -7 .554 565..08 565. 10 0 ..00 565.10 0 .5195 0 .00 -0 .48 0 .003 14 .700 518..70 12 .228 492.08
) 5.538 -7 .261 562. 94 562. 97 0. 00 562.97 0 .5174 0 .00 -0 .66 0 .004 1 4 .700
, 518..70 12 .245 492.28
10 4 .660 -6 .961 558..86 558..94 0..00 558. 94 0.5135 0 .00 -0 .95 0 .006 1 4 .700 518..70 12.278 492.66
il 3 .557 -6 .560 548. 52 548. 73 0.,00 548.73 0 .5037 0 .00 -1 .57 0 .008 14 .660 518..70 12.327 493.60
HUB 5 .557 -6 .560

VALUES OF PARAMETERS ON STREAMLINES AT STATION, 4, WHICH IS AN ANNUIUS »*

STÎlEr.MLIHE AXIAL AXIAL MF.RD. TAHG. ABS. ABS. ABS.r LOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NC. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IX. ) (IN.) (FT/EEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG. R.
TIP 10 .100 -6.810
1 10. 100 -6.810 500.72 500.72 0.00 500.72 0 . 4577 0.00 -0, 02 -0 . 001 14. 125 518.70 12.233 497.80
2 9 .623 -6.637 558.23 558. 23 0.00 558.23 0. 5129 0.00 -0.,06 -0..000 14. 670 518.70 12.259 492. 73
5 9. 152 -6.467 568.63 568.63 0.00 568.63 0.,5229 0. 00 -0 ..06 0..001 14.,700 518.70 12.199 491 . 75
4 8.656 -6.287 566 .72 566 .72 0.00 566.72 0 .,5211 0. 00 -0..04 0.,002 14.,700 518.70 12.215 491. 93
5 8. 129 -6.096 566.62 566 . 62 0.00 566.62 0..5210 0. 00 -0..01 0..003 14..700 513.70 12.216 491 ., 94
6 7.,565 -5.892 565.43 565.43 0. 00 565.43 0..5198 0. 00 0..03 0..004 14..700 518.70 12.225 492. 05
7 6 . 95 -5.671 563.80 563.,80 0.,00 563.80 0,.5182 0, 00 0 .08 0 .005 14,.700 518.70 12.239 492,.20
8 6 .,233 -5.428 561.14 561., 14 0. 00 561.14 0 .5157
, 0.,00 0 .15
, 0,.008 14..700 518.70 12.260 492.,45
9 5 .527 -5.154 557.03 557.,03 0.,00 557.03 0 .5117 0. 00 0 .22 0 .010 14,.700 518.70 12.294 492.,84
10 4 .,644 -4.835 549.46 549.,46 0.,00 549.46 0,.5044 0. 00 0,.27 0,.014 14,.700 518.70 12.355 493.. 54
n 3 .526 -4 .430 532.69 532.,69 0 .,00 532.69 0 .4832 0.,00 0 .12 0 .019 14..660 518.70 12.454 495..05
Hua 3,.499 -4.420
Table 10.2 Continued

»* VALUES OF PARAMETERS ON STREAMLINES AT STATION, 5, WHICH IS AN ANNULUS **

ITr.EAMLZNE AXIAL AXIAL MERD, TANG. ABS ABS. ABS.FLOM STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEI1P. PRE: s s . TEMP.
IIH. 1 (IH. Î CFT'SEC) CFT/SEC) (FT/SEC) (FT/SEC) I DEG) (DEG) 11./IN.) IPSIA) CDEG.R.) tPSIA) (DEG.R.)
TIP 10.100 -3.500
1 10.037 -3.500 508.58 503.61 0 . 00 5 0 8 . 61 0 . 4652 0 . 00 -0.58 -0. 005 14.125 513.70 12. 177 497.13
•J . (. 1 !, - 1.300 565.85 555.85 0.00 5 6 5 . 85 0. 5202 0., 00 -0.25 0 .,000 14.670 518.70 12. 197 4 92.0 1
i •J . 1 S -i.bOO 575."Ji 575.93 0.00 5 7 5 . 93 0.,5300 0 . 00 -0.09 0 . 001 14.700 518.70 12. 139 4 91.05
-3.500 573.77 5 7 3 . 77 0.00 573.77 0 .,5279 0 ..00 0.07 0.,002 14.700 518.70 12. 157 491.26
5 5.134 -3.500 573.29 57 3. 30 0.00 573.30 0..5275 0..00 0.27 0..004 14.700 518.70 12. 161 491.30
6 7.5/5 -3.500 571.52 571. 54 0. 00 571.54 0..5258 0,.00 0.52 0..006 14.700 518.70 12. 175 491.47
7 6.969 -3.500 56;.84 568. 90 0.00 568.90 0..5232 0,.00 0.83 0..009 14.700 518.70 12. 197 491.72
b t.iOi -3.500 564.31 564..44 0. 00 564.44 0 .5189 0 .00 1.20 0,.014 14.700 518.70 12..234 492.15
9 5.551 -3.500 556.99 557..21 0. 00 557. 21 0 .5119 0 .00 1.61 0,.020 14.700 518.70 12..292 492.82
10 1.667 -3.500 544.04 544..36 0. 00 544. 36 0 .4995 0 .00 1.96 0,.031 14.700 518.70 12.,396 494.00
11 3.539 -3.500 521.16 521 .35 0..00 521. 35 0.4774 0.00 1.53 0 .033 14.660 518.70 12 .541 496.05
HUB 3.500 -3.500

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 6, WHICH IS AN ANNULUS **

STREAMLINE AXIAL AXIAL MEP.D. TANG. ABS. ABS. ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
N O . RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN. ) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (I./IN.) CPSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 10.100 -2.200
1 10.078 -2.200 516.06 516.06 0. 00 516.06 0 . 4723 0. 00 0.20 0 .,026 14.,125 518.70 12. 123 496.50
2 9.60' -2.200 569.02 569.02 0.00 569.02 0.,5233 0. 00 -0.24 -0.,003 14..670 518.70 12. 171 491.71
3 9. 143 -2.200 580.41 580.42 0.00 580.42 0 ..5344 0 ..00 -0.32 -0..006 14..700 518.70 12.,102 490.62
4 8.655 -2.200 580.65 580.66 0.00 580.66 0..5346 0.,00 -0.21 -0..007 14..700 518.70 12.,100 490.60
5 a. 138 -2.200 582.84 582.84 0.00 582.84 0 ..5367 0..00 0.09 -0..005 14,.700 518.70 12.,081 490.38
6 7 .587 -2.200 583.39 583.42 0.00 583.42 0 .5373
, 0..00 0.58 -0,.000 14,.700 518.70 12..07 6 490.33
7 6.993 -2.200 582.13 582.27 0. 00 582.27 0,.5362 0,.00 1.26 0 .006 14,.700 518.70 12..086 450.44
8 6.340 -2.200 577.53 577.93 0. 00 577.93 0,.5320 0,.00 2.15 0,.015 14 .700 518.70 12,.122 490.86
9 5.606 -2.230 567.73 568.67 0.,00 568.67 0 .5230 0 .00 3.29 0 .027 14 .700 518.70 12,.199 491.75
10 4.742 -2.200 548.02 549.92 0..00 549.92 0 .5048 0 .00 4.77 0 .046 14 .700 518.70 12 .351 493.49
11 3.622 -2.200 497.21 500.67 0 ..00 500.67 0 .4576 0 .00 6.74 0 .105 14.660 518.70 12 .697 497.80
KL'3 3.560 -2.200
Table 10.2 Continued

** VALUES OF PARAMETERS OH STREAMLINES AT STATION, 7, WHICH IS AN ANNULUS **

ETFEAMLINE AXIAL AXIAL MERD. TANG. ABS ABS, ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
HO. .RADIUS COORD. VEL. VEL VEL VEL MACH NO, ANGLE SLOPE CURV. PRESS. TEMP PRESS. TEMP.
(IN. ) (IN. J (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEC) (DEG) Cl./IN.) (PSIA) (DEG.P.. ) (PSIA) (DEG.R.)
TIP 10.100 -1 .824
1 10.074 -1 - 824 504.89 504 .90 0.00 504,90 0.4617 0 ,00 0.30 -0. 017 14.125 518.70 12.204 497 .45
? 9 .too -1 .824 566.70 565. 72 0. 00 566 .72 0. 5211 0.00 -0.45 -0.023 14.670 518. 70 12.190 491.93
: •) . 1 10 -1 .824 i&2.4B 562.51 0.00 582.51 0 .5364 0.00 -0.61 -0.020 14.700 518. 70 12.084 490.42
4 8.648 -1 .824 585.94 585.96 0.00 585.96 0.5398 0.00 -0.49 -0.015 14.700 518.70 12.055 490.08
5 8.lis -1 .824 590.18 590.18 0.00 590.18 0.5439 0.00 -0.10 -0. 009 14.700 518.70 12.020 489.67
6 7.589 -1.824 591.36 591.88 0.00 591.88 0. 5456 0, 00 0,51 -0.002 14.700 518.70 12.005 489.50
7 7 .001 -1.824 590.87 591. 03 0. 00 591. 03 0. 5447 0.00 1,35 0. 006 14.700 518.70 12.012 489.58
S 6 , 357 -1.824 585.6 9 586.21 0.00 586. 21 0. 5400 0.00 2,43 0.017 14.700 518.70 12.053 490.06
9 5.611 -1.824 574.36 575.66 0. 00 575.66 D. 5298 0. 00 3,85 0.030 14.700 518.70 : 12.141 491.05
10 4.778 -1.824 551.73 554. 51 0.00 554.51 0. 5093 0.00 5.74 0.050 14.700 518.70 12.314 493.07
11 3.674 -1.824 500.07 506 , 16 0, 00 506, 16 0.4629 0. 00 8.90 0.093 14.660 518.70 •• 12.656 497.34
HUB 3,601 -1. 824

** VALUES OF PARAMETERS OH STREAMLINES AT STATION, 8, WHICH IS AN ANNULUS NX

STREAMLINE AXIAL AXIAL MERD, TANG, ABS, ABS, ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL, MACH HO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IN. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG .R.) (PSIA) (DEG.R.:
TIP 10.100 -0 .940
1 10.065 -0.940 481.80 482.52 0 ,00 482,52 0,4404 0.00 -3.14 -0.118 14.125 518.70 12.362 499,29
2 9.582 -0.940 563.17 563 ,93 0 ,00 563 ,93 0 ,5184 0 .00 -2.99 -0.062 14.670 518.70 12.213 492,19
1 9.116 -0 .940 588.17 588.75 0 ,00 588.75 0 ,5425 0 .00 -2.55 -0.046 14.700 518.70 12.032 489.81
4 8.634 -0.940 598.59 598.91 0 ,00 598,91 0 ,5524 0.00 -1.87 -0.032 14.700 518.70 11,946 488,80
5 8 .129 -0 .940 607.48 607 ,57 0 ,00 607,57 0.5609 0 ,00 -0.97 -0.020 14.700 518.70 11,871 487.93
6 7.595 -0 .940 612,19 612,19 0.00 612.19 0 ,5654 0.00 0.14 -0 .009 14.700 518.70 11,831 487.46
7 7.023 -0 .940 613.01 613 ,21 0.00 613,21 0 ,5664 0 ,00 1.46 0.001 14.700 518.70 11,823 487.36
8 6.398 -0 .940 608.85 609 .72 0 .00 609 .72 0 ,5630 0 ,00 3.07 0.012 14.700 518.70 11,853 487.71
9 5.699 -0. 940 598.10 600.48 0 .00 600,48 0 ,5540 0. 00 5.10 0.024 14.700 518.70 11,932 488.64
10 4 .883 -0 .940 575.35 581 .00 0 ,00 581 ,00 0 ,5350 0 ,00 8.00 0.041 14.700 518.70 12,097 490,56
11 3.847 -0.940 525.75 539.52 0 ,00 539 ,52 0 ,4 948 0.00 12.97 0.065 14.660 518.70 12,400 494.44
liU3 1.752 -0.910
Table 10.2 Continued

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 9, WHICH IS THE INLET OF ROTOR NUMBER. 1 »»

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN., (IN. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IN.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 10.051 0. 170
1 10.DOS 0. 158 533.13 538.79 0. DO 538.79 0. 4941 0..00 -8.31 -0. 045 14.125 518.70 11.953 494.50
2 9.562 0.,039 594.80 598.18 0.00 598.18 0. 5517 0..00 -6.10 -0..038 14.670 518.70 11.927 488.87
J 9.118 -Q ..031 613.03 615.04 0.00 615 .04 0 ..5683 0..00 -4.64 -0,.031 14.700 518.70 11.807 487.17
4 8.053 -0 .099
, 618.53 619.49 0. GO 619.49 0..5726 0..00 -3.19 -0..024 14.700 518.70 11.768 486.71
5 8.162 -0 .172
. 623.18 623.46 0. 00 623.46 0..5765 0.,00 -1.71 -0..017 14.700 518.70 11.733 486.30
6 7 .640 -0 .250 624.36 624.36 0. 00 624.36 0.,5774 0..00 -0.16 -0..010 14.700 518.70 11.725 486.21
7 7.077 -0,.340 622.12 622.33 0..00 622.33 0 ,5754
. 0..00 1.49 -0.,003 14.700 518.70 11.743 486.42
8 6.458 -0 .450
, 615.37 616.43 Q..00 616.43 0 .5696 0 .00 3.36 0..004 14.700 518.70 11.795 487.03
9 5.761 -0 ,556 602.82 605.69 0..00 605.69 0 .5591 0 .00 5.59 0.,014 14.700 518.70 11.887 488.12
10 4.942 -0 .697 578.50 585.09 0..00 585.09 0 .5389 0 .00 8.61 0 .029
. 14.700 518.70 12.062 490.17
11 3.892 -0 .839 527.46 542.10 0..00 542.10 0.4973 0.00 13.35 0 .060 14.660 518.70 12.380 494.21
HUB 3.771 -0 .855

