Liquid Rocket Engine Axial Flow Turbo Pumps
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Liquid Rocket Engine Axial Flow Turbo Pumps as PDF for free.
More details
- Words: 31,377
- Pages: 129
A
Z
_-.I
r-I-I
m
0 Z ),
r-n r._ ..,,., G..)
"o :Z:,, c-J
_3 "_1 m
_-c'-3 I_-
I'1"1
;]_:,, r._ ;:Z:,,
c-_ r-"0 =0
I-
),
mmm
rn
m
m
mmmm
=0 0 Z
t--|
mmmm m
c -I m
0 (n ), z Imm
Im
(n -0 _..dl
o m
mm V_dl
0
i: m
Z m
(n
Z _m r.S'J
=D -I
m Imm
"U
atom
0 Z
mm_ mb
,,,..j
r.J.I
FOREWORD
NASA
experience
Accordingly,
has
criteria
indicated
a need
are being
for uniform
developed
in the
criteria
following
for the
areas
design
of space
vehicles.
of technology:
Environment Structures
Individual are
components
completed.
of this
This
Guidance
and
Chemical
Propulsion
work
document,
will be issued
part
of
monograph. A list of all monographs of this document. These
monographs
except
as may
these
be
design
revised
monograph,
direction
of
project
"Liquid Howard
of the
was
Lewis
International of
participated
this
in interviews,
King
of Rocketdyne
Lewis
Research
Comments
concerning
National
and
W.
the and
Ohio
technical Space
44135.
1978
*Currently **Currently
with
Aerojet
Liquid
with Westinghouse
Rocket
Co.
Electric
Corp.
and
and
The J.
as NASA
as they
is one
such
final
pages
on the
requirements,
It is expected,
however,
eventually
that
will provide
and
monograph
was
of Rocketdyne of
review
of the
Corporation;
of
NASA
written
Aerojet the text.
under
Research
To
D.
assure
Rocket technical
community
In particular,
D. M. Sandercock
Headquarters
D.
Rockwell
Liquid
technical
the
Center; by
Division,
Jr. of Lewis.
throughout
critical
Wilcox
Lewis
B. Keller,
engineers
was prepared
Office,
Farquhar**
International
W.
not
to be desirable,
Criteria
by Russell
Rockwell
the monograph
Aeronautics Cleveland,
edited
Propulsion,
Turbopumps",
Design
Viteri
consultations,
Center;
reviewed
F.
and
specifications. indicate
M. C. Huppert*
scientists
Division,
collectively
April
was
document,
to design
W. Schmidt.
Center;
Chemical
as soon
to this one can be found
Axial-Flow
Chief,
Harold
on
monographs
vehicles.
Engine
and
Themonograph
accuracy
Office),
by
Research
Corporation;
Company.
project
space
Rocket
series
prior
may
W. Douglass,
management
Scheer
in formal
for NASA
as separate
as guides
as experience
practices
the
issued
to be regarded
specified
documents,
uniform This
are
Control
individually
Austin of the and
in detail. content Administration,
of this
monograph
Lewis
Research
will be welcomed Center
(Design
by the Criteria
GUIDE
The
purpose
of this
significant
monograph
experience
programs firm
TO THE USE OF THIS MONOGRAPH
to date.
guidance
product,
is to organize knowledge
It reviews
and
achieving
greater
for
and
and
greater that
are
preceded
The
the
Art,
section
of
current
by
2,
present,
for effective
in
design
consistency
in the
major sections references. State
assesses
efficiency
and
accumulated
practices,
in design,
design
effort.
a brief
and
discusses
in design,
and from
them
establishes in the
is organized
and
complemented
the
total
into by
design
the
operational
reliability
monograph
introduction
reviews
and
increased
The
use
development
end two
a set
problem,
of
and
identifies which design elements are involved in successful design. It describes succinctly the current technology pertaining to these elements. When detailed information is required, the best available references are cited. This section serves as a survey of the subject that provides background
material
Recommended
design. manager
The
have
been
displays
can be followed design
loosely
criteria
in section
provided.
base
effectively
for
the Design
Criteria
and
The
as a checklist
hove when
Recommended to
the
and briefly wha__._trule, design element to
guide, assure
of rules
for the
its adequacy.
3, state
is described;
guidance
organized
should
this
to satisfy this
each
cannot
Practices, practicing
body
decimally
numbered
subsections
correspond
from
continuity
of subject
in such
through
manual. of existing
be judged
into
both
monograph
or a design
organized
its merit
serve
or in assessing
procedure
positive
numbered
specifications,
to the
are
can
a design also
best
provide
similarly
Contents
design
technological
of the
be done
criteria. concisely,
in conjunction
designer
on
with
how
the
to
achieve
the
subjects
design.
sections
within
Criteria
Practices, the
references
Criteria,
successful Both
Design
to use in guiding
possible,
appropriate
the
The
Recommended
Whenever Design
a proper
Criteria, shown in italics in section 3, state clearly or standard must be imposed on each essential
successful
The
prepares
Practices.
The Design limitation, project
and
on how
sections is not
as a discrete intended
It is a summary successful
to and
design
effectively
iii
so that
to section.
a way
that
The
a particular
format
for
aspect
of
subject. be
a
design
a systematic techniques
it makes
designer.
subsections section
that
handbook,
ordering and
material
a
set
of
of the large
and
practices.
Its value
and
available
to and
useful
CONTENTS
.
INTRODUCTION
2.
STATE OF THE ART
3.
DESIGN
CRITERIA
APPENDIX
A - Conversion
APPENDIX
B - Glossary
REFERENCES
Page 1
............................. ........................... and Recommended of U.S. Customary
3 Practices
68
.................
Units to SI Units
93
.............
95
.............................
105
.................................
NASA Space Vehicle Design Criteria Monographs SUBJECT
Issued to Date
111
..............
STATE OF THE ART
DESIGN CRITERIA
2.1
3
3.1
68
Turbopump Speed
2.1.1
8
3.1.1
68
Turbopump
2.1.2
9
3.1.2
68
OVERALL
TURBOPUMP
DESIGN
Rotor Dynamics
2.2
10
3.2
68
Realm of Operation
2.2.1
10
3.2.1
68
Stage Hydrodynamic Design Blade Loading, Stall Margin, and Efficiency Velocity Diagrams Blade Angles
2.2.2
11
3.2.2
69
2.2.2.1 2.2.2.2 2.2.2.3 2.2.2.4
16 24 25 26
3.2.2.1 3.2.2.2 3.2.2.3 3.2.2.4
69 70 71 71
2.2.2.5 2.2.2.6 2.2.2. 7
26 31 34
3.2.2.5 3.2.2.6 3.2.2. 7
71 72 72
2.3
35
3.3
73
2.3.1 2.3.1.1 2.3.1.2
35 35 39
3.3.1 3.3.1.1 3.3.1.2
73 73 74
STAGE DESIGN
Solidity Cavitation Off-Design Clearances PUMP ROTOR
Performance
ASSEMBLY
Blades Profile Types Mechanical Design
STATE OF THE ART
DESIGN
2.3.1.3
43
3.3.1.3
80
2.3.2 2.3.2.1 2.3.2.2
43 43 45
3.3.2 3.3.2.1 3.3.2.2
81 81 81
2.3.3
45
_3.3
82
2.3.3.1 2.3.3.2
45 47
3.3.3.1 3.3.3.2
82 82
2.3.4 2.3.4.1 2.3.4.2 2.3.4.3
47 47 50 53
3.3.4 3.3.4.1 3.3.4.2 3.3.4.3
82 82 83 84
2.4
53
3.4
85
2.4.1 2.4.1.1 2.4.1.2
54 54 54
3.4.1 3.4.1.1 3.4.1.2
85 85 85
2.4.1.3
54
3.4.1.3
85
Vane Attachment Methods
2.4.2 2.4.2.1
54 54
3.4.2 3.4.2.1
85 85
Mechanical
2.4.2.2
55
3.4.2.2
86
2.4.3 2.4.3.1 2.4.3.2
57 57 58
3.4.3 3.4.3.1 3.4.3.2
87 87 87
2.4.3.3
59
3.4.3.3
87
2.4.4 2.4.4.1 2.4.4.2
60 60 60
3.4.4 3.4.4.1 3.4.4.2
88 88 88
2.4.5 2.4.5.1 2.4.5.2
61 61 63
3.4.5 3.4.5.1 3.4.5.2
89 89 90
2.5
63
3.5
91
3.5.1
91
SUBJECT
CRITERIA
Profile Tolerances, Surface Finish, and Fillet Radii Blade Attachment Methods Mechanical Design Rotor Configuration Mechanical Design Axial Thrust Balance System Types of Systems Mechanical Design System Stability PUMP STATOR ASSEMBLY Vanes Profile Types Mechanical Design Profile Tolerances, Surface Finish, and Fillet Radii
Design
Stator and Volute Housings Housing Types Hydrodynamic Design Mechanical Design Bearing Housings Types Mechanical
Design
Housing Interfaces and Static Sealing Interface and Seal Types Mechanical Design MATERIALS Property
Data
vi
SUBJECT
STATE OF THE ART
DESIGN CRITERIA
Ductility
3.5.2
91
ImpactStrength
3.5.3
92
Endurance Limit
3.5.4
92
SAFETYFACTORS
2.6
vii
64
LIST OF FIGURES
Title
Figure 1
Head coefficient
2
Cross-sedtional
3
Cutaway
4
Approximate
5
Typical
6
Nomenclature
7
Relationship of stage ideal head coefficient, efficiency, diffusion factor, and flow coefficient ....................
21
Variation incidence
23
8
9 10
11
and efficiency
Page
vs flow coefficient
view of Mark 15-F turbopump
view of M-1 fuel turbopump
for axial-flow
pumps
.......
.................
6
..........
..........
7
Ns - Ds diagram for single-disk pumps and low-pressure-ratio
flow model for an axial pump stage for axial pump blading
of total-pressure-loss for NACA-65-series
compressors
..................
....................
and efficiency
Comparison of theoretical and experimental cavitation for double-circular-arc profiles .......................
14
..............
27
breakdown
parameter 29
Correlation of pump cavitation parameter with ideal head coefficient, flow coefficient, stage efficiency, and diffusion factor ..............
12
Potential
13
Three types of rotor assemblies
14
Typical profile nomenclature
15
Comparison
16
Typical modified
17
Typical Campbell diagram for identifying
18
Modification
30
stall points of Mark 15-F pump during start of the J-2 engine
Goodman
.......
......................
distributions
of Mark 15-F first-stage
37
for three profiles
diagram for blade stress blade resonant
blade to eliminate
°..
Vlll
33 36
.......................
of basic thickness
12 13
parameter with diffusion factor at reference and double-circular-arc blades ...........
Effect of solidity on head coefficient
5
...........
38
.............. conditions resonance
40 ........ .........
41 42
Figure
Title
Page
19
Cam!?bell diagram for Mark 15-F first-stage
20
Design details for lq-1 dovetail
21
Fabrication
22
Thrust-balance
23
Typical
24
Thrust-balance
system for the M-1 fuel pump
25
Thrust-bearing
assembly on M-1 fuel pump
26
Rotor-stator
27
Stator
28
Rotor-stator
29
Volute types showing various degrees of foldover
30
Types of static seals used in axialpumps
31
Modified
32
Effect of base load on blade natural
33
Effects of fluid virtual mass on Mark 15-F vane natural
blade
attachment
of M-1 fuel pump rotor
................
..................
of a series-flow
assembly
segments
46 ................
thrust balance system
for Mark 15-F pump
Goodman
for M-1 pump
49 ............
48
.................
51
..................
52
..................
55
and rotor disks for Mark 9 pump
assembly
44
.....................
system used on Mark 15-F pump
performance
42
...............
56
....................
diagram illustrating
56 ................
58
................... safety factors
frequency
ix
62 ..............
76
and damping (M-1 dovetail) frequency
.........
.......
78 78
LIST OF TABLES
Title
Table I II
Chief Design Features
of Axial-Flow
Stage Design Data for Axial-Flow
III
Design Parameters
IV
Materials
Liquid-Hydrogen
Pumps
Page Pumps
4
...................
17
for Blade and Vane Profiles in Axial-Flow
Used for Major Components
...........
on Axial-Flow
X
Pumps
Pumps ...........
........
18 65
LIQUID ' ROCKET AXIAL-FLOW
ENGINE
TURBOPUMPS
1. INTRODUCTION An
axial-flow
pump
the
periphery,
that
fixed
blades
tips
and
(stator
by
fluid;
surrounds Both
axial-
and
By far,
with
and
greater
based
on
the
the small
fluid.
In
with
the
experience
important
to note,
that however,
herein,
axial-flow
pumps.
centrifugal
even Some pumps.
(primarily
volutes)
the
criteria
of the In most
are discussed
saw
of the
of
is
each
set
rotor
velocity
90 ° to the
diffuser
that
acquired
treated
instances,
on centrifugal
as such
reflects
of axial-flow
made
to incorporate
subsequent
technology in this
monograph
generally because
pumps
to the
may
(ref.
applicable the 1).
subject
the
J-2 is
based
hydrogen
extensive
design
on
technology
liquid
particularly
blading,
In five
This monograph
pump
involved
axial-flow
of development
service
V vehicle.
in the
areas herein,
levels
for
use of
of staging.
operational
Saturn
The
of the
has been that
is well suited
engines.
ease
various
technology,
evolved
areas
which
rocket
relative
extensive
of the
propellant-feed
consideration
all of which
no attempt that
to
through
and
this
of
type,
and
Apollo
design
some
led
weight,
five pumps
briefly
headrise
roughly
by the
in rocket-engine
of current
carried
15-F)
stages
technology though
monograph
and
Mark
been
that
fluid
spiral
(headrise)
blade
of a bladed
pumped
a high-speed
centrifugal
logically lower
mechanical
had
and
of
operating
traverses
consists
the
all
shaft, fluid
on
sets
at the
under
pump
pump
utilized
of the
of configurations, the
clearances
as the
increase
requirements
designed
these
been
made
development
and
axial-flow-compressor
discussed
design
been
S-IVB
number
design
applicable
and
(the
and
the
sweeps
imparts
have
however,
pump
small
blades
contains
fluid.
efficiency, were
experience
hydrodynamic
any
has
pumps S-II
action
pressure/flow
higher
one
on
a relatively
pumped
the
with
impeller
that
maintained
produced
to pressure
the
as a propellant,
axial-flow used
This
pumps
use
and
production;
engine
shaft.
blades;
A centrifugal-flow of the
is converted
and collects
its potential
instances,
vanes.
airfoil-shaped
or housing
are
parallel
of pressure
rotation
centrifugal-flow
hydrogen
type
stator
rotor
vanes
nearly
carrying
a casing
the
stator
flows
the
impeller
impeller
propellants
liquid
and
of this velocity the
systems.
on
the
and
of increases
so that
from much
fluid
blades
shaped
outward
dense
pumped
or a cylinder,
within
between
blades
summation
of rotor
(impeller)
speed
positioned
rotor
The
produced
of a set of disks at high
vanes)
between
conditions. (stage)
consists rotates
as the
in
the
area
of
use
was made
of
compressors.
It is
in this document design
be useful
are applicable to both matter
of the
in the
pumps
design
to both
of axial
types
of pumps
is covered
fully
in
The axial-pump designprocessis directed toward achievinga hydrodynamic andmechanical design configuration that will meet the requirements of the engine system within the constraintsimposedby other componentsin the turbopump assembly.Thus, the axial-pump design cannot be divorced from the design of components such as the turbine, inducer, bearings, and seals.In instanceswhere components or systems influence the axial-pump design, appropriate reference is made to other monographs in the Chemical Propulsion series. The hydrodynamic design of an axial pump involves basically (1) the selection of appropriate fluid-velocity diagrams and (2) the design of blading that will achieve fluid turning per the diagramswith the predicted loss. Major problems in hydrodynamic design include (1) failure to achieve pump headrise because of improper fluid turning or low efficiency, and (2) failure to maintain adequatestall margin during transient or steady-state pump operation. Structural adequacy of the axial pump is achievedby keeping operating stresseswithin material limits and by maintaining adequateclearancesbetweenrotating and stationary components. Major problems in structural (mechanical) designinclude blade and vane fatigue failures, excessive housing and rotor deflections, and failure of the thrust-balancesystemto keep rotor axial loadswithin manageablemagnitudes. The monograph beginswith a section that provides a brief background on the axial-pump applications and views the pump in terms of the total turbopump assembly.The remaining sectionstreat the pump designin the order in which a pump designerwould proceed.These sectionsdeal with stagehydrodynamic design,pump rotor assembly,pump stator assembly, pump materials for liquid-hydrogen applications, and safety factors as utilized in state-of-the-art pumps. In each of these areas, the monograph establishesthe basis for successfulaxial-flow pump design.
2
2. STATE OF THE ART
2.1
OVERALL
The
use
of
applications relatively
axial-flow
are
and carried through of these pumps and in figure
abrupt
included on
adding
the.
stages
M-l),
operating short.
integral
with
the
which
on
The
the
*Factors Terms,
been
limited
the
front a
This
drives
for
the
first
9. The highly
head
on
(NPSH)
for converting symbols,
axial
U.S.
materials,
M-l,
of the
and
26,
the
blading
pumps and customary
units
and abbreviations
life the
(size).
requirement
relative
simplicity
requirement
i.e.,
has
pump and
been
in
a significant
pumps
on the Mark to achieve
increased
to operate pressure System
or identified
assembly
of Units in Appendix
the inducer
that
includes
turbine.
overhung
from
on separate
t 5-F was identical
was utilized
sufficient
with
to be developed.
version
designed
in this
turbopumps of the other
a direct-drive mounted
of the
to
(SI Units) B.
the
rear
With
Mark
the
utilized 15-F
in
per stage.
values
prevent
a The
bearings.
to that
basic
headrise
at low
been
a service
discussed
stage
were
25 were
have
sec, with
pumps
an inducer
turbopumps
Mark
axial-flow
to the International
pumps
was 500
an uprated
are defined
the
packaging
15-F and M-1 liquid-hydrogen In general, configurations
26
9 and
blading
develop
and
for
designed the axial
a design
of
axial-flow
assembly,
been
curves pump
have been favoring
axial-flow
was
have to
because
an axial-flow
engine
in turn,
Chief design characteristics
axial
was
for example,
shown;
Mark
rocket
stage,
Mark
loaded
two
housing,
for the Mark
an
and
uprating
favored
fixed-speed that were
characteristic
of
weight,
thrust
short
rear-bearing-housing
15-F,
the
of the
for state-of-the-art
components
bearing
over a wide configurations
condition.
life
to the
liquid-hydrogen
and production. I*. Head-vs-flow
choice
efficiency, was
sec.
slope
The
engine
uprated
10 000
to
a centrifugal pump could pressure. Considerations
in
service
similar
of an additional Mark
inducers
suction
steep
for the M-1 pump, of
are
Mark
9 was
more
incisive; discharge
configuration
and
of
the points.
where
duration
system,
whereas
exception on
are
stall**
necessary
duration
turbopumps
Mark
has
and headrise were required. Additionally, the of axial pumps has restricted their use to
and cutaway views of the Mark in figures 2 and 3, respectively.
thrust-balance
The
axial
development
Cross-sectional are presented
bearing,
engines
advantages
the
overhauls) the
axial-flow
the
Design
(between
factor in monograph.
turbines
the
applications
to achieve
relatively
stator
apparent at
potential in those
Required life
in head
noted in table I was not the flow at the required
Additionally, (e.g.,
rocket
various levels of development their use are tabulated in table
1 ; readily
drop
applications to deliver pump
in
that did not require significant throttling or operation As noted, the state of the art is reflected in five
shown
the
pumps
DESIGN
in which high volumetric flowrate narrow operating-range capability
applications flow range. designed features
TURBOPUMP
of net
cavitation
are given in Appendix
positive in the A.
Table
I. -
Chief
Design
Features
of Axial-Flow
Liquid-Hydrogen
Pumps
Delivered Pump
Speed,
gpm
Headrise,* ft
10 230
51 500
32 800
flow,
Mark 9
Application
Number of stages
rpm Inducer plus six main stages
Phoebus (Development)
Mark 15-F
9 062
28 266
40300
Inducer plus seven main stages
J-2 engine (Operational)
.Ca
18 500
Mark 25
Mark 26
9 000
34 000
62 000
40 000
24 000
Tandem inducer
plus four main
Phoebus
stages
(Development)
Inducer plus seven main stages
J-2 engine (Experimental)
62 300
M-1
13 225
56 500
Inducer plus transition eight main stages
Overall
headrise
-
inducer
inlet to volute
discharge.
plus
M-1 engine (Development)
m nmSTALL LINE _HEAD/FLOW CHARACTERISTIC
2.8
*OVERALL HEAD COEFFICIENT INDUCER INLET TO VOLUTE DISCHARGE
2.6
26
2.4 z w
-
c,.)
2.2
o
J
2.0
.s
r_
I
"r
1.8
K 15-F
MARK 9
1.6
1.4
o_
80-
>_z _J
60-
la_
L
I • 04
I •06
I .O8
I .10
I .04
I .06
I .08
I .10
I
I .06
I .08
I .10
I .12
I .06
I .08
for axial-flow
pumps.
INDUCER INLET FLOW COEFFICIENT Figure
1.-
Head coefficient
and efficiency
vs flow
coefficient
I .10
I .08
I .10
.12
Ol FFUSER PUMP VOLUTE
_
TURBINE
PUMP STATOR
ROTORS
FLANGE REAR BEARING INLET
BOLT
FLUI O FLOW c_
INE NOZZLE | NE STATOR
FRONT IIF,,ARI NG
PUHP ItOTOit-
INOUCER STATOR AND FRONT DF..ARI NG HOU51NG
ilIRUSTDAL,ANCE PISTON
TUItB I NE INLET HANIFOLD REAR BEARING NOUS I NG
I NSULATI ON
Figure
2. -
Cross-sectional
view
of
Mark
15-F
turbopump.
REAR
BEARING
HOUSING EXHAUST
HOUSING
PUMP OUTLET
FRONT HOUS
BEARING
INDUCER HOUSING-_ i: TURBINE -.a
SECOND-STAGE
STATOR NE FIRST-STAGE TURBINE
NOZZLE
INE INLET MANIFOLD
BALANCE PUMP VOLUTE PUMP STATOR
CER
PUMP ROTOR
Figure
3. -
Cutaway
view
of
M-1 fuel
turbopump.
PISTON
ROTOR
ROTOR
following
blade
cooled
by
limits
the
for
rows
(ref.
liquid
2).
Both
hydrogen;
relatively
roller
bearing
short-life
and
ball bearings
DN values
pumps
(ref.
3).
Labyrinth
employed to control internal flow. Face-riding turbine end of pump to prevent propellant conditions
and
operated
control
successfully
2.1.1
minimize
at surface
applications,
initial
design
as possible
with
the
total DN
(ref.
vehicle
are given
phase
design
3),
4).
the
rubbing
of speed
parameters
pump
usual
hydrodynamic These
speed
(ref.
4),
and
at high turbine
include
turbine
stress
(ref.
of speed
seals
speed
and
speed (ref.
pump
in establishing specific
turbopump speed
N_ -
critical
turbopump
and
Ss.
speed These
are pump
parameters
rotational are
related
speed by
NQ1/_
(1)
H_
NQ _ as
where
Q = volume H = headrise,
speed, flowrate,
rpm
(2)
-
(NPSH)
N = rotational
2),
7.
suction
speed,
as
associated
expressions
Ns = specific
to
During
cavitation 5),
been
in order
the design
on the
been at the static
have
constraints
inducer
been
have
efficiencies.
is to set
mechanical
the effects
seals
face-riding
and
practice
and
constraints
restrictions
and have
state-of-the-art
shaft-riding
is operated
turbopump, the
(ref.
turbopump
of a new
seal
and
ft/sec
increased
in reference
Ns,
the
to achieve
with
utilized
seals have been utilized the turbine area during
operation;
up to 400
assembly.
Details
during
the
been
and
and lift-off leakage into
and
turbopump
6).
speed
weight
consistent
(ref.
speeds
specific
speeds
rocket-engine
high
Important
leakage
Speed
turbopump
bearing
the
Turbopump
In normal the
to
have
of 2 x 106 are considered
- (gpm) v2/ft 3/_
rpm gpm
ft
8
3_
N, the
Ss= suction specificspeed,rpm
The
NPSH
= net
design
speed
suction
speed
speed
forcing
functions.
design
constraints
change
in blade the
Mark usually
strict
of the
Mark
sequential
dynamic
for a new
and
stability
aspects
machines
are
of the
shaft
designed
and
so
discussed
the
which
have
Axial
loads
blades
limits of the
blading
26 were
not
and
type
of
rotors
has
An
received
extensive
have
both
to the study
systems
applicable
to
here,
In the
Mark
speeds
operating 9, 15-F, were
centrifugal
however,
of the and
to discuss
axial
briefly
the
with rotor dynamics in the state-of-the-art been used in the design of the rotor/bearing
In the M-1 turbopump, maximum
both
in rocket
discussion
the
speed 25, and
above
and disadvantages
shaft-and-bearing
was below
the
26 turbopumps,
system
criticals
of each
of these
system
first critical the
but
was speed
systems
below
design
(operation
high
radial reacted
below
in order load
and
the
system
to achieve
a high
stiffness
by a triple
first first
characteristics,
set of ball bearings
critical critical
speed), speed.
the
were rotor
approaches
were encased
used
high Thus,
are
bearing
speed
for the
Limited
testing
housing
so-called was
flexibility
M-1 Mod
performed
1 (initial with
9
was this
performed test
(ref.
configuration)
turbopump,
rotor/bearing roller
to support
in a radially
flexible
rear rpm.
akin
6.
were
critical
in a
previously
attention and
rates
16 000
resulted
were
done to ensure that radial shaft support would be at the intended Extensive analysis along with experimental effort to determine
system
by
speed-selection
considerable bibliography
6. It is appropriate
advantages
is desired
(turbine)
in
established
It was noted
this being locations. and
limited
speed-dependent
were could
sequential, the
on a design
were
with
and
to resonance.
precluded
based
26 pumps
design.
rotor
operating
The
approach
stiffness
Mark
programs.
turbopump
in reference
M-1
system
and
cavitation
Mark
pump
proximity
and problems associated Two approaches have
the
mode.
redesign
the
in reference
its supports.
that
first-flexural
since
and
turbopump
turbopumps.
the
of the
in effect
turbopump
presented
so that
15-F
and
operating-speed
of turbopump
for axial-flow
designed
15-F
Rotor Dynamics
overall design concepts axial-flow configurations. systems
sense,
development of
was inducer
Mark
resonance were
relation
behavior
design
ft
The
avoid these
frequency
Turbopump
engine
In
to that
in the
v2
on the M-1 turbopump
order
Note
conducted
head,
Ss of 43 000.
natural
This
2.1.2 The
in
designs
9.
suction
constraint
specific
operating
that
positive
- (gpm) ft3/,
because
bearings, the
shaft.
housing,
roller-bearing bearing spring
8). The
predicted
turbopump the
was
M-1 engine
program was terminated. However, during one of the tests an inadvertent overspeedto 15 500 rpm occurred.At this speedlevel, the bearingandsupport-strut accelerometertraces indicated that the critical speed was being approached, thus lending credibility to the analytical model, which is describedin reference8. The axial turbopumps that have been designedto operate above the rotor/bearing system critical speedshavehad ball bearingsat the support locations, a configuration that achievesa relatively low radial spring rate and low systemcritical speed.A rotor operating abovethe first critical speedmay developnonsynchronouswhirl (ref. 6); this phenomenonoccurred in varying degreesof severity on the Mark 9, 15-F, 25, and 26 turbopumps (refs. 9, 10, and 11). Nonsynchronous whirl and axial oscillations that occurred with the Mark 9 were examined extensively both analytically and experimentally (ref. 10). In the Mark 15-F, nonsynchronous whiff was identified as a major source of alternating stressin the turbine disks and disk-to-pump shaft coupling (the turbine disks resist plane-of-rotation changes associated with the whirl); in addition, rotor radial displacement as high as 0.030 in. peak-to-peakwas measured.These problems were solved by increasingthe axial preload on the ball bearings, which increased the threshold of shaft stability and suppressedshaft deflections to tolerable magnitudes.Severenonsynchronous whirl alsowas observedduring development testing of the Mark 25 pump (ref. 11). In this pump, singleball bearingswere used at the rotor support locations; a design changeto duplex ball bearingseliminated the severewhirl problem in the pump operating-speedrange. In general,the rotor dynamics problems that have occurred in axial-flow pumpshavebeen difficult to diagnose and solve. Suitable analytical models for the prediction of nonsynchronous whirl were not available at the time the pumps treated herein were designed.In the designprocess,an attempt wasmade to avoid whiff by considering those factors then known to be related to whirl problems and by designingso that the operating speedwas not near a critical speed.However, asindicated in the precedingparagraph,these measureswere not always adequate. Effort to solve nonsynchronous-whirl problems in centrifugal turbopumps was exerted subsequentto the design of the axial-flow pumps discussedin this monograph. This effort has provided improved analytical methods that permit a more thorough treatment of rotor instability during the designphase(refs. 12 and 13).
2.2
STAGE
2.2.1 The
DESIGN
Realm of Operation
considerations
discussed the specific
involved
in reference speed
range
7. The
in the
selection
discussion
in which
axial
of the
herein stages
have
10
type
of pump
is limited been
essentially
used.
for a given
applici_tion
to an identification
are of
Figure 4 (adptd. from ref. 14) showsdesign-point performance for various types of pumps in terms of specific speedNs and specific diameter Ds (D_ = DHla/QY2, where D =.rotor diameter, ft). This kind of figure is useful in the pump selectionprocess,in that it identifies specific areas where different types of pumps are suitable and givesan estimate of pump efficiency and diameter. As shown oll the figure, axial-flow-pump stageshavespecific speedsrangingfrom approximately 3200 to 11 000. It should be noted that these are stage characteristics; considerabledifference occurs when the entire pump is examined. For example, the figure indicates that for the M-1 mainstagespecific speedof 4470 an axial flow configuration is suitable; however, the specific speed of the entire M-1 pump is approximately 900, and examination of this region on the N_-Dsdiagram indicates that a centrifugal pump could havebeenselected,somedecreasein efficiency being anticipated.