ETRElAMLINE P.EL. FLOW REL. REL. REL..MACH WHEEL FLOW L.E:,RAD. MAX.TH. MAX.TH. TRAN.PT. SEGMENT LAYOUT
,M0. R/PTIP ANGLE TANG. VEL. VEL. NUMBER SPEED COEF. /CHORD /CHORD PT.LOC. LOCATION IN/OUT CONE ANG
(DEG) (FT/SEC) (FT/SEC) (FT/SEC) /CHORD /CHORD TURN.RATE (DEG)
TIP 1 . 0000 1407..19
1 0. 9956 68. 57 1401.08 1501.10 1. 3766 1401..08 0..3789 0.0018 0. 0290 0.6400 0.7000 0..0750 -10.92
2 0 , 9513 65. 92 1338.73 1466.29 1. 3524 1338..73 0..4227 0.0019 0..0300 0.6300 0.6474 0..1800 -8.46
3 0..9071 64..27 1276.47 1416.92 1. 3091 1276.,47 0.,4356 0.0020 0..0323 0.6200 0.6042 0..4300 -6.67
4 0..8608 62..91 1211. 36 1360.57 1.2577 1211..36 0..4395 0. 0022 0..0361 0 .6100 0.5627 0 ..6600 -4.87
5 0 .8120
. 61,,38 1142..70 1301.72 1. 2038 1142 ,70 0 .4429 0. 0025 0..0412 0 .6000 0.5193 0.,7900 -3.05
6 0 ..7601 59..73 1069.62 1238.51 1. 1454 1069.,62 0.,4437 0, 0028 0..0474 0.5800 0.4705 0..8300 -1.15
7 0 .704 1 57 ,87 990..76 1170.00 1..0816 990 .76 0 .4421 0,.0032 0.,0548 0.5600 0.4180 0 .8600 0.81
8 0 .6425 55 .71 904..14 1094.28 1..0112 904 .14 0 .4373 0..0036 0 .0630 0.5400 0.3592 0 .9600 2.95
9 0 .5732 53 .10 806 .59
. 1008.69 0..9310 806 .59 0 .4284 0..0041 0 .0717 0.5000 0.2862 0 .9800 5.44
10 0 .4917 49.78 691.,88 906.11 0..8346 691 .88 0.4111 0.,0045 0.0799 0.5000 0.2243 1 .0000 8.65
11 0 .3872 45 .14 544.,84 768.59 0..7051 544.84 0.3748 0.,0048 0 .0849 0.5000 0.1629 1.0000 13.69
HUH 0 .3752

INLET STREAMLINE ++++t+t+++++++++++++t++++++++t+++ LAYOUT CONE +++++++•++•++•++++++++++••+++++++++<•+++

STREAMLINE inc. S.S.IHC. IN.BLADE IN.BLADE TRAN.PT. BLD.SET 1ST SEG. MACH NO. SH.LOC. COV.CHAN. MIN.CHK. MIN.CHK. L.E.EDGE
KO. PCT. ANGLE ANGLE ANGLE ANGLE BL.ANGLE ANGLE S.S.CAM. AT SHOCK AS FHACT AS FRACT AREA PT.LOC.IN CIR.CENT.
PASS. (DEC) (DEC) (DEC) (DEG) (DEG) (DEG) (DEG) LOCATION OF S.S. OF S.S. MARGIN COV.CHAN. R*DO/DR

1 0. 70 2.76 0. 45 66. 21 66. 34 64.40 63. 39 4.46 1..4581 0..6941 0.,3059 0..0508 0 . 3942 -0.6973
2 7..79 2.,93 0 ..51 62..99 63. 04 60..43 59.,87 5..10 1.,4474 0..6421 0.,3579 0..0555 0..4757 -0. 3472
3 14..87 2.,94 0..40 61..33 61. 32 57..77 57.,75 5. 92 1.,4184 0..6009 0.,3991 0..0632 0..4782 -0. 1689
4 22 .27 3,.02 .
0 ,37 59..89 59..90 55..44 55..53 6..54 1..37 92 Q.,5607 0..4393 0.,0631 0.,4064 -0.,1323
5 30,,08 3 ,17 0 ,35
. 58..21 58..24 53..06 52..92 6..97 1.,3354 0..5183 0..4817 0.,0581 0 ,3238
. -0.,1497
6 38 ,39 3,,54 0 ,26 56..18 56..23 50..12 49.,35 7..92 1.,2976 0..4705 0 ..5295 0.,0582 0 .2413 -0 ,1736
7 47 ,36 4 .10 0 .20 53.,77 53..82 46.,89 45,,06 8.,81 1 .2586 0.,4186 0,,5814 0,.0576 0 ,2029 -0.,2022
8 57 .21 4 .84 0 .17 50.,88 50.. 93 42.,79 39,,44 10,,08 1 .2291 0.,3607 0,,6393 0 ,0636 0 .1892 -0..2463
9 68.31 6 .85 0.00 46 .25 46 .27 38 .03 31 .85 11 .96 1.1667 0 .2892 0 .7108 0 .0664 0 .2489 -0 ,2370
10 81 .35 6 .89 0.00 42 .89 42 .88 33 .80 23.10 11 .47 1 .0398 0 .2278 0 .7722 0 .0542 0 .3013 -0,,1856
11 98.08 5.41 -0.00 39 .74 39 .67 30 .51 13 .05 8.78 0 .8456 0 .1654 0 .7592 0 .0586 0.3661 -0 .1268
Table 10.2 Continued

VALUES OF PARAMETERS ON STREAMLINES AT STATION,, 10, WHICH IS THE OUTLET OF ROTOR NUMBER, 1 »»

STRE AMLINE AXIAL AXIAL HERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IH.) (IN. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEGl (DEGl (l./IN.) (PSIA) tDEG.R.l tPSIA) CDEG.R.)
T l ? 9.7'.8 1.765
1 9.696 1 .777 479.17 483.59 449.83 660.46 0.5576 42.93 -7.76 0.057 23.896 620.30 19.348 584.05
2 9.290 1.870 510.40 512.86 423.31 664.99 0.5665 39.54 -5.61 0.039 23.896 610.33 19.221 573.56
3 fi.r.8 9 I . 920 519.52 520.83 430.62 675.80 0.5775 39.58 -4.07 0.033 23.896 607.89 19.062 569.92
<i 8.«,76 1 .976 523.93 524.48 440.03 684.62 0.5868 40.00 -2.63 0.026 23.896 605.60 18.928 566.63
i 8.0<.<< 2.042 529.35 529.47 453.73 697.29 0.5993 40.60 -l.ZO 0.018 23.896 603.75 18.744 563.32
i 7.592 2.143 533.87 533.88 475.68 715.05 0.6162 41.70 0.26 0.010 23.896 602.85 18.495 560.33
7 7.113 2.254 5Î9.Î2 539.59 502.43 737.28 0.6374 42.96 1.80 0.002 23.896 601.99 18.178 556.77
8 6.604 2.383 547.62 548.57 536.41 767.25 0.6660 44.36 3.38 -0.010 23.896 601.26 17.747 552.29
9 6.058 2.555 559.88 562.15 578.34 806.53 0.7039 45.81 5.15 -0.024 23.896 600.35 17.166 546.23
10 5.454 2.666 576.64 580.68 638.47 863.04 0.7590 47.71 6.76 -0.052 23.896 599.85 16.312 537.88
n 4.759 2.720 608.61 616.14 729.02 954.51 0.8507 49.80 8.96 -0.106 23.896 599.55 14.886 523.72
HUB 4.658 2.728

STREAMLINE REL.FLOM REL. REL. REL.MACH WHEEL FLOW HEAD IDEAL HEAD ADIAB. DIFFUSION LOSS SHOCK DEGREE
NO. R/RTIP ANGLE TANG.VEL. VEL. NUMBER SPEED COEF. COEF. COEF. EFF. FACTOR COEF. LOSS REACTION
(DEG) (FT/SEC) (FT/SEC) (FT/SEC) COEF.
TIP 1. 0000
1 0. 9946 61.95 907.54 1028.35 0.8682 1357.38 0.3405 0. 2551 0. 3084 0.8271 0.4282 0.1:41 3.0400 0.7569
2 0. 9530 59.69 877.31 1016.21 0.8657 1300.62 0.3627 0. 2354 0. 2780 0.8466 0.4112 0.1128 0.0391 0. 7906
3 0 . 9119 57.38 813.89 966.28 0.8258 1244.52 0.3692 0. 2343 0.,2706 0.8658 0.4232 0.1007 0.0359 0..7890
4 0..8695 54.91 746.56 912.38 0.7820 1186.59 0.3723 0..2343 0..2637 0.8887 0.4361 0.0863 0.0313 0 ..7787
5 0..8252 51.79 672.47 855.89 0.7357 1126.21 0.3762 0 .2343
. 0,,2581 0.9080 0.4517 0.0743 0.0259 0..7624
6 0 .7783
. 47.72 5:7.19 793.61 0.6839 1062.87 0.3794 0..2343 0,,2553 0.9177 0.4729 0.0705 0.0215 0.,7361
7 0 .7297 42.44 493.44 731.19 0.6321 995.87 0.3833 0 .2343 0 .2527 0.9273 0.4941 0.0670 0.0171 0 .6998
8 0.6775 35.28 388.17 672.02 0.5833 924.59 0.3892 0 .2343 0 .2505 0.9355 0.5120 0.0652 0.0142 0 .6473
9 C .6214 25.63 269.71 623.50 0.5442 848.05 0.3979 0 .2343 0 .2477 0.9460 0.5169 0.0615 0.0076 0 .5740
:0 0 .5594 12.15 125.02 593.99 0.5224 763.49 0.4093 0.2343 0 .2462 0.9517 0.4930 0.0652 0.0002 0 .4621
n 0 .4:52 -5.82 -62.78 619.33 0.5520 666.24 0.4325 0 .2357 0.2453 0.9611 0.3657 0.0698 0.0000 0 .2714
iiua 0.4773

--- OUTLET STREAMLINE ** LAYOUT CONE

SlP.EnnLlUE PRESS. TEMP. AERO. ELEMENT LOCAL BLADE FORCES T.E;.RAD. DEV. OUT.BLADE OUT. BLADE MAX.CAMB. T.E.EDGE
•NO. PCT. RATIO RATIO CHORD SOLIDITY RADIUS FOR.AXIAL TANG. /CHORD ANGLE ANGLE ANGLE PT.LOC. CIR.CENT
SPAN (IN. ) (IN.) (LBS/IN) (LBS/IN) (DEG) (DEG) (DEG) /CHORD R»30/DR

1 1 .03 1.6917 1. 1959 3.6642 1. 3023 9.852 19.6015 -10.9508 0.0018 8.00 53.95 53.32 0.7273 0.3422
2 9. CO 1.6289 1. 1766 3.6725 1. 3642 9.426 17.7714 -10.3378 0.0019 6.80 52.89 52.55 0.6727 0 .1640
J 16 .87 1.4255 1. 1720 3.6701 1. 4273 9.004 16.5638 -10.3487 0.0020 4.80 52.58 52.36 0.6052 0.1109
<1 25.00 1.6255 1. 1675 3.6677 1. 4995 8.564 15.4306 -10.1285 0.0023 4.50 50.41 50.21 0.5511 0.1417
5 13.4"/ 1.6255 1. 1640 3.6658 1..5840 8.103 14.2440 -9.9200 0. 0026 4.60 47.19 46.99 0.5292 0.1910
6 42.36 1.6255 1. 1622 3.6645 1.,6847 7.616 12.8940 -9.7834 0. 0030 5.70 42.02 41.84 0,5192 0.2285
7 51.76 1.6255 1,, 1606 3.6644 1..8083 7.095 11.4312 -9.6067 0. 0034 6.63 35.81 35.65 0.5117 0.2617
a 61.76 1.6255 1..1592 3.6666 1,,9657 6.531 9.7984 -9.3923 0..0038 7.52 27.76 27.67 0.5006 0.3191
9 72.50 1.6255 1 .1574 3.6741 2.1769 5.909 7.9601 -9.0242 0.,0042 8.64 16.99 17.08 0.4964 0.3865
10 84.36 1.6255 1..1565 3.6951 2.4891 5.198 5.6774 -8.4343 0.,0046 10.39 1.76 2.66 0.4914 0.4775
11 98.01 1.6300 1 .1559 3.7588 3.0428 4.325 2.8398 -8.6529 0.,0048 12.52 -18.34 -15.47 0.4833 0.5900
Table 10.2 Continued

»» VALUES OF PARAMETERS ON STREAMLINES AT STATION, 11, WHICH IS AH AHHULUS **

STREAMLINE AXIAL AXIAL HERD. TANG. AGS. ACS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IM.) (111. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEC) (DEC) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9 6)0 3.000
1 9 575 3.000 540.90 541.86 455. 51 707.89 0 6005 40.05 -3.42 0 065 23.896 620.16 18.728 578.52
9 207 3.000 558 59 559.21 427. 13 703.67 0 6017 37.37 -2.70 0 047 23.896 610.33 18.709 569.17
3 8 eii 3.000 560 93 561.24 433. 38 709.08 0 6080 37.67 -1.90 0 036 23.896 607.91 18.617 566.10
4 8 443 3.000 560 78 560.88 441.75 713.95 0 6137 38.22 -1.06 0 027 23.896 605.62 18.532 563.23
8 033 3 000 561 90 561.90 454 37 722.62 0 6228 38.96 -0.14 0 019 23.896 603.77 18.398 560.34
6 7 600 3.000 561 72 561.78 475 16 735.78 0 6355 40.22 0.85 0 013 23.896 602.87 18.207 557.84
7 7 158 3.000 562 29 562.62 500 70 753.15 0 6523 41.67 1.96 0 005 23.896 602.00 17.955 554.82
S & 640 3.000 564 93 565.79 533 58 777.70 0 6759 43.32 3.15 -0 003 23.896 601.27 17.596 550.96
9 6 096 3.000 570 87 572.66 574 76 811.35 0 7086 45.11 4.53 -0 017 23.896 600.35 17.094 545.59
10 5 491 3.000 580 39 583.33 534 22 861.69 0 .7577 47.39 5.75 -0 027 23.896 599.86 16.333 538.08
11 4 .799 3.000 608 65 613.53 722 95 948.20 0 .8443 49.68 7.24 -0 107 23.896 599.55 14.986 524.73
HUB 4 .686 3.000

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 12, WHICH IS AN ANNULUS »*

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH HO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN. ) (IN. ) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEC) (DEC) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9.608 3.485
1 9.554 3.485 550.45 550.76 456.52 715.36 0. 6073 39.66 -1.90 0.044 23.896 620.16 18.628 577.63
2 9.190 3.485 570.20 570.40 427.90 713.06 0..6104 36.88 -1.52 0.037 23.896 610.33 18.582 568.06
3 8.822 3.485 573.48 573.57 433.92 719.21 0. 6173 37.11 -1.01 0.030 23.896 607.91 18.479 564.90
4 8.438 3.485 573.75 573.76 442.00 724.27 0..6233 37.61 -0.37 0.024 23.896 605.62 18.390 562.00
5 8.035 3.485 574.81 574.83 454.24 732.64 0,.6321 38.32 0.36 0.019 23.896 603.77 18.259 559.13
6 7.610 3.485 574.22 574.35 474.54 745.02 0 .6441 39.56 1.19 0.014 23.896 602.87 18.077 556.70
7 7.156 3.485 573.78 574.17 499.42 760.98 0,.6596 41.02 2.11 0.010 23.896 602.00 17.843 553.83
8 6.666 3.485 574.41 575.27 531.41 783.16 0 .6810 42.73 3.13 0.006 23.896 601.27 17.517 550.25
9 6.131 3.485 576.44 577.97 571.41 812.75 0 .7099 44.67 4.17 0.001 23.896 600.35 17.074 545.40
10 5.533 3.485 578.62 580.98 629.34 856.50 0 .7526 47.29 5.16 0.001 23.896 599.86 16.411 538.82
11 4.841 3.485 587.21 589.80 716.69 928.17 0 .8240 50.55 5.37 -0.027 23.896 599.55 15.301 527.86
HUB 4.731 3.485
Table 10.2 Continued

»« VALUES OF PARAMETERS OK STREAMLINES AT STATION, 13, WHICH IS THE IHLET OF STATOR NUMBER, 1, OF STAGE NUMBER, 1 «»

AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
MO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IK.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9. 600 4.116
1 9. 539 4.116 555.73 555.79 457.24 719.70 0.6113 39.44 -0.83 0.016 23.896 620.11 18.569 577.06
z 9 ..179 4.114 579.52 579.54 428.45 720.72 0.6174 36.48 -0.47 0.020 23.896 610.34 18.478 567.16
3 B . SIS 4.120 585.19 585.19 434.28 728.73 0.6261 36.58 -0.10 0.020 23.896 607.92 18.348 563.77
<. 5. 436 4. 124 587.04 587.06 442.07 734.89 0.6331 36.98 0.37 0.017 23.896 605.63 18.243 560.72
5 S. 041 4. 129 589.21 589.29 453.90 743.84 0.6425 37.61 0.94 0.014 23.896 603.78 18.102 557.76
6 7 .625
, 4. 135 589.49 589.72 473.62 756.37 0.6548 38.77 1.61 0.011 23.896 602.88 17.917 555.29
7 7 .182 4. 143 589.47 589.99 497.65 771.84 0.6699 40.15 2.42 0.010 23.896 602.00 17.687 552.45
8 6 .705 4. 151 589.61 590.63 528.39 792.49 0.6899 41.82 3.36 0.010 23.896 601.27 17.381 549.03
9 6 .184 4. 162 588.94 590.70 566.52 818.46 0.7154 43.80 4.43 0.014 23.896 600.36 16.989 544.63
10 5 .603 4. 175 584.18 587.08 621.54 854.97 0.7511 46.63 5.70 0.024 23.896 599.86 16.435 539.04
n <,
.920 4..190 563.20 567.04 705.13 904.84 0.8006 51.19 6.67 0.090 23.896 599.55 15.665 531.43
HJB 4 .799 4 . 192

STPEAtlLINE FLOW BEL.FLOW L.E:.RAD. MAX.TH. MAX.TH. TRAN.PT. SEGMENT LAYOUT

NO. B/p.rip COEF. ANGLE /CHORD /CHORD PT.LOC. LOCATION IK/OUT CONE AHG
(DEG) /CHORD /CHORD TURN.RATE (DEG)
TIP 1. 0000
1 0. 9936 0. 3949 57.67 0.0129 0. 0797 0.5000 0. 3310 1.0000 -0.20
Z 0 ..9561 0..4118 55.92 0.0123 0..0782 0.5000 0..3088 1.0000 0.27
3 0..9182 0..4159 53.81 0.0117 0..0767 0.5000 0 .3035
. 1.0000 0.58
4 0 .8788 0 .4172
. 51.54 0. 0111 0..0752 0.5000 0..2975 1.0000 0.92
5 0,.8377 0 .4187 48.75 0. 0104 0..0735 0.5000 0 .2917
. 1.0000 1.33
6 0 .7943 0 .4189
, 45.20 0 . 0097 0 ..0718 0.5000 0 .2873 1.0000 1.82
7 0 .7481 0 .4189 40.72 0 . 0090 0..0699 0.5000 0 .2821 1.0000 2.46
8 0 .6984 0 .4190 34.79 0 .0082
. 0.. 0 6 7 9 0.5000 0 .2754 1.0000 3.27
9 0 .6442 0 .4185 26.87 0 .0073
. 0 .0658 0.5000 0 .2670 1.0000 4.34
10 0 .5836 0 .4151 15.50 0..0063 0 .0633 0.5000 0 .2562 1.0000 5.78
11 0 .51Z5 0 .4002 -1.65 0 .0052
. 0 .0605 0.5000 0 .2406 1.0000 7.95
HUB 0 . 4999

INLET STREAMLINE •• + tt + 4- ++ 4t + tti + t++i + ttt + t + t + t + + + LAYOUT CONE ••••+•+••+«••• ++«••+•• +•+••••+••• +•••••••
fJTPEAMLINE INC. S.S.INC. IN.BLADE IN.BLADE TRAN.PT. BLD SET 1ST SEG. MACH NO. SU.LOC. COV.CHAN. MIH.CHK. MIH.CHK. L.E.EDGE
NO. PCT. ANGLE ANGLE ANGLE ANGLE BL.ANGLE ANGLE S.S CAM. AT SHOCK AS FRACT AS FRACT AREA PT.LOC.IH CIR .CENT
PASS. (DEG) (DEG) (DEG) (DEG) (DEG) (DEG) (DEG) LOCATION OF S.S. OF S.S. MARGIN COV.CHAN. R*DO/DR

1 1 28 2 94 -3 00 36.50 36 48 19.01 10 15 21 38 0.8314 0.3314 0.5198 0. 2368 0.0000 -0.0584

2 8 78 2 97 -3 00 33.50 33 52 19.39 10 62 17 78 0.8016 0.3080 0.5775 0. 1991 0.0000 0.0383
J 16 36 2 94 -3 00 33.64 33 68 20.29 11 60 16 95 0.8039 0.3026 0.6016 0. 1865 0.0000 0.0813
4 24 23 2 89 -3 00 34.09 34 12 21.10 12 23 16 47 0.8079 0.2967 0.6174 0 . 1775 0.0000 0.0826
5 32 46 2 82 -3 00 34.78 34 80 22.00 12 88 16 13 0.8163 0.2910 0.6312 0.1673 0.0000 0.0961
6 4 1 13 2 75 -3 00 36.02 36 03 23.14 13 65 16 12 0.8319 0.2870 0.6410 0 . 1580 0.0000 0. 1106
7 50 36 2 66 -3 00 37.49 37 49 24.45 14 49 16 .15 0.8514 0.2823 0.6508 0.1475 0.0000 0. 1201
• 60 30 2 55 -3 00 39.27 39 26 25.95 15 .30 16 26 0.8783 0.2762 0.6593 0. 1356 0.0000 0. 1332
1 71 14 Z 42 -3 00 41.39 41 38 27.80 16 .29 16 .33 0.9116 0.2685 0.6699 0 1219 0.0000 0. 1518
10 83 25 2 23 -3 00 44.40 44 40 30.18 17 .24 16.72 0.9620 0.2590 0.6769 0. 1091 0.0000 0. 1766
11 97 .47 1 95 -3 00 49.25 49 28 33.62 17 .94 17 .79 1.0395 0.2460 0.6771 0 1097 0.0000 0. 20 39
 Table 10.2 Continued

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 15, WHICH IS AN ANNULUS

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOW 1STREAM. STREAM. TOTAL TOTAL STATIC STATIC
110. RADIUS COORD. VEL. VEL. VEL. VEL. MACK NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) CI./IK.) CPSIA) CDEG.R.) (PSIA) (DEG.R.)
TIP 9.600 7.340
1 9.531 7.340 580.30 580.30 0.00 580.30 0.4871 0.00 0.01 -0.000 23.421 618.99 19.915 591.01
R 9.194 7.340 579.12 579.13 0.00 579.13 0.4896 0.00 0.16 -0.001 23.458 610.26 19.912 582.38
i S.S<I6 7. 3'IO 579.07 579.08 0.00 579.08 0.4906 0.00 0.35 -0.003 23.468 607.94 19.909 580.07
4 S.4:6 7.340 579.57 579.59 0.00 579.59 0.4920 0.00 0.55 -0.005 23.478 605.69 19.898 577.76
5 a.110 7.340 580.24 . 580.29 0.00 580.29 0.4934 0.00 0.75 -0.008 23.480 603.87 19.882 575.87
6 7.7ia 7.340 580.45 580.53 0.00 580.53 0.4940 0.00 0.97 -0.010 23.466 602.97 19.862 574.95
7 7.i05 7.340 579.27 579.39 0.00 579.39 0.4933 0.00 1.18 -0.013 23.415 602.10 19.828 574.19
S 6.865 7.340 577.01 577.18 0.00 577.18 0.4916 0.00 1.39 -0.016 23.340 601.37 19.786 573.67
9 6.191 7.340 572.13 572.35 0.00 572.35 0.4877 0.00 1.56 -0..020 23.223 600.46 19.737 573.22
10 5.869 7.340 557.50 557.72 0.00 557.72 0.4749 0.00 1.61 -0..023 22.954 599.96 19.670 574.10
11 5.264 7.340 520.82 520.89 0.00 520.89 0.4424 0.00 0.99 -0..016 22.407 599.66 19.588 577.08
H'J3 5.137 7.340

VALUES OF PARAMETERS; ON STREAMLINES AT STATION. 16, WHICH IS AN ANNULUS KK

STREAMLIHE AXIAL AXIAL MERD. TANG. , ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH HO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) Cl./IN.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9.600 7.446
1 9.531 7.446 582.82 582.82 0.00 582.82 0.4893 0.00 0.01 -0.000 23.421 618.99 19.886 590.77
2 9.195 7.446 581.58 581.58 0.00 581.58 0.4918 0.00 0.15 -0.001 23.458 610.26 19.884 582.14
i S.848 7.446 581.41 581.42 0.00 581.42 0.4926 0.00 0.33 -0.003 23.468 607.94 19.681 579.84
C| S.489 7.446 581.70 581.72 0.00 581.72 0 .4939 0.00 0.52 -0.004 23.478 605.69 19.873 577.55
5 2.115 7.446 582.02 582.07 0.00 582.07 0.4949 0.00 0.71 -0.006 23.480 603.87 19.861 575.70
6 7.724 7.446 581.73 581.80 0.00 581.80 0.4951 0.00 0.91 -0.007 23.466 602.97 19.847 574.83
7 7.311 7.446 579.84 579.95 0.00 579.95 0.4938 0.00 1.11 -0.009 23.415 602.10 19.821 574.14
8 6.871 7.446 576.60 576.75 0.00 576.75 0.4913 0.00 1.30 -0.012 23.340 601.37 19.791 573.71
9 6.397 7.446 570.38 570.56 0.00 570.56 0.4861 0.00 1.45 -0.014 23.223 600.46 19.757 573.39
10 5.874 7.446 553.84 554.03 0.00 554.03 0.4716 0.00 1.48 -0.016 22.954 599.96 19.712 574.44
11 5.265 7.446 515.48 515.55 0.00 515.55 0.4377 0.00 0.90 -0.015 22.407 599.66 19.643 577.55
HUB 5.139 7.446

I
Table 10.2 Continued

1, OF STAGE NUMBER, 1 »»

S
i
VALUES OF PARAMETERS ON STREAHLIHES AT STATION. 14. WHICH IS THE OUTLET OF STATOR

STREAHLItlE AXIAL AXIAL HERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
HO. RADIUS COORD. VEL. VEL. VEL. VEL. rjACH HO. AHGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
CIH.) (IH.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEC) (DEC) (l./IH.) (PSIA) (DEG.R.1 (PSIA) (DEG.R.)
TIP 9.400 6.353
1 9.531 6.352 571.06 571.06 0.00 571.06 0.4789 0..00 0.06 -0.002 23.421 619.10 20.020 592.01
2 9.159 6.347 570.33 570.35 0.00 570.35 0..4819 0 .00
. 0.40 -0.005 23.458 610.26 20.013 583.22
JS 8..137 6.)A5 570.86 570.89 0.00 570.89 0.4833 0 .. 00 0.67 -0.006 23.468 607.94 20.003 580.85
<1 S.'.72 fc . 3'.'. 571.79 571.86 0.00 571.86 0 ..4851 0.00 0.93 -0.007 23.478 605.68 19. . 988 578.49
'J
6
8.093
7.696
6. 3'.3
6.343
572.68
572.96
572..81
573..17
0.00
0.00
572.81
573.17
0 ..4867
0..4874
0..00
0..00
1.22
1.57
-0.008
-0.011
23.480
23.466
603.85
602.96
19.969 576.58
19..948 575.65
7 7.276 6.342 571.94 572..28 0.00 572.28 0..4870 0.00 1.99 -0.015 23.415 602.09 19..910 574.86
S 6.830 6.342 570.40 570..95 0.00 570.95 0..4861 0.00 2.51 -0.022 23.340 601.36 19..857 574.26
9 6.350 6 . 342 567.78 568..65 0.00 568..65 0..4844 0.00 3.17 -0.034 23.223 600.45 19..779 573.56
10 5.822 6.343 558.83 560,.18 0.00 560..18 0..4771 0.00 3.99 -0.053 22.954 599.95 19..643 573.86
U 5.221 6.347 544.64 546..62 0.00 546..62 0..4652 0.00 4.88 -0.120 22.407 599.65 19..318 574.79
HUB 5.092 6..348