2.2.2 The
Stage Hydrodynamic
procedures
design that
used
practices exist
that
stations
in the
hydrodynamic
for axial-flow
compressors
in the
assumed
pump
the
for
figure
6. In the condition
flow
are
(i.e.,
Weight and start
and
usually as noted
density are
being
pump
(or stage)
of the
propellant
pumping
rise
used
weight
flowrate
process.
as the
(e.g.,
in
and
are dictated
by the
with
and
pressure
beginning
pressure the
In reference
angles
engine
thrust
rise
end
engine flow
7, a method
from
is described
11
in
average in the
by
applying
propellant
at
the
and
inlet
pressure the
operating
is
the
stage
turbopump
range
and
of the
and
in which
flowrate
density
In and
changes
requirement
of
state
or an increment the
desired
pump. propellant
thermodynamic
process
exit).
requirements,
to volume
headrise
first
total
straightforward, heating
the
weight
of the
inducer
requirements
propellant
pumping
for
at the
are converted
pumps), of the
the
radial-equilibrium
include
with
conversion
When
is shown
model.
associated
rise is determined and
blading
flow
variations
conditions
required
radial
A typical
represent
accounted
and
off-design
factor.
the
it is
at discrete
and
stages
fluid
constraints
propellants, conversion
flow
velocities,
designer
are
axial-flow
high-pressure-hydrogen
for a given at the
the
for
followed conditions
in which
plane.
diagrams
addition,
to the
Additionally, the
plane
streamline-to-streamline)
location
speed,
by
provide incompressible
significant
of
temperature,
previously.
analysis, For
designer
have flow
model
(hub-to-tip)
velocity
energy
efficiences
turbopump
pressure
(i.e.,
streamline
is prescribed
characteristic pump
the
rise,
pressure,
flowrate
headrise.
to
pressure fluid
speed
assembly, the
given
and
Radial
or estimated
flow
5 ; nomenclature 5), the
continuity,
generally
(blade-to-blade)
in the meridional (fig.
streamline.
pumps
three-dimensional
a two-dimensional
in figure
each
of axial
18). The
circumferential
illustration
at
loss data
by
the flow
from
Losses
Specifications
in the
is illustrated
model
at a given
experimental
stage
flow
determined
considerations.
flowrate
stage
flow
design (ref.
approximated
to describe
a pump
flow
are
average
can be used
model
Design
required
the
points of the
headrise
is
N
= NQI/2H -3/4 S
_=
.30
-I12HI14
Ds = DQ 1.0
..... _._ .40 " TIP CLEARANCE _,_¢.", .40 RATIO 2 X 10-4 _ .*"- T..._ - _ _-... ",,._",.. 50 T/=.20
. 30
H -
DISCHARGE MINUS HEAD
.."--_-'_-._.._.. \\r'- 77=. 60 .40 _---. __.._-..'..'_,"
STATIC
INLET
HEAD
TOTAL
-N.-N--N-PITOT
]._.u .... "C "_'_ _ " .""''-"'... _--_,,_ "- "...."''".30._
"_L'-''-- _oRAGRy DISPLACEMENT ..... PARTIAL EMISSION --CENTRIFUGAL OR AXIAL
.._ _ -. :. _.- ....,r-_ .5o"_ v \ __
0.1
__I
I
J
;
/
_O_otoL,
_
_x'-
_,_\\_".6o___
.---_ _._'._\'-__o__-.____
.io
ba
TiP
,, 8o ,'--.
CLEARANCE
\
\ _
_
",
" _'\X\_._
"_O
_
___
P, AT,O _x ,o-3, -,., \\_,,,--:---.--.)_ ___ .80
¢)
M I O M-I <>MARK OMARK QMARK _"A" "B" Q"C"
O.O
i
I
NSITION STAGE TRA MAIN STAGE 9 AND 15-F 25 AND ALSO "E" STAGE 26 AND ALSO "D" STAGE STAGE STAGE STAGE
l f I lO2
\
Ik NASA rH/r •
--\\_'_C_. -_'_'-------"_ -\x _-¢_\\_
T = 0.4
(REF. !.5 )
NASA rH/r T = 0.7
(REF. 16 )
41,NASA
I
\
AXIAL_
I
rH/r T = 0.8
I I
I
_
--'_
-_/)
(REF. 17)
I
I
i I
i
i
104
103
I I I 105
N S
Figure 4. -
Approximate
N s -- D s diagram
for single-disk
pumps and low-pressure-ratio
compressors
(adptd.
from
ref. 14).
ROTOR
MERIDIONALI, STREAMLINES
STATOR TIP
_--
HUB (A) VIEW
N HERIDIONAL
I
PLANE
I
I
I
I
w _2
/Y AXIAL
/
\
TANGENT CAMBER
ul
CL
/
--
"_'\
)
U
Vlu I
(B) VIEW IN CIRCUMFERENTIAL AT A GIVEN STREAMLINE
Figure
5. -
Typical
flow
model
13
PLANE
for an axial pump stage.
TO MEAN LINE
MEAN
CAMBER
CHORD
L INE
!' \/ _
1¢/2 .
AXIAL
O[ STAGGER _
CAMBER
ANGLE ANGLE
_
DEVIATION
Wl
FLUID
ANGLE
INLET
(RELATIVE)
VELOCITY
(RELATIVE) _12 FLUID FLUID i
INLET ANGLE ANGLE EXIT
INCIDENCE
ANGLE
w2 C FLUID CHORD S
EXIT VELDCITY LENGTH
TANGENTIAL
SPACING
Figure 6. - Nomenclature for axial pump blading.
determined
from
point
the
at
the
summation
end
of
of incremental
each
pumping
isentropic-enthalpy
increase
by
then
be determined
and
in sizing
With
inputs
parametric evolve
of volume study
an overall
requirements for flowrate
used
the
flowrate,
usually pump
efficiency. local
normally
densities areas
at
will satisfy
enthalpy.
determined
the both
by at the
in the
speed,
conditions that
studies
Propellant
turbopump
is conducted,
in isentropic is
flow-passage
headrise,
configuration
(ref. 19). The and headrise:
increases
increment
and
off-design
50%
streamline
using
state
state
the
points
can
pump. requirements, being
hydrodynamic
are conducted
The
dividing
the
and
a
used
to
mechanical
dimensionless
forms
Va
= -6
gcH ¢'-
(3)
r/n (V2u
U2
-
-- Vlu)
U
(4)
where _b = flow g a =
fluid
coefficient absolute
U -- blade
tangential
qJ -- head
coefficient
gc = gravitational
H = headrise,
riri = stage
velocity
(axial),
velocity,
conversion
ft/sec
ft/sec
factor,
32.17
lbm-ft* -lbf-sec 2
ft-lbf* -lbm
hydraulic
efficiency
(sec.
2.2.2.1.2)
V2u = tangential
component
of absolute
velocity
at rotor
exit,
Vxu -- tangential
component
of absolute
velocity
at rotor
inlet,
ft/sec ft/sec
lbm-ft *The
use of gc in units
make
equation
of 32.17
(4) and similar
lbf_sec2
equations
instead
of g in units
dimensionally
correct
15
of ft/sec 2 and the use of H in units under
all environments.
of
ft -lb f -Ibm
instead
of ft
Typically the procedure involves a preliminary selection of blade and vane profile type (sections 2.3.1.1 and 2.4.1.1), velocity diagram type (section 2.2.2.2), hydrodynamic loading (section 2.2.2.1.1), and solidity (section 2.2.2.4), all of which areinterrelated with flow coeffcient, head coefficient, and efficiency. Blade tip speed(which is related to rotor diameter) and flow coefficient (which is related to blade tip speedand hub/tip ratio) are then studied over a range of values to determine stagehead coefficients and the number of stagesnecessaryto produce the required headrise.This procedureallows a number of pump configurations with different diameters and lengths to be examined in terms of stress, weight, and turbopump critical speed. Changesin the preliminary selections of profile type, velocity diagram type, etc., may be necessaryin the processof arriving at a suitable configuration. When the configuration has been established, velocity diagrams, hydrodynamic loading, and efficiency at each streamline in the meridional plane are determined, and the blade and vane angles(section 2.2.2.3) are selected to produce the desired diagrams. Off-design performance is then estimated, iterations againbeing madeif necessaryto compromiseproperly the design-point andoperating-rangerequirements. Basic design data for the state-of-the-art stages are given in tables II and III. In the multistage pump configurations, the stageshavebeen essentiallyidentical within the given configuration. The M-1 pump had a lightly loaded "transition" stagebehind the inducer to provide a uniform head distribution to the first mainstageand to provide better cavitation characteristics than were possiblewith the more heavily loaded mainstage.Additionally, the M-1 pump had a linearly decreasing outside diameter between the third-stage stator dischargeand the fifth-stage rotor dischargeto account for hydrogen compressibility. Stages for the Mark 9, 15-F, 25, and 26 pumps were identical within each configuration with the exception that none of the last stagesin thesepumpsutilized a stator. The flow path on the Mark 25 was adjustedlinearly from the first rotor inlet to the last rotor dischargeto account for hydrogen compressibility. Adjustments for hydrogen compressibility were not incorporated in the designof the Mark 9, 15-F,and 26 pumps. 2.2.2.1
BLADE
2.2.2.1.1 Blade High
Blade Loading
loading.
increase
-
in the
blade
in order
LOADING,
The
to
obtain
and Stall
energy
tangential
loading,
STALL
velocity stage
AND
EFFICIENCY
Margin
added
i.e., large high
MARGIN,
to the
fluid by the
of the
fluid
changes
between
in tangential
headrise
and
pump the
velocity
therefore
rotor rotor
(V2u-
a small
blade inlet Vtu
number
an increase in blade loading could turning in the direction of rotation,
be achieved However,
would
relative
(w2),
the
fluid
velocity
at the
16
rotor
exit
and
the
of the
rotor
exit.
in fig. 5), are desirable
observed in figure 5 that that would increase fluid decrease
is a function
of stages.
thereby
It can
be
by a blade shape this configuration increasing
flow
Table
II. -
Stage
Design
Data
for
Axial-Flow
Pumps
Diffusion
Factor
Rotor _T
_T
Ns
Ds
Stator
Hub
Tip
Hub
Tip
0.037
0.392
0.333
rH
rT
6.80
8.00
0.850
0.396
0.126
7640
0.0400
0.916
0.352
6.80
8.00
0.850
0.420
0.258
4470
0.0478
0.894
.448
.598
.477
.477
Mark 9
2.99
3.61
0.829
0.294
0.226
4450
0.0511
.52
.41
.57
.54
Mark 15-F
Same
as Mark 9
Mark 25
Same as "E"
Mark 26
Same as "D"
"A"
stage
3.13
3.63
0.861
0.390
0.279
4000
0.0511
0.87
.58
.48
.58
.56
"B" stage
3.13
3.63
0.861
0.390
0.316
3650
0.0533
0.87
.64
.54
.64
.62
"C"
stage
3.13
3.63
0.861
0.390
0.338
3460
0.0538
0.86
.68
.57
.68
.66
"D"
stage
3.13
3.63
0.861
0.390
0.35
3380
0.0542
0.84
.72
.61
.72
.70
"E"
stage
3.13
3.63
0.861
0.465
0.35
3220
0.0500
0.85
.53
.47
.55
.50
v = 0.4
1.78
4.53
0.393
0.284
0.135
10650
0.0272
0.800
.593
.223
.577
.373
NASA v = 0.7
3.15
4.50
0.700
0.294
0.282
4760
0.0429
0.937
.693
.426
NA
NA
3.60
4.50
0.800
0.466
0.391
385O
0.0436
0.955
.631
.664
NA
NA
Stage
M-1 transition M-1 main
NASA
stage
stage
(rotor
only)
NASA
v = 0.8
(rotor
only)
NA = notapplicable
Table III.-
Design Parameters
Chord, Pump
Profile
for Blade and Vane Profiles in Axial-Flow
Maximum thicknessto-chord ratio
in.
Camber, deg
Stagger,
deg
Profile type Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Blade Vane
C-4 C-4
1.430 0.991
1.315 0.991
1.200 0.991
O. 120 0.08
0.095 0.08
0.070 0.08
1.037 1.19
0.872 1.08
0.745 0.99
19.30 14.60
13.85 17.50
3.40 43.60
47.30
53.10
59.20
33.50
32.72
39.90
M-I mainstage
Blade Vane
C-4 C-4
1.302 1.057
1.176 0.968
1.070 0.882
0.120 0.t5
0.097 0.15
0.075 0.15
1.43 1.42
1.13 1.18
0.915 1.00
27.86 26.04
21.86 29.57
18.50 37.56
36.39 36.83
45.02 35.27
51.17 34.12
Mark 9
Blade Vane
1.37 0.87
........ ........
1.36 0.98
0.09 0.117
........ .......
0.05 0.065
1.21 1.9
........ ........
1.05 1.8
18.5 40.0
........ ........
12.0 40.0
61.9 38.0
........ ........
67.9 35.3
Mark
Blade
M-1 transition stage
15-F
Nonstandard Nonstandard
Vane
Oo
Solidity
Pumps
.......
Same as Mark 9 ...............................................
Mark 25
Blade Vane
......
Same as "E" --
Mark 26
Blade Vane
......
Same as "D"--
"A" stage
Blade Vane
Modified double-circular-arc
0.923 1.00
........ ........
0.892 0.924
0.140 0.120
........ ........
0.051 0.050
1.37 1.4
........ ........
1.14 1.3 l
38.7 36.9
........ ........
21.36 34.8
40.35 37.38
........ ........
54.39 40.35
"B"
stage
Blade Vane
0.912 0.99
........ ........
0.872 0.860
0.138 0.121
........ ........
0.052 0.054
1.34 1.31
........ ........
1.11 1.3
43.1 39.4
........ ........
24.9 39.4
39.05 35.12
........ ........
53.75 38.99
"C"
stage
Blade
0.899 0.980
........ ........
0.852 0.850
0.14 0.122
........ ........
0.054 0.054
1.33 1.30
........ ........
1.09 1.29
46.0 43.0
........ ........
27.1 43.1
38.0 34.1
........ ........
53.15 37.85
Vane Blade Vane
Modified double-circular-arc
0.909 0.975
....... .......
0.836 0.874
0.132
........
0.062
1.33
........
1.06
50.0
........
28.9
36.5
........
52.22
0.123
........
0.058
1.33
........
1.29
47.8
........
46.1
34.1
........
36.85
"E" stage
Blade Vane
Nonstandard Nonstandard
1.667 1.538
....... .......
1.268 1.394
0.104 0.085
........ ........
0.081 0.089
1.9 1.7
........ ........
1.28 1.61
45.9 35.1
........ ........
29.1 36
30.0 41.1
........ ........
41.0 38.9
NASA
v = 0.4
Blade Vane
Double-circular-arc
1.50 1.50
1.50 1.50
1.50 1.50
0.10 0.08
0.085 0.08
0.07 0.08
2.5 2.37
1.44 1.36
1.0 0.95
61.35 62.20
13.84 51.40
5.43 44.15
19.97 20.34
........ ........
70.19 9.61
NASA
v = 0.7
1.52
1.52
1.52
0.085
0.0775
0.070
1.44
1.19
1.01
27.6
19.8
0
52.2
60.5
67.1
1.49
1.49
1.49
0.09
0.08
0.07
1.25
1.11
1.00
43.4
27.0
39.8
46.2
55.0
"D" stage
Blade (no stator)
NASA
v = 0.8
Blade (no stator)
42.7
diffusion large
in the
amounts
in total
blade
row.
Large
of diffusion
headrise.
tend
In reference
velocity
gradients
to produce 20,
thick
a blade-loading
on
the
blade
or
angles
and
was used for correlating loading, i.e., the loading axial-flow head
and
In the over
pump
work,
loading
limits.
notation
the
stage
solidity.
experimental at which the the
of figure length,
blade-row
diffusion
boundary
layers
and
based
on diffusion
5, diffusion
are given
In axial-compressor
total-pressure-loss data boundary layer separates factor
has been
factors
for
associated
parameter
row was developed for axial-flow compressors. The parameter, the index of local diffusion on the blade suction surface in terms velocities
surfaces
used
the
similarly
rotor
and
relatively
diffusion of fluid work,
and from
the
high losses in the blade
factor DF, is an inlet and outlet diffusion
factor
for indicating limiting the suction surface. In
in estimating
stator,
with
with
loss in total
constant
radius
by the expressions
(DF) R = 1 -
w2 Awu -- + -W 1
(5)
2OWl
where (DF)R
= diffusion
factor
Wl
= fluid
inlet
W2
= fluid
exit
Aw u = change ft/sec
for rotor
relative
velocity,
relative
velocity,
in tangential
cr = solidity
ft/sec ft/sec
component
of relative
velocity
--- wl sin/31R - w2 sin/32R,
= C/S (fig. 6), dimensionless
and
V3 (DF) s = 1 -
V--_+ 2aV----_
where (DF)s
= diffusion
VE = fluid
inlet
V3 = fluid
exit
factor absolute absolute
for stator velocity, velocity,
AV_
ft/sec ft/sec
19
(6)
AVu
= change
in tangential
e = solidity
values
given
II.
Values
Stall
is the
table
of absolute
velocity
= V2u -
V3u, ft/sec
as above
Diffusion-factor in
component
for hub
and
tip
for the
generally
fall
state-of-the-art
within
the
axial-flow-pump
range
blading
appropriate
to
are
axial-flow
compressors. Stall
margin.-
stage
when
headrise in
the
flow
drops
with
separation abruptly.
state-of-the-art
blading (table factor reached the
as the
of
in the Three
axial
pumping rotor
or stator
different pumps.
capability
factor,
ratio
the
with
been the
used
Mark
in an
axial-flow-pump
to the
point
to define 9 and
"A"
where
the
is an index
relative
of blade-passage
the
stall point
through
when the rotor-hub or stator-tip factor (RF) dropped to a value
factor
of fluid outlet-to-inlet
occurs
progresses
have
Experience
retardation
that
passage
conditions
II) indicated that stall occurred a value of 0.75 or the retardation
diffusion
defined
loss
"E"
diffusion of 0.5. As
diffusion.
It is
velocity:
W2
(RF)R
-
(rotor)
(7)
(stator)
(8)
Wl
V3
(RF)s
In the
M-1
which
the
equivalent
design,
stall
equivalent diffusion
prediction
diffusion factor
was ratio
blade suction surface ratio of 0.10, the
and stator
constant
radius
COS/32R (DFeq)R
-
based
_
on
or factor
is approximately
velocity on the thickness-to-chord with
- V2
the
DFeq
equal
reported
is used
to the
to the fluid outlet equivalent diffusion
ratio
an
an indicator
of the
relative factors,
in reference
fluid
21, in
of stall.
maximum
The
relative
velocity. For a blade with at minimum loss, for rotor
are
0.61
.12 + -_
[Vla]
Wl sin/31R
(9)
[Wl ]
cosmos{06, (DVoq s-
method
72O
[V uV3u]
IV2] [v2a
(10)
In the M-l, stall was consideredto occur when either DFeq had a value of 2 (ref. 19). Figure 7 showsthe relation of ideal head coefficient, diffusion factor, efficiency, and flow coefficient for an axial pump stagewith a reaction of 0.5 (sec.2.2.2.2) and a solidity of 1.5. The curves are based on analysis at the 50% streamline (pitchline) and, as noted on the figure, the efficiencies do not include tip clearanceloss or other secondary flow losses. Examination of the figure indicates that a design point could be selected to obtain maximum stage efficiency. However, in the state-of-the-art pumps, the design point selection has been made consistentwith maintaining a stall margin. The trend has been to designfor higher flow coefficients to take advantageof the increasedideal head coefficient at a given diffusion factor. However, for a given flow, pump diameter, and speed,a limit is reached, since as the flow coefficient is increased the blade heights become small and the tip leakagelossesbecome increasinglysignificant. Note also from the figure that increased stall margin (decreaseddiffusion factor) for a given flow coefficient, reaction, and solidity can be achieved
only
at the
expense
of ideal
head
coefficient. 7/ .80
1.4
1.2
EFFICIENCIES ARE BASED ON LOSS DATA SHOWN IN FIGURE 8 AND DO NOT INCLUDE TIP CLEARANCE OR OTHER SECONDARY FLOW LOSSES. REACTION R - 0.5 SOLIDITY _1.5 OF - DIFFUSION FACTOR - STAGE EFFICIENCY
.84
.86
1.0 .88
b-
0.8
.9O
¢9
w O ¢9
.e DE
.gT 0.6 .915
OF 0.8
0.4 .91 .90 .88
0.2
.86
FLOW COEFFICIENT
Figure 7.-
Relationship of stage ideal head coefficient, diffusion factor, and flow coefficient.
21
efficiency,
2.2.2.1.2 Stage
Efficiency
hydraulic
efficiency
is defined
as the
ratio
H
Hi
of actual
headrise
H to the
ideal
headrise
Hi:
r/H
where
Hloss
is the
Stage
head
loss
losses,
and
can
flow
have
state-of-the-art estimated, M-1
broken
efficiencies
methods
the
of the head
be
secondary
Design-point Two
sum
the
pump,
reference reproduced
here
losses
in the into
to
Mark
by
pump
were
based
on
as figure
8 (adptd.
the
losses, boundary
25,
and
and
in table
hydraulic
26 pumps,
hub-to-tip
end-wall layers
are given
losses
from the
profile stages
predict
9, 15-F,
(11)
stage.
blade
loss distribution
loss) incidence
Hloss
Hi
produced
used
In the
radial
losses
losses
down
been
(minimum
Hi
for the axial-flow
pumps. and
-
-
was
friction and
for axial-flow
from
18). The
ref.
efficiency
an average assumed
compressors.
total-pressure-loss
clearance.
II. in
efficiency
was In
correlation
at
The
correlation
parameter
COS/_exit
(12a)
2(/
using
the
conventions
= total-pressure
of figures
(head)
5 and
6,
- loss coefficient
Hloss (rotor) w_/2g
(12b)
c
Hloss
(stator) Vz2/2gc
_exit
=
fluid
angle
(12c)
at the rotor
or stator
exit
22
(/3z_ or /33s in figure
5), deg
is
is given
by
where,
the
to be constant.
diffusion-factor/total-pressure-loss
developed
tip
(annulus)
PERCENT
OF
ROTOR
BLADE HEIGHT T P TO HUB
FROM
.14
r_ p-
SHADED .12
AREA
IS REGION
OVER
WHICH
TOTAL-PRESSURE-LOSS PARAMETERS FELL FOR THE IO-PERCENT STREAMLINE
lO
.I0 •08
I r_
oq U
_13 MJ D_ I J
O
•06 , 50, •04
90
O. 02
AND
70, ALL
STREAMLINES
I 0. I
.2
.3 DIFFUSION
I .4
1
I
.5
.6
FACTOR
Figure 8. -- Variation of total-pressure-loss parameter with diffusion factor at reference incidence for NACA-65-series and double-circular-arc blades (adptd. from ref. 18).
.7
In the
design
methods
of
given
determine based
in
the
tip,
than
lower
incidence
portion
of the
for the
streamlines
reference
16
that
parameter
for
axial-flow
correlating
near
minimum-loss
18. The
given
total-pressure-loss 16 indicated and in view
blading,
loss parameters
results
especially
M-1
in reference
the on
the
those
shaded
at and
determined
angles
near
area
the
indicated
were
pump
in figure
rotor
that
by the
8 was used
tip. This
the
rotors
for axial-flow
determined practice
magnitudes generally
compressor
to was
of
were
rotors.
the
lower,
Reference
that a specific explanation for the lower magnitudes was not readily apparent, of limited number of rotors that were tested, no generalization of the results
was attempted. It is important in the
to note
design
the that
correlation likely
care
be
discussed
2.2.2.2 any
the
selection
stage
the
margin. has
also
blade
varies location
parametrically diagram (ref.
and
28).
It
correlation. to
the
22, which,
the
of ref.
the
Effort
the
rotors
total-pressure-loss
in which
flow
to
design
of
are reported
static
headrise
stall
extend the
in 16),
parameter
outlet
angles
are
range
of
the
axial-flow
in references
in the
pumps
23 through
that
maximum
the
axial
26.
would
radius;
(i.e., that
was
concluded in using
the
velocity
fluid
and in
of
stage
by of the
sets In
been since
the
extent
stage. has
desirable,
the
is equal
then
the
stall margin for
ratio
to a great
reaction
is therefore stage
efficiency
of the
selection
stator
attainable
R = 0.5)
profile
This
and
selection
a maximum
been used is constant
a symmetrical pump
design
if a forced-vortex offer
benefit
stage).
established
as the
the stall
equal
can be obtained. occurs
to one-half
the
with
a
velocity
of
27).
selected
determine
in
is in effect
is defined
rotor
magnitudes
factor
diagram
diagram
R (reaction in the
margin
implies
with A
of velocity
reaction
flow pattern has fluid axial velocity
only.
type
stage
stator
(ref.
inversely to
in the
dominant
when
element
at all radii
appreciable
and the
shown
A free-vortex radial free-vortex flow the velocity
the
indiscriminately
in reference
(including
applying blading
to be accomplished
velocity
rotor
diagram
the rotor
for
to the
efficiency
been
symmetrical
radial
rotor
pumps,
in the
when staggered
of this effort
pump,
value
diffusion
A symmetrical
diffusion
be used
is emphasized rotors
subsequent
Results
8 cannot
DIAGRAMS
in the
of velocity
restriction
used
conducted
in an axial-flow
state-of-the-art
It
was
monograph.
in figure
axial-pump
exercised
range
of a design
headrise
influences
be
given
This three
18 to highly the
VELOCITY
For
of
should
data in this
degree
pumps.
results
outside
experimental
static
test
of reference
to
the correlation
of axial-flow
summarizing cautions
that
lower that
the
loading, for
selected
hub/tip
in all state-of-the-art axial pumps. In the from hub to tip while the fluid tangential diagram with flow
better ratios
forced-vortex
24
therefore
a hub-tip
ratio
pattern
that
efficiency, of flow
can be achieved of
imposed or higher
0.8
and
pattern.
higher,
0.8
was
at one studied
a symmetrical head
coefficient
there
Additionally,
was
no
it was
concluded that with the free-vortex flow pattern, the radial location of the symmetrical diagramwasnot critical aslong asit wasnearthe meanradius. Blockagefactors are included in the determination of velocity diagramsto account for the reduced flow areasresulting from end wall and bladesurface(profile) boundary layers.The magnitudes selectedfor these factors are dependent on the particular designaswell as the designmethod being used. A designmethod employing appropriate experimental loss data (usedon M-1) automatically accountsfor profile boundary-layer blockageand it is necessary only to account additionally for end-wall boundary-layer blockage. In the M-1 pump, this blockage was estimated as 4% of the annulus area.A designmethod in which averagestage efficiencies are estimated must utilize blockage factors that take into account the area reduction due to both end-wall and blade-surfaceboundary layers. In the Mark 9, 15-F, 25, and 26 pumps, these factors were estimated at approximately 10%of the annulus areaon the basis of compressor blade-surfaceand end-wall boundary-layer information obtained from reference 18. Analytical and experimental investigations of stages with impulse blading have been conducted (refs. 29 and 30), but to date this type of blading hasnot been utilized in rocket enginepumps.
2.2.2.3 With
BLADE the
ANGLES
velocity
diagrams
established
in the
meridional
plane,
the
blade
angles
and
blade
shape are selected to turn the flow in accordance with the desired diagrams. This selection involves determination of the incidence angle, camber angle, and deviation angle at each of the
hub-to-tip
and,
for
streamlines
the
pumps
considerations. turning prediction For angle
)
rotor 18, 27,
deviation
angles.
were
determined
basis
of experimental
Mark good
15-F selected
26),
the
design
the
6).
angle
the
and
row
M-l,
of the
data a design basis
deviation
would
that
reduce
methods
incidence In the
of achieving angle
the
angles
angle
selected, deviation
stage
stage
established
from
of
work
angle Mark
25
compressors testing
of +3 ° was and
selected
26 pumps,
low loss. In all of these
was determined
from
25
the
loss)
pumps
relationship
Accurate diagrams.
and vane
camber
of deviation
8%. angles
and
and
deviation
in reference in air.
In the
on the
basis
a design
of fluid
6.
desired
blade
by about
designer cavitation
amount
angle
1° in prediction
minimum
for
or
the
with
of incidence
low-Mach-number
by the
the
in achieving
an error
(at
is selected minimum-loss
0 and
for selection
incidence
obtained
angle important
31 give procedures
In the
angle either
of a 50%-reaction
of 1.0 indicated
and
on
the incidence
is extremely
analysis
incidence
based
camber
a solidity stator
The
was with
on
performance. on
6 that,
27,
by use
pumps,
cavitation
was
figure
deviation
of 30 ° and
References
on
in reference
for the
5 and herein,
is dependent
of the
example,
angles
of
Note
(/31 -/32
(figs.
discussed
incidence (Mark
(ref.
prediction angles 18 on the Mark
of achieving angle
9, 15-F, 31)
9 and of 0 °
25, and
0 _6-
(13)
where a = distance
flexit
fluid
=
to the point
exit
angle,
of maximum
camber
from
leading
edge,
in.
deg
flexit
- dimensionless
ratio
.50
Subsequent fluid to
to the
were
conducted
design slotted
2.2.2.4
SOLIDITY
in general,
increases increase the
as the
solidity increasing
desired the
length.
Here,
weight,
and
2.2.2.5
CAVITATION
Rocket
engine
provide high
the
suction
is sufficient
the
turbopumps
capability performance to preclude
cavitation
test
the
flow
skin
data
test
applicable
obtained of
fluid
(high
with
problems
limits
friction
number increased
with
incorporate at low
inlet
the
pressure following
26
the
data.
solidity)
included inlet
angle,
These
values
area).
pressures. of the stage
as possible,
as the
reached,
because coefficient
since for
that
conflict
may
an attempt and
row
a given with
can be made
increasing
length,
losses
the
chord
increased
rotor
rotor.
initial
Inducers pumped (ref.
head
Additionally,
a case,
of blades
turbopump
inducers
hydrodynamic a hydrodynamic
the ideal
are soon
blade
III.
Both
From
a requirement In such
the
in table
coefficient
blades,
standpoint.
in the
as the
multiple-circular-arc,
selection.
blades
and
thin
by reducing
increase
data
range
of available
9). However,
is confronted
and
wide
given
into
(increased
for operation
The
are
as many
(fig.
as a rule
water
of experimental
18.
stages
limits
require
critical-speed
using
double-circular-arc,
a relatively
factor,
a structural
solidity
range
tests
26).
enter
to use
cascade
reference for
over
the
diffusion
stages
designer
potential
within
solidity from
desired
in
axial-pump
is increased
high-solidity
thickness
to achieve
profiles
desirable
of reaction,
with
given
considerations
values
length,
an extended
23 through
selected
herein,
performance
(refs.
design
it is usually
fixed
those and
state-of-the-art
been
mechanical
chord
of
discussed
to provide
as
solidity
(o)
standpoint for
and
solidities
have,
pumps
double-circular-arc
angle,
and
such
characteristics
camber
The
of the
in order
methods
fluid-turning and
design
are designed fluid
2).
pumping
element
to achieve
to a magnitude
In order
to
to achieve
that high
].00 n-
0.90 __
0.6
-JO.80
0.5
0.4
% I-z w
---O.3 w O i_)
_0.2 //
] REACTION R = 0.5
//
0.1
I 0.5
Figure
suction
performance,
therefore
must
pumping
elements
conditions inlet
All the
In
inlet a
flow
The with with
2. Discusssion
9,
design here
initial
more criteria
area
25,
and
were
inlet from
the
inlet loaded
have pumps,
the
necessary. pump
The
inlet
inducer
to the
Additionally, must
and
achieved
inlet
the
be compatible
inducers
headrise, ill the
exit
axial
as the _low
stage of the
exit with
the that
cavitation
for inducers types
initial
area
inducer
A discussion
and mainstage
27
and
stage
better blading.
practices
to transition
the stage
loaded
mainstage
utilized totally
inducer
recommended
will be limited
are
hub-to-tip)
accomplished
a lightly
heavily
and efficiency.
mainstage. from
26
was
mainstage
I 2.5
flow the
mainstage.
between
stage
and
angles
turbopumps
15-F,
stage
for the
at the
higher-flow-coefficient and
I 2.0
on head coefficient
in flow
requirements
transition the
of solidity
coefficients
axial-flow
Mark
conditions
possible
the
of the
"transition"
utilized.
along
the
flow
velocities
state-of-the-art
mainstage
flow
of fluid
Effect
a transition
requirements
element. pump,
low
provide
(i.e.,
flow
9.-
I I 1.0 1.5 SOLIDITY G
pumping
transition, stage.
mainstagc provided
performance or" inducer is presented of axial-flow
and
In the
M-1
inlet
was
acceptable than
that
cavitation ill rc['el-ellCe stages.
Analyses are conducted on the initial axial-flow stagesfollowing the inducer to ensurethat sufficient pressureis availableto prevent headrisedegradationcausedby cavitation. This can be accomplishedby comparisonwith cavitating test results of similar designsor by analysis of fluid velocities on the bladesurface.The cavitation parameter r r NPSH TR
(14)
--
u 2/2go
and cavitation
number
K
Pf
Pv
-
K -
(15)
pfw_/2gc
where NPSH
= net
positive
suction
static
pressure,
lbf/ft 2
Pv = fluid vapor
pressure,
lbf/ft
Pf = fluid
Of = fluid
which
are
density,
commonly
characteristics
lbm/ft
used
of the
references
15 through
in
figure
10
correlation
parameter
represents
the
flat-plate
were
lightly
0.426,
0
highly
32 through
cambered
the
NASA
for
parameter
according
inducer
also
been
to the
little
used
inducers
modified
head
parameter (ref.
magnitudes
to evaluate
the
cavitation
loss data
r r and 35).
at which
The
head
two,dimensional
for these
a cavitation solid
line
breakdown theory
rotors
are
breakdown in the will
figure
occur
of reference
for
35. The
(i.e., head breakdown) data for the NASA v = 0.4 and v with the inducer theory of reference 35. These rotors
or no
As might
v = 0.8 axial
34. Cavitation
cavitation
developed
with
have
on axial-flow blading are relatively limited. Several axial pump blade profiles have been tested and the results are reported in
the high-head-loss compare favorably
loaded
design,
blading.
of
-- 0 °, respectively).
fiat-plate
3
terms Z
inducers
shows that axial rotors
2
17 and
cavitation
figure = 0.7
the
in
ft-lbf/lbm
in inducer
axial-flow
Cavitation performance data rotors with double-circular-arc given
head,
camber
at the
be expected, rotor
(DF
theory.
28
tip (DF data
= 0.664,
from
= 0.223, the more
0
= 5.43°,and heavily
0 = 27 °) did not correlate
loaded well
DF
=
and with
.5
• :Iz
[]
(M
z II
w F,-
. 4
.Z
.3
/
o_
F-
1-2Z
.x9
g
(MOD,FIED TWO-
/
.2
•
DIHENSIONAL THEORY AS GIVEN IN PART 2 OF
/ /
/k/ 0.1 0.02
REFERENCE
.04
.06
BREAKDOWN
.08
CAVITATION
PARAMETER,
10.
-- Comparison for
In view velocities
were
static
pressure
pressure
(_T/2)
_l"
--rH/r T
REF.
17.1 °
0.4
15
• 0% 034%
22.9 °
0.7
16
• 5_ []46%
21.5 °
0.8
17
B T AS USED HERE IS THE BLADE TIP ANGLE MEASURED FROM A PLANE NORMAL TO THE AXIS OF ROTATION.
theoretical
double-circular-arc
and
experimental
cavitation
breakdown
parameter
profiles.
of the lack of cavitation performance data on similar designs, analysis on the blade surface has been used to evaluate the cavitation characteristics
state-of-the-art surface
of
.12
• 3% A40%
NOTE:
Figure
.I0
CORRELATION
Z = _ TAN
HEAD LOSS
35)
blading. on
to ensure
for predicting
In the
determined
Mark
by use
the
blade
that
incipient
the
9, 15-F,
25, and
26 pumps,
of a stream-filament
surface
was then
pumps
cavitation
would
method
determined
be
free
and
from
of airfoil-type
fluid (ref.
compared
cavitation.
blading
was
velocities 18). The
used
of fluid of the
on the blade minimum
with
the
local
fluid
vapor
An approximate
method
in the
method
M-1.
The
utilizes the equivalent diffusion factor DFeq (sec. 2.2.2.1) along with the cavitation parameter TR and the cavitation number K. An approximate value for the ratio of maximum fluid the
velocity equivalent
blade method solidity
on the
blade
diffusion
by use of the cavitation for
blade
= 1.5 are
profiles shown
surface
factor.
to fluid
This number
with in figure
ratio
velocity was then
K and
the
at the related
cavitation
a thickness-to-chord 11. For
a blade
29
blade
inlet
to the
NPSH
parameter ratio
row
with
of
derived
by using
requirement
for the
7R.
0.10,
a given
was
The
results
reaction design
flow
of this
= 0.5,
and
coefficient
1.4 REACTION
(R)
-- 0.5
SOLIDITY (a) NPSH
=
1.5
/ /
"/'R-- "-u2/2gc 1.2
DF
/ /
/
_'R
/ o--
l.O
•88 /
z
/
UJ
/ /
I.¢.
.8 UJ
o
.91 "-r
.6.915 DF
0.8 .4 --
0.2
--
•4 1
,
,
.3
5
"rR= 0"351 0
7
f
0.2
.4
OF
EFFICIENCIES FACTOR
FIGURE END-WALL Figure
11. -
Correlation flow
AT
ARE
8 AND
THE
DO
NOT
l 5
.8
1.0
COEFFICIENT
TOTAL-PRESSURE-LOSS
SION
1.3
I
VALUES FOR CAVITATION PARAMETER ON BLADE WITH THICKNESS-TO-CHORD STAGE
l.l
.6
FLOW NOTES:
.
I
BASED
( _R ) ARE RATIO OF ON
THE
PARAMETER 50%
INCLUDE
CORRELATION
VERSUS
STREAMLINE
BASED O.IO.
AS
DIFFU-
SHOWN
TIP-CLEARANCE
ON OR
LOSSES. of pump
coefficient,
cavitation
stage efficiency,
3O
parameter and
with
diffusion
ideal factor.
head
coefficient,
and
ideal
head
parameter (inducer) 2.2.2.6
coefficient,
rR. stage
required
NPSH
can
With the required NPSH then can be determined.
known,
the
OFF-DESIGN
Analysis
of
predicted
for
engine
the
Reference
19
which
transients,
gives
losses
Mark
and angle
values. Off-design component that the test
are given
be
mixture-ratio
made
the
requirement
so that
excursions,
predictions
was
was
Mark and
the 25
for
36
assumed
and
to
to vary
more
were
blade
the
(2)
be
pump
and
cavitation
of the
previotls
M-1
performance
based
on
the
(1)
a one-dimensional at
to results
can
chamber-pressure
as to how
constant
according
reliable
conditions of predicted
in reference
prediction
predicted
channel
with
losses
loss was assumed resulted from the
off-design flow results. Curves
pump
from
the
excursions.
deviation
the point
angle
and
two-dimensional
mean-line
design
be
analysis*
value
of low-Mach-number
in
and
blade
air tests
at the
at a higher level to account for tip leakage and end-wall boundary-layer of predictions with M-1 test results indicated that the one-dimensional
analysis
15-F
deviation
must
reference
assumed
50% streamline but losses. Comparision mean-line
of
angle were
determined
headrise
methods involve assumptions with incidence angle.
off-design
method
deviation
channel
performance
prediction losses vary
blade-element
be
PERFORMANCE
off-design
In general, blade-element
the
method.
Off-design
a one-dimensional
were
assumed
mean-line
to be constant
to be equal to the velocity variation in incidence angles
(see sketch below). This and measured headrise
method versus
performances analysis at their
of the in which
design
point
head of the normal velocity on the rotors and stators at correlated favorably with flowrate for the Mark 25
9. j_
- /THIS •
_
IS THE NOAMAL VELOCITY C_PO_ENT
'NVt_ i
THE _ REPRESENTS THE
_ ",.
OFF-DESIGN
CONDITION
I\
In general, the off-design performance engine mixture-ratio or chamber-pressure any
problems.
distinct Analysis
However,
operating based
levels
on values
there during
was the
at the 50% streamline
required of the excursion (at a tendency
engine
transients
(fig. 5).
31
for
state-of-the-art design point
the in the
Mark
15-F
J-2 engine
pumps in regard to thrust) has not caused pump start.
to stall
at three
Start problems are not unique to engineswith axial-flow pumps but are related to the interaction of the pump (whether axial or centrifugal) and the thrust chamber.For example, the RL10 engine, which employed centrifugal pumps for both oxidizer and fuel, had start problems similar to those on the J-2. Both the J-2 and the RL10 areregenerativelycooled enginesin which the hydrogen flow from pump dischargeis routed through tubes around the thrust-chamber walls and servesas a coolant before it is injected into the combustion chamber.Start anomalieshavebeen associatedwith reduced fuel flow coincident with rapid increases in thrust-chamber coolant-circuit pressure or in chamber pressure. The consequenceof the reduced fuel flow (which in the axial pump may be a stalled condition) is somewhat dependent on the enginesystem cycle, but in severecasesthe typical result is damageto the thrust chamberas a result of inadequatecooling. Figure 12 illustrates the three potential stall points of the Mark 15-F pump during the J-2 engine start transient; these potential stall points were termed spin-down stall, LOX-dome-primestall, andhigh-speedstall. Spin-down
stall.
discharge) by
and
the
tubes.
prior
pump
and
During
vaporized. zone and
- Spin-down to main was
this
stall
propellant
discharged
interval,
the
undershoot.
stall
initial
ignition.
was
The
fuel-pump
fuel
unrestricted
warmed
by
the
was
flowrate dropped. by an initial
avoided
on
the
acceleration
flowrate
was
thrust-chamber
the
and
accelerated cooling
a portion
was
of the upper combustion The head demand on the
Thus flowrate by
rapidly
chamber,
resistance quickly.
J-2
(start-tank
into thrust
encountered the high pressure drop increased
increased, and was characterized
Spin-down
during
relatively
fuel
The warm fuel then injector, and injector
fuel pump quickly chamber with fuel
occurred
the process overshoot
thermal
of priming the followed by an
preconditioning
of the
chamber. Several
methods
pre-start
purge
development, different a
from
technique
was
the combustion
on
the
cooling
tubes from
chamber stall.when
point
during
On
the
pre-start
purge
and
and
start
fuel J-2X
employed
standpoint
an index
during
dump
cold of the
the
low-level
second
stall
point,
indicated
the
oxidizer
flow
primed
the
dome
degree
some
produced of chill
a
from
preconditioning have
The hydrogen
been
the
is
employed.
high resistance
line
around
the
cooling-tube-bypass as a film coolant
in
operation. as
LOX-dome-prime
manifold
of the
to over 150 psi. This increase in chamber fuel-pump head. Since the speed change
32
to 8 sec.), With
each
a bypass
successfully.
(up
helium.
to avoid
engines,
The
chamber pressure quickly increased an immediate demand for increased
lead
However,
alternatives
of utilizing
and
fuel
Thrust-chamber several
an overboard been
with starts.
so that
uncertain.
J-2S
a long
satisfactory
of view,
the
has the
utilized:
chamber, was
incorporates start.
attractive
12, occurred
and
were
produced
measurement
during
LOX-dome-prime figure
gradient
system
injector
thrust-chamber
nitrogen, methods
an operational
RL 10 engine the
preconditioning
cold
of these
temperature
undesirable of
with each
temperature
single
The
of chamber
stall injector,
on and
pressure caused could not occur
50× 103
FLOW STALL
REQUIREMENT
LINE _/
CONSTANT j___j
AT
O/F
4O
_'_A"_s_° / l\
_o,._
3O w
Q.
/,
a-
2O i,i h
.'/¢1_
"\\ 2s oooRP.
\\\
SP,N_DOWN_J_/_2\_ORIGINAL
lO
S/ALL
_'_.,
2 FUEL Figure
12.-
START
Potential
stall
\
17 000
RPM
1
1
[
1
4
6
8
IO
PUMP points
INLET of
Mark
FLOW, 15-F
33
pump
O3 xl
GPM during
start
of
the
J-2
engine.
instantaneously,
pump
shown
12, the
high
in figure flow
and
Control achieved most
of
fuel-pump
by
regulating
engines,
engine
start.
Initial
J-2 by
two
system
the
pressure
in the
value;
the
in the
if the
main
sufficiently
high
level.
CLEARANCES
As
hub/tip
increasingly
high
ratios
are
clearances
high, to keep
stator-vane running
so
of
Mark
9 pump
tip clearances. tip clearance
effect
changing
higher decreased
than
that
head from
rotor
is not
as the
main
as
speed
line
from
to
stall)
was
prime.
The
dome
valve.
In
flow
during
the
dome
prime
main
start
was
both
oxidizer
the
early
were
J-2 and
portion
relieved
valve
increased
of
on the
was reduced;
by utilizing
the
control valve.
annular
a higher
has been
running
seeks
this O/F
achieved
by
became
thrust
built
up to a
had
area tip
becomes
approximately
was
designed
hub/tip small
on the 0.005
to operate
of
losses
stages
to maintaining
was
an
clearance
tip clearance
clearance
of
stall
the
given radial
M-1 pump
a vane
and
the
(O/F)
was
In state-of-the-art
rotor-blade
clearance
and
to
ratio
high-speed
clearance
area,
work.
nature
the engine
engine
tip
in
mixture
necessary
before
flow
stage
The
in. The
with
similar
so-called
the
attention low.
running
0.015
was
full open
of the
was
buildup,
The
increases,
considerable the
thrust
O/F
total
losses
stall,
for an oxidizer/fuel
oxidizer
pump
in air to determine
showed
the range
amount of was increased
of
of the
of the
12). During
axial
in.
tested
results
within
9 pump
Thus,
0.049
tip
Mark
in.;
the
at a blade
in. (this
was
a
- see fig. 3).
appreciable clearance Mark
at LOX
high-speed
reached
that was
was
The
flow.
nearness
following
oxidizer
oxidizer
is orificed
valve
these
0.020
configuration
point,
percentage
and
clearance
clearance
shrouded The
tip
set
of
pressure
main
opening
paragraph,
of
high
in order was
portion
the
an
percentage
relatively
pump
initial
level (fig.
with
of
an increasingly
the
fuel
a constant
contol
chamber
restrict
this
along
therefore
stall problems
stall
oxidizer
ratio
to
reduced
high pressure.
of the
J-2 engine
previous
flow
2.2.2.7
15-F
third thrust
a problem
become
(1)
The
design-point
a transient
(and
used
of
spin bottle.
The
oxidizer
the
was
during
stall.
and
expense
a throttleable
15-F pump
hydrogen-gas
as noted
flow
magnitude
changes:
restricting
underwent
utilized
speed
at the
coefficient
method
stall.-
achieved
to low
Mark
LOX-dome-prime 5.0 at the
pump
flow
this
pump
High-speed
fuel
technique
RL10
(2)
was
low pressure
effective
and
head
tested
clearance stator.
rotor
tip
the
(0.95%
pump
the
effect
of variations
in rotor
insensitive
to changes
was rather
to 3.25%
of vane
height)
was lost, particularly near the stall 1.58% to 3.57% of the blade height.
symmetrical
in the
that
The
clearance
was
not
unexpected,
and
the
static
air tests was
also
increased.
34
pressure showed (This
since
the
(ref.
the
clearance
stator
in stator
However,
an
point, as the rotor tip The more pronounced velocity
rise in the that
37).
and
stall
diagram
rotor margin
effect
for
the
is substantially of the
on stall
pump
also
was
observedin tests with liquid hydrogen but wasnot systematically investigated.)Attempts at determining the effects of tip clearanceon efficiency during the tests were unsuccessful becauseof the relatively smallrise in air temperature. Axial clearancesbetween blades and vanesin generalhavebeenselectedto minimize overall pump length while maintaining adequate protection against blade and vane axial interference or vibration during pump operation. In the Mark 25 and 26 pumps, the mechanical designof the bladesand vanesfrom a vibration standpoint wasbasedin part on maintaining a specified axial clearance between rotor and stator rows. The analysis for establishingacceptableaxial clearancesconsidersthe forced vibration amplitude of a blade or vane to be a function of the wakevelocity fluctuation of the upstreamrow. The velocity fluctuation in turn is a function of the relation of axial spacingto upstreamcord length and the proximity to resonance of the blade natural frequency with the wakes from the upstreamrow. The analysisis basedon information presentedin references38 through 42.
2.3
PUMP ROTOR
Tile
pump
and
the
rotor
assembly
thrust
(Mark
rotor
15-F
and
clamped consisting
2.3.1
Blades
setting
in section
(i.e.,
Research
and
so-called
series,
and
standard
the
"non-standard" profiles excessive axial
surface
pumps
will
rotor
have
blades,
been
the
Mark blades
rotor
ill the
structure
with
consisting
(Mark 9 and and dovetail
pump
utilized
rotor
assembly
through-bolts rotor structure
be
profiles
double
These arc
C-4) that
achieve
diffusion.
are listed
turn
in table
was were a
Values
design
the
turned
circular
(British to
will
fluid
tile
profiles
of
disks
25), and (M-l).
structure,
design
of the
integral
blading
(with
integral
a builtup
used. either
The
for the
III: typical
remaining
blade significant profile
35
the thickness
modified
prescribed
involves
desired angle
compressors
included
maximum
the
desired
for axial-flow with
procedure
so that
through
of blading
profiles
designed
of the
a one-piece
the hydrodynamic
that
fluid
standard
profile
concepts 13):
a builtup
2.2.2,
profiles
development
the
Three
consists
rotor
TYPES
of blade
achieved on
26),
here
(fig.
together by of a welded
PROFILE
As indicated
system. assemblies
Mark
blading) assembly
2.3.1.1
as discussed
balance
state-of-the-art
ASSEMBLY
with
the
yielded
NACA-65 at
profile
nomenclature
design
loss). data
series,
the
British-C
In the
M-l,
pumps profiles
parameters
a
utilized or
distribution
is illustrated
are
predicted
state-of-the-art velocity
and
diagrams
considerable
midchord.
double-circular-arc surface
the selection
velocity
special
to
avoid
employed in figure
14.
in
A--l
BLAOES .AC.,NEB ,NTE_RALLY W.THROTO. 0.0.----_
_.LANCE /--P,STON
I
THRUST-I_LANCE-__-_=
.
4-.- BEAR ING COOLANT-FLOW
I
RETURN
A-_ (A)
I SECTION
MARK 15-F
BLADES HACHINED INTEGRALLY
WITH
/---
DISK
,__
L/_A_ STUB
SHAFT
SHA T _TIE
BOLT
(B) MARK 9
BLADES, TSI_
IT' ON
DOVETAI LED ---_
BALANCE
_
P' STON
._,,_,_,M_
..
,_1_
e LEAKAGE-FLOW
RETURN
(c) M-i
Figure
13. -- Three
types
36
of rotor
assemblies.
A-A
STAC K I NG I
AXIS
/
ILING LOWER (PRESSURE) SURFACE UPPER
\
(SUCTION) SURFACE
\ CENTER
OF
GRAVITY
\ \
MAXIMUM-MOMENT-OFINERTIA AXIS
/ /
/
LEADING
EDGE SECTION
J_x'_---
M IN IMUM- MOMENT-OFINERTIA
AXIS Figure 14. - Typical profile nomenclature.
3?
A-A
EDGE
Standard
profile
camber
line
(fig.
compressors the
w z
are
thickness
surface
shapes
are
6).
Basic
compared on the
by
a specified
thickness
in figure
distribution
velocity
defined
is that blade
thickness
distributions
15 (adptd. which
for
from
ref.
is required
distribution
some
of
about
the
a mean
profiles
used
18). On the non-standard
to satisfy
the
desired
in
profiles,
distribution
of
profile.
I
65-SERIES .....
C_
BLADE PROFILE I • DOUBLE CIRCULAR ARC
w
I
I
10
ZO
0
1 30
I
I
I
I
40
50
60
70
80 ,
90
100
PERCENT CHORD
Figure
The
15.-
selection
Comparison
of
considerations.
a blade
of basic
thickness
profile
is based
According
to
imposed by distribution
stress considerations, in order to reduce
distribution
would
maintained good
over
diffusion
on
and
the
9, 15-F,
Mark
exceed
the and
times thickness
double-circular-arc to obtain used
for
the
M-1
of the
upper
portion. were
distribution
the
inlet at
blade
with
0.6
for stress
axial-flow
a fiat
(with
the
resulting
the
chord. radius
larger
than
As previously
Research
38
has
had
Mark
26
surface
a thin
been
velocity velocity
velocity
commensurate
method velocity leading
conducted
with used
on
of ref.
18)
did not edge
utilized
the British
is
maximum
profiles
trailing-edge
noted, also
a fiat ideal
distribution,
pump
design
requirements
have the
suction-surface
profiles The
18).
mechanical
non-standard
stream-filament
the maxhnum
The
local
velocity The
fromref.
thickness
as is possible,
well back.
that
reasons). stages.
maximum
surface
achieving
leading-edge
and
maximum
(limited)
designed
of
profiles(adptd,
hydrodynamic
within
(suction)
such
velocity.
about
the In
three
for axial-flow pumps would to a minimum. Further,
are positioned
25 pumps
profile
a stiffer
which
for
both
43,
profiles cavitation
camber
velocity
on
reference
in
rear
maximum
the 1.2
maximum
one
as much
thickness to control
be
distributions
with
a modified
radius
(primarily
C-4 profile on
was
standard
double-circular-arc profiles (refs. 15 through 17, 23, and 32 through 34), on multiple-circular-art and slotted double-circular-arc profiles (refs. 24 and 25) and on variations of profiles similar to thoseusedon the Mark 26 (ref. 44). From a mechanical-designstandpoint, the structural merit of the profile is reflected in the camber angle, the chord length, the thickness distribution, and the maximum thickness-to-chordratio. The camberangleis establishedby the hydrodynamic design.For a given chord and thickness distribution, increasedblade strength is achievedby increasing maximum thickness-to-chord ratio. From the hydrodynamic standpoint, however, it is desirable to maintain low blade thickness. Thus, a mechanically desired maximum thickness-to-chordratio may not be achievable(note in table III that a maximum value of 0.15 was usedin the state-of-the-artblading). Increasedblade strength alsomay be achieved by increasing the chord, maximum thickness-to-chordratio being held constant. However, with a given solidity (sec. 2.2.2.4), this procedure requires a decreasein the number of bladesso that, again,a limit may be reachedfrom the hydrodynamic standpoint.
2.3.1.2
MECHANICAL
DESIGN
The predominant requirement in the mechanical design of axial pump rotor blades the blade withstand the combined steady-state and vibratory stresses for the required the kept
pump.
Basically,
within
vibratory stress stress condition. compromises
blade The
in the
vibratory of
due
to
to
centrifugal
magnitudes
load
Blade
the
are
provide
the
operation diagrams
not
accurately
vibratory
stress
due
is and
is highly
to gyroscopic a
the
recovery
loads
pump
analysis
of
on
and
stacking
the
blade
untwist
line
moment
steady-state
the
to consider only the (fluid force), and
forces
discussed
and
and
design.
dependent
been necessary hydrodynamic
about
to centrifugal
As a result,
in axial-flow
in the blading combined
stress
strength,
hydrodynamic
be estimated.
areas
moment
(to
adequacy,
only
due
utilized
steady-state
or ultimate
stresses
unsteady
is subject
stresses
tilt
the
it usually has steady-state
hydrodynamic
to occur during pump of modified Goodman in general
blade
in which yield
limit corresponding to the steady-state must be met without unacceptable
intractable
shear
has not been
structural
that
at best,
the
and
and
moment)
appraising
predicted the use
normal
magnitudes.
hydrodynamic
assuming
The
can,
blading, centrifugal,
one
on either
steady-state
45);
loads
which
The
(ref.
of the more
state-of-the-art due to the
based
material fatigue considerations design.
to these
stresses
is an iterative limits
predictable
is one
loads.
negligible
In
are
stress
geometry. In the normal stresses vibratory
procedure property
hydrodynamic
loads
of the blading
type
design
material
is kept below the These structural
hydrodynamic response
the
specified
is that life of
have to
forces been
of
counteract
in this monograph. and
vibratory
stresses
are compared with blading material properties by (fig. 16). As indicated above, vibratory stress predictable. was
This
proportional
39
uncertainty to
the
has been hydrodynamic
handled
by stress
(Tcf
= STEADY-STATE STRESS DUE TO CENTRIFUGAL LOAD
mff
= STEADY-STATE DUE TO FLUID
°_s
= COMBINED
STRESS FORCES
STEADY-STATE
STRESS
_alt
= VIBRATORY STRESS - ASSUMED EQUAL TO off k = STRESS CONCENTRATION FACTOR
II.
.a
ALLOWABLE
ALTERNAT
', -N/---
% o'alt
k alt
"
-I
%s MEAN
Figure 16. -
(proportionality blade
root
plotted
factors fillet
on
have
ranged and
Goodman
STRESS
o"m,
PSI
Typical modified Goodman diagram for blade stress.
is estimated
the
ING
from
applied
diagram
0.3
to
1). A stress-concentration
to the vibratory
along
with
the
stress;
calculated
identify
resonant
frequency, for rotor
with blade
designed blade rows
axial-flow
have
axial-flow by
flutter
the stiff
has
been
diagrams
Rotating
no
stall
of the
known
discussed chord
blades checked
margin
magnitude
is then
steady-state
stress.
(plots
of
failure
pump
been used. A typical figure 17. Typically,
on operating
speed
been
instances in this
length with
has
at the
little
observed
axial-pump
blade of
root)
by the empirical
blading
of flutter.
4O
given
not
has
Adequacy in reference
been
low
of the 39.
between
preceding blade of excitation in identified
as an
occurred.
aspect
been
blade
Campbell diagram blades have been
vibration
the
and
vs
was maintained
have
blade
In general,
of the rules
has
that
self-excited
monograph.
likelihood
but
failures
envelope
speed
and known sources of excitation. Wakes from in the flow stream are the predominant sources
for any
been pumps
relatively
a 15-percent
frequencies obstacles
source
There divided
at least
pumps.
excitation
Campbell
forcing function as a variable) have resonant conditions is shown in
so that
natural or other
conditions,
for the
this stress maximum
The design is deemed acceptable if the point falls within the material a blade resonant condition is known not to exist at the design speed.
To
factor
(flutter) ratio and
in
(blade has
blading
the
length
resulted
in
in regard
to
3RD
NATURAL
FREQUENCY
_I
'*zzzJ_Izlz_zltI*_IIzr_tttt_z*_ttt'''zzr'_''r*_z1_/'_-_l_-z'DUE BLADEFREQUENCY TO BAND TOTHISRESONANT BE SPEED FREEcoNDITIONS_.I...oFRANGE /
TOLERANCE
2ND
NATURAL
FREQUENCT,,,
---l
/
f
. ,,,_,_
...............
:_{:_::z _:z_i
_
J
/
RESONANT j
IST
NATURAL
Jill
la_
FREQUENCY
tiill,
//
iii
llllt
J
'_IIIill
•
z:z::__/1
I
_:/,._\_/
CONDITION
J
_2
--D,,.