STREAMLINE FLOU HEAD IDEAL HEAD STATOR STAGE STAGE DIFFUSION STATOR SHOCK ELEMENT AERO. DEGREE
NO. R/RTIP COEF. COEF. COEF. PO.RATIO PO.RATIO AD.EFF. FACTOR LOSS COEF. LOSS SOLIDITY CHORD REACTION
COEF. (IN.)
TIP 1.0000
1 0.9928 0..4058 0..2446 0..3084 0..9801 1,.6581 0..7933 0.,4531 0.0891 0.0000 1.2888 2.2708 0.1485
2 0.9572 0..4053 0..2258 0..2780 0..9817 1..5990 0..8123 0..4307 0.0808 0.0000 1.3380 2.2708 0.1664
3 0 ..9205 0..4057 0 .2250 0..2706 0..9821 1..5965 0..8314 0..4303 0.0771 0.0000 1.3923 2.2709 0..1800
4 0..8825 0..4063 0..2252 0..2637 0..9825 1..5971 0..8542 0..4283 0.0739 0.0000 1.4536 2.2711 0.. 1897
5 0.8430
. 0.4070 0.2253 0 .2581
. 0.9826 1.5973 0.8730 0..4296 0.0717 0.0000 1.5236 2.2713 0..2031
6 0..8016 0.4072 0.2250 0..2553 0.9820 1..5963 0..8811 0..4364 0.0719 o.oooo 1.6048 2.2718 0..2208
7 0.7550 0.4064 0.2239 0.2527 0.9799 1.5929 0.8859 0.4468 0.0774 0.0000 1.7012 2.2727 0..2417
8 0.7115 0.4053 0.2222 0..2505 0.9767 1.5877 0.8872 0.4612 0.0853 0.0000 1.8183 2.2740 0.2693
9 0.6614 0.4035 0.2196 0.2477 0.9718 1.5798 0.8867 0.4790 0.0974 0.0000 1.9655 2.2764 0.3034
10 0.6065 0.3971 0.2137 0.2462 0.9606 1.5615 0.8679 0.5098 0.1262 0.0000 2.1603 2.2806 0.3489
11 0.5439 0.3870 0.2028 0.2453 0.9377 1.5284 0.8267 0.5507 0.1809 0.0002 2.4424 2.2887 0.3955
.530-
— — — OUTLET• STREAMLINE — •• LAYOUT COHE +t+

STREAMLINE LOCAL BLADE FORCES T.E.RAD. DEV. OUT.BLADE OUT.BLADE MAX.CAMB. T.E.EDGE
NO. PCT. RADIUS FOR.AXIAL TANG. /CHORD ANGLE ANGLE ANGLE PT.LOC. CIR.CENT
SPAN (IN.) (LBS/IN) (LBS/IN) (DEC) (DEG) (DEG) /CHORD R*DO/DR
1 1.54 9.535 2.8504 8.7334 0.0129 16.20 -16.20 -16.17 0.5001 -0.1101
2 9.12 9.184 2.4286 8.1020 0.0123 12.30 -12.30 -12.29 0.4999 -0.0873
1 16.92 8.826 2.4295 7.9809 0.0116 10.50 -10.50 -10.50 0.4998 -0.0602
4 25.02 8.454 2.4461 7.8106 0.0110 9.70 -9.70 -9.70 0.4997 -0.0433
5 33.44 6.067 2.4872 7.6654 0.0103 9.10 -9.10 -9.10 0.4995 -0.0413
6 42.25 7.660 2.5829 7.5812 0.0096 8.80 -8.80 -8.81 0.4993 -0.0412
7 51.55 7.229 2.6633 7.4781 0.0089 8.60 -8.60 -8.61 0.4990 -0.0381
i 61.45 6.767 2.7640 7.3726 0.0081 8.80 -8.80 -8.82 0.4985 -0.0374
9 72.10 6.267 2.8617 7.2158 0.0072 9.00 -9.00 -9.03 0.4978 -0.0326
10 83.81 5.713 2.9351 6.9995 0.0063 10.30 -10.30 -10.28 0.4968 -0.0064
11 97.14 5.071 2.9569 7.3620 0.0052 14.20 -14.20 -14.03 0.4949 0.0287
Table 10.2 Continued

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 17, WHICH IS AN ANHULUS

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. B,'.DIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9.1,00 11 .010
1 9.bll 1 1 .010 589.27 589.27 0.00 589.27 0.4951 0.00 -0..00 0..000 23.421 618.53 19.809 589.68
2 9.198 11 . 010 588.04 588.04 0.00 588.04 0.4976 0.00 0..01 -0..000 23.458 610.21 19.808 581.47
i 5.854 11..010 587.58 587.58 0. 00 587.58 0.4981 0.00 0..01 -0..000 23.468 607.94 19.810 579.24
8.498 11..010 587.26 587.26 0.00 587.26 0.4988 0.00 0..00 -0..000 23,478 605.71 19.809 577.04
5 8.127 11..010 586.64 586.64 0.00 586.64 0.4990 0.00 -0..00 -0..000 23.480 603.90 19.808 575.29
6 7.738 11,.010 584.97 584.97 0.00 584.97 0.4979 0.00 -0,.01 -0..000 23.466 603.01 19.810 574.56
7 7.328 11,.010 581.18 581.18 0.00 581.18 0.4949 0.00 -0..04 -0..000 23.415 602.14 19.807 574.06
S 6.889 11.010 575.29 575.29 0..00 575.29 0.4899 0.00 -0.07 0,.000 23.340 601.41 19.808 573.90
9 6.415 11.010 565.40 565.40 0..00 565.40 0.4815 0.00 -0.13 0.001 23.223 600.50 19.817 573.92
10 5.887 11 .010 543.72 543.72 0..00 543.72 0.4625 0.00 -0.20 0.002 22.954 600.01 19.824 575.42
11 5.266 11.010 497.78 497.78 0..00 497.78 0.4220 0.00 -0.22 0.004 22.407 599.70 19.822 579.09
HUB 5.140 11.010

«* VALUES OF PARAMETERS ON STREAMLINES AT STATION, 18. WHICH IS AN ANHULUS ^

STREAMLINE AXIAL AXIAL MERD. TANG. ABS. ABS. ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
NO. RADIUS COORD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IM.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9.600 11.350
1 9.531 11.,350 589.87 589.87 0.00 589.87 0.4957 0.00 -0.00 0.000 23.421 618.53 19.803 589.62
2 9.198 11..350 588.61 588.61 0.00 588.61 0.4981 0.00 0.01 0.,000 23.458 610.21 19.802 581.41
3 8.855 11.,350 588.10 588.10 0.00 588.10 0.4986 0.00 0.,01 0..000 23.468 607.94 19.803 579.19
4 8.499 11..350 587.70 587.70 0.00 587.70 0.4992 0.00 0.,00 0..000 23.478 605.71 19.804 576.99
5 8.128 11,.350 586.98 586.98 0.00 586.98 0.4993 0.00 0..00 0..000 23.480 603.90 19.804 575.25
6 7 . 740 11 .350 585.18 535.18 0.00 585.18 0.4981 0.00 -0..01 0.,000 23.466 603.01 19.808 574.54
7 7.329 11 .350 581.21 581.21 0.00 581.21 0.4949 0.00 -0..02 0..001 23.415 602.14 19.807 574.05
8 6.891 11.350 575.11 575.11 0..00 575.11 0.4898 0.00 -0,,05 0,.001 23.340 601.41 19.810 573,91
9 6.416 11 . 350 564.97 564.97 0..00 564.97 0.4811 0.00 -0,.09 0,.002 23.223 600.50 19.821 573.96
10 5.888 11.350 542.96 542.96 0..00 542.96 0.4618 0.00 -0,.14 0.003 22.954 600.01 19,832 575.48
11 5.266 11.350 496.84 496.84 • 0..00 496.84 0.4212 0.00 -0 ,15 0.003 22.407 599.70 19.832 579.17
HUB 5.140 11.350
Table 10.2 Continued

** VALUES OF PARAMETERS ON STREAMLINES AT STATION. 19, WHICH IS AN ANNULUS «*

STREAMLINE AXIAL AXIAL HERD. TANG. ABS. ABS. ABS.FLOW STREAM. STREAM. TOTAL TOTAL STATIC STATIC
HO. RADIUS COORD. VEL. VEL VEL VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
(IN.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IH.) (PSIA) (DEG.R.) (PSIA) (DEG.R.)
TIP 9.600 14.440
1 9.531 14.440 588.97 588.97 0.00 588.97 0.4950 0.00 0.00 -0.000 23.421 618.17 19.811 589.35
2 9.198 14.440 587.87 587.87 0.00 587.87 0.4974 0.00 -0.00 0.000 23.458 610.17 19.810 581.45
J 8.85<i 14 .440 587.41 587.41 0.00 587.41 0.4980 0.00 -0.00 -0.000 23.468 607.94 19.812 579.26
4 8.498 14.440 587.04 587.04 0.00 587.04 0.4986 0.00 -0.00 -0.000 23.478 605.72 19.811 577.07
5 8.127 14.440 586.39 586.39 0.00 586.39 0.4988 0.00 -0.00 -0.000 23.480 603.93 19.811 575.34
6 7.738 14.440 584.68 584.68 0.00 584.68 0.4976 0.00 0.00 -0.000 23.466 603.04 19.814 574.62
7 7.327 14.440 580.90 580.90 0.00 580.90 0.4946 0.00 0.00 -0.000 23.415 602.17 19.810 574.12
& &.888 14.440 575.08 575.08 0.00 575.08 0.4897 0.00 0.01 -0.000 23.340 601.45 19.810 573.95
9 6.414 14.440 565.44 565.44 0.00 565.44 0.4815 0.00 0.02 -0.000 23.223 600.54 19.816 573.95
10 5.886 14. 440 544.29 544. 29 0. 00 544.29 0.4629 0.00 0.04 -0.001 22.954 600.05 i9.<SlV 575.40
•n 5.266 14.440 499.70 499.70 0.00 499.70 0.4237 0.00 0.05 -0.001 22.407 599.74 19.803 578.97
HUB 5.140 14.440

** VALUES OF PARAMETERS ON STREAMLINES AT STATION, 20, WHICH IS AN ANNULUS xx

STPF.inLINr. AXIAL AXIAL HERD. TANG. ABS. ABS. ABS.FLOU STREAM. STREAM. TOTAL TOTAL STATIC STATIC
1,0. HADIUi, coonD. VEL. VEL. VEL. VEL. MACH NO. ANGLE SLOPE CURV. PRESS. TEMP. PRESS. TEMP.
tlli.) (IN.) (FT/SEC) (FT/SEC) (FT/SEC) (FT/SEC) (DEG) (DEG) (l./IN.) (PSIA) (DEG.R.) (PSIA) (DEG • R.)
TIP 9.600 16.000
1 9.531 16.000 588.85 588.85 0.00 588.85 0.4950 0.00 -0.00 -0.000 23.421 618.01 19.811 589.20
2 9.198 16 .1)00 587.82 587.82 0.00 587.82 0.4974 0.00 -0.00 -0.000 23.458 610.16 19.811 581.44
Î 8.854 16.000 587.39 587.39 0.00 587.39 0.4979 0.00 -0.00 -0.000 23.468 607.95 19.812 579.26
4 8.498 16.000 587.03 587.03 0.00 587.03 0.4986 0.00 -0.01 -0.000 23.478 605.73 19.ri? 577.08
5 8.127 16.000 586.38 586.38 0.00 586.SB 0.4988 0.00 -0.01 -0.000 23.480 603.94 19.Six 575.36 '
6 7.738 16.000 584.71 584.71 0.00 584.71 0.4977 0.00 -0.01 -0.000 23.466 601.06 19.&14 574.63
7 7 .327 16.000 580.94 580.94 0.00 580.94 0.4947 0.00 -0.01 -0.000 23.415 602.19 19.810 574.13
8 6.888 16.000 575.14 575.14 0.00 575.14 0.4898 0.00 -0.02 -0.000 23.340 601.46 19.810 573,96
9 6.414 16.000 565.49 565.49 0.00 565.49 0 .4816 0.00 -0.02 -0.000 23.223 600.56 19.816 573.97
10 5.886 16.000 544.32 544.32 0.00 544.32 0.4630 0.00 -0.03 -0.001 22.954 600.06 19.817 575.41
11 5.266 16.000 499.62 499.62 0.00 499.62 0.4236 0.00 -0.04 -0.001 22.407 599.75 19.804 578.99
HJB 5.140 16.000
Table 10.3 Design code stage and overall performance predictions

»«•« COMPUTED COMPRESSOR DESIGN PARAMETERS FOR A ROTRTIOHAL SPEED OF, 16042.RPM «**

"• THE CORRECTED UEIGHTFLOW PER UHIT OF CASING ANNULAR AREA AT THE INLET FACE OF THE FIRST BLADE ROW IS 38.78 LBS/SEC/FT SQ «•

«* MASS AVERAGED ROTOR AMD STAGE AERODYNAMIC PARAMETERS **

TOTAL TOTAL FOR. AX. GAS BENDING MOMENTS
STAGE BLADE FLOW HKAD ID. HF.AO PRESS. TEnp. ADIA. POLY. ASPECT SHAFT FOR EACH BLADE TORQUE POWER
MO. TÏI'E COKF. COEF. COEF. RATIO RATIO ERF. EKF. RATIO THRUST FOR. AX. TANG.
(LBS) (FT-LBS) (FT-LBS) (FT-LBS) (HP)
1 BOTOR 0.4280 0.2355 0.2600 1.6293 1.1652 0.9060 0.9122 1.55 1048.80 17.698 -12.650 698.10 2132.37
1 STATOR 0.4152 0.2229 0.2600 1.5900 1,1652 0.8575 0.6665 2.05 -242.80 2.310 6.067

** CUMULATIVE SUMS OF MASS AVERAGED ROTOR AND STAGE AERODYDAttIC PARAMETERS **

TOTAL TOTAL FOR. AX.
STAGE BLADE WEIGHT TOTAL TOTAL PRESS. TEMP. HEAD IDEAL HEAD ADIA. POLY. SHAFT TORQUE POWER FRACT
NO, TYPE FLOW PRESS. TEtlP. RATIO RATIO COEF. COEF. EFF. EFF. THRUST ENERGY
(LBS/SEC) (PSIA) (DEG. R.) • (LBS) (FT-LBS) (HP)
1 INLET 73.30 14.666 518.70
1 ROTOR 73.30 23.896 604.38 1.6293 1.1652 0.2355 0.2600 0.9060 0.9122 1043.80 698.10 2132.37 1.0000
1 STATOR 73.30 23.318 604.38 1.5900 1.1652 0.2229 0.2600 0.8575 0.8665 806.00
 Table 10.4 Rotor blade manufacturing coordinates generated by NASA design code

** BLADE SECTION PROPERTIES OF ROTOR HO. 1 «•

NUMBER OF BLADES = 22.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 0.941 IN.
BLAÛE SECTION STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
RAD. COORDINATES SETTIHG C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
NO. LOC. L H ANGLE L H IMIH IMAX ANGLE CONSTANT STIFFNESS
(IN.) (IN.) (IN.) (DEC.) (IN.) (IN.) (IN.)**2 (IN.)**4 (IN.)»*4 (DEC.) (IH.)»»4 (IH.)»-6
1 10.100 1.9107 -0.0010 69.835 1.9107 -0.0010 0.27895 0.0002066 0.215474 69.339 0.0007290 0.1777493
2 9.750 1.9210 0.0330 64.181• 1.9210 0.0330 0.27792 0.0001906 0.211467 64.201 0.0007287 0.1748674
1 9.200 1.9160 0.0521 59.444 1.9160 0.0521 0.29232 0.0002576 0.219638 59.552 0.0008587 0.1837134
4 a.450 1.9225 0.0560 56.711 1.9225 0.0560 0.32400 0.0003674 0.240418 56.493 0.0011897 0.2030173
SECTION NO. 1 COORDINATES SECTION NO. 2 COORDINATES SECTION NO. 3 COORDINATES SECTION HO. 4 COORDINATES
L HP HS L HP HS L HP HS L HP HS
(IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IH.) (IN.) (IN.) (IN.) (IH.)
0.0000 0.0073 0.0073 0.0000 0.0070 0.0070 0.0000 0.0071 0.0071 0.0000 0.0078 0.0078
0.0069 99.9999 0.0146 0.0064 99.9999 0.0139 0.0063 99.9999 0.0142 0.0068 99.9999 0.0156
0.0070 0.0000 99.9999 0.0070 0.0000 99.9999 0.0071 0.0000 99.9999 0.0081 0.0000 99.9939
0.1000 -0.0002 0.0222 0.1000 0.0011 0.0229 0.1000 0.0019 0.0242 0.1000 0.0029 0.0271
0.2C00 -0.0010 0.0292 0.2000 0.0018 0.0315 0.2000 0.0037 0.0343 0.2000 0.0057 0.0388
0.1000 -0.0025 0.0351 0.1000 0.0022 0.0393 0.1000 0.0052 0.0416 0.3000 0 no?2 0.0497
0.4000 -0.0046 0.0400 0.4000 0.0021 0.0464 0.4000 0.0064 0.0524 0.4000 0.0101 0.0598
0.5000 -0.0072 0.0440 0.5000 0.0017 0.0526 0.5000 0.0075 0.0605 0.5000 0.0120 0.0692
0.6000 -0.0102 0.0471 0.6000 0.0011 0.0582 0.6000 0.0083 0.0679 0.6000 0.0115 0.0779
0.7000 -0.0135 0.0494 0.7000 0.0002 0.0631 0.7000 0.0089 0.0748 0.7000 0.0146 0.0859
0.8000 -0.0172 0.0510 0.8000 -0.0010 0.0674 0.8000 0.0093 0.0811 0.8000 0.0153 0.0932
0.9000 -0.0212 0.0519 0.9000 -0.0023 0.0712 0.9000 0.0095 0.0868 0.9000 0.0158 0.0999 03
1.0000 -0.0254 0.0523 1.0000 -0.0037 0.0744 1.0000 0.0096 0.0920 1.0000 0.0160 0.1058 rvj
1.1000 -0.0297 0.05"21 1.1000 -0.0053 0.0771 1.1000 0.0095 0.0966 1.1000 0.0160 0.1111
].2000 -0.0340 0.0515 1.2000 -0.0069 0.0793 1.2000 0.0093 0.1037 1.2000 0.0156 0.1158
I.ICOO -0.0385 0.0504 1.3000 -0.0086 0.0812 1.3000 0.0090 0.1043 1.3000 0.0150 0.1198
1.4000 -0.0428 0.0490 1.4000 -0.0103 0.0826 1.4000 0.0087 0.1075 1.4000 0.0142 0.1232
1.5000 -0.0471 0.0474 1.5000 -0.0119 0.0838 1.5000 0.0082 0.1101 1.5000 0.0132 0.1261
1.t 00: -0.0512 0.0455 1.6000 -0.0134 0.0846 1.6000 0.0077 0.1124 1.6000 0.0120 0.1283
1.7000 -0.0552 0.0436 1.7000 -0.0148 0.0852 1.7000 0.0073 0.1142 1.7000 0.0106 0.1300
1.8000 -0.0588 0.0415 1.8000 -0.0161 0.0856 1.8000 0.0068 0.1157 1.8000 0.0091 0.1312
1.9000 -0.0621 0.0393 1.9000 -0.0171 0.0859 1.9000 0.0062 0.1167 1.9000 0.0075 0.1319
2.0000 -0.0649 0.0173 2.0000 -0.0179 0.0860 2.0000 0.0058 0.1175 2.0000 0.0058 0.1321
2.1000 -0.0673 0.0359 2.1000 -0.0183 0.0862 2.1000 0.0059 0.1180 2.1000 0.0039 0.1316
2.2000 -0.0691 0.0349 2.2000 -0.0183 0.0863 2.2000 0.0061 0.1182 2.2000 0.0021 0.1305
2.3000 -0.0705 0.0330 2.3000 -0.0182 0.0864 2.3000 0.0051 0.1181 2.3000 0.0007 0.1286
2.4000 -0.0711 0.0108 2.4000 -0.0177 0.0866 2.4000 0.0051 0.1174 2.4000 -0.0008 0.1256
2.5000 -0.0696 0.0323 2.5000 -0.0158 0.0874 2.5000 0.0058 0.1158 2.5000 -0.0021 0.1216
2.6000 -0.0658 0.0388 2.6000 -0.0141 0.0889 2.6000 0.0059 0.1130 2.6000 -0.0012 0.1167
2.7000 -0.0611 0.0407 2.7000 -0.0125 0.0872 2.7000 0.0057 0.1091 2.7000 -0.0041 0.1109
2.800: -0.0567 0.0387 2.8000 -0.0110 0.0846 2.8000 0.0055 0.1041 2.8000 -0.0048 0.1043
2.9000 -0.0522 0.0419 2.9000 -0.0097 0.0814 2.9000 0.0054 0.0980 2.9000 -0.0052 0.0969
1.0000 -0.0474 0.0429 1.0000 -0.0085 0.0767 3.0000 0.0052 0.0908 3.0000 -0.0054 0.0886
3.1000 -0.0421 0.0404 3.1000 -0.0073 0.0704 3.1000 0.0048 0.0824 1.1000 -0.0051 0.0795
3.2000 -0.0362 0.0369 3.2000 -0.0062 0.0626 3.2000 0.0044 0.0728 3.2000 -0.0050 0.0696
3.3000 -0.0295 0.0329 3.3000 -0.0050 0.0535 3.3000 0.0038 0.0619 3.3000 -0.0043 0.0589
3.4000 -0.0217 0.0277 3.4000 -0.0037 0.0428 3.4000 0.0030 0.0497 3.4000 -0.0034 0.0475
3.5000 -0.0127 0.0213 3.5000 -0.0022 0.0305 3.5000 0.0019 0.0361 3.5000 -0.0023 0.0352
3.6000 -0.0021 0.0139 3.6000 -0.0005 0.0166 3.6000 0.0007 0.0210 3.6000 -0.0008 0.0222
3.6182 0.0001 99.9999 3.6247 0.0000 99.9999 3.6447 0.0010 99.9999 3.6491 0.0000 99.9999
1.6211 99.9999 0.0122 3.6273 99.9999 0.0125 3.6465 99.9999 0.0135 3.6505 99.9999 0.0153
3.6253 0.0063 0.0063 1.6319 0.0064 0.0064 3.6519 0.0068 0.0068 3.6569 0.0077 0.0077

i
Table 10.4 Continued

»* BLADE SECTION PROPERTIES OF ROTOR NO. 1 **

NUMBER OF BLADES = 22.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 0.941 IN.
BLADE SECTION STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
RAD. COORDINATES SETTING C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
HO. LOC. L H ANGLE L H IMIN IMAX ANGLE CONSTANT STIFFNESS
(IN.) (IN.) (IN.) (DEG.) (IN.) (IN.) (IN.)**2 (IK.)*«4 (IH.)**4 (DEG.) (IH.)»»4 (IN.)'"6
9 6.150 1.Î8<.9 0.2149 35.522 1.8849 0.2149 0.64225 0.0045910 0.475851 35.118 0.0090665 0.4015445
10 5.650 1.8439 0.2731 28.553 1.8439 0.2731 0.71637 0.0074531 0.532403 28.683 0.0123591 0.4387047
11 5.150 1.8368 0.3554 21.417 1.8368 0.3554 0.77157 0.0118065 0.567879 22.015 0.0153858 0.4638299
12 4 «.50 1.8388 0.4473 14.489 1.8388 0.4473 0.81317 0.0175734 0.594534 15.523 0.0178553 0.4915950
SECTION NO. 9 COORDINATES SECTION NO. 10 COORDINATES SECTION NO. 11 COORDINATES SECTION NO. 12 COORDINATES
L HP HS L HP HS L HP HS L HP HS
(IN.) (IN.) (Hi.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.)
0.0000 0.0140 0.0140 0.0000 0.0153 0.0153 0.0000 0.0164 0.0164 0.0000 0.0172 0.0172
0.0095 99.9999 0.0273 0.0090 99.9999 0.0292 0.0083 99.9999 0.0306 0.0076 99.9999 0.0315
0.0161 0.0001 99.9999 0.0180 0.0002 99.9999 0.0205 0.0005 99.9999 0.0231 0.0010 99.9999
0.1000 0.0164 0.0631 0.1000 0.0189 0.0748 0.1000 0.0254 0.0871 0.1000 0.0337 0.0994
0.2000 0.0344 0.1003 0.2000 0.0405 0.1218 0.2000 0.0551 0.1445 0.2000 0.0740 0.1678
0.3000 0.0508 0.1353 0.3000 0.0607 0.1655 0.3000 0.0832 0.1979 0.3000 0.1118 0.2314
0.4000 0.0657 0.1679 0.4000 0.0796 0.2061 0.4000 0.1096 0.2475 0.4000 0.1472 0.2906
0.5000 0.0792 0,1983 0.5000 0.0971 0.2436 0.5000 0.1343 0.2935 0.5000 0.1800 0.3456
0.6000 0.0913 0.2265 . 0.6000 0.1133 0.2781 0.6000 0.1573 0.3360 0.6000 0.2104 0.3966
0.7000 0.1021 0.2524 0.7000 0.1282 0.3101 0.7000 0.1784 0.3748 0.7000 0.2381 0.4436
0.8000 0.1115 0.2761 0.8000 0.1419 0.3396 0.8000 0.1974 0.4103 0.8000 0.2635 0.4868
0.9000 0.1197 0.2977 0.9000 0.1543 0.3656 0.9000 0.2151 0.4434 0.9000 0.2868 0.5261
1.0000 0.1268 0.3171 1.0000 0.1648 0.3880 1.0000 0.2313 0.4729 1.0000 0.3080 0.5614
1.1000 0.1326 0.3344 1.1000 0.1747 0.4096 1.1000 0.2456 0.4988 1.1000 0.3273 0.5930
1.2000 0.1372 0.3496 1.2000 0.1836 0.4284 1.2000 0.2584 0.5216 1.2000 0.3444 0.6210
1.3000 0.1408 0.36Z7 1.3000 0.1910 0.4439 1.3000 0.2696 0.5415 1.3000 0.3595 0.6454
1.4000 0.1434 0.3735 1.4000 0.1971 0.4566 1.4000 0.2792 0.5584 1.4000 0.3726 0.6663
1.5000 0.1452 0.3820 1.5000 0.2020 0.4671 1.5000 0.2871 0.5721 1.5000 0.3835 0.6836
1.6000 0.1461 0.3883 1.6000 0.2058 0.4750 1.6000 0.2933 0.5829 1.6000 0.3924 0.6974
1.7000 0.1463 0.3924 1.7000 0.2083 0.4805 1.7000 0.2979 0.5907 1.7000 0.3991 0.7077
1.8000 0.1456 0.3943 1.8000 0.2095 0.4833 1.8000 0.3007 0.5954 1.8000 0.4037 0.7144
1.9000 0.1441 0.3941 1.9000 0.2096 0.4837 1.9000 0.3019 0.5972 1.9000 0.4059 0.7175
2.0000 0.1418 0.3917 2.0000 0.2084 0.4816 2.0000 0.3012 0.5958 2.0000 0.4059 0.7170
2.1000 0.1387 0.3872 2.1000 0.2059 0.4769 2.1000 0.2988 0.5913 2.1000 0.4034 0.7127
2.2000 0.1349 0.3806 2.2000 0.2022 0.4696 2.2000 0.2944 0.5835 2.2000 0.3985 0.7045
2.3000 0.1303 0.3719 2.3000 0.1972 0.4598 2.3000 0.2882 0.5725 2.3000 0.3910 0.6923
2.4000 0.1249 0.3611 2.4000 0.1909 0.4472 2.4000 0.2800 0.5580 2.4000 0.3807 0.6758
2.5000 0.1188 0.3481 2.5000 0.1832 0.4320 2.5000 41.2697 0.5399 2.5000 0.3676 0.6549
2.6000 0.1120 0.3330 2.6000 0.1743 0.4139 2.6000 0.2573 0.5181 2.6000 0.3515 0.6293
2.7000 0.1045 0.3158 2.7000 0.1640 0.3930 2.7000 0.2427 0.4923 2.7000 0.3322 0.5986
2.8000 0.0962 0.2963 2.8000 0.1522 0.3690 2.8000 0.2257 0.4623 2.8000 0.3094 0.5625
2.9000 0.0873 0.2746 2.9000 0.1391 0.3418 2.9000 0.2063 0.4278 2.9000 0.2831 0.5204
3.0300 0.0777 0.2506 3.0000 0.1244 0.3114 3.0000 0.1843 0.3885 3.0000 0.2527 0.4718
3.1000 0.0675 0.2243 3.1000 0.1082 0.2774 3.1000 0.1594 0.3435 3.1000 0.2181 0.4159
3.2000 0.0566 0.1956 3.2000 0.0905 0.2396 3.2000 0.1317 0.2932 3.2000 0.1788 0.3517
3.3C00 0.0451 0.1644 3.3000 0.0711 0.1978 3.3000 0.1007 0.2362 3.3000 0.1344 0.2783
3.4000 0.0329 0.1306 3.4000 0.0499 0.1516 3.4000 0.0663 0.1718 3.4000 0.0843 0.1940
3.5000 0.0202 0.0941 3.5000 0.0270 0.1005 3.5000 0.0281 0.0992 3.5000 0.0279 0.0969
3.6000 0.0068 0.0548 3.6000 0.0021 0.0442 3.5659 0.0008 99.9999 3.5428 0.0016 99.9999
3.6490 0.0001 99.9999 3.6068 0.0003 99.9999 3.5814 99.9999 0.0330 3.5613 99.9999 0.0300
3.6566 99.9999 0.0312 3.6184 99.9999 0.0331 3.5892 0.0181 0.0181 3.5671 0.0172 0.0172
3.6669 0.0162 0.0162 3.6278 0.0176 0.0176
Table 10.4 Continued