15_
|5_
MARGIN
-.,,._---MECHANICAL
DESIGN
sP EED ,..._
.,,I-
PUMP
SPEED,
the
effort
blade
aspect
natural
ratio
the
models
compared
reasonably
lower
accurate been
cases,
blade
rather
failures tapering frequency resonant diagram
on
the
the
the
and
all the
natural
Mark
leading
of the condition in figure
15-F edge
were to the
results
frequencies.
of analytically
flutter
analyses.
computer
predicted
(ref.
47).
Other
fatigue
In
pumps has
failures
unpredicted
determining the
solutions natural
approach
Blade in that
that
natural
the
lumped
frequencies
that
have
not
M-l,
of had
been have
blading
sufficiently occurred
frequencies
have were
in
of excitation.
it. For
to failure
problems
has
example,
instances
of
eliminated hub
by cutting
as shown
blade to a magnitude that during pump operation 19. A course
46)
cantilevered-beam
solution
redesign
problem 1, and
(ref.
inadequacy,
sources
expedient to
beams
SPEED
blade resonant conditions.
and
approximately
the
to this
difficult
Campbell-diagram
experimental
ratios,
upstream
than
to the
cantilevered
largely
In most
for was
well with
to predict
attributed with
devoted
blading of
aspect
resonance
been
frequencies
of
parameter with
has
DESIGN
RPM
Figure 17. - Typical Campbell diagram for identifying
Considerable
--NOMINAL
of action
in figure was was
involving
41
back
the
18. This
above the eliminated, modification
been
to modify
first-stage chord change
the
existing
rotor-blade at the
fatigue
tip by ¼ in. and
increased
the
natural
forcing frequency, so that the as shown by the Campbell of existing
parts
rather
than
I
Figure
18. -
(A)
ORIGINAL
BLADE
(B)
MODIFIED
BLADE
Modification
of Mark
BLADE
15-F first-stage
blade to eliminate
resonance.
NATURAL
IFIED
FREQUENCY----,,,.,,_
BLADE
8OOO ORIGINAL BLADE
m
FORCING
6000
FREQUENCY 19 CYCLES/REV--_
Z w D
4000 OPERATING SPEED RANGE 2000
10 OO0 PUMP
Figure
19. -
Campbell
20 SPEED,
diagram
42
000
30 000
RPM
for Mark
15-F first-stage
blade.
redesign
can
be taken
is at a tolerable rotor
blades
the
were
not
modification
pump
when
the
from
tested
performance
the
engine
to determine
described.
the
However,
penalty
system
the
associated
standpoint.
degradation
with
The
in stage
modification
did
FINISH,
AND
not
the modification
Mark
15-F
first-stage
performance
caused
noticeably
affect
by
overall
performance.
2.3.1.3
PROFILE
It is necessary the
profile
has
fairing
in both
fluid
desired
surface
specified the
of the
the
better
63
than
finishes Fillet
as small factor.
Fillet
thickness.
both
of 63 microinches blading. the
been
and (gin.)
In
practice,
tolerances
usually
M-1
blades
to
the
The
strength
of the
specified
methods
as a
necessary
produced
as manufactured
desired
+ ¼°.
has been
the
to
blade
continuous the
within
fatigue
rms
have
and
to achieve
held
the performance
dimensional
order
in order
on the basic
of a smooth In
have
limits
tolerance
requirement
typically
RADII
tolerance
the
directions.
can affect
juncture
of the
consistent
radii
have
ranged
Stress-concentration
2.3.2
the
roughness
example,
strict
Typically,
in. with
angles
fairly
FILLET
to
a surface
typically
had
finish surface
rms.
possible
estimated
on the basis
blade
profile
and
with
maintaining
from
approximately
factors,
(root
a reasonably
applied
of information
its support 30%
to the
such
as that
section) small
to 60%
vibratory given
have
been
kept
stress-concentration
of the
profile
stress
(sec.
in reference
48.
maximum
2.3.1.2),
have
Blade Attachment
2.3.2.1
METHODS
Axial-pump
blading
state-of-the-art attachments
principally
pump
rotor
20). on
machined
carry
centrifugal
required, and
configurations.
The
machined
with
the
The
selection
the
blades
blades.
This
load
it may this
been
cost
of
of the
are
and
However, a lower-cost
benefit
may rotor,
override for
the
example,
43
blades,
if large rotor
weight had
376
drum
which for
had
an axial
will be heavier
to the heavier
attachment.
a rotor
considerations.
attached
is due
or
method
assembly
to produce
M-1 pump
disks
attachment
mechanically
difference
with
M-1 mainstage
a blade
manufacturing,
of the
be possible
integrally
exception
weight,
in which
integrally the
has
pumps (fig.
based
blades,
blade
within For
within
longitudinal
state-of-the-art
gin.
at the as
and
profile
blade
of 32/_in. radii
the
the
profiles performance.
as + 0.002
surface
for
manufacture
been
6),
A maximum
requirement
blade
transverse
(fig.
finish
SURFACE
hydrodynamic
been
angles
blade.
TOLERANCES,
to manufacture
achieve
are
only
magnitude
rotor
benefit mainstage
of
dovetail is
In general,
a
than
one
with
required lots
by using integrally blades
all
pump
structure
manufacturing assembly
on
to
of blades individual machined
in comparison
BEND UP END AFTER
AS_
_
S._A_ _ P,N--I_ /_-
_----------_-'/ \.. "_/
/_
o_SHEAR .'N
ROTOR
ALL IN
BLADE
DIMENSIONS
BLADE
ASSEMBLY
I
INCHES
O. 305
-o. 3o5 F
0.024
O. 2587
R
0.025
0.2748
t 0.025 0.026
R
0.1445
0.2587-1P-
SECTION
THROUGH
SECTION
ROTOR SLOT
Figure 20.-
THROUGH
BLADE
Design details forM-1 dovetail attachment.
44
DOVETAIL
I
with
102
for
the
integral-blade because
design
The
because
MECHANICAL
critical blade
and
strength
attachment
outward
so that
the
situation pins
Stresses beam
the
lock
in the the
kept
2.3.3
an
replacement
in selecting pumps.
centrifugal
load
of the
the
below
could during
operation.
as leaf that
were
calculated
in the
direction
would
however,
were
with
incorrect
by
assuming
parallel
blades a force
assenbly
that
to the
resistance in figure
pin
dovetail
is, the
based
fatigue
on limit
positioning
and
the
blade
radially
in the
same
dovetail
were
met
with
bottom
possible
acted
slot
section). that
limits
on the
was not
the
of the
neck
requirements
to exert
load
material
axial
M-1 at the
airfoil;
property
to position
These
springs
to the
the
the
The
to
analysis
centrifugal
for
desired
be tip ground
Stress
for the
provisions
additionally
20. Note
discharge)
a load,
used
material
include
the
and
structure.
20).
(transposed
as that
specified
also
exist
on
maintained
it was
(designed
loads
rotor
(fig.
was
must
tabs
to the
blade
of the
with
this
as a cantilever
acting
at the
tip
of
to sliding that could occur at the 20 that reverse load (i.e., load in a
be counteracted
by the
bentup
tab. The
source
and
not definable.
CONFIGURATION
Three
rotor-structure
rotor
structure,
flow
made
Rotor
2.3.3.1
pumps
blade
engine
within condition.
would
force
pump
of such
individual
based
manner
stress
M-l,
in figure
blade
toward
probability
of blades
in rocket
loads
was
airfoil
the pin. No allowance was made for frictional blade dovetail and rotor slot interface. Note direction
number
a consideration
method
section)
stress
assembly
attachment
bending
vibratory
was
the
attachment
in the same
vibratory
rotor
as illustrated method.
with
and
In the
that
and
dovetail) attachment
neck
steady-state
blade.
vibratory
(the
configuration
of the
for been
is rare
to carry
blade
stress the
to the
retention contact
and
was achieved
and
corresponding
a large
has not
of damage
designed
dovetail
steady-state
tensile
such
requirement
objects
as the
steady-state
adequacy
maximum
be
dovetail
of the
design
Structural
this type
steady-state
tang
section
The
by foreign
must
airfoil
a single
of machining
DESIGN
attachment the
utilized
cost
expensive.
method,
transmit
shear
very damage
blade
The
9; the
of potential
attachment
2.3.2.2
Mark
(fig. path
machined
13). for
The
M-1
rotor
and
machined
concepts from
A number
thrust ring
was
components
been
a single
of axial
balance
structure
have
system
and
in the
forging,
was
were
machined
holes
a one-piece TIG
utilized
bearing (hollow)
welded
used
coolant
on
the
in the flows
configuration
together
45
state-of-the-art
as shown
pumps. Mark
forging and
15-F
in figure
and
Mark
to provide
to lighten
fabricated
A one-piece
the
from 21. The
26
a return structure.
four Mark
forged 9 and
INCONEL FORGING
718
_-------CONSUHABLE
LLER
INSERT
WELD
(INCONEL
BEFORE
7|8)
(IHCQNEL AFTER
WELDING
ROTOR
718)
WELDING
DRUM STUB
FRONT
STUB
SHAFT
(2
PC)
CIRCUHFERENTIAL. WELD
Figure 21. -
Mark
25
together the
pump with
torque
The
choice
integral
considered. weight
of
configuration,
(fig.
and
design
than
in
concern forging the
welded
cost
from and
construction
rotor
46
shafts
clamped
radial
positioning,
the
turbopump
the
for the
during
such the
weight
forging noted
stub
relative
is made
that (ref.
In In
from and
M-1 was was
manufacturing the
the
M-1
selected
design
pump,
forging
in reducing to eliminate (ref.
with a
(a one-piece
a single
difficulty
sufficient
initial design
standpoint
machined quality
size,
a one-piece
49).
critical-speed
previously
as
selection.
indicated
a rotor
required
and
to attain
influence and
achievable size
disks
tie bolts.
types
desirable
over
used
pump
speed
lowest
the
rotor.
Considerations
rotor
rotor),
of the
in the
builtup
of
with
were
phase. critical
was
a builtup
However, and
or
a comparison result
concept
by shear
design
methods,
of M-1 fuel-pump
13). Rabbets
a one-piece
would
a single
builtup
transmitted
one-piece
is stiffer
the
preliminary 15-F,
blades
lightweight rotor
being
assembly Mark
used bolts
between or
methods, the
through
loads
conceptual of
rotors
Fabrication
(3)
50).
The
was the this two
pumps that have usedthe builtup concept (Mark 9 and Mark 25) weredesignedfor ground application. In both cases,the relative easeof stagingand the capability to test singlestages during development were the primary considerationsin selection.
2.3.3.2
MECHANICAL
Reference
6 presents
various
types
to the
special
Pump
a complete
of rocket
rotors
of
subjected loads
the
designs used
the
pump
less
than
input
have in the
the
superimposed are load
analysis
(transient
the
that
attach
have
was
been
were the
employed, discussion
the
torque
optimized
in
and
been
used
used
along have
above
turbine
since
the
of rotor
Mark
to the
drive
15,F
pump.
for these
couplings
been
rotors
and
Mark
pumps
is presented
was
2.3.4.1
TYPES
Turbopump
Gleason
of the
mounted
has
the torque
power torque shaft
and stress
speed
model
used
25, a ball spline
on separate
the
of the
bearings.
with Mark
tests
analysis
of the inducer. were
and
in defining
determined case of the
evaluations
splines
and
at a mechanical
design
disk
been
estimated
calculated
nominal
from
steady-state
in the
Mark
maximum
unbalanced
attachment and
of all
in the
(ref. of the
Curvic* M-1 to
coupling
was
A complete
6.
OF SYSTEMS
rotors
pressure-times-area
for 9 and
in reference
Axial Thrust Balance System
used
limit since
nonuniform been
typically have been methods. In the
26
In the Mark
with
The
shutdown
has to
been the
of photoelastic have
2.3.4
*Copyright,
used
due
DN,
as determined
from
pump,
and
torque
turbine.
resulting the
Loads
on all state-of-the-art
in the
drive
means
The bearing
and
of 5 percent
10 percent
typically
and
startup
moments
section.
by
methods
for the is limited
bending,
loads.
magnitude
torque,
bending
determined
pump
during
(magnitudes
and
has
the
dissipation
torque
inertial
steady-state
alternating
torque-limited
generally
speeds
therefore
by permissible
Average stresses in one-piece type rotors or thin-shell theory and finite-difference rotor
couplings
An
been
critical
herein
centrifugal,
and
between
been
power
load
have
51). Finite-difference builtup-type rotors. Splines
value). and
torque,
established
support
steady-state
at the
speed
turbopump. finite-element 15-F,
has
forces
condition
design
process
turbine
and
discussion
thermal
been
a bearing
Centrifugal
hydrodynamic
possible
had
on the
typical).
with
design
steady-state the
stresses,
The
simultaneous
has, in effect,
hydrodynamic
from
rotors.
designs. to
along
rotors
of loads,
turbopump
of axial-pump
are
state-of-the-art
torque
discussion
engine
features
differential-pressure the
DESIGN
are forces
Works,
subjected and
Rochester,
fluid New
to
high
momentum
York.
47
axial changes
thrust in the
loads pump
originating and
turbine.
from These
loads
must
be known
accurately,
bearings,
back
pistons,
or some
combination
form
of compensating
and
some
The
Mark
concept balance axial
9, Mark
shown system gap
at the
15-F,
centrifugal
25, and
to counteract drums,
In axial-flow
piston Mark
be made balance
methods.
balance
Mark
must
impellers),
of these
(resulting
on both
piston provide characteristics
(on
provisions
have
been
26 pumps
them
by thrust
compensating
turbopumps,
balance
thrust
bearings
used.
incorporated
the so-called
series-flow
in figure 22. High-pressure fluid from pump discharge is introduced into the and flows through two variable orifices in series to a low-pressure area. Shaft
movement
orifice
that
vanes
and
a are
nominal
preload
designed
package
to minimize
from
sides
changes
of the
in
piston.
pump
Resultant
or turbine changes
thrust)
in pressure
force change to counteract the unbalanced shown in figure 23. All these thrust-balance pump into
operating
the
bearing
or prevent
point
the
package.
rubbing
only
Axial
of the
axial
stops
causes differential
load. Typical systems were
load
on the
were
into
orifices.
Xl
l x2 .
.
BALANCE-PISTON
.
FORCE
THRUST-BALANCE
(F)
FLOWRATE
(WB)
J/I/Z/i
/
¢.
\
/
X I
I
I
I
i
I
i
o5. GAP
RATIO
23. -
Typical
i
_i
I.o Xl X I +
Figure
i
X2
performance
of a series-flow
48
thrust-balance
in
across
the
performance designed such
bearings
incorporated
balance-piston
a change
system.
would the
be the bearing
I VOLUTE
ASSY MANIFOLD ,o
VARIABLE PRESSURE
HIGHORIFICE
VARIABLE
LOW-
PRESSURE
ORIFICE
STATOR
HIGHPRESSURE
_-_
SEAL
ASSY
SPACER
BEARING RETAINING
NUT
;EAL PISTON
BALANCE
/
RING
SECONDARY
PIN LOCK
/I SEAL
PIN MATING
LOW-PRESSURE
RING
SEAL FIRST-STAGE
ROTOR
PRIMARY
ASSY
LABYRINTH
SEAL.
SEAL
SPACER
Figure 22. -
Series-flow thrust-balance system used on Mark 15-F pump.
_
DISK
Balance-piston of the 0.015
such
that
+ .001
spacer load
on the
until
the
bearing
from
0.015
rotor
to
suction)
The
turbine
was
zero
assembly
to
system
tandem direction.
pump
Additional
the
flow
of the
on
system,
is presented
2.3.4.2
MECHANICAL
thrust
magnitude
standpoint, aerodynamic such
are
and
to
load.
was
orifice
was
(fig.
Preload
22).
the
a specified
bearing
exercised
seal).
gaps) machining
in the
preloading)
were
a on axial
retaining
to
Preload
load applied
on
nut
prevent the
load
pump
opposite
end
direction.
accomplished
with
in figure
reversal load was
26)
and
line (Mark
bearing/balance-piston
(i.e.,
axial
bearing of the
Mark
arrangements,
at the support
previously across
the
toward
pump
one
bearing
in the
for load
sharing
in one
to a lower
externally
9 and
turbine
load
of
designed
routed
in figure a single
differential
capability
has been
Mark
the
front
in pressure
to
the
toward
25. As in the case
a change
necessarily
and
load
pump is shown flowed through
to the
Thrust
up
suction
1 and
a bias
transmitted
system
15-F
pump
was
caused
package
(M-l).
pressure The
flow
area
of
also has
25).
including
a parallel-flow
3.
DESIGN
predict
accurately
extremely
high
thus small.
for example,
by applying be
operation
by custom
assembly
torquing
had
axial
shown
only
in references
analyses as the
springs
(Mark
information
involves
end
bearing
load
thrust-balance
to the
usually
support
was achieved
to provide
movement
bearing
internally
to
This
unbalanced
externally
It is difficult
15-F pump, high-pressure
assembly
low-pressure
and
designed
accommodated
because
both
routed
and
proper
temperature.
and
axial the
be
discharge
the
with the
setting
point.
shaft
counteract
set,
Mark
during
(care
the
manner
was
operating
could
and
to achieve
thrust-balance system used in the M-1 hydrogen fluid was introduced from pump discharge and
concept,
piston
seal the
(gap
in order
on the
in fig. 22)
gap
a set of ball bearings
discussed
been
toward
piston
was achieved
2 bearing
orifice
limits
of low-pressure
low-pressure
(no.
in a similar
the
nominal
through
the
at liquid-nitrogen
orifice.
pump
in. setting
procedures
rotor
variable
areas
(sum
the
assembly
pump
The design for 24. High-pressure
the
balance
assembly
was achieved
These
The
The
low-pressure
transmittal
to strict
travel
bearing
rotor
be held
axial
in. The end
must
system. the
to fit between
turbine
the
gaps
thrust-balance
designed
the
orifice
can
be
Thrust
the
significant loads
drops in flow passages, and system and in the pump
of
thrust
pressure-times-area
are in themselves
magnitude
axial
orifice
fluid rotation and turbine
even
though from
5O
and
in the
rotor.
The
analysis
the
variation
in predicted
inaccuracies,
from
a percentage
pump
to inaccuracies.
coefficients effects proper)
a turbopump
forces,
obtained subject
of
hydrodynamic Additionally, thrust
balance
on pressure distribution are involved. In order
and
turbine
assumptions system,
in
pressure
(both in the balance to cope with these
"dLund
lan_ t-lAI aqJ. Joj_ ws;sAs
eOUeleq-_.sm4.
L -- "#E aJnfi!-I
3NI7dS
_07_ 39V)IV37 inN 9NINIVI3_ NOISId 33NV7VB f3S HIN I 'dAgV7 tr_
gVl
N307 1V3S HINI_ABV7
WIHS NOiSId
33NV7VB AlgN3SSV _OIVIS
9NI_V3B
ONISNOH 3NIggfll
9NISflOH 31n70A
THRUSTCOOLANT
BEARING ._/BELLEV
I LLE
SPRINGS
i
-: ....
_--'_-
SEAL
_/
F---
I
-"--
ROTOR
!
SEAL
I
ROLLER BEARING
TRIPLE THRUST BEARING
ROLLERBEARING HOUSING
--
FLEXIBLE
COOLANT
BEARING
SUPPLY
SUPPORT
Figure
25.
-
Thrust-bearing
assembly
52
on M-1
fuel
pump.
problems,
design
correction
capability
balance
pistons
calculated force
margins
the
during
on
axial
of
in
the
the
Mark
thrust;
balance
the
thrust-balance
systems
development
9,
15-F,
i.e.,
if the
piston
would
phase
25,
and
have
been
calculated
by
pressure the
no
either
where
2.3.4.3
SYSTEM
In
state-of-the-art
system
due
or
piston
to
differential
with
and
balance
the
mechanical
pumps,
is unstable,
considerable systems.
can
that
systems
original
cause
make have
contact been
balance-piston
without
instability
Hardware
damage involved
balance-piston the
The
axial
damage
to
during
the
examined
Essentially
in the
due
orifices. of the
to
heavily This
or heavily
cause
the
has
The
loaded
erratic
the
form
oscillations
pump
se.
Stresses
have
maximum speed
the
oscillations
worn
or
condition bearings,
thrust
been
calculated
being also
stator
housings,
assembly the
volute,
to
used
have
the
rotor,
for been
same
bearings
and
Dynamic
characteristics
either
design
by
for
the
was
adopted
changes
normally
broken
analog
to
Mark
resulted behavior
was
for
not
by
and of the of the
simplified
functioned
Mark
15-F,
cavity
of a catastrophic
inserts
that
25,
in bearing
explained
pressure. nature
and
(originally)formed
of the bearing
instances, not
rotating
pump the
if
the Mark 9, 25, and 26 had 15-F, there were numerous
in contamination
in several
9
of
occur
simulation
in balance-piston
was
carbon
stability
which
motion (refs. 52 and 54) or solution (refs. 52 and 55). designed
and
given of the
oscillations.
analytically
of abrupt
balance-piston
PUMP STATOR
bearing
system
rotor
been
operation was achieved through a series of changes involving orifice insert material, and the method of orifice retention.
2.4
of thrust-balance
growths
Aside from scuffing on balance-piston surfaces, problems. During the development of the Mark
primarily overheated
at design
centrifugal
attention
configuration
change.
and 26 pumps. no operational of
the
thrust
the
design
or
the
twice
pressure-times-area
pump
per
techniques,
thermal
nonlinear equations describing balance-piston linear equations programmed for digital-computer
occasions
pistons
example,
to handle
the
permit
STABILITY
surfaces
properly
closed,
calculated
to
For
sized
investigations 52 and 53.
finite-element
the
thrust-balance
thrust-balance
The
the
provided
necessary.
self-compensating stationary
problems
analyses
across
Stresses
included
the
disk
differential
analysis.
the
structural
were
were
to twice
been
turbopump.
pumps
orifice
be equal
operating conditions. Analytical and experimental and pressure distribution are reported in references There
of the
26
high-pressure
have
coolant
failure.
completely,
clearances,
the and
Although trouble-fxee
flow
restrictions,
ASSEMBLY as discussed and
herein
the cylindrical
53
consists housing
of the that
vanes,
encases
the
the vanes.
front
and
rear
2.4.1
Vanes
2.4.1.1
PROFILE
The types considered 2.3.1.1.
of profiles used in their selection
The
2.4.1.2 With
TYPES
significant
VANE the
vibration
for the vanes in the state-of-the-art are the same as those for the blades
profile
design
MECHANICAL
exception
of
of stator
vanes
parameters
are listed
pumps and the factors are discussed in section
and
in table
III.
DESIGN
centrifugal
force
in axial
flow
considerations, pumps
the
is identical
analysis
to the
of load,
rotor
blade
stress,
and
analysis
(sec.
2.3.1.2). Vane
development
During
the
caused
by
with
stress
excitable During design figure
the of
design the
the
vanes
3 that
the
calculated
surfaces
steady-state
the
of the
stress tooling
to the had
TOLERANCES, associated as those
Vane Attachment
2.4.2.1
METHODS 15-F segments
and
Mark that
vane.
This
the
natural
blades. cracking
frequency
by increasing
change
the
decreased
to magnitudes
the
above
larger
those
were
radii
on
the
a profile the
and
on was
portion
of the
shroud
raised
redesign.
pressure ratio
leading
Note
the vane
profile
across
a thickness-to-chord
FINISH,
tolerances,
discussed
26 pumps,
problem.
designed,
the
fabricated,
hydrodynamic
Since
it
and
suction
of 0.15
instead
trailing
edges
and
for
axial
loss for the pump.
SURFACE with
on
pressure
necessitated
been
the
stress
As initially loading
that
to obtain
2 to 3% performance
same
a vane
steady-state
differential
already
required
are the
of
to compromise
hydrodynamic
to a magnitude that
2.4.2
cylindrical
due
action
considerations
Mark
the
stress
necessary
configuration.
This
Design
with
of vane
was corrected
frequencies
a potential
0.10.
PROFILE
In the
alleviate
apart
2.4.1.3
vanes
to
moved
in an estimated
occurred instances
result
of the
natural
it became
were
original
that were
the problem
section
vane
profile
resulted
pump
the
to withstand
the
root
those there
as the
rotor;
was a shrouded
additive
to use
diagnosed
at the
M-1 pump,
in order
the
to
15-F,
frequencies.
M-1 vane
adequate
However,
was desirable of
of the
Mark
off the
increased
forcing
similar
the
was
wakes ratio
and
been
of
cracking
forcing
by known
structurally vane.
The
the
thickness-to-chord
steady-state
have
development
fatigue.
resonance profile
problems
early
for blades
the
AND
surface
were
around
the
54
finish,
in section
vanes
assembled
FILLET and
fillet
radii
2.3.1.3.
machined rotor
RADII
integrally and
encased
on
three
120 °
in a one-piece
volute/stator housing(fig. 26). In the Mark 9 andMark 25 pumps, the assemblyconcept was similar. The vanes, however, were integral with segmentedrings (three 120° segments comprising one stator row) with cylindrical spacersusedbetween stator rows (fig. 27). The M-1 mainstagesincorporated individual vaneswith mounting lugs that, when assembled, were captive in cylindrical retaining rings(fig. 28). The rings and vaneswerethen encasedin cylindrical housings.
ROTOR ASSEHBLY
STATOR SEGMENT (THREE SEGMENTS PER ASSEMBLY)
Figure
As in the
case
of rotor
weight,
manufacturing,
pumps,
the
methods
housings.
These
distortion
associated
2.4.2.2 The loads
attachment
to the
support
Rotor-stator
blades,
the
and
with
assembly
selection
assembly
selected
cylinders
MECHANICAL
vane
26.-
have
Mark
considerations. the
split
15-F
pump.
of an attachment
permitted
preclude an axially
for
the
potential
method
Note use
that,
in
of continuous
propellant
for vanes
leak
the
is based
on
state-of-the-art
cylinders paths
for and
stator thermal
housing.
DESIGN must structure
be
designed and
to transmit
additionally,
55
the
to position
steady-state and
retain
and
vibratory
the vane
both
airfoil axially
ROTOR
STATOR
Figure
SEGMENT
27.-
DISK
(TYP)
(TYP)
Stator
segments
and
rotor
disks
for
Mark
(57
9 pump.
VANES) /_VANE
(TYP)
KEYS (6 PLACES)
_IyNSTNAGEBoSL_ TOR
BLADE
(TYP)
ROTOR (24
RETAINING
Figure
28.--
Rotor-statorassembly
56
for
M-1
pump.
PLACES)
RING
and
circumferentially.
used
in the
has been
There
state-of-the-art
transmitted
Individual
stress
Additionally,
diameter
gives
the
configuration.
Thus, the
length,
the
resulting
and
1 presents
turbopumps.
The
pumps;
associated
HOUSING
stator
and
the
axial
one
of the
two
housings
pumps,
flow
volute
major
of
a
load
and
volute
stress
single-vortex particularly stabilizing
and
of flow
diffusion
on the
design
bending
and
a the
outside
a
suitable
moments over
collectors
is applicable
is limited
to both
essentially
have a as
noted
effective performance.
than
to the
structure
propellant
turbopump
housing
together,
as in the
as in the
Mark
housing
and
weight, housings
diffuse
tie the
volute
some
degree overall
when
an appreciable
in rocket
engine
centrifugal
or axial
particular
features
and
fabrication since
flow
walls
together.
reference
1,
a
folded
the
efficiency
57
design
the
can volute;
of a volute-exit
of
Mark
25
phase
pumps
be
have
had
path
is
vanes
in
radial
of the
9, the
in utilized
this
housing
leak
hydrodynamic
in figure results
involves
Single
propellant
the exception
volute in
of It is
consist
9, and
considerations.
which
vanes line.
26 pumps.
the
as illustrated
the
It may
Mark
as to minimize
envelope,
motion
M-I,
a potential
With
of "foldover" housing
assembly.
state-of-the-art
as well
encases
to the discharge
Mark in
pumps), for the
the
double-vortex
in maximizing
15-F and
configuration
that
flow
bolted
in production
in
the
of the
deflection,
had
delivers
volute) housing,
smaller
rather
pressure-containing
and
stator/volute
to structurally
Additionally,
path to
be distributed
members and
to turn
permits
flow which
present
through
low.
generally
herein
is the
(especially
proper
sections
volute
not
designs.
stator/volute
All of the
volute
in
does
be transmitted
vibratory
can
are relatively
reference
collects
be a single
are preferred
the
and
attachment
assembly
28).
usually
must
length
on the stator (fig.
a continuous
arc
discussion
housing and
stator
or it may
eliminated.
pump)
methods
Housings
discussion
structural
(i.e.,
hydrodynamic,
load
attachment
TYPES
stages
selection
units
the
the axial-flow
M-1
the
in shear
loading
steady-state
therefore
in the
acting
of forming
of the
a complete
in the
appreciable
airfoil
loads
therefore with
2.4.3.1
The
necessity
section
material
that
with
(torque)
by keys
hydrodynamic
Stator and Volute
Reference
The
lug
(e.g.,
an
the
to
2.4.3
the
problems
circumferential
housing
only
designer
transposed
flow
vanes
since
no structural
The
adjacent
of
problem,
attachment.
been
pumps.
to the
attachment
critical
have
kind
conical
Mark
29.
Folding
lower to of
the
weight. obtain motion
diffuser
and
a is in
_SER
MARK
(A)
9
(CAST)
NO
rIARK
FOLDOVER
(B)
Figure 29. -
Two
standard
techniques housings:
cast
is less than
that
obtained
only
housing
which For
can
be
example,
steel
a
flightweight
Inconel
718
structures
was
encountered
core
breakdown
complete
core
completely achieve
in the heat,
the
steel
structures.
original
to the been
axis utilized
together discharge
in the
structurally. section
of
volute
housings
stage
of pump
enters
rotation exit
the
of
core the
Mark
diffuser
metal.
erosion
25,
housing
vanes
and
welded
difficulty
in
because
of
Problems
of
problems
housings
15-F,
vanes
liquid
diffuser
volute/stator
achieved.
volute
(flightweight
of the
was
were
not
necessary
to
for the Mark
26
for a
advantages,
Considerable
of the
However,
grinding for
the
and
diffusion
on axial-flow an exit prior
passage
In the and
and
area
pump
cost
always
and
requirement
pressures
these
not
stator
and
pumps
9 pump
were
builtup
DESIGN
vohlte
the last axial
in the
The
are
M-1
axial
time
in practice
configurations).
fluid
resolved.
lead
weight,
initial
immediate
housing and
passages.
design)
discussion 1. The
volute
considerable
HYDRODYNAMIC
reference
an
for the later
FOLDOVER
the state-of-the-art
However,
the
CONSIDERABLE
fabrication
of additional for
(CAST)
degrees of foldover.
the
structure.
not
erosion, were
and
than
A complete
was
various
in fabricating
expense
planned
(C)
Ideally,
selected
hydrodynamic
welded
from
design
resolved,
(other
2.4.3.2
at the
breakdown
acceptable
employed welding.
were
were
from
FOLDOVER
for a welded
castings
because casting
and
M-I
types showing
been
casting
(WELDED)
LIMITED
Volute
have
stator/volute
15-F
M-l, 12-in.
system pumps
passage
and
to its being to guide
and
additional discharge
58
have
been
is gradually collected
diffuse
diffusion pipe
hydrodynamic
the
designed
turned
in the flow
design
volute and
such
toward
in the
that
to tie other
the
a plane
proper.
was accomplished
attachment;
is given
the
in
flow
normal
Vanes
have
volute
walls
between
the
10-in.
axial
pumps,
the
mean velocity in the volute was the sameas that in the dischargepipe (i.e., there was no conical diffusion section). The diffuser vaneshavebeen designedfor zero incidence at the design flowrate, the vane angle and area distributions being chosen to minimize friction losseswhile maintaining a specifiedvane loading (maximum diffusion factor). The volutes proper havebeen designedon a one-dimensionalbasiswith approximately constant velocity (ref. 56).
2.4.3.3
The
MECHANICAL
assembled
of the
volute,
turbopump.
designed flange
to and
DESIGN
stator From
withstand mount
housing, a stress
internal
reaction
and
bearing
standpoint,
pressure,
loads,
and
the to
pressure
hydrodynamic vehicle
design both
of
inertial
and
loads
volute
magnitude,
growths, housing;
that
usually
is the
an
finite-difference
solved
cause critical
deflections
axisymmetric
The second approach with test results.
Proof-pressure
testing
state-of-the-art
volutes
designs generally have deflection analyses.
from
a structural
has
been
and been
shell
model
subjected
at In the
revolution, are used
loading
analysis
is used
be sized
59
standpoint. first,
the
part
Because
of
structural
rotating
most
of rings,
and
the
and
manufacturing
volute
complexity, to verify
to
have section
beams
on
and
the
volute
shell
is
/'or either
stresses
of the
testing
cross
programs
accuracy
are
designed
matched,
the volute
may be stator
in volutes
plates,
approach,
line
components.
critical
are
computer
(an
to react
generally
be
Stresses
the
junctions
second
loads
must
and
reasonable
as a normal to complete
Mount
stationary
due
a volute
misalignment,
points
to detennine
demonstrated
housings.
assembly. mount
from
surges,
must
installation
consisting
rotations of
has
specified
stator
line
In the
stresses.
analysis
and 58, resp.). were compared
pressure
so
orifice
are determined
hydrodynamic
be
loads,
be provided
and
connections
the
between
approaches.
and
thin
or finite-element
at
rubbing
and
for moments
must
oscillations
the
housing
structure
by a simplified
The
internal
line
must
balance-piston
for increased
possible
on the
the
by two
is represented
are as
more
analytically
foundations.
equations
could
rigidity
loads
from
assembly
circumferential
alignment,
pressure
foundation
and line inertia forces. Turbopump mounts in this case, rotor thrust loads and turbine
imposed and
Sufficient
for propellant from
and
circumferential
and
determined loads
are additionally
determined
of the volute
treated
that Flanges
asymmetric
deflections
to account
structural
housing
axial
bearing
and
the
volute/stator
affect
axial
form
assembly loads.
excursions,
differential thermal the volute/stator
minimize
elastic
than
is typical).
considerable
been
In order
on
assembly
stator
engine
greater
20%
symmetric
The
pump
analysis.
pressure
pressures, located
the
acceleration,
increase
of
and
the
stator
that housing deflections do not adversely clearance, or blade and vane tip clearances. Internal
housings
when
(refs. solutions
process new the
57
ol]
volute
stress
and
2.4.4
Bearing Housings
2.4.4.1
TYPES
The
primary
function
positioning
and
turbopumps
have
cradled
between
and
turbine
the
have
assemblies
for
either
(i.e.,
type,
with
state-of-the-art design
2.4.4.2 As to
axial
and
can
or turbopump
be
severe
and blade
desired
clearances
and
6; the
its effect
on
9,
loads
tip the
the
reference
turbopump
25,
degree itself
being
mounts
attained
and or
axial axial
pump
end)
stage.
is
bearing
bearing
housing
Additionally,
axial
bearing
for the
volute
radial
locally
their
stops,
M-1 bearing
at
be the
bearing
been
bearing
speed.
utilized
housing,
stator
the same
carrier.
on critical
have
and
stiffness
in the
influence
M-1 rear
M-1
of
will in general
forgings
of the the
assembled
of the
end
In the
housings.
proper that
in
which
housing,
all was
flightweight
housings.
also
bearing be
includes
rotating-assembly
and deflections
achieved. piston
26
pumps
was
the
and,
external
structural
are subjected
thermal loads
is given
gradients
reacted
to radial
be minimized
rigidity
in those
the
housings
pressure,
must
Axial
form
bearing
attention
so that
is necessary
designs
with
at the and
axial the
to maintain
axially
preloaded
load. reacted
by
discussion critical
the
such
6O
that
bearing
of bearing speed.
housings are determined during the flight inertia loads of the turbopump and
components the internal
particular
Radial are
balance
must
rotor,
housing),
design,
clearances thrust
stationary
of this assembly,
turbopump
turbine
of the
vane
loads
by the include 15-F,
the
mounts.
for
radial
reference
Mark
and
axial
front
inducer
of the
exception
As a part
in the
to maintain
be reacted 2.3.4) and
the (pump
seal packages,
housing
castings
requirement
2.4.3,
deflection
desired
rotor
that front
All of the
of the
radial
state-of-the-art
DESIGN
radial
flanges
The
row
packages,
stiffness
As with
turbopump.
(which
bearings,
such
bearing
CRES
the
casting.
in section of the
alignment
All
in terms
of bearing
with
a single
an immediate
indicated the
maintain
of the
bearing.
stator
spring
The
design
300-series
MECHANICAL
foundation
locations
outboard end)
the
classified
desired
housings not
be
or flexible).
the
of
from was
for
bearing
and
assembly.
bearing
inducer
(turbine
vanes
a discussion
structures
machined
can
rigid
6 presents
Welded
two
the
rear
included
types
mount
Reference
the
is to provide rotating
mounts.
housing
bearing
at
with
of the
have
housings
turbopump
supported
incorporated
turbopump
bearing the
bearings,
outboard
housing
Bearing
the to
been the
designs and
of
support
axial
Rotor
housings housing axial
radial thrust
thrust-balance-system rotor assembly. thrust
toward
are
discussed
spring loads
The pump
rate that
in and must
analysis (sec. design of the suction
was
transmitted to the turbopump housing assemblyby the front bearinghousing and thrust toward the turbine was transmitted by the rear bearing housing. Axial thrust in the M-1 pump was reacted only at the front bearinghousingthrough a triple set of ball be/_ringsand spring system,asshown in figure 25. In general, stress
the
complexity
analysis
calculating
of the
local
of
total
stresses
inducer-vane stresses flange stresses).
the
bearing
structure
that,
iv, the
in the
front
housing
difficult. opinion
bearing
structure
Stress of the
has
analyses designer,
housing,
stress
Housing Interfaces and Static Sealing
2.4.5.1
INTERFACE
The
components
Structural
making
continuity
assembly
and
SEAL up
must
operation.
positioning
the be
pump
housing
maintained
Particular
in general
In
this
concentricity mainstage
was stator
machined
been
accomplished
types
59
of seals
design at the manufacturer.
the
has
been
each
must
by using
presents that
been
with
a monitoring
port
M-l,
leaks
developed
in some
prevent problems
seal hydrogen were
and with possible period of time.
leaked,
a helium
leakage
to
associated creep
with of
the
the
final
in (e.g.,
and
the
at
the
local
interfaces.
interfaces
during
diametral
tolerances
pump
components
an
and
was
interference
are
difficult
volute
and
(2)
fit on
the
seals
for
61
support
and
housing
stator
in general
rocket
engines.
in figure
and secondary
was
and
rear bearing
liquid
illustrated
cases 60
stackup subjected
30. The
has
seals).
During both
the indicates of the to high
The
interface
is specified by seals (i.e., double
where
through
Reference
which
and
to ease and rear
clearance
bearing
to the
and
assembly.
in those
tolerance
stator
front
relative
pump static
and
the
the
rotor
introduced
atmosphere.
material,
rotor
seal type and in general pump utilized redundant
joints,
purge the
of pumps
the primary
of the
the
to (1)
using
the and
of the
in axial
between
bolted
to stationary
(i.e.,
as a unit,
discussion
used
seal is dictated by the Note that the M-1 axial
seals
secondary
during
a detailed
have
affected
positioning
shims
of
by
subassemblies
machined
Axial
are
be given
accomplished
that
on
were
as a unit).
reaction,
the parts assembled at different temperatures three housing interfaces between the front
machining
accomplished
housing
were
Reference
case,
be of significance
to mount
assembly
across
attention
diameters of the mating flanges with the buildup problems. The M-1 had bearings.
analytical primarily
TYPES
concentricities that provide proper alignment of rotor sealing that must reliably prevent propellant leakage. Radial
a precise
consisted
could
due
2.4.5
AND
made
have
the seal conical
testing
of the
the primary
monitoring that
and
port
to
the
leakage
double-sealed
joints
loading
for
a long
•sdLund le!xe u! pasn sleos op, eJ,s _.o sadA_L -
'OE oJnS!-I
_;z:_!vw (G)
6 _w
(3)
ONl_dS 3177V13N G31Vfl13V-3_nSS3_d
9N I _1-0
>
3177V.l.3
(]3xvnJ.3V-3_lnss3ua C',l ',D
9z :w_vw (]NV3-St :_VW
////////__!_ (]3,:IVHS-FI 31llVl3W
(B)
_-. (v)
.... ,_
_ii_ 7V3S
A_IVONOg3S
I
_///"_'/
031vn13v-3_nss3_a 7V3S
1_10(:1 9N I _OJ. I NOW "
I_VWI_a
2.4.5.2 A
MECHANICAL
tight
interface
assembly
in the been to
separating the
the
loads
at any
the
that
to the
joint
is analyzed
a spring
model
Stresses
and
is used
the
at
under
all
allowable
loads
typically
steady-state
interface the
loads
preload
preload
Interface
in
inertia
bolt
exceeding
pump
Each
is included and
system.
The
environmental
in determining
expansion
joint
without
combinations. for
The
operating
must
be
analysis
that
examined
(e.g.,
must
pressure
be transmitted
mount).
as a spring
for the
adequate
effects
conditions.
misalignment,
turbopump
calculated
probable
loading
conditions.
an
thermal
conditions,
all potential line
all load
to ensure and
temperature
propellant
are
for
of the
assembly
transient
the
or contractions
elastic
stiffness
temperature loads
and
resulting
of the
condition stresses
from
in each
thermal
parts
being
of the
effects
making
investigated interface
are included
in
analysis.
Measurement of applied required bolt preload normally each
result
part
in the preload.
Preload
range
parts.
This
been
joint
can
method
necessary
2.5
torque (torque wrench) at assembly. However,
in a preload
required
that
assembly
can must
be minimized is time
vary
by a factor
be capable
by
measuring
consuming
to use this method
is the most common the uncertainties
but
may
method of achieving in friction coefficients
of 3 or 4. This
of withstanding
deflection
of
be warranted
in the stationary
condition
a load
the
the
or compression
assemblies.
of axial
that
3 or 4 times
tension
in critical
components
means
a
It has not
pumps.
MATERIALS
The
materials
stated the
enough
be
interface
interface
parts.
great
of separating for
ensure
the joint
and
maintained
or flange
for
load,
through
up
bolt
and
carefully
be
must
examined
conditions,
The
must
combinations
stresses have
joint
therefore
probable
DESIGN
utilized
previously, selection
Relative
of
strength-to-weight susceptibility
steady
steady-state
major
components
pumps
were
based
on
materials
ductility,
ratios must
conditions. and
designed
has,
fatigue
in order
have
adequate
ductility
vibratory-stress
components
such
conditions,
been
expansion
in order
the
to avoid
as blades
that
strength
the
basis
characteristics,
weight
alloys low.
fracture-type are
Thus
temperature. on
designs,
pump
IV. As
applications.
evaluated
In flightweight to keep
in table
at liquid-hydrogen
thermal
fatigue
63
are noted
for liquid-hydrogen
in general,
failure.
desirable
pumps
properties
strength,
are In
of axial
material
to hydrogen-embrittlement
additionally stress
was
candidate ratio,
strength-to-weight alloys
all of these
of materials
merit
in the
exposed
is a consideration.
and
with
high
Candidate
failure to
of
under
significant Material
thermal-expansionrates must be consideredin those components having a critical interface (e.g., bearings in a bearing housing), since a prescribed fit at both assemblyand operating conditions is required. Consideration additionally is given to fabrication processesand operating environment in which hydrogen could be absorbedinto or otherwisecontaminate the material and result in hydrogen embrittlement andsubsequentfailure. Considerablematerial property data have been obtained during rocket enginedevelopment programs. Much of the work done in support of cryogenic pump and other component development is reported in references61 through 66. Thus, the discussionherein is limited to someof the more significant material problemsthat haveoccurred with axial pumps. As noted previously, the M-1 rotor structure consisted of four forgings that were TIG welded to form a one-piecerotor. Extensive development work wasconducted to establish welding and inspection procedures for the weldments (ref. 64). One pump rotor was fabricated and utilized in liquid-hydrogen turbopump testing. This rotor had known weld defects prior to the test program. Post-test examination of one of the rotor weldments indicated that nearly all of the defects propagated during testing (ref. 60). The effect that thesedefects would have on long-time operation of the rotor wasnot established,however, becausepump testing wasdiscontinued when the M-1 engineprogram was terminated. As indicated in table IV, titanium alloy A110-AT-ELI was utilized for the M-1 transition rotor. Mechanical-property testsconducted on the forgingsindicated the elongation to be an unacceptably low 1 percent at liquid-hydrogen temperature. It was determined that excessivehydrogen content in the forgings wasresponsiblefor the low ductility. Elongation of 10 percent at liquid-hydrogen temperature was achievedby degassingthe forgings in a hard vacuumto lower the hydrogen content (ref. 67). Carbon was used initially as the orifice insert material on the Mark 15-F thrust-balance system.This choice was madein order to avoid galling during contact of the balancepiston and orifice insert. As noted previously, contact during pump operation was sufficient in someinstances to break the carbon. The carbon particles contaminated the bearingcoolant flow and at times resulted in bearing failures. To prevent these impact-type failures, the material waschangedto leadedbronze, asnoted in table IV.
2.6 The
SAFETY various
organizations
propulsion design either
system the
design
of
for safety
policies
pumps
therefore factors
for
employ on
organization
occasionally
is appropriate values
and
design
but
responsible
component
instructions
structural practice,
FACTORS
the
definitions
to define that
have
the been
structural
individual
safety
or in general
the
the
factors.
manuals Values
responsible have
been
have terms
64
of
that
for safety
contracting consistent
differed
as used
utilized
design
from
with
contain
of axial
or
other
comprehensive
factors
are specified
agency.
Terms
aeronautical
organization
in this document
in the design
a turbopump
by
used
engineering
to organization. and
to indicate
pumps.
in
typical
It
Table
IV.
-
Materials
Used
for
Major
Components
Pump Component
Mark 9
Mark
15-F
on
Axial-Flow
Pumps
Configuration
Mark
25
Mark
26
M-1
Material Rotor
K-Monel
310
Blades
310
K-Monel
K-Monel
Same
as
Mark
15-F
lnconel
K-Monel
718
Mainstage: Inconel 718; transition: Ti A110-AT-ELI
Volute
310
Stator
310
310
housing
(Integral with
310
304
310
304
ELC
volute) Vanes
Front
bearing
310
310
310
Inconel
310
310
310
347
310
310
310
304
718
housing Rear bearing housing Balance
piston
Balance orifice
piston
K-Monel
Inconel
Flame-plated
Leaded
tungsten carbide on
bronze
Silver-plated 310
A1 2024
718
A1 7075-T73 304
310
Limit
load.
or service under
the
limit
operating
there
all
factor
maximum
load.
operating
The
specified
hydrostatic-proof-test
environmental
limits
(a
maximuln
limits
defined
(3-sigma) physical
of the
by
in the
multiplying following
and
engine
a combination
specified factor
factors
have
or calculated
pressure)
3-standard-deviation
is uncertainty
limit-load
is the
(excluding
maximum
vehicle limits.
load
lnaximum
including
specified
When
The
pressure
(1)
vehicle
-
load >
1)
been
65
that that
or vehicle,
or
lack
used
of
limits
limits
3-sigma to
in axial-pump
the
of the
influence
or (3) the
of 3-sigma
of a service
can be expected
operating
variables
is applied
value
loads,
maximum
and
specified
data specified design"
on
load
to occur engine (2)
or the
engine
conditions, or
or
operating
calculated
a
Type Centrifugal
load
to rotational
speed;
speed)
Load
on blades
Structural
load
induced load
Design
safety
than
to fluid
due
to internal
due
factor.
load
in
-
of the
the
stress.
the
load
slightest,
may
be
prevents sometimes
thrust
design
mechanical
design
safety
factor for
design
design
load
multiplier
(or divider)
uncertainties,
e.g.,
within
structure.
distributions
(or pressure)
is the
stress,
or combination
-The
failure
is an arbitrary the is the
variations
product
greater
in material
of the limit
load
(o1
from
the
factor.
stress
load
stress).
produces
defined
- The
1.05
gimballing
and load
safety
1.0
contraction
to account
design
(or
1.2
and and
design
quality,
The
application
Allowable
called
is applied
pressure
- The
the design
stress.
highest
factor
1.1
expansion
(or pressure).
and
Design
load
forces
to engine
fabrication
pressure)
(limit
is sometimes
due
1 applied
Design
result
by thermal
Inertial
properties
to rotation
factor
1.1
design
Load
due
Limit-load
of load
allowable
of the
as buckling,
structural
of design
loads,
load
pump
yielding,
the component from referred to as criterion
in any
(or
stress)
structural or
resulting
whichever
is the
element
ultimate,
performing its load or stress;
element,
fatigue
intended allowable
condition
load
that,
results
if exceeded
under
consideration.
failure,
whichever
in
ill
Failure condition
function. Allowable load is stress is equivalent to material
strength. Margin stress
of safety. exceeds
the
- The
margin
design
load
of safety or stress.
(MS) The
MS-
where
R is the
Material maximum
ratio
endurance alternating
of the design limits. stress
load
is the
margin 1
-The
material
that
the material
of safety
to the
the
is defined
allowable
endurance can
66
by which
allowable
load
or
as
1
R
or stress
fraction
sustain
limit
(also
load
or stress.
called
for an infinite
fatigue number
limit)
is the
of cycles.
Proof
pressure.
prove
the
-Proof
adequacy
pressure of design
is the and
pressure and the proof-pressure temperature is not feasible, the difference
in material
strength
test
pressure
quality.
The
factor. proof-test
in the
Proof-pressure
design
factor.
of
at operating
-
the
temperature
pressure to obtain the proof pressure. state-of-the-art axial pump components. In practice
in the 0.2%
design
offset)
of axial
factor
pumps,
of 1.2 typically
a typical
on
factor stress,
for fatigue, expressed as a ratio of material typically has had a value of 1.33 ; the fatigue
values
of 4X
predicted
cycles
typical
for low-cycle
of cycles
to failure
the
at to
to partially
product
of the
the design compensate
temperature
to ensure
is a multiplying
A value
1.1 ; the
and
fixture
(based
fatigue; i.e., the number operating cycles.
has been
a component is the
limit
operating for the
at which
the
are to be subjected to proof-pressure tests are do not occur during the proof test. Care is
proof-pressure
Proof-pressure
to
pressure
When proof testing pressure is adjusted
proof testing is conducted. Components that designed such that detrimental deformations exercised simulated.
applied
proof
fatigue, should
67
value
factor
that
factor has
is properly
applied been
for design
for ultimate
loading
to the
used
in testing
safety-factor has
been
design
for
1.5. The
the
yield safety
endurance limit to allowable alternating factor based on cycles to failure has had and
10X predicted
be 4 or 10 times
the
cycles number
for high-cycle of predicted
3. DESIGN
CRITERIA
Recommended 3.1
OVERALL
3.1.1
Practices TURBOPUMP
Turbopump
Design
criteria
presented
and
unacceptable
critical
occur
exist
13 provide
between
operating
3.2 3.2.1
turbopump
rotor
or speeds
(for
rotors
at which
speed
are
dynamics
shall
self-excited
verify
that
nonsynchronous
whirl
range.
designed
recommended
to operate
analytical
and critical
above
modeling
speeds
are given
the
first
techniques. in reference
critical).
References
Recommended
6,
margins
6.
DESIGN
selection
o.f a pump
requirements speed
shall
verstts
capabilities, Reference
O,'pe to satisfy based
on
configuration
7 presents
engine
be
upratotg
recommended
a complete
practices
and stage
head
and ./low
and
pump
speco"ic
and probable
of the design the
off-design total
relationships,
weights,
discussion with
of
efficiency
probable
associated
design
examination
and
potential,
given
selection
operating
considerations, of
the
ralzge
costs. design
various
types
criteria, of
pumps
and for
application.
It is recomlnended above
turbopump
Realm of Operation
The
rocket
speeds
speeds
STAGE
selecting
in
of the complete turbopump rotating assembly and support system and utilized to predict critical speeds and the threshold speed of
whirl
12, and
of
in the operating-speed
An analytical model should be formulated non-synchronous
involved
Rotor Dynamics
predictions
do not
practices
7.
Turbopump
Analytical
DESIGN
Speed
recommended
in reference
3.1.2
and
approximately
that
an axial
3000
configuration
and when
be considered
throttleability
68
and
wide
when
stage
fixed-speed
specific flow
speeds
range
are
are not
required.
Additionally,
it is recommended
that
a detailed
capabilities be made whenever pump uprating is a design is appropriate to both the axial pump and the centrifugal the
addition
of stages
centrifugal
3.2.2
pump
is relatively
requires
straightforward,
an additional
stage
Stage Hydrodynamic
The
stage
design
requirements,
shall
of the
requirement and the specific speed pump. The axial pump, in which
is recommended
to satisfy
axial-pump
the
when
uprating
the
competing
requirement.
Design
reflect
an
mechanical
examination
acceptable
compromise
requirements,
and
among
hydrodynamic
pump
configuration
overall
requirements. The the
flow
model
used
in the
three-dimensional
axial-flow
real
compressors
blade-to-blade to final
flow
coefficient,
flow
be
planes
Prior
selection hub/tip
guides
in the parametric
ratio,
of axial-flow
blade
should
approximate
that
design
practices
is recommended approach,
average
flow
flow in the hub-to-tip
hydrodynamic
design,
tip speed
is achieved.
stages
should
conditions
plane
(ref.
a parametric
be made
It is recommended
that
the
18).
study
so that the
for in
involving
an acceptable
tbllowing
be used
as
should
be
study:
•
The
stage
design
•
The
stage
hub/tip
For
stages
less
than
with
flow
high
1700
coefficient
ratio
ft/sec
for high-strength
3.2.2.1.1
It this the
stage
and
configuration
BLADE
in
to represent
of the detail
mechanical
design
condition.
followed;
are used
pump
3.2.2.1
hydrodynamic
LOADING,
should
should
hub/tip
not
be greater
ratios alloys.
STALL
MARGIN,
be less than than
(_> 0.8),
for high-strength
nickel-base
not
the
AND
0.9. blade
titanium
0.25.
alloys
design and
tip
speed
less than
1500
ft/sec
EFFICIENCY
Blade Loading and Stall Margin
Design-point
blade
loading
shall
reflect
an acceptable
compromise
of stall
margin
and efficiency. It is recommended loading. influence
For the
that
a given selection
the
diffusion
application, of the
factor the
(eqs.
(5) and
stall-margin
design-point
diffusion
69
(6))
requirement factor.
be used of Stages
as a measure the
pump
designed
will
of blade greatly
for optimum
efficiency should have a maximum design-point diffusion factor between0.45 and 0.55 (at any radius on either the rotor or stator). For pumpsin which a minimum number of stages is desired, a design-point diffusion factor between 0.55 and 0.60 may be selected if stall-marginrequirementspermit.
3.2.2.1.2
Stage Efficiency
Predicted
stage
friction,
efficiency
and secondary
It is recommended predicting particularly to highly
with
high
stagger
Predictions
3.2.2.2
radial
due
to profile,
end-wall
8. Data
care
experimental for
should
in references
be estimated
and
results
be
used
with
shown
be exercised
on
highly
in applying
8,
figure-8 blades
26.
secondary-flow
loss to obtain by methods
in
figure
and multiple-circular-arc
23 through
loss
to the profile
data
axial-flow-pumps
with
for double-circular-arc
(annulus)
can
pump
the
loss,
overall
presented
stage
including
tip
efficiency.
in references
The
68 and
69.
DIAGRAMS
pattern
suitable
losses
favorably
therefore
Data
are reported
losses
and
compare
and
blading.
be added
VELOCITY
account
in figure
not
region,
should
of these
The
do
end-wall-friction
loss,
magnitude
tip
angles
of
clearance
22)
staggered
into
compressor,
as illustrated
(ref.
in the
data
cascade,
losses,
blades
take
flow.
that
profile
staggered
shall
of flow
and
compromise
of
flow
with
the
stage
type
of velocity
headrise,
diagram
stage
shall
efficiency,
be based
and
on a
stall-margin
requirements. A
free-vortex
recommended
for designs
desirable
to use
alternate
flow
In
the
that,
patterns of and
due
whose magnitude The factor should
offer
the
ratios
pumps,
0.8,
a more
velocity
blade-surface
by linear taper by the method
hub/tip
less than
coolant
for liquid-hydrogen
reduction
a symmetrical
having ratios
might
of bearing
to end-wall
preferably estimated area
hub/tip
preparation
recirculation due
pattern
and
velocity greater
the
suitable
diagrams,
than
design
boundary density
layers increases
layers
normally
depends on the particular be selected by the designer
0.8.
should
at
the
mean
If it becomes
be examined
radius
is
necessary
or
to determine
if
compromise. the
compressibility
thrust-balance-system
in the flow path; the presented in reference
to boundary
diagram
should
flow,
and
of
be considered.
in excess
the
channel
of 6 percent
pump area
fluid,
reduction
It is recommended be accounted
for,
magnitude of the density change can be 7 (pp. 99-102). As indicated previously, is accounted
for by using
a blockage
factor
design as well as the design method being used. on the basis of experience with similar designs.
70
Caution should be exercisedbecause,as pointed out in reference 18, indiscriminate use of correction factors can lead to a design that is as poor or worse than one in which boundary-layercorrections are ignoredcompletely.
3.2.2.3
BLADE
Fluid
The
turning,
deviation
angles,
selection
of
incidence
angle
from
Thus,
a cavitation
and
guides
References
23
minimum-loss for
Accurate
prediction
design.
calculating
In view
that
deviation-angle provide
and
available
or which
exact have
solidity are
rule been
3.2.2.5
for
be the angle
Mark in
extend
the
references
range also
and
slotted
range
of
range
68,
and
be
III.
used
The
for
and
may
23 through
deviation-angle and
it
deduced
profiles
performance
70.
26 pumps,
2.2.2.3
References
double-circular-arc
of
an acceptable
25, and
in table
deduced
provide
the
31,
NACA-series
profiles.
of
18, 27, 68, and
27,
9, 15-F, section
for
C-series
Acceptable
extend
18,
optimum
for minimum
in achieving
to those
recommended
e.g.,
as that
that
important
given
diagrams.
in references
in references
similar and
same
blade
is inappropriate.
are given
of the
rule
parameters 18 are
these
be
within have
selecting
the been
be within for
achieving
degradation
for
a range
can
ideal with
headrise, increasing
0.75 should solidity
CAVITATION
The pump
mainstage
shall not
which
26
rules
for
turning-angle
profiles.
be subject
to cavitation.
71
experimental
in axial-pump
be specified.
applications
of approximately
high
¢br
demonstrated
solidity
in axial-pump
associated
values
successiVEly
a value
demonstrated
selected
desirable
efficiency
design
velocity
application;
correlations
given
incidence,
SOLIDITY shall
No
necessarily
are
double-circular-arc
profiles;
specific
is extremely
deviation-angle design
the
angles
in the
the
incidence
18.
angle
that
design
in reference
for multiple-circular-arc
Solidity
that
for
angles
reference
reflect
and
use
correlations
double-circular-arc
3.2.2.4
of
for standard
data
correlations
the
by fluid
data
deviation
having
rules
be used
deviation
of its successful
profiles
will not
provide
given
as defined
on
incidence
26
of the
recommended
nonstandard
is dependent
for selecting rules
and
properly
magnitude
through
incidence
Methods
shall
standpoint
a recommended
procedures
also
of radius
and
loss.
is
as a function
camber,
incidence
70.