•» BLADE SECTION PROPERTIES OF ROTOR NO. 1 **

NUMBER OF BLADES = 22.0
AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 0.941 IN.
BLADE SECTION STACKING POINT SECTION BLADE SECTION
RAD. COORDINATES SETTING SECTION MOMENTS OF IHERTIA IMAX SECTION SECTION
NO. LOC. C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION
L H ANGLE L H TWIST
(IN.) (IN.3 (IN.) (DEG.) (IN.) (IN.) ANGLE CONSTANT STIFFNESS
5 a.150 1.9332 0.0695 53.964 (IH.)**2 (IN^"»f»4 (inf)*<I4 (DEG.) (IN.)*"'4 (IN.)*»6
6 7.650 1.9332 0.0695 0.36699 0.0005709 0.268899
7 7.150
1.9294 0.0922 50.491 1.9284 0.0922 0.42462 0.0009668 0.308470
53.591 0.0017605 0.2288616
1.9178 0.1233 46.376 1.9178 0.1233 49.993 0.0027434 0.2622730
A 6.650 1.9090 0.1633 41.511 0.49100 0.0016445 0.355827 45.871 0.0042279 0.3017055
1.9090 0.1633 0.56203 0.0027611 0.4077:3 40.951 0.0062901 0.3447527
SECTION NO. 5 COORDINATES SECTION NO. 6 COORDINATES
L HP HS SECTION HO. 7 COORDINATES SECTION NO. 8 COORDINATES
(IN.) (IN.) (IN.) , !• HP HS L HP HS
(IN.) (IN.) (IN.) (IN.) (IN.) (IN.) L HP HS
0.0000 o.ooss o.ooas 0.0000 0.0100 0.0100 (IN.) (IN.) (IN.)
0.0075 99.9999 0.0175 0.0083 99.9999 0.0198 0.0000 0.0113 0.0113 0.0000 0.0127 0.0127
0.0092 0.0000 99.9999 0.0089 99.9999 0.0223 0.0093 99.9999 0.0249
0.0106 0.0000 99.9999 0.0123 0.0000 99.9999
0.1000 0.0046 0.0312 0.1000 0.0068 0.0371 0.0142 0.0001 99.9999
0.2000 0.0091 0.0452 0.1000 0.0092 0.0440 0.1000 0.0128 0.0526
0.2000 0.0137 0.0549 0.2000 0.0188 0.0666
0.3000 0.0131 0.0584 0.3000 0.0198 0.0716 0.2000 0.0265 0.0815
0.4000 0.0165 0.0706 0.4000 0.0253 0.0873 0.3000 0.0275 0.0878 0.3000 0.0390 0.1086
0.5000 0.0195 0.0820 0.4000 0.0353 0.1077 0.4000 0.0504 0.1341
0.5000 0.0300 0.1019 0.5000 0.0422 0.1262
0.60CO 0.0220 0.0926 0.6000 0.0341 0.1154 0.5000 0.0605 0.1579
0.7000 0.0240 0.1023 0.6000 0.0483 0.1435 0.6000 0.0695 0.1801
0.7000 0.0374 0.1279 0.7000 0.0535 0.1595
0.8000 0.0255 0.1112 0.8000 0.0402 0.1394 0.7000 0.0774 0.2006
0.9000 0.0266 0.1193 0.8000 0.0579 0.1743 0.8000 0.0841 0.2194
0.9000 0.0422 0.1498 0.9000 0.0614 0.1877
I.0000 0.0272 0.1266 1.0000 0.0437 0.1593 0.9000 0.0897 0.2367 CD
1.1000 0.0274 0.1331 1.0000 0.0642 0.2000 1.0000 0.0942 0.2524
1.1000 0.0445 0.1677 1.1000 0.0661 0.2110
1.2000 0.0271 0.1388 1.2000 0.0448 0.1752 1.1000 0.0976 0.2666
1.3000 0.0265 0.1438 1.2000 0.0673 0.2208 1.2000 0.0999 0.2790
1.3000 0.0444 0.1817 1.3000 0.0677 0.2294
1.4000 0.0255 0.1481 1.4000 0.0436 0.1873 0 1012 0.2898
1.5000 0.0241 0.1516 1.4000 0.0674 0.2369 1.4000 0.1015 0.2996
1.5000 0.0421 0.1919 1.5000 0.0663 0.2432
1.6000 0.0224 0.1545 1.6000 0.0402 0.1957 1.6000 0.0647 0.2483 1.5000 0.1011 0.3073
1.7000 0.0205 3.1567 1.7000 0.0379 0.1986 1.7000 0.0627 0.2519 1.6000 0.1002 0.3127
1.8000 0.0182 0.1582 1.8000 0.0351 0.2005 1.8000 0.0604 0.2537 1.7000 0.0987 0.3163
1.9000 0.0155 0.1591 1.9000 0.0322 0.2011 1.8000 0.0967 0.3181
2.0000 0.0129 0.1591 1.9000 0.0578 0.2539 1.9000 0.0943 0.3179
2.0000 0.0292 0.2003 2.0000 0.0549 0.2525
2.1000 0.0103 0.1582 2.1000 0.0262 0.1981 2.0000 0.0913 0.3159
2.2000 0.0078 0.1560 2.1000 0.0518 0.2496 2.1000 0.0879 0.3120
2.2000 . 0.0232 0.1947 2.2000 0.0486 0.2451
2.3000 0.0054 0.1528 2.3000 0.0202 0.1900 2.2000 0.0840 0.3064
2.4000 0.0033 0.1485 2.3000 0.0451 0.2392 2.3000 0.0798 0.2989
2.4000 0.0172 0.1841 2.4000 0.0415 0.2317
2.5000 0.0013 0.1432 2.5000 0.0143 0.1769 2.5000 0.0378 0.2228 2.4000 0.0752 0.2897
2.6000 -0.0005 0.1368 2.6000 0.0116 0.1686 2.5000 0.0702 0.2786
2.7000 -O.OC20 0.1296 2.6000 0.0341 0.2124 2.6000 0.0649 0.2659
2.7irOO 0.0090 0.1592 2.7000 0.0303 0.2007 2.7000 0.0593 0.2514
2.8000 -0.0033 0.1213 2.8000 0.0066 0.1487 2.8000 0.0265 0.1875
2.9000 -0.0042 0.1122 2.9000 0.0044 0.1371 2.8000 0.0535 0.2351
3.0000 -0.0049 0.1022 2.9000 0.0227 0.1730 2.9000 0.0475 0.2171
3.0000 0.0025 0.1245 3.0000 0.0190 0.1572
3.1000 -0.0052 0.0913 3.1000 0.0009 0.1108 3.0000 0.0413 0.1974
3.2000 -0.0051 0.0796 3.1000 0.0154 0.1400 3.1000 0.0349 0.1759
3.2000 -0.0003 0.0963 3.2000 0.0120 0.1216 3.2000 0.0284 0.1526
3.3000 -0.0047 0.0671 3.3000 -0.0011 0.0808 3.3000 0.0088 0.1018
3.4000 -0.0039 0.0539 3.4000 -0.0014 0.0644 3.3000 0.0219 0.1276
3.5000 -0.0026 0.0398 3.4000 0.0058 0.0809 3.4000 0.0154 0.1009
3.5000 -0.0013 0.0472 3.5000 0.0031 0.0586
3.6000 -0.0009 0.0251 3.6000 -0.0005 0.0292 3.5000 0.0089 0.0723
3.6462 -0.0000 99.9999 3.6000 0.0008 0.0352 3.6000 0.0025 0.0418
3.6420 0.0000 99.9999 3.6413 0.0000 99.9999 3.6385 0.1016 99.9999
3.6478 99.9999 0.0178 3.6443 99.9999 0.0209 3.6448 99.9999 0.0243
3.6552 0.0090 0.0090 3.6527 0.0106 0.0106 3.6437 99.9999 0.0279
3.6540 0.0124 0.0124 3.6537 0.0143 0.0143
Table 10.4 Continued

«« BLADE SECTION PROPERTIES OF ROTOR HO. 1 "«

NUMBER OF BLADES = 22.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 0.911 IH.
BLADE SECTION STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
RAD. COORDINATES SETTING C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
NO. LOC. L H ANGLE L K iniN iriAX ANGLE CONSTANT STIFFNESS
CIH. ) (IN.) (IN.) (DEG.) (IN.) (IN.) (IH.)**2 (IH.)«»4 {IN.)**4 (DEG.) (IN.]M4 (IK.)««5
1Î <,.200 1.SÎ88 0.5192 8.880 1.8388 0.5192 0.83726 0.0231597 0.613628 10.392 0.0191505 0.5182821
14 3.700 1.8317 0.5610 3.836 1.8317 0.5610 0.84587 0.0294368 0.630241 6.240 0.0190949 0.5192583
15 4.302 1.8392 0.5048 10.083 1.8392 0.5048 0.83301 0.0219077 0.609606 11.466 0.0189553 0.5122277
SECTION NO. 13 COORDINATES SECTION NO. 14 COORDINATES SECTION NO. 15 COORDINATES
L HP US L HP HS L HP HS
(IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.)
0.0000 0.0177 0.0177 0.0000 0.0180 0.0180 0.0000 0.0176 0.0176
0.0071 99.9999 0.0319 0.0066 99.9999 0.0320 0.0072 99.9999 0.0318
0.0251 0.0016 99.9999 0.0267 0.0022 99.9999 0.0247 0.0015 99.9999
0.1000 0.0414 0.1095 0.1000 0.0487 0.1183 0.1000 0.0397 0.1073
0.2000 0.0916 0.1874 0.2000 0.1076 0.2036 0.2000 0.0877 0.1832
0. 3000 0.1382 0.2598 0.3000 0.1616 0.2870 0.3000 0.1325 0.2536
0.4000 0.1809 0.3270 0. 4000 0.2155 0.3672 0.4000 0.1733 0.3188
0.SCCO 0.2212 0.3920 0.5000 0.2543 0.4214 0.5000 0.2124 0.3825
0.6000 0.2589 0.4505 0.6000 0.2906 0.4943 0.6000 0.2497 0.4395
0.7000 0.2901 0.5025 0.7000 0.3351 0.5622 0.7000 0.2789 0.4892
0.8000 0.3203 0.5521 0.8000 0.3722 0.6152 0.8000 0.3078 0.5375 09
0.9000 0.3491 0.5974 0.9000 0.4035 0.6620 0.9000 0.3356 0.5820
1.0000 0.3751 0.6377 1.0000 0.4334 0.7067 1.0000 0.3607 0.6212
1.1000 0.3985 0.6736 1.1000 0.4621 0.7474 1.1000 0.3831 0.6561
1. 2000 0.4196 0.,7056 1.,2000 0..4882 0.7828 1.2000 0.4032 0.6873
I.3000 0.,4385 0..7337 1.,3000 0..5115 0.8134 1.3000 0.4212 0.7146
1..4000 0..4549 0.,7577 1.,4000 0..5323 0.8399 1.4000 0.4369 0.7380
1.• fiopn 0..4690 0..7778 1.,5000 0,.5505 0.8624 1.5000 0.4502 0.7575
1..6000 0..4805 0..7940 1..6000 0..5659 0.8808 1.6000 0.4611 0.7732
]..7000 0..4896 0.8063 1..7000 0..5785 0.8951 1.7000 0.4695 0.7850
1..8000 0..4961 0.8147 1,.8000 0.5882 0.9053 1.8000 0.4755 0.7930
1..9000 0..4999 . 0 .8191 1.9000 0.5949 0.9113 1.9000 0.4789 0.7971
2.0000 0 .5009 0.8194 2.0000 0.5984 0.9130 2.0000 0.4796 0.7972
2.1000 0 .4990 0.8155 2.1000 0.5986 0.9103 2.1000 0.4775 0.7932
2.2000 0.4942 0.8072 2.2000 0.5954 0.9031 2.2000 0.4726 0.7849
2 .3000 0.4862 0.7944 2.3000 0.5885 0.8910 2.3000 0.4646 0.7721
2.4000 0.4748 0.7769 2.4000 0.5777 0.8739 2.4000 0.4533 0.7547
2.5000 0.4599 0.7545 2.5000 0.5627 0.8514 2.5000 0.4387 0.7324
2 .6000 0 .4412 0.7262 2.6000 0.5433 0.8231 2.6000 0.4205 0.7048
2.7001 0.4185 0.6924 2.7000 0.5190 0.7885 2.7000 0.3984 0.6714
2.8000 0.3914 0.6521 2.8000 0.4894 0.7470 2.8000 0.3722 0.6319
2 . 9000 0 .3595 0.6048 2.9000 0.4539 0.6977 2.9000 0.3415 0.5856
3.0000 0.3225 0.5497 3.0000 0.4119 0.6395 3.0000 0.3058 0.5317
3.1000 0.2797 0.4856 3.1000 0.3624 0.5713 3.1000 0.2648 0.4692
3.2000 0.2305 0.4114 3.2000 0.3043 0.4906 3.2000 0.2177 0.3970
Î.3000 0.1742 0.3250 3.3000 0.2362 0.3949 3.3000 0.1640 0.3133
3.4000 0.1096 0.2243 3.4000 0.1558 0.2796 3.4000 0.1027 0.2161
3.5000 0.0354 0.1057 3.5000 0.0596 0.1363 3.5000 0.0327 0.1023
3.5397 0.0030 99.9999 3.5473 0.0072 99.9999 3.5392 0.0026 99.9999
3.5602 99.9999 0.0237 3.5688 99.9999 0.0125 3.5593 99.9999 0.0255
3.5633 0.0147 0.0147 3.5688 0.0111 0.0111 3.5631 0.0154 0.0154
Table 10.5 Stator blade manufacturing coordinates generated by NASA design code

«« BLADE SECTION PROPERTIES OF STATOR NO. 1 FOLLOWING ROTOR NO. 1 **

NUMBER OF BLADES = 34.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 5.200 IN.
BLADE SECTION STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
J-.HD. COORDINATES SETTING C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
NO. LOC. L H ANGLE L H IMIN IMAX ANGLE CONSTANT STIFFNESS
(IH.) (IN.) (IN.) (DEC.) (IN.) (IN.) (IN.)"*2 (IH.)*«4 (IN.)**4 (DEC.) (IH.)»»4 CIN.)*'>6
9 5.075 1.1349 0.2522 17.152 1.1349 0.2522 0.23510 0.0016577 0.072133 17.569 0.0010299 0.0241935
10 1.750 1.1341 0.2791 16.929 1.1341 0.2791 0.23051 0.0019127 0.070614 17.521 0.0009634 0.0236715
11 9.600 1.1352 0.2291 10.180 1.1352 0.2291 0.32621 0.0023691 0.10*565 10.189 0.0025315 0,0372960
SECTION NO. 9 COORDINATES SECTION NO. 10 COORDINATES SECTION HO. 11 COORDINATES
L HP HS L HP HS L HP HS
(IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (m.) (IN.) (IN.)
o.cooo 0.0126 0.0126 0.0000 0.0114 0.0114 0.0000 0.0297 0.0297
0.0053 99.9999 0.0229 0.0044 99.9999 0.0204 0.0134 99.9999 0.0545
0.0179 0.0012 99.9999 0.0167 0.0013 99.9999 0.0405 0.0020 99.9999
0.1030 0.0372 0.0853 0.1000 0.0428 0.0896 0.1000 0.0247 0.1086
0.2000 0.0771 0.1438 0.2000 0.0880 0.1536 0.2000 0.0596 0.1638
0.3000 0.1127 0.1952 0.3000 0.1285 0.2097 0.3000 0.0905 0.2120 CO
0.4000 0.1443 0.2401 0.4000
0.5000
0.1641
0.1954
0.2586
0.3010
0.4000
0.5000
0.1176
0.1409
0.2536 a<
0.5000 0.1717 0.2789 0.2892
0.6000 0.1952 0.3118 0.6000 0.2218 0.3365 0.6000 0.1608 0.3191
0.7000 0.2148 0.3390 0.7000 0.2441 0.3661 0.7000 0.1773 0.3436
0.8000 0.2306 0.3606 0.8000 0..2622 0.3897 0.8000 0.1897 0.3626
0.9000 0.2426 0.3769 0.9000 0..2760 0.4075 0.9000 0.1997 0.3770
1.0000 0..2509 0.3879 1.0000 0..2856 0.4196 1.0000 0.2058 0.3861
1.1000 0.2553 0.3935 1.1000 0..2910 0.4261 1.1000 0.2087 0.3903
1.2000 0.,2559 0.3940 1.2000 0,.2920 0.4270 1.2000 0.2082 0.3897
1.3000 0..2527 0.3891 1.3000 0.2887 0.4221 1.3000 0.2045 0.3841
1.4000 0.2454 0.3789 1.4000 0,.2808 0.4115 1.4000 0.1973 0.3736
1.5000 0.2341 0.3630 1.5000 0.2684 0.3948 1.5000 0.1869 0.3580
1.6000 0.2186 0.3415 1.6000 0.2511 0.3718 1.6000 0.1729 0.3372
1.7000 0.1987 0.3138 1.7000 0.2288 0.3421 1.7000 0.1556 0.3111
1.8000 0.1742 0.2796 1.8000 0.2010 0.3051 1.8000 0.1347 0.2795
1.9000 3.1449 0.2385 1.9000 0.1676 0.2601 1.9000 0.1102 0.2421
2.0000 0.1105 0.1896 2.0000 0.1279 0.2062 2.0000 0.0819 0.1985
2.1000 0.0705 0.1321 2.1000 0.0813 0.1419 2.1000 0.0499 0.1483
2.2000 0.0246 0.0647 2.2000 0.0271 0.0655 2.2000 0.0139 0.0908
2.2460 0.0012 99.9999 2.2426 0.0015 99.9999 2.2305 0.0020 99.9999
2.2582 99.9999 0.0203 2.2546 99.9999 0.0177 2.2576 99.9999 0.0542
2.2626 0.0114 0.0114 2.2580 0.0102 0.0102 2.2709 0.0296 0.0296
Table 10.5 Continued

** BLADE SECTION PROPERTIES OF STATOR NO. 1 FOLLOWING ROTOR NO. 1 «»

NUMBER OF BLADES = 14.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 5.200 IN.
BLADE SECTION STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
HAD. COORDINATES SETTING C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
NO. LOC. L H ANGLE L H iniN IMAX ANGLE CONSTANT STIFFNESS
(IN. ) UN.) (IN.) (DEC.) (IN.) (IN.) (IN.)*»2 (IN.)**4 (IN.)*«4 (DEC.) (IN.)o»4 (IH.)«»6
5 7.325 1.1352 0.1:81 14.229 1.1352 0.1881 0.27622 0.0013371 0.037780 14.265 0.0016379 0.0297528
6 6.750 1.1353 0.1952 15.211 1.1353 0.1952 0.26515 0.0013258 0.083507 15.267 0.0014606 0.0282438
7 6.175 1.1354 0.2042 16.253 1.1354 0.2042 0.25415 0.0013281 0.079247 16.356 0.0012973 0.0267262
8 5.625 1.1354 0.2213 16.960 1.1354 0.2213 0.24414 0.0014169 0.075432 17.170 0.0011559 0.0253639
SECTION NO. 5 COORDINATES SECTION NO. 6 COORDINATES SECTION NO. 7 COORDINATES SECTION NO. 8 COORDINATES
L HP HS L HP HS L HP HS L HP HS
(IN.) (IN.) (IN.) (IK.) (IN.) (IN.) (IN.) (IN.) (IH.) (IN.) (IN.) (IN.)
0.0000 0.0209 0.0209 0.0000 0.0188 0.0188 - 0.0000 0.0167 0.0167 0.0000 0.0146 0.0146
0.0110 99.9999 0.0394 0.0096 99.9999 0.0352 0 . 0 0 8 2 99.9999 0.0310 0.0068 99.9999 0.0270
0.0270 0.0009 99.9999 0.0246 0.0009 99.9999 0.0222 0.0009 99.9999 0.0200 0.0010 99.9999
0.1000 0.0219 0.0847 0.1000 0.0242 0.0828 0.1000 0.0267 0.0813 0.1000 0.0307 0.0817
0.20C0 0.0483 0.1302 0.2000 0.0523 0.1299 0.2000 0.0569 0.1302 0.2000 0.0644 0.1340
0.3000 0.0719 0.1704 0.3000 0.0775 0.1715 0.3000 0.0839 0.1734 0.3000 0.0946 0.1802
0.4000 0.0927 0.2056 0.4003 0.0997 0.2078 0.4000 0.1077 0.2111 0.4000 0.1213 0.2205
0.5000 0.1108 0.2359 0.5000 0.1190 0.2391 0.5000 0.1284 0.2437 0.5000 0.1444 0.2553
0.6000 0.1261 0.2615 0.6000 0.1354 0.2656 0.6000 0.1460 0.2712 0.6000 0.1642 0.2848
0.7000 0.1388 0.2826 0.7000 0.1489 0.2875 0.7000 0.1606 0.2939 0.7000 0.1806 0.3091
0.8000 0.1489 0.2993 0.8000 0.1597 0.3047 0.8000 0.1722 0.3119 0.8000 0.1937 0.3285
0.9000 0.1564 0.3116 0.9000 0.1677 0.3174 0.9000 0.1809 0.3252 0.9000 0.2036 0.3428
1.0000 0.1613 0.3196 1.0000 0.1729 0.3257 1.0000 0.1866 0.3339 1.0000 0.2103 0.3523
1.1000 0.1636 0.3233 1.1000 0.1754 0.3296 1.1000 0.1893 0.3380 1.1000 0.2136 0.3571
1.2000 0.1633 0.3228 1.2000 0.1751 0.3291 1.2000 0.1892 0.3377 1.2000 0.2137 0.3570
1.3000 0.1604 0.3180 1.3000 0.1721 0.3242 1.3000 0.1860 0.3328 1.3000 0.2105 0.3521
1.4000 0.1549 0.3089 1.4000 0.1662 0.3150 1.4000 0.1799 0.3233 1.4000 0.2039 0.3423
1.5000 0.1467 0.2956 1.5000 0.1576 0.3012 1.5000 0.1708 0.3092 1.5000 0.1940 0.3275
1.6000 0.1360 0.2778 1.6000 0.1462 0.2829 1.6000 0.1586 0.2903 1.6000 0.1805 0.3077
1.7000 0.1226 0.2555 1.7000 0.1319 0.2599 1.7000 0.1433 0.2665 1.7000 0.1635 0.2824
1.8000 0.1064 0.2285 1.8000 0.1147 0.2321 1.8000 0.1249 0.2376 1.8000 0.1429 0.2516
1.9000 0.0876 0.1967 1.9000 0.0945 0.1992 1.9000 0.1032 0.2034 1.9000 0.1184 0.2149
2.0000 0.0659 0.1599 2.0000 0.0714 0.1610 2.0000 0.0782 0.1635 2.0000 0.0900 0.1718
2.1000 0.0414 0.1177 2.1000 0.0451 0.1172 2.1000 0.0498 0.1177 2.1000 0.0575 0.1219
2.2000 0.0140 0.0699 2.2000 0.0157 0.0674 2.2000 0.0178 0.0653 2.2000 0.0206 0.0643
2.2444 0.0009 99.9999 2.2467 0.0009 99.9999 2.2485 0.0009 99.9999 2.2487 0.0010 99.9999
2.2601 99.9999 0.0382 2.2613 99.9999 0.0337 2 . 2 6 2 ? 99.9999 0.0293 2.2614 99.9999 0.0248
2.2707 0.0204 0.0204 2.2704 0.0181 0.0181 2.2696 0.0158 0.0158 2.2675 0.0136 0.0136
Table 10.5 Continued