ANGLES
(table
On the Ill),
data applications.
basis
of magnitudes
it is recommended
to 1.9. High-solidity be
are
analyzed
is at a tolerable
to
that
stages, ensure
level (fig.
that 9).
which the
It is recommended that the inducer be designed to provide sufficient head to avoid cavitation in the initial mainstagefor all anticipated pump operating conditions. Adequacy of the initial mainstage as free from cavitation should be determined from cavitation-test data for similar designsor from analysisof fluid velocities on the bladesurface.
3.2.2.6
OFF-DESIGN
The
pump
PERFORMANCE
stall
anticipated
point
during
at any
either
operating
transient
speed
shall
or steady-state
be at a flowrate pump
less than
that
operation.
For designs with hub/tip ratios greater than 0.8, it is recommended that a diffusion factor 0.75 or a retardation factor of 0.50 at any radius on either the rotor or stator be assumed
of as
the
of
condition
0.70
or
a
at which retardation
corresponding
3.2.2.7
to the
will occur.
factor
of
It is further
0.55
be
recommended
used
minimum-flow-coefficient
as
a
that
a diffusion
permissible
requirement
factor
operating
condition
of the pump.
CLEARANCES
3.2.2.7.1
Radial
Radial
tip clearances
It is recommended vane height of deflection rotor
stall
and
thermal
that
Axial
blade
centrifugal
growth,
and
clearances
rotor
shall that
succeeding
vane
row.
clearance
(or
stackup,
blade
tilt
vibratory
loads,
rotor
component
tip
and vanes clearance
shall minimize
of not
more
head
than
losses.
2 percent
of the
clearance. The clearance analysis should (and hydrodynamic pressure imbalance frame
and
dynamics
housing
deflections,
blade
or
include effects if applicable),
component
differential
effects.
Axial
It is recommended The
a radial
be used as an operating due to rotor imbalance
contractions,
3.2.2.7.2
on the blades
differential
minimize
an operating
blade)
row
analysis in
wake
axial
axial
be at least should direction
thrust-bearing thermal
effects
clearance effects used),
deflection,
contractions.
72
blade
between
10 percent
include (if
on adjacent
tip frame
a blade
of the of
or vatze rows.
chord
(or vane) length
assembly
deflection deflections,
dimensional due rotor
to
row
of the
and
tolerance
steady-state
Poisson
the
upstream
effect,
and and
The axial clearancerecommendedaboveis consistentwith and is a necessarycondition in the blade designpractice outlined in section 3.3.1.2.3. Deviation from this clearancevalue may be desirableif, for example,the initial mainstagebladerow is precededby a long-chord inducer stator. A smaller axial clearance(i.e., lessthan 10%upstream chord) would increase the amplitude of the load fluctuation and would require that appropriate methods referenced in section 3.3.1.2.3 be utilized in designing the blade from the vibration standpoint.
3.3
PUMP ROTOR
3.3.1
ASSEMBLY
Blades
3.3.1.1
PROFILE
The
blade
TYPES profile
blade-surface
shall
(1)
velocities
produce
and
(2)
the
desired
provide
the
fluid
blade
turning
with
with
required
adequate
structural
strength. The
selection
cavitation
of
a
profile
to a minimum,
high
blade-surface
velocities;
the
relative
velocity
profiles
inlet
designed
mainstages standard the
to
where
(the
thickness-to-chord
As
a
general
with
ratios
should held
of 0.10
and
imposed
by
trailing-edge
a maximum the
Thus, be
maxim,
or 0.11 by
(within
trailing-edge structural thickness
preferred.
the
best
the
chord
the
length
are
and
is
bending with
the
the
thinnest
of 0.13
be used
stresses
British
chord, maximum
by
exist,
maximum
1.25
times
nonstandard for
Of the common
If the NACA-65-series defined
reduce
recommended
the
camber,
are utilized
profile
or
consideration.
same
that
than
profiles
double-circular-arc),
ratio
If excessive
To
excessively
avoid
C-4 offers
and
maximum
blade
strength
profile
is utilized,
standard
thickness
(ref.
43).
as an upper structural
It
is
limit,
adequacy
thickness-to-chord
ratio
limits). radii
and up
over
thickness-to-chord
increasing solidity
profiles
is recommended. thickened
a maximum
having
application. losses, no greater
distribution
predominant
C-4, and
profiles
if standard
C-4 profile
velocity
velocity
is the
British for
profile
Double-circular-arc
prescribed
of cavitation 65,
particular
suction-surface
is recommended.
should
be achieved
Leading-
reduce
aerodynamic that
constant
the
modulus
ratio. edge
recommended
on and
NACA
the British
the trailing distribution.
depend margin,
achieve
section
is required,
will stall
avoidance
profiles
maximum
type improve
should
be kept
manufacturing to
about
as small
considerations.
one-quarter
73
of
the
as possible At blade
within
subsonic maximum
the
limitations
speeds, thickness
a total (i.e.,
trailing-edgeradius equal to one-eighth of maximum thickness) should havelittle effect on aerodynamic performance (ref. 43). It is cautioned that a specified fluid outlet angle is demanded,and excessivetolerance within the above limit may not fulfill the outlet-angle requirement.
3.3.1.2
MECHANICAL
3.3.1.2.1
DESIGN
Structural
The
Strength
mechanical
effects
design
of centrifugal,
It is recommended mechanical
design be
(fluid
force)
load
(as
of
magnitude
in accordance
pump)
should
stress
is at least with
the
determined
the
stress
due
an
the criteria
above
and
of
be determined
nominal stress
maximum
the
design
due nominal
speed.
and stress
this maximum
in section
at a This
to hydrodynamic
steady-state
with
practices
loads. loads
the
examination
combined
on the combined
vibratory
steady-state
the
and
be based
to centrifugal
percent
establish
be determined with
10
and
maximum
from
to
shall
off-design magnitude.
steady-state
stress
3.3.1.2.3.
Stress Distribution
The stress distribution Stresses
analysis shall in the blade.
should edge,
blades,
the
root
and
at the
include
outermost
steady-state
radius.
The
for a particular
the
point stress
stresses design
on the usually
listed because
the
stress
due
to centrifugal
•
Normal
stress
due
to hydrodynamic
•
Shear
•
Normal (on
and
twisted
to direct
shear airfoils)
stresses
the
stress
surface
(see
section
where
vary
with
and
the
stacking
stress
axis)
at the blade
sections,
leading
edge,
fig. 14); for cantilevered
be considered,
magnitudes
the blade although
blade
is tangent some
to
may
be
geometry:
load bending
hydrodynamic due
convex should
condition
(along
of the
is at the
below
Normal
due
stress
of longitudinal
an examination
•
stress
maximum
at a number
should
greatest
fillet
negligible
identify
be determined
this determination
trailing the
that
blades
hydrodynamic,
steady-state
combined
Vibratory
and
the
speed
should
requirements
steady-state
that
stress
3.3.1.2.2
of axial-flow-pump
to
and hydrodynamic
74
moment
load untwist
forces
moment
resulting about
the
from airfoil
centrifugal stacking
load line.
• Normal stressdue to bending moments resulting from blade stacking-line tilt or offset. If the direct and torsional shear stressesfor the particular design are appreciable, it is recommendedthat the principal stressesbe determined and that the Mises-Henckytheory of failure (distortion-energy theory) (ref. 71) be used to calculate an "effective" stressfor comparision with uniaxial material property data.
3.3.1.2.2.1
Blade
The
Tolerances
stress
analysis
magnitudes
and
The blade maximum
stress stress.
tolerance
and
recommended
shall
natural
the
tip
section
that
the
frequency
be used
3.3.1.2.3
Vibratory
Stresses
The predicted stress.
diagrams strength,
stress
steady-state should
data
state
be
as
based
endurance
is
at
the
range
of the
vibratory defined on
limit.
by
adequate
(1)
of the
uncertainties
could
blade
shall
stress
on
Assume
a
safety
stress-concentration blade. design
Section of fillets.
load.
tolerances
otz
stress
factor 3.3.1.3.2
result
the
vibration
analysis,
from
maximum
and
the allowable
alternating
the
blade
should
Goodman
diagrams.
applied
that
the
31, with additional
safety factors conditions
equal
to
the
should to
the and
is
of
Goodman ultimate
stress
line
of
of 1.33 on fatigue, 1.5 set forth in sections
proceed
as follows:
steady-state
then fillet
limits
strength,
alternating
stresses,
the
Modified
to yield
allowable
vibratory
criteria
be within
it
minimum
be less than
appropriate
75
In
blade.
magnitude
provides
tolerance condition that gives hub section is at the minimum
for a nominal
magnitude This
that the
factors
in predicting
vibratory-stress
hydrodynamic
blade
frequencies
modified
It is recommended
involved
for when
natural
the diagram be constructed as shown in figure on ultimate, and 1.1 on yield. Note the 3.3.1.2.3.1 through 3.3.1.2.3.5. In view
of
maximum.
that
in lieu of the
and
property and
effects
analysis should be conducted This condition usually occurs
conditions
material
the
frequencies.
tolerance
Predicted
consider
be
at the
recommended
stress multiplied
root
section
practices
due
to
by
a
of
the
for
the
F = MATERIAL ENDURANCE e F = MATERIAL ULTIMATE tu
LIMIT STRENGTH
B
Fty
= MATERIAL
YIELD
STRENGTH
(0.2%
OFFSET)
Ftu
F
a-
(SF) e = FATIGUE
SAFETY
FACTOR
4-J
Fe
(SF) u = ULTIMATE
_
(SF)y
SAFETY
FACTOR
b
w e¢I--
_
_
ALLOWABLE
__'_ z
: YIELD
SAFETY
FACTOR
ALTERNATING
STRESS
(_F)e
_
_
orm
_-
Z
=-'
7_
__
_
MEAN STRESS O-m, PSI
Figure
The
31.
stress
forces)
-
state
stress
Designs are
considered
The
preceding
vibratory
can
(1)
(2)
the
diagram
stress
illustrating
predicted
stress
falls
If the
factors.
steady-state
the
falls
predicted
(centrifugal
in (1) above
in figure
below
point
the
safety
determined
as shown
state
to reduce
and
the
the
ratio.
neglects be
load,
row,
the
vibratory
acceptable.
practice
stress
frequency met:
by
diagram
Ftu
plus
should
fluid
be plotted
16.
allowable
above
the
alternating line,
steady-state
the
stress
stress
blade
until
line
geometry
an acceptable
is achieved.
hydrodynamic blade
the
be changed
design
Goodman
Goodman
in which
should
Modified
defined
and
on the modified
(3)
(SF) Y
_o j
(2)
ty
LOWER 0F(-T TT '
hi b--1 ,:K
determined
vibration
axial
row
should
The
first
wakes
spacing
natural
from
the
is based
the
between
the
frequency upstream
of blade
of
row
than the
at
outlined,
blade
row
analyzed
design between
forcing
frequency
natural
76
of the
occur
15 percent above the mechanical on speed should be maintained of the blade
given
be
the
steady-state from
and
below
the
due
the
upstream
that
resonance
a pump
(fig.
17).
speed
to
natural must
upstream blade
speed. Additionally, the second harmonic frequencies
stress
conditions
upstream
such
nonresonant
wake-to-blade
following
should
will not
that
to wakes
the
being
at least margin
and any
the
the
10 percent
blade
premise
due
factor
practice
to or greater
the
fluctuation
magnification the
on
product
of load
for
be equal
and by
amplitude
Specifically,
The
damping
be
blade
chord. due
to
which
is
a 15-percent of the wake
It is recognized that it will not always be possible to apply the above practice- for example, in an axial pump with a wide operating range. In such cases,it is recommended that vibration amplitudes (and stresses)be estimated by the methods outlined in reference 41 or 42. Additionally, it is recommendedthat the designerconsult references72 and 73 to assist in the solution of designproblems that might arise in a specific application. These referencescontain extensivebibliographieson the subjectof blade stressandvibration.
3. 3.1.2.
3.1
The
Fabrication
stress
Effects
analysis
shall
include
the
effects
of manufacturing
processes
on material
properties. The
material
blades
ultimate
should
processes
3. 3.1.2.
be
and
3. 2
surface
Calculations
as
the
models
following
should
Base
used
for the
natural
include
in
The of
degree
the
for
of the
results
hydrogen
shall
include
(i.e.,
be
employed
74
the
M-1
Mark the
Fluid
75.
15-F
of the
depends
be
used
dovetailed
to nominal
Methods
and
fixity
should
fluid:
of vibration.
references
10-times-size the
used
in the
effects
of
design
of the
manufacturing
blade.
effects
should
of base
designer
data
mass
mode
the
Effects
frequencies
force
that at base loads equivalent "builtin" at the base.
the
limit
reflect
the
effects
of
blade
taper,
pre-twist,
(e.g.,
ref.
and 72).
camber)
as well
Additionally,
the
be considered:
Experimental
Virtual
endurance
that
manufactured
geometric
centrifugal
fixity:
judgment
blade
and
environment. that
of
strength, specimens
and Environmental
of and
effect
yield from
finishes
Geometric
geometry Analytical
strength, obtained
In
stator
virtual-mass
will
effects
vane
33,
be small,
in air,
with
and
(from water
of reference denser
fluids
geometry
of this effect
data
75. the
32.
be assumed
blade
ref.
frequency
that
Vibration natural
Verification
of Natural
reduction
testing
of" prototype
Frequencies or actual
frequencies.
77
blades
shall
verify
the
calculated
to be
and for
a
with
in liquid
significant. 3.3.1.2.3.3
Note
are given 74)
are compared Note
the
analyzed.
in figure
could
on the
and
being
shown
magnitude oil,
attachment,
design
are
depend the
method
of
the blade
experimental
vibrating but
type
specific
blades
speed,
for determining figure
the
the
rotor
design
analytical-prediction
effect
on
for
blade
is
I
F
I
_FREQUENCY
FIRST
FLEXU_L
sooo__t__.__----l---.--------/
NOTE:
SEE
FIGURE
20
FOR
M-|
--_
DOVETAIL .O5
25001_
DIMENSIONS
_2ooo_
,--LOGAR.T. .C 4
iooo-
"___ -CENTRIFUGAL AT NOMI_L
5°° t
Effect
.03
°
• 02
_
z
.01
LOAD DESIGN
SPEED
500
I
I
IOOO
1500. BASE
Figure 32. -
D-
I_-IE° °'VALENT I
O
.04
LOAD,
of base load on blade natural
I
I
2000
2500
I 3000
LBF
frequency
and damping
(M-1 dovetail).
TEST MODE FLUID
O
1.0--
O
\
•
WATER
\
OIL
A D
r-
PREDICTED
FIRST
a.
'*0.5 =, u.
w
_v
'
500
NATURAL
Figure
33.-
Effects
of fluid
I • 1500
1000 FREQUENCY,
virtual mass on Mark
78
HZ
15-F vane natural
frequency
(data from
ref. 74).
In view
of the
uncertainties
blading,
it is recommended
through
a frequency
within
the
effect,
3. 3.1.2. 3.4
Resonance
Campbell frequencies
with
for
frequencies
in the and
of low-aspect-ratio
on prototype
known
or actual
potential
vibration
operational
of elasticity,
shall
under should
and
forcing analysis,
blades
frequencies experimental
environmental
effects
fluid virtual-mass
effects).
potential
frequencies
(i.e.,
be used
to forcing for other ports).
3.3.1.2.3.5
Self-Excited
from
forcing
by
the proximity
frequencies.
Figure
to resonance 17
of blade
shows
the
natural
recommended
applicable.
frequencies sources
to determine
forcing
margins
be examined return-flow
be separated
all conditions.
potential
In addition
size shall
the
frequencies
to account
in modulus
natural
be conducted
to encompass
of these
frequencies
margin
diagrams
testing
the
Margin
proximity-to-resonance
Blade
use
change
natural
adequate
bench
be modified
centrifugal
in predicting
sufficient
In the
should
Blade
that
range
pump.
magnitudes
involved
due
to wakes
of excitation
from
(e.g.,
adjacent
blade
thrust-balance
rows,
system
the
pump
should
or bearing-coolant
Vibration preclude
self-excited
vibration.
It is recommended that the empirical frequency-parameter ref. 39) be used to avoid self-excited vibration:
2zr
rule
noted
below
(adptd.
from
ft C
_t-
_> 1.6
(16)
_>
(17)
Wl
2rr fb C _b W1
where
_t = torsional ft = first
frequency
torsional
parameter
frequency,
Hz
79
0.33
fb = first flexural frequency,Hz C = blade chord length, ft wl = fluid relative velocity at stall mid-radius, ft/sec _b = flexural frequency parameter
3.3.1.3
PROFILE
3.3.1.3.1
TOLERANCES,
Tolerances
Profile
and Surface
tolerances
performance
the
longitudinal The
that of
direction.
specification
used
polish
"Out-of-spec" that
local
rotor
stator
hydrodynamic
be held
finish
should
consider
Surface
-There that
blades,
A fillet radius equal information (ref. 48) for be used
FILLET
RADII
affect
blade
hydrodynamic
finishes
do
in. be specified fairing
within
in
on the
both
the
basic
profile
transverse
and
¼°.
the manufacturing
of 63 /_in. rms
technique
or better
that
will be
are recommended.
be permitted.
not
meet
On
the
especially In no
in case
from
the
J-2
specifications other the
hand,
engine ("out
should
(Mark
of spec")
small
trailing-edge
program
usually
deviations region,
an out-of-spec
15-F
that can
seriously
condition
are
prevail that
pump) of small
in all the affect
would
the affect
be accepted.
Fillet radii shall considerations.
outlined
continuous
is evidence
performance.
Fillet Radii
should
should
not
3.3.1.3.2
practice
angles
should
integrity
of -+ 0.002
Blade
blades.
structural
applicable
tolerance and
hydrodynamically.
or
not adversely
smooth
the
conditions
consequence
shall
a
marks
parts.
AND
adequacy.
of surface
in producing
Transverse
finish
a maximum
restriction
FINISH,
Finish
and surface
or structural
It is recommended with
SURFACE
the
be as small
within
to the maximum thickness indicates a stress-concentration
recommended in
as possible
section
in assessing
fillet-to-blade 3.3.1.2.3.
The
of
the
stress-concentration
thickness factors
80
imposed
by structural
the blade is recommended. factor of approximately
reference
ratios.
limits
ratio noted for
for use above
other
in the
or other fillet-to-blade
Available 1.1 would be blade suitable
design data
thickness
3.3.2
Blade Attachment
3.3.2.1
METHODS
The
blade
attachment
manufacturing, An
appropriate
method
designed.
A
study
which
in
single
should
be made mechanically
large
practice
during
3.3.2.2 3.3.2.2.1
Single-tang
corresponding in sizing
that
determined
The
method
retention
of
considerable will be reacted
The
data
to retain blade
the
particular
be recommended. design
the
pump
phase
are
of the
evaluated
turbopump.
in applications
expense
of
to prevent
being
A configuration
considerations
consideration
be made
those
in pure
must
attachment
airfoil
weight).
If
incorrect
The
requiring mechanically
assembly.
on
in the
airfoil
axially
all
probable
be
applied,
withstand
loading
M-1
pump
are
recommended
and vibratory-load that
condition in section
blade
the
A steady-state
stress
shall
failure.
used
state
defined
under
the
cause
This
the
margin
would
is shown
cause
on figure
if
condition airfoil
failure
16 and
is the
at the should same
as
3.3.1.2.3.
in the load
dovetail
slot
conditions.
because
should If
it is difficult
provide
shear
pins
to ensure
that
positive are the
used, load
shear.
predicted
stress
as defined
to
practice
Stress
predicted
blades,
to a stress
Vibratory
alternating The
the
at
on
(cost)
receive
cost
are utilized.
dovetail.
by the
safety
properly
should
wouM
blades
the
used
lower
similar
attached
be used
3.3.2.2.2
which
dovetails
dovetail
depends
manufacturing
should
provision
attached
to that
mechanically
of weight,
Strength
mechanically
equivalent
compromise
DESIGN
Structural
For
blades
or preliminary
blades
(potential
MECHANICAL
an acceptable
cannot and
conceptual
are selected,
reflect
the
therefore weight,
the
lots
blades
shall
considerations.
attaching
attached
production
attached
for
assembly,
use
of
method
and assembly
state
in the
attachment
shall
be
less
than
the
allowable
stress. steady-state by
modified
and vibratory Goodman
stresses diagrams.
81
should
be compared
Modified
Goodman
with
material diagrams
property should
be
constructed with adequatesafety factors applied to yield strength, ultimate strength, and endurancelimit. It is recommendedthat the diagrambe constructed in accordancewith the practice defined in section3.3.1.2.3. It is recommended that the maximum stress in the neck section of the dovetail be determined by methods basedon the photoelastic test results(ref. 76). The vibratory stress magnitude should include an appropriate stress-concentrationfactor (ref. 48). Generous fillets should be used.
3.3.3
Rotor
3.3.3.1
CONFIGURATION
The
basic
rotor
compromise
of weight,
A recommendation cannot
configuration
for
properly
be
size,
a basic
made.
(one-piece critical
or builtup)
speed,
configuration
Both
builtup
and
shall
cost,
and assembly
that
would
one-piece
interface, builtup
from
a single
tie bolt,
and
concpet,
one-piece
3.3.3.2
welded
forging bearing
and
MECHANICAL
recommended
Axial Thrust
3.3.4.1
TYPES
should
be examined
phase and a suitable choice Size permitting, a one-piece
this
construction
problems
problems
that
that
may
may
be
made rotor
precludes
(1) disk
be associated
with
associated
with
a a
practices
for mechanical
design
of the
rotor
are presented
Balance System
OF SYSTEMS
thrust-balance
It is recommended type
weldment-quality
for all applications
DESIGN
3.3.4
system
because
misalignment
optimum
configuration.
Design criteria and in reference 6.
The
(2)
is preferred, journal
an acceptable
considerations.
configurations
during the turbopump conceptual or preliminary design after evaluation of assembly methods, weight, and cost. machined
be
reflect
(e.g.,
system that
shall preclude
excessive
a self-compensating
so-called
series-flow
thrust
thrust-balance or
82
double-acting)
loads system
on the bearing. be used.
depends
on
The
choice
the
particular
of
turbopump design. Each type should be examined during the conceptual or preliminary design phase of the turbopump to determine compromisesin terms of recirculating-flow requirements(pump performancepenalty), net thrust load magnitude anddirection over the pump operating range,and potential instability.
3.3.4.2
MECHANICAL
3.3.4.2.1
Design Basis
The
design
the the In view both
of the
pump and turbopump. of the
at
balance
DESIGN
thrust-balance
turbine
over
uncertainties
design pistons
point
system
the
total
involved
and
over
be designed
system,
3.3.4.2.2
the axial
Structural
The
excess
load
design
effects
It is recommended design This
speed stress
of
that
that and
3.3.4.2.2.1
Balance
the
10 percent
the
development not possible
the
thrust-
and
net
axial
thrust
operating
and
range,
of
range
turbine
of
axial
thrust
it is recommended
with
provision
in the
that
system
program. During the initial to counteract thrust with
by thrust
stress
Piston
due
with
caused
balance
differential
to
10 percent
stress
Axial deflection differential shall
than
capability
the
pump
operating
be reacted
centrifugal,
be combined
the piston interface.
It is recommended
of
is at least
should
should
on
transient
to
phase of a thrust
bearings.
Strength
mechanical
combined loads.
load
and
predicting
turbopump
permit trimming during the turbopump the start transient, where it is normally balance
be based
in accurately
the
with
shall
steady-state
the
by
pressure,
centrifugal
above
the
stress
of the
and
load
nominal
due
differential
based
on
dijJ'eretztial
be determined design
to maximum thermal
be
thermal
at a mechanical
speed
of the
differential
contraction
the
turbopump.
pressure
at the
across
piston/shaft
Deflection
at the outer diameter of' not adversely affect the flow that
shall
piston
the
total
piston axial
be
piston
sized travel.
83
the balance system.
so that
outer
piston
diameter
due
axial
to
pressure
deflection
is less
3.3.4.2.3
Balance Piston/Pump
Contact
at the
operating
practices
are presented
piston
and
that
stops
It is recommended rubbing
during
interface
in reference
the
Orifice
start
stationary
or gall with
SYSTEM
STABILITY
3.3.4.3.1
shaft
shall be positive
under
all
orifice
the
shall not
Reference
should
mating
make
in the bearing
transient.
orifices
6.
Contact
be incorporated
on impact
3.3.4.3
pump
stationary
the turbopump
a precaution,
shatter
and
Balance Piston/Stationary
The balance condition.
As
piston
Contact
conditions.
Recommended
3.3.4.2.4
balance
Shaft
contact
package
3 presents
be fabricated
rotating
surfaces
from
(sec.
at any
to avoid
operating
balance
recommended a material
piston
practices. that
will not
3.5).
Range of Stable Operation
The
thrust-balance
Dynamic
analysis
that
will be stable
given
in references
system
should for
shall
be conducted
all turbopump
52, 54, and
for liquid-hydrogen
systems
at high pressures
•
Increasing
cavity
•
Decreasing
cavity
•
Increasing
total
The
Two-Phase thrust-balance
operating
increased
Operating
over
the
to establish
55. From
•
3.3.4.3.2
be stable
these stability
(increased
turbopump
conditions. is achieved
Methods by
modulus)
volume drop.
Flow system
shall
not
be sub/ect
84
to two-phase/'low.
configuration
of dynamic
it can generally
area
pressure
range.
a thrust-balance-system
references,
bulk
operating
analysis
be concluded
are that
It is recommended that the thrust-balance-systemreturn flow be introduced into the pumping system at a point where the pressurelevel is greaterthan the vapor pressureof the recirculating fluid. Conditions within the thrust balancesystemshould be examinedand the local static pressureof the fluid kept abovefluid vapor pressureat all points within the flow circuit.
3.4
PUMP STATOR
3.4.1
Vanes
3.4.1.1
PROFILE
Design profile
TYPES
criteria and recommended types (section 3.3.1.1).
3.4.1.2 With
MECHANICAL the
for vane
3.4.1.3 Design
criteria
fillet
of
radii
3.4.2
3.4.2.1
are the
design
TOLERANCES,
and
for
centrifugal-load
mechanical
PROFILE
practices
vane
profile
types
are
the
same
as for blade
DESIGN
exception
practices
SURFACE
recommended
same
effects,
design
as for blades
FINISH,
practices
as for blades
the
are the same
for vane
(section
AND profile
criteria
and
recommended
(section
3.3.1.2).
FILLET
RADII
tolerances,
surface
finish,
and
3.3.1.3).
Vane Attachment
METHODS
The
vane
attachment
manufacturing,
particular
pump
configuration assembly
method
and assembly
As is the case with
weight,
ASSEMBLY
the
blades,
being
study
difficulty,
reflect
in and
an acceptable
compromise
of weight,
considerations.
an appropriate
designed,
made
shall
and selecting cost.
method
a
single a suitable
The
use
85
for attaching practice method
of individual
cannot should vanes
the vanes be
depends
on the
recommended.
The
include
evaluation
will in general
require
of a
stator housing with a greater envelope diameter than that required for vanesmachined integrally on segmentedrings or cylinders. Thus, from a weight standpoint, the latter method is preferable. Individual vanesshould be considered when largeproduction lots are required.
3.4.2.2
MECHANICAL
3.4.2.2.1
Structural
The
vane
cause
shall
withstand
loading
equivalent
to that
which
would
failure.
and vibratory load condition at the attachment that would cause airfoil failure should be used
exception
figure
Strength
attachment
airfoil
A steady-state in the airfoil the
DESIGN
of centrifugal
load
considerations,
corresponding to a stress in sizing the attachment.
this condition
is the
same
as that
state With
shown
on
16.
3.4.2.2.2 The
Vibratory
Stresses
predicted
stress
alternating The
attachments in
should
a blade
The
practice
with
include
3.4.2.2.3
attachment
shall
dovetail
be below
to the
vane
Goodman
It is recommended
the
allowable
process
should projection (sec.
should diagrams.
factors that
compared
Modified
applied the
be
Goodman
to yield
diagram
With
strength,
be constructed
3.3.1.2.3. be fillet
3.3.2.2.2).
stress-concentration
platform
stress
safety
in section
to the
design
an appropriate
modified
projections
applied
vibratory
adequate
limit. defined
lug-type
be
and
by with
endurance
the
should
the projection
as defined
constructed
and
with
concentration used
be
strength,
in accordance
vane
steady-state
data
should
ultimate
Vane
attachment
property
diagrams
in the
stress.
predicted
material
state
factor
should
be used.
stator
assembly
examined
to determine
steady-state The (ref.
stress
vibratory 48).
if a stress
similar
stress
A generous
to that
magnitude fillet
Load Transmittal method
the housing conditions.
used shall
to transmit provide
positive
load
86
circumferential
transmittal
under
and all
axial probable
loads
to
load
from
Keys acting in shear have been successfullyutilized to transmit stator torque loads on all state-of-the-art configurations and are therefore recommended.With regard to axial load, the stator assembly should be designed to be captive in the pump housing assembly. Differential thermal contraction of the housing and stator assemblyin the axial direction should be matched to the extent that excessivelooseness(or conversely excessivestress)is avoided.
3.4.3
Stator and Volute
3.4.3.1
HOUSING
The
TYPES
stator/volute
housing
hydrodynamic, Reference and
be examined
avoid
should
excessive
diffuse
the
and
castings carefully
flow
cost have
are not
and
considered.
3.4.3.2
HYDRODYNAMIC
3.4.3.3 Design
criteria
and
and
portion
compromise
considerations. discussion
of that
tie the
from
walls
axial
passage together
noted,
achieved
either
last
flow
a weight
along
stator/volute would
or the
cast
stage
radially
the Thus,
cast
all
lead the
Welded time if the
and
proper
vanes
structures cost
guide should
total
pump
be an integral
are presented
in reference
1.
practices
are presented
in reference
1.
lead
advantages
methods
practices
87
volute
if fabrication
fabrication
housing
standpoint.
these
DESIGN
recommended
For for
design phase, considerations
the
structures
structures,
stator/volute
into
DESIGN
recommended
design
housings.
are recommended; and
however,
in practice.
welded that
with
be optimum
and hydrodynamic
structurally.
consideration,
As
been
both
the
volute
of
configuration. from
in the
it is recommended
MECHANICAL criteria
flow vanes
With
permits,
Design
the
predominant.
concept
state-of-the-art configuration
at a suitable
is a predominant
always
acceptable
and fabrication
volute
housing
is preferred
Diffuser
if weight
the
an
be made. During the conceptual or preliminary and deflection, weight, and fabrication (cost)
29)
losses,
be gradual.
be selected time
(fig.
reflect
weight,
for
basic
in arriving
volute
shall
for a detailed
practice
cannot properly load, stress
A "folded"
and
be consulted
a recommended
applications hydrodynamic
To
design
and deflection,
recommended
pumps,
should
stress
1 should
criteria axial
Housings
should assembly unit.
of be
3.4.4
Bearing Housings
3.4.4.1
TYPES
The
bearing
turbopump
rotor
requirements
imposed
Radial
and
axial
conceptual themselves if
specially same
required
by the
satisfy
radial
stiffness
critical
speed
considerations
thrust
balance
requirements
from
requirements
should
bearings
provided by
a housing
piston
orifice
(ref.
be used
housing. with
housings
preload
axial
stiffness
The Low
should
be
achieved
limit
radial
to
axial
should
will be established
assembly. constants.
spring
rate
be achieved
Rotor
axial
movement
axial
spring
constant
bearing
housing
is desired.
beyond
the
stop
the
housings for radial
locally
movement
by spring
to preclude
during
bearing values
in a
(ref.
10).
If, for example, loading
with
position
contact
axial
should
with
the
be
balance
DESIGN
Strength
mechanical
design
of
provided
if a specified
the
high
Structural
effects
standpoint,
stops
and
by
3).
MECHANICAL
The
with
are used,
in the
reacted
3.4.4.2.1
carrier
imposed
system.
of the bearing
a critical-speed
bearing
principal ball
3.4.4.2
system
stiffness
designed
preloaded stops
shall
or preliminary design phase of the turbopump should be rigid structures with high spring
stiffness, The
housing
pressure
of
the
loading,
internal
shall
loading,
be
based
external
on
the
loading,
combined
and
thermal
gradients. Internal
and
dependent (common
external on
to
the
loads specific
all designs)
determined
from
determined
in accordance
loads
at
the
flanges
thermal
may
be
turbopump propellant due
contractions,
assembly from
to thermal
be
to mount
a hot-gas
to be
and points). must
by
It
is recommended to
Rotor
will be subjected
1.2
loads
(common
practices
defined
line
installation
inertia
forces
turbopump Turbine-end be designed
gradients.
88
bearing
the
internal
to all designs) in reference components
reaction housings
to withstand
pressure
maximum
misalignment,
of attached mount
will be partially
that
times
and
criteria
caused
housing
equal
analysis. the
consider
turbine
bearing design.
assumed
with can
necessary
the
turbopump be
the hydrodynamic
differential also
to which
loads that
as
should
be
6. External line
pressure,
(sec.
3.4.5).
(dependent
separate
the stresses
pressure
and
cryogenic deflections
It on
3.4.4.2.2
Clearances
The
bearing
housing
turbopump The
pump
housing)
and Tolerances design
shall
preclude
rubbing
housing
assembly
(i.e.,
be examined
the
as a unit
front
and
rear
in establishing
bearing
the
radial clearance. As a part of this assembly, the bearing and stator alignment. Interference joints with mating tolerances
and
concentricities
suitable
stator clearances. Absolute magnitudes will dimensional tolerances should be controlled can conveniently The
radial
rotor
be achieved
deflection
(high
Adequate
the
safety
be utilized. and
spring below
factors
that
shall
The
safety
movement to
points
should
would
the
stator
axial
"built-up"
due
limit
cause
bearing
radius
the
stress
and
rotor
and
design. Axial axial clearance
to radial
radial
load
on the
deflection).
be sufficiently
rigid
The to limit
rubbing.
housing
against
total
3.4.5
Housing Interfaces
3.4.5.1
INTERFACE
AND
SEAL
design
If
analysis
will dictate
webs
ultimate
and
are
are
and the
1.1 on
method
recommended
used
in
the
0.2
percent
to be used for analysis structure,
structural housing
testing
of
bearing
housings
assembly.
and Static Sealing TYPES
Alignment housing-to-housing
and axial
alignment
interfaces of" the rotor
and
seals
relative
shall
to the
89
in of
web
are recommended.
by
turbopump
strength
methods
of 1.0 or greater
deflection of the
1.5 on ultimate
Finite-element
structures.
ratios
and
or as part
of
of the specific
analysis.
thick-shell
The
rotor
desired
protect
factors
complexity
deflection
thickness-to-fillet
3.4.5.1.1
desired
on the particular that the desired
is that
and
stator
contribute strongly to rotor should be dimensioned with
the
mount those
and
by shimming.
the
rate
volute,
rotor
housings housings
to achieve
assembly
housings,
probable
be dependent to the extent
consider
at turbopump
complex-shape,
component
stationary
failure.
stress
Verify
and
Factors
It is recommended yield
pump
should
to magnitudes
Safety
yield
analysis
structure
deflection
3.4.4.2.3
during
bearing-housing
bearing-housing local
rotating
components.
must
diametral
of
provide
turbopump
for
and
housing
maintain assembly.
radial
as a
It is recommendedthat interference-fit pilot diametersbe usedon housing-to-housingjoints and that some degree of interface fit be maintained under all interface environmental conditions. Note that the recommendedpractice here does not apply to those interfaces where an extreme temperature differential may exist (e.g., between the pump housing and turbine manifold).
3.4.5.1.2
Leakage
The
housing-to-housing
throughout Available of
seal
the
types
sealing
possible.
should and
throughout
the turbopump
up the
joint
The
In critical used.
assembly determining
type
the
conceptual
selected.
with
the
guidance
exist
in
operating
or preliminary
A seal
same
type
fluid
that
should
design
has
phase
demonstrated
be utilized
wherever
on seals.
such
steady-state
the required
bolt
loads
the joint
interface
and
remains
for symmetric operating
be calculated
in determining
housing-to-housing
that
be examined
pump
should
all
joints
range.
be preloaded should
the
assembly used
leakage
and
conditions,
preload,
the
for each
of the
tight
all operating
asymmetric and
elastic
under
loads
at the
transient-temperature
stiffness
of the parts
environmental
conditions
reliably
the preload
making and
a
stresses.
Bolt Preload method
without
probable
interface
In determining
model
3.4.5.2.2
shall
should
conditions,
conditions. spring
joint The
propellant
Strength
continuity
assembly
prevent
DESIGN
Structural
conditions.
during
seal
design
reliably
range.
applications
59 provides
Structural
interface
operating
a suitable
MECHANICAL
3.4.5.2.1
and seal shall
be examined
in previous
Reference
3.4.5.2
be
turbopump
turbopump
reliable
The
the
interface
for
preloading
exceeding joints,
In those torque
bolt
joints and
where
the
that torque
minimum
stress. of friction
at assembly
shall
induce
stresses.
it is recommended
maximum coefficient
the
allowable
The
bolt
elongation
measurement
or other is specified,
probable
coefficient
minimum
permissible
should
be used
90
of friction assembly
in determining
positive the
preload
maximum should
torque
minimum
indicators permissible
be employed and
joint
the
maximum
preload.
in
3.4'5.2.3
Safety
Design yield.
Since
the
Factors
safety
joint
factors
design
shall
is based
assembly, it is important factors of 1.5 on ultimate However,
these
safety
allowable
stresses
are
on
that
the
the joint
elastic
will
stiffness
factors within
should the
be
linear
stress
examined
range.
The
not
of
to stay within the linear range strength and 1.1 on 0.2 percent
used in comparing bolt design failure (ref. 71) is recommended.
3.5
ensure
fail
the
in either
ultimate
components
or
making
up
the
of the stress-strain diagram. Safety yield generally are recommended.
for
each
"effective"
to material
property
given
are applicable
design stress
data.
The
to
ensure
that
of the
bolt
should
be
theory
of
Mises-Hencky
the
MATERIALS
Criteria and recommended practices with liquid-hydrogen propellant.
3.5.1
Property
Selection
successfully are given treatment
3.5.2
to axial-flow
pumps
for use
Data
of materials
for
components
be based on guaranteed minimum reflect probable minumum property Typical property references such
here
in liquid-hydrogen
properties values.
or typical
data at liquid-hydrogen temperature as 61 through 66. Recommended
in liquid-hydrogen
axial
pumps
principally in references 62, for the heat-treatable alloys.
noted
64,
and
63,
are
axial-flow property
pumps
for various materials materials that have in Table 66
and
IV. Data include
shall
data adjusted
are been
for these
the
to
given in utilized materials
appropriate
heat
Ductility
Materials It is recommended
shall possess that
adequate
materials
ductility with
at liquid-hydrogen
an elongation
at liquid-hydrogen temperature be utilized yielding under steady load conditions.
for
91
temperatures.
of at least
components
that
4 percent may
in four be
subject
diameters to local
3.5.3
Impact Strength
Materials
for
adequate
impact
If impact
components
loading
strength
that
strength
be subject
at liquid-hydrogen
is anticipated,
(Charpy
may
to
of
at
least
loading
shall
possess
temperature.
it is recommended
V-notch)
impact
12
that
ft-lbf
the
(or
materials
possess
equivalent)
at
an impact
liquid-hydrogen
temperature. In
particular,
materials
rubbing should combinations
for
thrust-balance-system
components
not shatter on impact have demonstrated
or gall with non-shattering
thrust-balance-systems
and
liquid-hydrogen
mating and
that
may
surfaces. The anti-galling
hydrostatic
be
subject
to
following material characteristics in
bearings
and
are
therefore
Stationary
Component
recommended: Rotating
Component
K-Monel Inconel
718
Titanium
(tungsten-carbide
plated)
(Ti-A 1 10-AT-ELI)
3.5.4
Endurance
Materials
for
vibratory
stress
Experimental
data
components
defining
the
only
it may
be necessary
turbopump.
However,
blades
or
subject
shall possess
selected, of
on a few selected
vanes,
experimental
data
comparable
to those
bronze
Leaded
bronze
Leaded
bronze
(ref.
77)
Limit
obtained of the
Leaded
for with
to
adequate
combined
endurance
endurance
limit
at liquid-hydrogen
alloys.
If endurance-limit an endurance
the
significance
specimens
of the production
clearly that
of the indicates
reflect
92
data ratio
appreciable
that
temperature
have
available
for the
are not
in the preliminary
endurance-limit
manufacturing
component.
and
limit.
to assume
example,
steady-state
final
magnitude designs
should
processes
and
been alloy
design
phase
in the
design
be surface
based
on
finishes
APPENDIX Conversion
Physical
of U.S. Customary
A Units
Conversion
U.S. customary unit
quantity
to SI Units
SI unit
factor a
Angle
deg
rad
1.745 x 10 -2
Flowrate
gpm
m3/sec
6.309x10
Force
lbf
N
4.448
Headrise
ft
m
3.048x10
ft-lbf/lbm
J/kg
2.989
ft-lbf
J
1.356
ft
m
3.048x10-1
in.
cm
2.54
Load
Ibf
N
4.448
Mass
Ibm
kg
4.536x10
-1
NPSH
ft
m
3.048x10
-1
ff-lbf/lbm
J/kg
2.989
N/cm 2
6.895x10
N/m 2
4.788x101
rpm
rad/sec
1.047x10
-1
psi (lb f/in. 2)
N/cm 2
6.895x10
-1
/_in.
/_m
2.54x10
oF
K
ft 3
m 3
2.832x10
-2
gal
m 3
3.785x10
-3
Impact
energy
Length
Pressure
psi (lbf/in. psf(lbf/ff
Rotational
speed
Stress Surface
finish
Temperature
Volume
aExcept conversion quantities, Revision,
for
temperature, factor see
NASA
to
where obtain
Mechtly, SP-7012,
the
equivalent
E.
A.:
The
2) 2)
conversion value International
is in
SI
made unit.
System
as For of
1973.
93
shown,
Units.
Physical
-1
-1
-2
5 K
multiply
a complete
-s
value
listing
of
Constants
given
in
9 (°F +459'67)
U.S.
conversion
factors
and
Conversion
customary for
basic
Factors.
unit physical Second
by
APPENDIX
B
GLOSSARY Definition
Term
allowable
load (or stress)
the load that, if exceeded in the slightest, structural element under consideration. buckling, yielding, ultimate, or fatigue
produces failure of the pump Failure may be defined as failure, whichever condition
prevents the component from performing its intended Allowable load is sometimes referred to as criterion load allowable
to material
ratio of blade height (or length)
aspect ratio balance dram (balancing
stress is equivalent
drum)
strength.
to chord length
special balancing device used to balance axial thrust pumps; it can be used in combination with an automatic or alone (seldom)
base fixity
index of the relative tightness in the mounting or the vane in the vane support
blockage
decrease in effective and end wall
blockage
factor
function. or stress;
in multi-stage balancing disk
of the blade in the rotor
flow area due to the boundary
layer on the blades
the fraction or percentage by which design flow area is increased to account for blockage; conversely, the ratio of flow area corrected for blockage to design flow area
cavitation
formation
of vapor
bubbles
in a flowing
liquid whenever
the static
pressure becomes less than the fluid vapor pressure chord length
linear distance between the end points of the blade-profile leading and trailing edges as measured on the chord line (a line joining the points of intersection of the blade profile leading edge and trailing edge with the mean camber line)
creep
permanent deformation than the load necessary
of material caused by a tensile load that is less to yield the material; some time is required to
obtain creep critical speed
cryogenic
shaft rotational
speed
system coincides
with a possible forcing frequency
fluids or conditions
95
at which
a natural
at low temperatures,
frequency
of a rotor/stator
usually at or below -238°F
Definition
Term
trade
curvic coupling
name
generated
of the
Gleason
in a manner
Works
for a face-gear
type
of coupling
similar to that used for bevel gears
design load (or pressure)
product
of the limit load (or pressure)
and the design safety factor
design stress
the stress, in any structural element, that results from the application of the design load or combination of design loads, whichever condition results in the highest stress
deviation
angle between fluid outlet direction camber line at the trailing edge
angle
an index of local diffusion
D-factor (diffusion
and the tangent
on the blade suction surface:
factor) w2
end wall
(DF)R
=
1--
(DF)s
=
1 -
surface of the housing
endurance (fatigue
limit limit)
forced-vortex
free-vortex
flow
flow
hydrogen
ratio embrittlement
W1
V3 -V2
Awu +
2owl
AVu +
2oV2
and rotor hub between
maximum alternating stress at which an infinite number of cycles flow in which the fluid tangential other than inversely with radius
adjacent
a material
presumably
can endure
is constant
velocity varies inversely
from hub to tip while
with radius
ratio of rotor radius at blade hub to rotor radius at blade tip loss of ductility newly formed
in a metal
as a result
of the exposure
of the metal
to
gaseous hydrogen
impulse stage
stage in which there is no change in static headrise
incidence
angle between
angle
blades
velocity is forced to vary in a manner
flow in which the fluid axial velocity the fluid tangential
hub/tip
to the blade mean
fluid-inlet
direction
and tangent
across the rotor to blade mean camber
line at leading edge limit load (or pressure)
maximum expected load (or pressure) that will occur in a structure under the specified conditions of operation, with allowance for statistical variation
96
Definition
Term
Mach number
ratio of the speed of fluid flow to the speed of sound in the fluid
magnification
ratio of the deflection produced by an alternating produced by a steady load of the same magnitude
factor
the fraction by which load or stress
margin of safety (MS)
the allowable
MS
net positive
suction
head
NPSH =
(NPSH) proof pressure
=
1 --R
load to the deflection
load or stress exceeds
the design
1
-
total fluid pressure - fluid vapor _uid density
design pressure multiplied by the proof-test is the reference from which the pressure
pressure
]
at inlet
safety factor (proof pressure levels for acceptance testing
are established) radial equilibrium
flow condition in an annular passage in which there is no radial velocity component; i.e., the fluid pressure forces in the radial direction.are in equilibrium
reaction recovery
with the centrifugal
the ratio of static headrise moment
retardation
factor
bending caused radial line
in the rotor to static headrise
by centrifugal
an index of blade-passage
forces
force in a blade
that is tilted
from
a
diffusion: w2
(RF)R
in the stage
wl
V3 (RF)s
V2
root
juncture
of blade and rotor hub
safety factor
an arbitrary multiplier (or divider) greater than 1 applied in design to account for uncertainties in design, e.g., variations in material properties, fabrication
quality, and load distributions
within the structure
solidity (blade)
ratio of blade chord length to blade spacing
stacking
imaginary line on which the centers of gravity of the prone stacked to form the blade or vane shape from hub to tip
axis (or line)
97
sections
are
Definition
Term
stagger angle
the angle between the chord line and a reference is the axis normal to the plane of the blade row
stall
loss of pumping capability surface of the blades
stall margin
margin between pump operation at the design-point flow coefficient and operation at the flow coefficient at which the pump will stall
thrust-balance
untwist
forces
flow
direction
as a result of flow separation
flow through the thrust balance system that area) force necessary to balance axial thrust that
provides
on the suction
the (pressure
forces acting on a twisted reduce the blade twist
blade
virtual mass
mass of fluid near a vibrating
blade that vibrates with the blade
volute
spiral-shaped portion stage of a pump
of the housing
produce
that usually
a torque
tending
X
to
that collects the fluid from the last
Definition
Symbol C
chord
C-4
designation
D
diameter
Ds
specific
diameter,
DF
diffusion
factor
DN
index to bearing
length for a family
of airfoil
shapes
D s = DH¼/Q _
speed capability,
in mm and rotational
the product
speed (N) in rpm
ELC
extra low carbon (content)
F
material
f
frequency
g
acceleration
due to gravity
ge
gravitational
constant,
strength
Ibm-ft
98
32.17
lbf-sec 2
of bearing
bore size (D)
Definition
Symbol H
headrise,
i
fluid incidence
k
stress-concentration
MS
margin of safety
N
pump rotational
H = H 2 -H
1 (stage)
angle factor
speed
specific speed, Ns = NQY_/H 3A NPSH
net positive suction head
O/F
ratio of mass flowrate
Q
volume flowrate
P
pressure
R
(1) reaction (2) ratio of design load or stress to allowable
RF
retardation
r
radius
S
blade tangential
of oxidizer
to mass flowrate
of fuel
load or stress
factor
spacing
suction specific speed, S s = NQ'A/(NPSH) 3A SF
safety factor
TIG
tungsten-inert-gas
T73
designation alloys
U
blade tangential
V
fluid absolute
velocity
W
fluid velocity
relative to blade
(welding method)
for a heat-treating
and
velocity
\
99
tempering
process
for aluminum
Definition
Symbol Z
cavitation-breakdown
correlation
parameter,
Z = q_tan (/3T/2)
65 series
NACA designation for a family of airfoil shapes stagger angle fluid angle deviation angle efficiency blade camber angle
P
hub-to-tip radius ratio, p = rH/r T frequency parameter
o.
_-R
(1) stress; (2) solidity, cavitation
parameter,
o = C/S z R = NPSH/(u2/2gc)
flow coefficient,
_ = Vm/u
head coefficient,
_ = gcH/u 2
total-pressure-loss
coefficient:
Hloss
Subscripts a
axial
alt
alternating
b
flexural
cf
centrifugal
e
endurance
eq
equivalent
exit
outlet
forces
"
100
c_R =
w12/2gc
H loss ;_s=
V22/2ge
Subscripts
f
fluid
ff
fluid forces
H
hub; hydraulic
i
ideal
1
liquid
m
meridional;
R
rotor
S
stator
SS
steady
T
tip
t
torsional
tu
tensile
ultimate
ty
tensile
yield
u
tangential
v
vapor
1
rotor
inlet
2
rotor
outlet
3
stator
(fluid) mean
state
outlet
or stator
inlet
or second
rotor
Material
Identification
CRES
corrosion-resistant
helium
pressurant
Inconel
inlet
718
trade
name
steel
helium of
(He)
per MIL-P-27407
International
5597A)
101
Nickel
Co.
for
nickel-base
alloys
(AMS
Identification
Material K-Monel
trade name
of International
alloy containing
Nickel Co. for a wrought,
age-hardenable
Ni, Cu, and A1
leaded bronze
copper alloy containing
LH2
liquid hydrogen
LOX
liquid oxygen,
polyurethane
any of various thermoplastic polymers that contain-NHCOO-linkages; produced as fibers, coatings, flexible and rigid foams, elastomers, and resins
Ti-A 110-AT-ELI
zinc and lead
(H2), propellant propellant
an extra-low-interstitial interstitial controlled
grade per MIL-P-27201
grade per MIL-P-25508
(ELI)
grade
notch toughness, and ductility - 423°F (LH 2 temperature) 300 Series
of Ti-5A1-2.5Sn
in which
elements O, N, and H and the substitutional element at lower-than-normal contents; strength-to-density
series of austenitic
remain
at acceptable
the
Fe are ratio,
levels down
to
stainless steels
(e.g., 304, 310,347) 304L (304 ELC)
extra-low-carbon variety of 304 austenitic steel; used in weldments for corrosive conditions where intergranular carbide precipitation must be avoided
2024
wrought
aluminum
alloy with Cu as principal
alloying element
7075
wrought
aluminum
alloy with Zn as principal
alloying element
ABBREVIATIONS Identification
Organization AF
Air Force
AIAA
American Institute
ASME
American
NAA
North American
NACA
National
Society
for Aeronautics of Mechanical
Aviation,
Engineers
Inc.
Advisory Committee
102
& Astronautics
for Aeronautics
(now NASA)
Identification
Organization NREC
Northern
PWA
Pratt & Whitney
WADC
Wright Air Development
Research
and Engineering
Aircraft
103
Center
Corporation
REFERENCES 1. Anon.: Liquid Rocket Engine Centrifugal Flow Turbopumps. Monograph, NASA SP-8109, December 1973. 2.
Anon.: Liquid Rocket Engine Turbopump NASA SP-8052, May 1971.
3. Anon.: Liquid Rocket Engine Turbopump NASA SP-8048, March 1971.
Inducers.
Turbines.
Bearings. NASA Space Vehicle
NASA Space
8. Severud,
L. K.; and Reeser, H. G.: Analysis
Rocket
Vehicle
Engines.
Design
Criteria
Monograph,
NASA
NASA Space Vehicle Design Criteria
NASA
Space
of the M-1 Liquid Hydrogen
Whirling Speed and Bearing Loads. NASA CR-54825, 9.
Design Criteria Monograph,
Seals. NASA Space Vehicle Design Criteria
6. Anon.: Liquid Rocket Engine Turbopump Shafts and Couplings. Monograph, NASA SP-8101, September 1972. 7. Anon.: Turbopump Systems for Liquid Monograph, NASA SP-8107, August 1974.
Design Criteria
NASA Space Vehicle Design Criteria Monograph,
4. Anon.: Liquid Rocket Engine Turbopump Rotating-Shaft Monograph, NASA SP-8121, February 1978. 5. Anon.: Liquid Rocket Engine SP-8110, January 1974.
NASA Space Vehicle
Aerojet-General
Vehicle
Turbopump
Design
Criteria
Shaft Critical
Corp., Dec. 20, 1965.
Gunn, S. V.; and Dunn, C.: Dual Turbopump Liquid Hydrogen Feed System Experience. presented at the Ninth Liquid Propulsion Symposium (St. Louis, MO), Oct. 25-27, 1967.
10. Graham, R. D.; Rowan, B. F.; and Shen, F. A.: Axial Pump Rotordynamic Study, NASA CR-72302, R-6982, Rocketdyne Div., N. Am. Aviation, Inc., Oct. 1, 1967. 11. Gunter,
E. J., Jr.: Analysis
Lab. for Engineering
of Rocketdyne
Nuclear Feed System Turbopump
Paper
Final Report.
Failure. W-12-229,
Res.
Sciences, Univ. of Virginia, June 1966.
12. Childs, D. W.: The Space Shuttle Main Engine High Pressure Fuel Turbopump Rotordynamic Instability Problem. Paper 77-GT-49, Gas Turbine Conference (Philadelphia, PA), March 27-31, 1977. 13. Choy, K. C.; Gunter, E. J.; and Allaire, P. E.: Rotor Transient Analysis 1 -Theory; Vol. 2 -Computer Program. Rep. 528144 (MAE 77/105), 1977. 14. Balje, O. E.: A Study on Design Criteria ASME, Series A, vol. 84, 1962, pp. 83-114.
and Matching
105
by the Modal Method. Vol. Univ. of Virginia, Sept. 8,
of Turbomachines.
J. Eng. Power,
Trans.
15. Crouse, J. E.;Montgomery, J.C.;andSoltis,R. F.: Investigation of thePerformance of anAxialFlow PumpStageDesigned by the BladeElementTheory- DesignandOverallPerformance. NASATN D-591,February1961. 16. Crouse,J. E.; and Sandercock, D. M.: Blade-Element Performance of 0.7 Hub-TipRadiusRatio Axial-Flow-Pump Rotorwith Tip DiffusionFactorof 0.43.NASATND-2481,September 1964. 17. Miller,M. J.; andCrouse, J. E.:DesignandOver-AllPerformance of anAxial-Flow-Pump Rotorwith aBlade-TipDiffusionFactorof 0.66.NASATND-3024,September 1965. 18. Johnsen,I. A.; andBullock,R. O., eds.:Aerodynamic Designof Axial-FlowCompressors. NASA SP-36,1965. 19. Farquhar,J.;
and Lindley, Aerojet-General
CR-54822, 20.
Lieblein,
S.; Schwenk,
B. K.: Hydraulic Design of the M-1 Liquid Hydrogen Corp., July 15, 1966.
F. C.; and
Broderick,
Limiting Blade Loading in Axial Compressor 21.
R. L.: Diffusion
Blade Elements.
Factor
for Estimating
NACA RME53D01,
Lieblein, S.: Analysis of Experimental Low Speed Loss and Stall Characteristics Compressor Blade Cascades. NACA RME57A28, 1957.
22. Miller, M. J.; Crouse, Axial-Flow 589-599.
J. E.; and Sandercock,
Pump Rotors
23. Taylor, W. E.; Murrin, 1 -Double-Circular-Arc December 1969.
D. M.: Summary
Turbopump.
NASA
Losses and
1953. of Two-Dimensional
of Experimental
Investigation
of Three
Tested in Water. J. Eng. Power, Trans. ASME, Series A, October
1967, pp.
T. A.; and Colombo, R. M.: Systematic Two-Dimensional Cascade Tests, Vol. Hydrofoils. NASA CR-72498, UARL-H910254-50, United Aircraft Corp.,
24. Taylor, W. E.; Murrin, T. A.; and Colombo, R. M.: Systematic Two-Dimensional Cascade Tests. Vol. 2 -Multiple Circular-Arc Hydrofoils. NASA CR-72499, UARL-J910254-56, United Aircraft Corp., April 6, 1970. 25. Colombo,
R. M.; and Murrin,
Double-Circular-Arc
Hydrofoils.
T. A.: Systematic NASA CR-72870,
Two-Dimensional United Aircraft
Cascade
Tests. Vol. 3 - Slotted
Corp., May 1, 1972.
26. Murrin, T. A.; and Taylor, W. E.: Systematic Two-Dimensional Cascase Tests. Vol. 4 - Cascade Test Data. NASA CR-121101, United Aircraft Corp., March 15, 1973. 27. Shepard, _28.
D. G.: Principles
Anon.: Parametric Inc., unpublished,
of Turbomachinery.
The Macmillan
Studies. Part V of Pump Hydrodynamic pp. 33-40.
Co. (New York),
Design. Rocketdyne
Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
106
1956. Div., N. Am. Aviation,
Unpublished. Collected
*29. Campbell,W. E.; Robertson,C. F.; and Yoshikawa,D. K.: The Designand Evaluationof an Axial-FlowImpulsePump.Aerojet-General Corp.,December 1962. 30. Bissell,W. R.; King,J.A.; Umemoto,G.A.; andWong,G. S.:Investigation of LiquidHydrogen Axial FlowImpulsePump(U). Bull.7th LiquidPropulsionSymp.,vol. 2,JohnsHopkinsUniv.,November 1967,pp.233-258.(CONFIDENTIAL) 31. Howell,A. R.: FluidDynamicsof Axial Compressors. Proc.Inst.Mech.Engrs.(London),vol. 153, 1945,p.445. 32. Crouse,J. E.; Soltis, R. F.; and Montgomery,J. C. Investigationof the Performanceof an Axial-Flow-Pump StageDesigned by the Blade-Element Theory-Blade ElementData.NASATN D-1109,December 1961. 33. Crouse, J.E.;andSandercock, D.M.: DesignandOver-AllPerformance of anAxial-Flow-Pump Rotor with aBlade-TipDiffusionFactorof 0.43.NASATND-2295,May1964. 34. Miller,M. J.; andSandercock, D.M.: BladeElementPerformance of Axial-Flow-Pump Rotorwith Blade-Tip DiffusionFactorof 0.66.NASATND-3602,September 1966. 35. Stripling,L. D.; andAcosta,A. J." Cavitationin Turbopumps. Parts1 and2, J. BasicEng.,Trans. ASME,SeriesD,vol.84, 1962,pp.326-350. 36. Serovy,G. K.; and Tysen,J. C.: Predictionof Axial Flow Turbomachine Performance by Blade ElementMethods. Paper61-WA-135, ASMEWinterAnn.Mtg.,Nov.26-Dec.1,1961. 37. King.J.A.: TestingPumpsin Air.Paper67-WA/FE4, ASMEWinterAnn.Mtg.,Nov.12-17,1967. 38. Lieblein,S.;andRoudebush, W. H.: Low SpeedWakeCharacteristics in Two-Dimensional Cascades andIsolatedAirfoil Sections. NACATN-3771,1956. 39. Armstrong,E. K.; andStevenson, R. E.: SomePracticalAspectsof Compressor BladeVibration.J. RoyalAeron.Soc.,vol.64, no.591,March1960,pp. 117-130. 40. Kemp,N. H.; andSears,W. R."Aerodynamic Interference BetweenMovingBladeRows.J. Aeron. Sci.,vol. 20,no.9,September 1953,pp. 585-597,612. 41. Smith,J. E.: Vibrationof Bladesin AxialTurbomachinery. Paper66-WA/GT-12, ASMEWinterAnn. Mtg.,NewYork,NY,Nov.27-Dec.1,1966. 42. Traupel,Walter:ThermischeTurbomaschinen. ErsterBand,SpringerVerlag(Berlin;Gottingen; Heidelberg), 1958. 43. Carter,A. D. S.: BladeProfilesfor Axial-FlowFans,Pumps,Compressors, etc.Proc.Inst. Mech. Engineers (London),vol. 175,no. 15,1961,pp.775-806. Dossier source
for
material
the
design
available
criteria
monograph
for inspection
"Liquid
at NASA
Lewis
Rocket Research
107
Engine Center,
Axial-Flow Cleveland,
Turbopumps." Ohio.