*« BLADE SECTION PROPERTIES OF STATOR NO. 1 FOLLOWING ROTOR NO. 1 *•

NUMBER OF BLADES = 14.0 AXIAL LOCATION OF STACKING LINE IN COMPRESSOR = 5.200 IN.
BLADE Si:CTIOK STACKING POINT SECTION BLADE SECTION SECTION MOMENTS OF INERTIA IMAX SECTION SECTION
RAD. COORDINATES SETTING C.G. COORDINATES AREA THROUGH C.G. SETTING TORSION TWIST
NO. LOC. L H ANGLE L H iniN IMAX ANGLE CONSTANT STIFFNESS
( IN. ) UN.) (IN.) (DEC.) (IN.) (IN.) (IH.)»'<2 (IH.)*»4 (IN.>''*4 (DEC.) (IN.)»>>4 {IH.)»»6
1 9.625 1.1352 0.2326 10.202 1.1352 0.2326 0.32714 0.0024396 0.109044 10.215 0.0025458 0.0374829
2 9.050 1 .1353 0.1894 10.969 1.1353 0.1894 0.31091 0.0016324 0.101485 10.969 0.0022673 0.0346121
: 8.475 1.1352 0.1845 12.177 1.1352 0.1845 0.29901 0.0014738 0.096707 12.185 0.0020401 0.0329079
4 7.900 1.1352 0.1841 13.138 1.1352 0.1841 0.28745 0.0013794 0.092145 13.159 0.0018305 0.0312922
SECTION NO. 1 COORDINATES SECTION NO. 2 COORDINATES SECTION NO. 3 COORDINATES SECTION NO. 4 COORDINATES
L HP HS L HP HS L HP HS L HP HS
(IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.) (IN.)
0.0000 0.0298 0.0298 0.0000 0.0274 0.0274 0.0000 0.0253 0.0253 0.0000 0.0231 0.0231
0.0133 99.9999 0.0546 0.0144 99.9999 0.0516 0.0135 99.9999 0.0476 0.0123 99.9999 0.0435
0.0408 0.0021 99.9999 0.0351 0 . 0 0 1 1 99.9999 0.0322 0.0010 99.9999 0.0295 0.0009 99.9999
0.1000 0.0252 0.1097 0.1000 0.0194 0.0954 0.1000 0.0195 0.0908 0.1000 0.0203 0.0872
0.2000 0.0609 0.1657 0.2000 0.0452 0.1412 0.2000 0.0445 0.1355 0.2000 0.0455 0.1318
0.3000 0.0925 0.2145 0.3000 0.0683 0.1816 0.3000 0.0668 0.1750 0.3000 0.0680 0.1713
0.4000 0.1202 0.2567 0.4000 0.0886 0.2168 0.4000 0.0865 0.2094 3.4000 0.0879 0.2057
0.5000 0.1441 0.2927 0.5000 0.1063 0.2471 0.5000 0.1037 0.2391 0.5000 0.1052 0.2354
0.6000 0.1643 0.3231 0.6000 0.1213 0.2727 0.6000 0.1182 0.2642 0.6000 0.1199 0.2605 CO
0.7000 0.1812 0.3479 0.7000 0.1337 0.2937 0.7000 0.1303 0.2848 0.7000 0.1321 0.2812 œ
0.8000 0.1938 0.3670 0.8000 0.1436 0.3103 0.8000 0.1398 0.3010 0.8000 0.1417 0.2975
0.LOOO 0.2042 0.3818 0.9000 0.1508 0.3224 0.9000 0.1469 0.3130 0.9000 0.1488 0.3095
1.0000 0.2104 0.3909 1.0000 0.1555 0.3303 1.0000 0.1514 0.3207 1.0000 0.1535 0.3173
1.1000 0.2133 0.3952 1.1000 0.1576 0.3339 1.1000 0.1535 0.3243 1.1000 0.1556 0.3209
1.2000 0.2129 0.3946 1.2000 0.1573 0.3333 1.2000 0.1532 0.3237 1.2000 0.1553 0.3203
1.3000 0.2090 0.3589 1.3000 0.1544 0.3284 1.3000 0.1504 0.3190 1.3000 0.1525 0.3156
1.4000 U.20ia 0.3783 1.4000 0.1489 0.3193 1.4000 0.1451 0.3100 1.4000 0.1472 0.3067
1.5000 0.1911 0.3625 1.5000 0.1409 0.3059 1.5000 0.1373 0.2969 1.5000 0.1394 0.2935
1.6000 0.1769 0.3415 1.6000 0.1303 0.2880 1.6000 0.1271 0.2794 1.6000 0.1291 0.2761
1.7000 0.1591 0.3151 1.7000 0.1171 0.2657 1.7000 0.1143 0.2575 1.7000 0.1162 0.2542
1.8000 0.1378 0.2831 1.8000 0.101J 0.2387 1.8000 0.0990 0.2311 1.8000 0.1008 0.2278
1.9000 0.1127 0.2451 1.9000 0.0829 0.2070 1.9000 0.0811 0.2000 1.9000 0.0828 0.1966
2.0000 0.0838 0.2009 2.0000 0.0618 0.1703 2.0000 0.0607 0.1640 2.0000 0.0621 0.1606
2.1000 0.0510 0.1500 2.1000 0.0379 0.1283 2.1000 0.0376 0.1229 2.1000 0.0387 0.1194
2.2000 0.0141 0.0916 2.2000 0.0113 0.0808 2.2000 0.0117 0.0764 2.2000 0.0126 0.0728
2.2301 0.0021 99.9999 2.2359 0 . 0 0 1 1 99.9999 2.2390 0.0009 99.9999 2.2418 0.0009 99.9999
2.2577 99.9999 0.0543 2.2565 99.9999 0.0513 2.2575 99.9999 0.0471 2.2587 99.9999 0.0427
2.2709 0.0297 0.0297 2.2708 0.0273 0.0273 2.2708 0.0250 0.0250 2.2708 0.0227 0.0227
 189

ROTOR
STACKING
POINT
LEADING
EDGE

BLADE SETTING
ANGLE

STATOR
LEADING
EDGE BLADE SETTING
STACKING ANGLE
POINT S

Figure 10.1 Rotor and stator blade manufacturing coordinate

system used by NASA design code
 190

XI. APPENDIX B: STATOR- AND ROTOR-RELATIVE TIME AVERAGES

Figure 11.1 shows a "snapshot" of the rotor and stator at a particular

rotor/stator orientation. The coordinate system fixed to the rotor frame

D
of reference Is denoted by R, Z, 6 * and the coordinate system fixed to

the stator frame of reference (I.e., measurement frame of reference) Is

S R
denoted by R, Z, e . The R, Z, e coordinate system Is related to the
S
R, Z, e coordinate system through the rotor/stator orientation, denoted

by R = 2if (n-1 )/(N|^N^p), where 1s the number of rotor blades, N^p

1s the number of rotor shaft positions per rotor blade passage, and n Is

the particular rotor shaft position. When the rotor trailing edge Is

drcumferentlally aligned with the stator leading edge, Q = 0. Point A,

shown on Figure 11.1, Is fixed to the rotor frame of reference, and points

B through K are fixed to the stator frame of reference. At each stator

survey point B to K, the absolute velocity has been measured at N^p = 50

consecutive rotor/stator orientations equally spaced over a rotor pitch.

Therefore, at any stator survey point (e.g., B to K), stator-relative

time-average quantities are readily calculated as

G = E ^ (11.1)
1=1 "^SP

where G 1s any flow-field parameter (e.g., V or V ) , and the

superscript — denotes the time average. Within the context of this

thesis, time Is not real time, but rather corresponds to the relative

position of the rotor and stator (I.e., t = Q/w).

Although the data were acquired In the stator-relative (absolute)

frame of reference, i t Is also possible to resolve a "time-averaged"

 191

(A) = Q/y-

,S
0"

Figure 11.1 Blade-to-blade view of the rotor and stator at a

particular Instant In time, t , showing the
relationship between the rotor- and
stator-relatlve coordinate systems
 192

description of the absolute flow field from the perspective of an observer

moving with the rotor blades. Again referring to Figure 11.1, the rotor-

relative time average of the absolute velocity at point A, 1n the rotor

frame of reference, can be determined from the time-resolved measurements

of the absolute velocities acquired at points B to K, 1n the stator frame

of reference, as follows. First, the rotor/stator orientations ( I . e . ,

rotor shaft positions) which correspond to circumferential alignment of

point A, In the rotor frame of reference, with survey points B to K, In

the stator frame of reference, are determined, see Figure 11.1(B). Next,

the above determined rotor/stator orientations are used to determine the

absolute velocities which an observer, sitting at point A 1n the rotor

frame of reference, would see as point A became coincident with points B

through K 1n the stator frame of reference. This Is accomplished by using

the above determined rotor shaft positions to Interpolate from the measured

distributions of the absolute velocities which were acquired at each point

B to K as a function of rotor shaft position. Finally, the arithmetic

average of these absolute velocities are used as an estimate of the

rotor-relative time-averaged absolute velocity at point A. At every axial

survey station, the above procedures were repeated at each of 50 equally

spaced points across the rotor pitch In order to generate a complete

description of the entire rotor-relative time-averaged flow field. The

equation for determining any rotor-relative time average parameter from

Its phase-lock averaged results is

N.
/V n 1 c
G(r,2,e^ = ^ £ 0" (r,z,e ^ , T . ) (11.2)
•^0 1=1 ^ ^

where = (N^p - (0^ - 0^g)NpN5p/2ir).

 193

XII. APPENDIX C: ESTIMATION OF BOUNDS ON TURBULENCE COMPONENTS

The turbulence 1s determined from the standard deviation of the

velocities, but from the manner In which the data were acquired, I t Is not

possible to calculate the standard deviation of the velocities In the

axial and tangential directions. However, I t Is possible to estimate,

from the measured data, the minimum and maximum bounds of the standard

deviations of the velocities In the axial and tangential directions. The

procedure for estimating these bounds are explained In this Appendix.

Figure 12.1 shows a sketch of the Z, e' measurement plane

Indicating the measured velocity components and the corresponding

estimates of their standard deviations. The two measured fringe angle

orientations are Identified as and The ensemble-average

absolute velocity and flow angle are Identified as Vj and T-j-.

respectively. The measured ensemble-averaged velocities and their

corresponding estimated standard deviations are Identified as V and

m
respectively, where m Is assigned the value 1 or 2 to Indicate the

particular velocity component. The shaded region In Figure 12.1

Identifies the estimated bounds, corresponding to one standard deviation

(iJ), of the fluctuations of the total absolute velocity and flow angle.

The bounds of the shaded region are used to estimate the i j bounds of the

standard deviations of the axial and tangential components of the total

absolute velocity, as follows.

The equations of the lines which are colllnear with the direction of

the measured velocity components are

V Vz = Ve "m <'2-1'
 194

MIN
MAX

"^MAXIMUM
ENVELOPE
OF V-j-/ Pj
FLUCTUATIONS

Figure 12.1 Fringe-plane view of the velocity components

showing the relationship between the
ensemble-average and standard deviations of the
axial and tangential velocity components and the
measured velocity components
 195

where the subscript m Is assigned the value 1 or 2 to denote the

particular measured velocity component, and the subscripts z and 0

denote the axial and tangential directions, respectively. The equations

of the lines which are perpendicular to lines 8.^, and which Intersect

lines at and (V^ * Jj,) are

(12.2)

where the subscripts m, z, t are defined as In Equation (12.1), the

subscript _L denotes the perpendicular to line

Now, assume that the true i J envelope of the fluctuations Is

defined by an ellipse, which would be the case I f the distribution of

measured velocities were gausslan and the turbulence were nonlsotropic.

For Isotropic turbulence the true i j envelope of the fluctuations

would defined by a circle. The above assumption 1s not a required

condition, but does facilitate the proceeding explanation. Further, assume

that the major axis of the true i j ellipse of fluctuations

contains points B and D, which are the vertldes of the shaded region

Illustrated In Figure 12.1. This ellipse could be. In the limit, a line to

the largest ellipse which inscribes the shaded region. Note that in order

to have measured the projections of the true i j ellipse of the

fluctuations, the i j ellipse must be tangent to a l l four sides of

the shaded region in Figure 12.1, but the i j ellipse does not need to

completely inscribe the shaded region. For example, the projection of a

line segment between points B and D in the and component

directions would define the extent of the and fluctuations,

respectively. Conversely, the projections of a line segment between points

 196

A and C 1n the and component directions would also define the

extent of the and fluctuations. Therefore, the major axis of the

true I j ellipse of the fluctuations could be In any orientation

between line segment A-C to line segment B-D, as long as the bounds of the

i j ellipse are tangent to a l l four sides of the perimeter of the shaded

region.

The ellipse with the smallest possible area ( I . e . , minimum overall

fluctuations) would have i t ' s major axis colllnear with line segment A-C,

and an ellipse with the largest possible area ( I . e . , maximum overall

fluctuations) would have I t ' s major axis colllnear with line segment B-D.

Therefore, the minimum and maximum bounds of the i J fluctuations of axial

and tangential velocity components are defined by the vertlcles of the

shaded region. The vertlcles of the shaded region In Figure 12.1 are

determined from the Intersections of lines and as follows.

(V^ ± J|)s1n ag - (Vg ± J^)s1n

s1n(a2 - A^)

(V2 ± J^)COS - iXj )cos "2

Sln(a2 - "'l)

+ +
+ cTj ) s 1n «2 ~ (VG ± J^)s1n "1
s1n(a2 - «^)

(V2 + )cos (*1 — (VI ± J-,) cos "2

The minimum and maximum bounds of the i J fluctuations of the axial

and tangential velocity components can, therefore, be determined as

 (~ I rj ,s1n
ttp ± X s1n a, "I
<4'max • - "ikh = H sInL ' )
m1n IB/ «J I J

( Il / r^9 cos ± J l cos a_l

'4)max = «n"; - vJ^Jz = Aas ' N
min IB/ ^ 1 J

from which the variances and the turbulence are determined.

 198

XIII. APPENDIX 0: WAKE ENHANCEMENT PROCEDURE

Contour plots of the t1me-resolved turbulent kinetic energy have been

previously used [42-45,89] to Identify the transport of wakes through a

downstream blade row. However, as these wakes progress downstream they

become harder to discern from the surrounding flow field, due to mixing,

and therefore plots of the turbulent kinetic energy may not be sufficient

to clearly delineate the wake fluid. In order to better resolve the rotor

wakes throughout the stator row of the present research compressor, a

procedure was developed for enhancing the rotor wake signature.

Figure 13.1 shows a typical plot of the total turbulence at two

points In the stator flow field; one upstream of the stator row, and one

near the stator exit. The total turbulence, Sj, Is defined as

y min min

where J and J are calculated from Equations 12.3 and 12.4,

^mln min
respectively. Upstream of the stator row the total turbulence within the

rotor wake is very high relative to the freestream. However, near the

stator exit the total turbulence within the rotor wake Is only slightly

higher than the freestream. Furthermore, the free-stream total turbulence

varies from inlet to exit and blade to blade. Therefore, in order to

enhance the rotor wakes throughout the stator passage the total turbulence

is normalized such that the freestream value Is zero and the maximum

value, which usually occurs in the rotor wake, is one. This normalization

is applied at each spatial survey point throughout the stator flow field.
 UPSTREAM OF STATOR DOWNSTREAM OF STATOR
in
'MAX
'MAXI

4 I
'MIN-^
'MIN

(A) ACTUAL ROTOR WAKE.

i -0.25

0 100 0 100
PERCENT ROTOR PITCH
(B) NORMALIZED ROTOR WAKE.

Figure 13.1 Illustration of wake enhancement procedure

 200

As a result of the enhancement procedure, the rotor wakes have equal

weight throughout the stator flow field as shown In Figure 13.1(B). Of

course, any information about the dissipation of the rotor wakes Is lost,

but the width of the rotor wakes and their phase Is preserved. Contour

plots of the enhanced rotor wakes are constructed from contour levels at

0.25 of the peak value In the rotor wake (arrived at through t r i a l and

error), as shown 1n Figure 13.1(B). Thus, most of the rotor wake Is

Identified within the shaded regions of Figures 5.10 and 5,11.