Unpublished.
Collected
44. Anon.:J-2Program,QuarterlyProgress Report,PeriodendingAugust31, 1963.NAA-RR-2600-12, Rocketdyne Div.,N.Am.Aviation,Inc. 45. Sawyer,J. W.,ed.: GasTurbineEngineering Handbook.GasTurbinePublications, Inc. (Stamford, CT),1966. 46. Severud,L. K.; and Chinn,T.: AnalyticalandExperimentalVibration Analysisof the Turbine Bucketsfor theM-1LiquidOxygenTurbopump.NASACR-54830, Aerojet-General Corp.,1965. 47. Regan,P. J.: MechanicalDesignof the M-1 Axial Flow Liquid HydrogenFuel Pump.NASA CR-54823, Aerojet-General Corp.,Feb.15,1966. 48. Peterson, R. E. StressConcentration Factors.JohnWiley& Sons,Inc.(NewYork),1953. 49. Anon.: J-2 Program,QuarterlyProgress Report,PeriodEndingOctober,1960.NAA-RR-2600-1, RocketdyneDiv., N. Am. Aviation,Inc., November,1960. 50. Anon.: Development of a 1,500,000-1b-Thrust (NominalVacuum)LiquidHydrogen/LiquidOxygen Engine.QuarterlyTech.Prog.Rep.4014-02Q-2, Aerojet-General Corp.,July25, 1963. 51. Lynn, E. K.: ExperimentalStressAnalysisin the Designof a Liquid HydrogenPumpRotor. Experimental Mechanics, vol.2,no. 12,December 1962,pp. 19A-23A. 52. Watters,W. E.; andLuehr,L.: Development of Steady-State andDynamicPerformance Prediction Methods for Turbopump Self-Compensating Thrust Balance Systems.NASA CR-72630, AGC-9400-19, Rev.1,Aerojet-General Corp.,March30, 1970. 53. Young,W. E.; and Due, H. F.: Investigationof Pressure PredictionMethodsfor RadialFlow Impellers,PhaseII, FinalReport.PWAFR-1276,Pratt& WhitneyAircraft Div. of UnitedAircraft, March8, 1965. 54. Connelly,R. E.: DesignStudyof the Mark9 Pump.WADC-TN-59-122, Rep.R-1354,Rocketdyne Div.,N. Am.Aviation,Inc.,March1959. *55. Anon.:DynamicStabilityStudyof a Series-Flow ThrustBalanceSystem.AppendixB, R6809P-1, Rocketdyne Div.,N.Am.Aviation,Inc.,1968,unpublished. 56. Stepanoff,A. J.: Centrifugal andAxialFlowPumps.Seconded.JohnWiley& Sons,Inc.(NewYork), 1957. 57. Radkowski,P. P.; Davis,R. M.; andBodul,M. R.: A NumericalAnalysisof the Equationof Thin Shellsof Revolution.ARSJ.,vol.32,no.1,January1962,pp.36-41. 58. Becket,E. B.; andBrisbane,J. J.: Applicationof the Finite-Element Methodto StressAnalysisof SolidPropellantRocketGrains.Rep.S-76,voi. I (AD 474031),November1965;vol. II, part 1 (AD 476515)andvol.II, part2 (AD 476735),Rohm& HaasCo.,January1966. * Dossier for thedesign criteriamonograph "LiquidRocketEngine Axial-Flow Turbopumps." Unpublished. Collected source material available forinspection atNASALewisResearch Center, Cleveland, Ohio. 108
59. Anon.:LiquidRocketDisconnects, Couplings, Fittings,FixedJoints,andSeals.NASAStihee Vehicle DesignCriteriaMonograph, NASASP-8119, September 1976. 0.
61.
Blakis, R.; Lindley, B. K.; Ritter, Ji A.; and Watters, W. E.: Initial Test Evaluation Hydrogen Turbopump Including Installation, Test Procedures, and Test Results. Aerojet-General Corp., July 20, 1966. Sessler, J. G.; and Weiss, V.: Aerospace Structural Metals (Syracuse University), Air Force Materials Lab., March 1967.
62. Williams, L. R.; Young, J. D.; and Schmidt, Thermal Expansion R-6981, Rocketdyne
Handbook.
Elevated
Janser, G. R.: Summary Corp., July 22, 1966.
64.
Inouye, F. T.; Hunt, V.; Janser, G. R.; and Frick, V.: Application Components. NASA CR-788, Aerojet-General Corp., June 1967.
of Alloy
65.
Inouye, F. T.: Properties of Large 7079 Aluminum NASA CR-513, Aerojet-General Corp., July 1966.
in a Cryogenic
Matin, Craig: MPR-3-251-369,
Technology
Engineering
and
63.
*66.
of Materials
2 vols. ASD-TDR-63-741,
E. H.: Design and Development
Properties of Aerospace Materials at Cryogenic Div., N. Am. Aviation, Inc., March 30, 1967.
of theM-1 Liquid NASA CR-54827,
of M-1 Engine. NASA CR-54961,
Alloy Forgings
Typical Low-Temperature Mechanical Properties of Several Rocketdyne Div., N. Am. Aviation, Inc., 1963, unpublished.
Handbook
of
Temperatures.
Aerojet-General
718 in M-1 Engine
Environment.
Materials.
Memo
67.
Anon.: Development of a 1,500,0004b-Thrust (Nominal Vacuum) Liquid Hydrogen/Liquid Oxygen Engine. Quarterly Tech. Prog. Rep. 2555-04Q-1, January - March 1965, Aerojet-General Corp., April 20, 1965.
68.
Carter, A. D. S.: The Axial Compressor. Sec. 5 of Gas Turbine Principles ed., D. Van Nostrand Co., Inc. (New York), 1955.
69.
Vavra, M. H.: Aero-Thermodynamics 1960.
70.
Horlock,
71.
Roark,
72.
Anon.: The Vibration of Blades in Axial Turbomachinery, Part 1: Theory and Practice of Design and Development. Rep. 1088-1 (AD 645156), Northern Research and Engineering Corp. (Cambridge, MA), April 29, 1965.
73.
Anon.: The Vibration of Blades in Axial Turbomachinery, Part 2: Design and Development Handbook. Rep. 1088-2, Northern Research and Engineering Corp. (Cambridge, MA)' April 29, 1965.
J. H.: Axial Flow Compressors. R. J.: Formulas
and Flow in Turbomachines.
Butterworth
for Stress and Strain.
Fourth
Scientific
and Practice,
John Wiley & Sons, Inc. (NY)
Publications
ed., McGraw-Hill
(London),
1958.
Book Co. (New York),
Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
109
Sir H. R. Cox,
1965.
Unpublished. CoUected
*74.
Turner,
J. D.: The
Rocketdyne 75.
Bisplinghoff, 1962.
76.
Heywood,
77.
Young, Pressure
Div.,
Effect
of Fluid
N. Am. Aviation,
R. L.; and Ashley,
R. B.: Designing W.
E.;
and
Cryogenic
AFRPL-TR-67-130,
Inc.,
On Stator
1963,
H.: Principles
Turbopumps, &
J.M.: Final Whitney
Blade
Resonant
Frequency.
SM 3111-8073A,
unpublished. of Aeroelasticity.
by Photoelasticity.
Reddecliff, Pratt
Density
Chapman
and Hall,
Investigation
of
Report,
March
Aircraft
John
Div.,
Wiley & Sons,
Ltd. (London),
Hydrostatic 1, United
Bearings
1966
-March
Aircraft,
Inc. (New
York),
1952. for 31, May
use
in High
1967 15,
(U). 1967.
(CONFIDENTIAL)
* Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
110
Unpublished.
Collected
NASA SPACE VEHICLE DESIGN CRITERIA MONOGRAPHS ISSUED TO DATE
ENVIRONMENT SP-8005
Solar Electromagnetic
SP-8010
Models of Mars Atmosphere
SP-8011
Models of Venus Atmosphere
SP-8013
Meteoroid Environment March 1969
SP-8017
Magnetic Fields-Earth
SP-8020
Surface Models of Mars (1975),
Revised September
SP-8021
Models of Earth's
(90 to 2500 km), Revised
SP-8023
Lunar Surface Models, May 1969
SP-8037
Assessment
and Control
of Spacecraft
SP-8038
Meteoroid Environment October 1970
Model-1970
SP-8049
The Earth's
SP-8067
Earth Albedo and Emitted
Radiation,
July 1971
SP-8069
The Planet Jupiter
December
1971
SP-8084
Surface Atmospheric Revised June 1974
SP-8085
The Planet Mercury
SP-8091
The Planet Saturn (1970),
June 1972
SP-8092
Assessment June 1972
of Spacecraft
Radiation,
111
(1974),
Revised December
(1972),
(Near Earth
and Extraterrestrial,
Atmosphere
1974
Revised September
Model-1969
Ionosphere,
and
Revised May 1971
1972
to Lunar Surface),
March 1969
Magnetic
1975 March 1973
Fields, September
(Interplanetary
1970
and Planetary),
March 1971
(1970),
Extremes
(1971),
Control
(Launch
and
Transportation
Areas),
March 1972
Electromagnetic
Interference,
SP-8103
ThePlanets Uranus,Neptune,andPluto(1971),November 1972
SP-8105
Spacecraft Thermal
SP-8111
Assessment
and Control
SP-8116
The Earth's
Trapped
SP-8117
Gravity Fields of the Solar System,
SP-8118
Interplanetary
SP-8122
The Environment
Control,
May 1973
of Electrostatic
Radiation
Charges, May 1974
Belts, March 1975 April 1975
Charged Particle Models (1974), of Titan (1975),
March 1975
July 1976
STRUCTURES SP-8001
Buffeting
SP-8002
Flight-Loads
SP-8003
Flutter,
SP-8004
Panel Flutter,
Revised June 1972
SP-8006
Local Steady
Aerodynamic
SP-8007
Buckling of Thin-Walled
SP-8008
Prelaunch
Ground Wind Loads, November
SP-8009
Propellant
Slosh Loads, August
SP-8012
Natural Vibration
SP-8014
Entry Thermal Protection,
SP-8019
Buckling of Thin-Walled
Truncated
SP-8022
Staging Loads, February
1969
SP-8029
Aerodynamic May 1969
SP-8030
Transient
SP-8031
Slosh Suppression,
During Atmospheric Measurements
Ascent, Revised November During
Buzz, and Divergence,
Launch
and Exit, December
1964
July 1964
Loads During
Launch
Circular Cylinders,
and Exit, May 1965
Revised August 1968 1965
1968
Modal Analysis,
September
1968
August 1968
and Rocket-Exhaust
Cones, September
Heating
Loads From Thrust Excitation,
112
1970
May 1969
1968
During Launch
February
1969
and Ascent,
SP-8032
Bucklingof Thin-Walled DoublyCurvedShells,August1969
SP-8035
WindLoads
SP-8040
Fracture
SP-8042
Meteoroid
SP-8043
Design-Development
SP-8044
Qualification
SP-8045
Acceptance
SP-8046
Landing 1970
SP-8050
Structural
SP-8053
Nuclear and Space Radiation
SP-8054
Space Radiation
SP-8055
Prevention 1970
SP-8056
Flight Separation
SP-8057
Structural 1972
SP-8060
Compartment
SP-8061
Interaction
SP-8062
Entry Gasdynamic
SP-8063
Lubrication,
SP-8066
Deployable
SP-8068
Buckling Strength
SP-8072
Acoustic
SP-8077
Transportation
During Ascent, June 1970
Control
of Metallic Pressure Vessels, May 1970
Damage Assessment,
May 1970
Testing, May 1970
Testing, May 1970 Testing, April 1970
Impact
Attenuation
Vibration
for Non-Surface-Planing
Prediction,
Protection,
of Coupled
October
Design Criteria Applicable
Venting, November with Umbilicals
Instability
(Pogo), October
1970 to a Space Shuttle,
Revised March
1970
and Launch
Heating,
January
Stand,
August 1970
1971
and Wear, June 1971 Deceleration
of Structural
Loads Generated
Systems,
June 1971
Plates, June 1971
by the Propulsion
and Handling
113
June 1970
June 1970
Structure-Propulsion
Aerodynamic
April
June 1970
Effects on Materials,
Mechanisms,
Friction,
Ganders,
Loads, September
System, 1971
June 1971
SP-8079 SP-8082
StructuralInteractionwith ControlSystems, November 1971 Stress-Corrosion Crackingin Metals,August1971
SP-8083
DiscontinuityStresses in MetallicPressure Vessels, November 1971
SP-8095
PreliminaryCriteria for the Fracture Control of SpaceShuttle Structures, June1971
SP-8099
Combining AscentLoads,May1972
SP-8104
Structural InteractionWith Transportationand HandlingSystems, January1973
SP-8108
Advanced Composite Structures, December 1974
GUIDANCEANDCONTROL 8P-8015
Guidance andNavigation for EntryVehicles, November1968
SP-8016
Effectsof StructuralFlexibilityon Spacecraft ControlSystems, April 1969
SP-8018
Spacecraft Magnetic Torques,March1969
SP-8024
Spacecraft Gravitational Torques,May1969
SP-8026
Spacecraft StarTrackers, July 1970
SP-8027
Spacecraft RadiationTorques,October1969
SP-8028
EntryVehicleControl,November1969
SP-8033
Spacecraft EarthHorizonSensors, December 1969
SP-8034
Spacecraft MassExpulsion Torques,
SP-8036
Effects
of Structural
February
Flexibility
December
on Launch
1969 Vehicle
Control
Systems,
1970
SP-8047
Spacecraft
Sun Sensors, June 1970
SP-8058
Spacecraft
Aerodynamic
SP-8059
Spacecraft 1971
Attitude
114
Torques, Control
January
During
1971
Thrusting
Maneuvers,
February
SP-8065
TubularSpacecraft Booms(Extendible,ReelStored),February1971
SP-8070
Spaceborne DigitalComputerSystems, March1971
SP-8071
Passive Gravity-Gradient LibrationDampers, February1971
SP-8074
Spacecraft SolarCellArrays,May1971
SP-8078
Spaceborne ElectronicImagingSystems, June1971
SP-8086
Space VehicleDisplaysDesignCriteria,March1972
SP-8096
Space VehicleGyroscope Sensor Applications, October1972
SP-8098
Effectsof StructuralFlexibility on Entry VehicleControlSystems, June1972
SP-8102
Space VehicleAccelerometer Applications, December 1972
CHEMICAL PROPULSION SP-8089
LiquidRocketEngineInjectors,March1976
SP-8087
LiquidRocketEngineFluid-Cooled Combustion Chambers, April 1972
SP-8124
Liquid RocketEngineSelf-Cooled CombustionChambers, September 1977
SP-8113
Liquid RocketEngineCombustionStabilizationDevices,November 1974
SP-8120
LiquidRocketEngineNozzles, July 1976
SP-8107
Turbopump Systems for Liquid Rocket
SP-8109
Liquid
SP-8052
Liquid Rocket
Engine Turbopump
SP-8110
Liquid Rocket
Engine Turbines,
SP-8081
Liquid Propellant
SP-8048
Liquid Rocket Engine Turbopump
Bearings, March 1971
SP-8121
Liquid Rocket
Rotating-Shaft
Rocket
Engine Centrifugal
Gas Generators,
Engine Turbopump
115
Engines, August 1974
Flow Turbopumps, Inducers,
January
December
1973
May 1971
1974
March 1972
Seals, February
1978
SP-8101
Liquid 1972
Rocket
SP-8100
Liquid
Rocket
Engine
SP-8088
Liquid
Rocket
Metal
SP-8094
Liquid
Rocket
Valve Components,
SP-8097
Liquid
Rocket
Valve Assemblies,
SP-8090
Liquid
Rocket
Actuators
SP-8119
Liquid
Rocket
Disconnects,
September SP-8123
Liquid
SP-8112
Pressurization
SP-8080
Liquid Disks,
Engine
Turbopump
Shafts
Turbopump Tanks
Gears,
and Tank
and
March
Couplings,
1974
Components,
August
1974
1973
May
Couplings,
May
1973
November
and Operators,
September
1973
Fittings,
Fixed
Joints,
and Seals,
1976
Rocket
Rocket
Lines,
Bellows,
Systems
Flexible
for Liquid
Pressure
and Explosive
Rockets,
Regulators,
Valves,
Selection
Relief
March
SP-8064
Solid
Propellant
SP-8075
Solid 1971
Propellant
SP-8076
Solid
Propellant
Grain
Design
SP-8073
Solid
Propellant
Grain
SP-8039
Solid
Rocket
SP-8051
Solid
SP-8025
Hoses,
and
October
1975
Valves,
Check
Factors
June
in Rocket
Motor
Design,
Ballistics,
March
Structural
Integrity
Analysis,
June
Motor
Performance
Analysis
and Prediction,
Rocket
Motor
Igniters,
Solid
Rocket
Motor
Metal
SP-8093
Solid
Rocket
Motor
Internal
Insulation,
SP-8115
Solid
Rocket
Motor
Nozzles,
June
SP-8114
Solid
Rocket
Thrust
Vector
Control,
SP-8041
Captive-Fired
Testing
116
March
of Solid
Valves,
1977
Burst
1971
and Internal
Cases,
April
1973
and Characterization,
Processing
Filters,
1971
April
1970 December
1976
1975
Rocket
December Motors,
1974 March
1971
October
1972 1973 May
1971
Z
_-.I
r-I-I
m
0 Z ),
r-n r._ ..,,., G..)
"o :Z:,, c-J
_3 "_1 m
_-c'-3 I_-
I'1"1
;]_:,, r._ ;:Z:,,
c-_ r-"0 =0
I-
),
mmm
rn
m
m
mmmm
=0 0 Z
t--|
mmmm m
c -I m
0 (n ), z Imm
Im
(n -0 _..dl
o m
mm V_dl
0
i: m
Z m
(n
Z _m r.S'J
=D -I
m Imm
"U
atom
0 Z
mm_ mb
,,,..j
r.J.I
FOREWORD
NASA
experience
Accordingly,
has
criteria
indicated
a need
are being
for uniform
developed
in the
criteria
following
for the
areas
design
of space
vehicles.
of technology:
Environment Structures
Individual are
components
completed.
of this
This
Guidance
and
Chemical
Propulsion
work
document,
will be issued
part
of
monograph. A list of all monographs of this document. These
monographs
except
as may
these
be
design
revised
monograph,
direction
of
project
"Liquid Howard
of the
was
Lewis
International of
participated
this
in interviews,
King
of Rocketdyne
Lewis
Research
Comments
concerning
National
and
W.
the and
Ohio
technical Space
44135.
1978
*Currently **Currently
with
Aerojet
Liquid
with Westinghouse
Rocket
Co.
Electric
Corp.
and
and
The J.
as NASA
as they
is one
such
final
pages
on the
requirements,
It is expected,
however,
eventually
that
will provide
and
monograph
was
of Rocketdyne of
review
of the
Corporation;
of
NASA
written
Aerojet the text.
under
Research
To
D.
assure
Rocket technical
community
In particular,
D. M. Sandercock
Headquarters
D.
Rockwell
Liquid
technical
the
Center; by
Division,
Jr. of Lewis.
throughout
critical
Wilcox
Lewis
B. Keller,
engineers
was prepared
Office,
Farquhar**
International
W.
not
to be desirable,
Criteria
by Russell
Rockwell
the monograph
Aeronautics Cleveland,
edited
Propulsion,
Turbopumps",
Design
Viteri
consultations,
Center;
reviewed
F.
and
specifications. indicate
M. C. Huppert*
scientists
Division,
collectively
April
was
document,
to design
W. Schmidt.
Center;
Chemical
as soon
to this one can be found
Axial-Flow
Chief,
Harold
on
monographs
vehicles.
Engine
and
Themonograph
accuracy
Office),
by
Research
Corporation;
Company.
project
space
Rocket
series
prior
may
W. Douglass,
management
Scheer
in formal
for NASA
as separate
as guides
as experience
practices
the
issued
to be regarded
specified
documents,
uniform This
are
Control
individually
Austin of the and
in detail. content Administration,
of this
monograph
Lewis
Research
will be welcomed Center
(Design
by the Criteria
GUIDE
The
purpose
of this
significant
monograph
experience
programs firm
TO THE USE OF THIS MONOGRAPH
to date.
guidance
product,
is to organize knowledge
It reviews
and
achieving
greater
for
and
and
greater that
are
preceded
The
the
Art,
section
of
current
by
2,
present,
for effective
in
design
consistency
in the
major sections references. State
assesses
efficiency
and
accumulated
practices,
in design,
design
effort.
a brief
and
discusses
in design,
and from
them
establishes in the
is organized
and
complemented
the
total
into by
design
the
operational
reliability
monograph
introduction
reviews
and
increased
The
use
development
end two
a set
problem,
of
and
identifies which design elements are involved in successful design. It describes succinctly the current technology pertaining to these elements. When detailed information is required, the best available references are cited. This section serves as a survey of the subject that provides background
material
Recommended
design. manager
The
have
been
displays
can be followed design
loosely
criteria
in section
provided.
base
effectively
for
the Design
Criteria
and
The
as a checklist
hove when
Recommended to
the
and briefly wha__._trule, design element to
guide, assure
of rules
for the
its adequacy.
3, state
is described;
guidance
organized
should
this
to satisfy this
each
cannot
Practices, practicing
body
decimally
numbered
subsections
correspond
from
continuity
of subject
in such
through
manual. of existing
be judged
into
both
monograph
or a design
organized
its merit
serve
or in assessing
procedure
positive
numbered
specifications,
to the
are
can
a design also
best
provide
similarly
Contents
design
technological
of the
be done
criteria. concisely,
in conjunction
designer
on
with
how
the
to
achieve
the
subjects
design.
sections
within
Criteria
Practices, the
references
Criteria,
successful Both
Design
to use in guiding
possible,
appropriate
the
The
Recommended
Whenever Design
a proper
Criteria, shown in italics in section 3, state clearly or standard must be imposed on each essential
successful
The
prepares
Practices.
The Design limitation, project
and
on how
sections is not
as a discrete intended
It is a summary successful
to and
design
effectively
iii
so that
to section.
a way
that
The
a particular
format
for
aspect
of
subject. be
a
design
a systematic techniques
it makes
designer.
subsections section
that
handbook,
ordering and
material
a
set
of
of the large
and
practices.
Its value
and
available
to and
useful
CONTENTS
.
INTRODUCTION
2.
STATE OF THE ART
3.
DESIGN
CRITERIA
APPENDIX
A - Conversion
APPENDIX
B - Glossary
REFERENCES
Page 1
............................. ........................... and Recommended of U.S. Customary
3 Practices
68
.................
Units to SI Units
93
.............
95
.............................
105
.................................
NASA Space Vehicle Design Criteria Monographs SUBJECT
Issued to Date
111
..............
STATE OF THE ART
DESIGN CRITERIA
2.1
3
3.1
68
Turbopump Speed
2.1.1
8
3.1.1
68
Turbopump
2.1.2
9
3.1.2
68
OVERALL
TURBOPUMP
DESIGN
Rotor Dynamics
2.2
10
3.2
68
Realm of Operation
2.2.1
10
3.2.1
68
Stage Hydrodynamic Design Blade Loading, Stall Margin, and Efficiency Velocity Diagrams Blade Angles
2.2.2
11
3.2.2
69
2.2.2.1 2.2.2.2 2.2.2.3 2.2.2.4
16 24 25 26
3.2.2.1 3.2.2.2 3.2.2.3 3.2.2.4
69 70 71 71
2.2.2.5 2.2.2.6 2.2.2. 7
26 31 34
3.2.2.5 3.2.2.6 3.2.2. 7
71 72 72
2.3
35
3.3
73
2.3.1 2.3.1.1 2.3.1.2
35 35 39
3.3.1 3.3.1.1 3.3.1.2
73 73 74
STAGE DESIGN
Solidity Cavitation Off-Design Clearances PUMP ROTOR
Performance
ASSEMBLY
Blades Profile Types Mechanical Design
STATE OF THE ART
DESIGN
2.3.1.3
43
3.3.1.3
80
2.3.2 2.3.2.1 2.3.2.2
43 43 45
3.3.2 3.3.2.1 3.3.2.2
81 81 81
2.3.3
45
_3.3
82
2.3.3.1 2.3.3.2
45 47
3.3.3.1 3.3.3.2
82 82
2.3.4 2.3.4.1 2.3.4.2 2.3.4.3
47 47 50 53
3.3.4 3.3.4.1 3.3.4.2 3.3.4.3
82 82 83 84
2.4
53
3.4
85
2.4.1 2.4.1.1 2.4.1.2
54 54 54
3.4.1 3.4.1.1 3.4.1.2
85 85 85
2.4.1.3
54
3.4.1.3
85
Vane Attachment Methods
2.4.2 2.4.2.1
54 54
3.4.2 3.4.2.1
85 85
Mechanical
2.4.2.2
55
3.4.2.2
86
2.4.3 2.4.3.1 2.4.3.2
57 57 58
3.4.3 3.4.3.1 3.4.3.2
87 87 87
2.4.3.3
59
3.4.3.3
87
2.4.4 2.4.4.1 2.4.4.2
60 60 60
3.4.4 3.4.4.1 3.4.4.2
88 88 88
2.4.5 2.4.5.1 2.4.5.2
61 61 63
3.4.5 3.4.5.1 3.4.5.2
89 89 90
2.5
63
3.5
91
3.5.1
91
SUBJECT
CRITERIA
Profile Tolerances, Surface Finish, and Fillet Radii Blade Attachment Methods Mechanical Design Rotor Configuration Mechanical Design Axial Thrust Balance System Types of Systems Mechanical Design System Stability PUMP STATOR ASSEMBLY Vanes Profile Types Mechanical Design Profile Tolerances, Surface Finish, and Fillet Radii
Design
Stator and Volute Housings Housing Types Hydrodynamic Design Mechanical Design Bearing Housings Types Mechanical
Design
Housing Interfaces and Static Sealing Interface and Seal Types Mechanical Design MATERIALS Property
Data
vi
SUBJECT
STATE OF THE ART
DESIGN CRITERIA
Ductility
3.5.2
91
ImpactStrength
3.5.3
92
Endurance Limit
3.5.4
92
SAFETYFACTORS
2.6
vii
64
LIST OF FIGURES
Title
Figure 1
Head coefficient
2
Cross-sedtional
3
Cutaway
4
Approximate
5
Typical
6
Nomenclature
7
Relationship of stage ideal head coefficient, efficiency, diffusion factor, and flow coefficient ....................
21
Variation incidence
23
8
9 10
11
and efficiency
Page
vs flow coefficient
view of Mark 15-F turbopump
view of M-1 fuel turbopump
for axial-flow
pumps
.......
.................
6
..........
..........
7
Ns - Ds diagram for single-disk pumps and low-pressure-ratio
flow model for an axial pump stage for axial pump blading
of total-pressure-loss for NACA-65-series
compressors
..................
....................
and efficiency
Comparison of theoretical and experimental cavitation for double-circular-arc profiles .......................
14
..............
27
breakdown
parameter 29
Correlation of pump cavitation parameter with ideal head coefficient, flow coefficient, stage efficiency, and diffusion factor ..............
12
Potential
13
Three types of rotor assemblies
14
Typical profile nomenclature
15
Comparison
16
Typical modified
17
Typical Campbell diagram for identifying
18
Modification
30
stall points of Mark 15-F pump during start of the J-2 engine
Goodman
.......
......................
distributions
of Mark 15-F first-stage
37
for three profiles
diagram for blade stress blade resonant
blade to eliminate
°..
Vlll
33 36
.......................
of basic thickness
12 13
parameter with diffusion factor at reference and double-circular-arc blades ...........
Effect of solidity on head coefficient
5
...........
38
.............. conditions resonance
40 ........ .........
41 42
Figure
Title
Page
19
Cam!?bell diagram for Mark 15-F first-stage
20
Design details for lq-1 dovetail
21
Fabrication
22
Thrust-balance
23
Typical
24
Thrust-balance
system for the M-1 fuel pump
25
Thrust-bearing
assembly on M-1 fuel pump
26
Rotor-stator
27
Stator
28
Rotor-stator
29
Volute types showing various degrees of foldover
30
Types of static seals used in axialpumps
31
Modified
32
Effect of base load on blade natural
33
Effects of fluid virtual mass on Mark 15-F vane natural
blade
attachment
of M-1 fuel pump rotor
................
..................
of a series-flow
assembly
segments
46 ................
thrust balance system
for Mark 15-F pump
Goodman
for M-1 pump
49 ............
48
.................
51
..................
52
..................
55
and rotor disks for Mark 9 pump
assembly
44
.....................
system used on Mark 15-F pump
performance
42
...............
56
....................
diagram illustrating
56 ................
58
................... safety factors
frequency
ix
62 ..............
76
and damping (M-1 dovetail) frequency
.........
.......
78 78
LIST OF TABLES
Title
Table I II
Chief Design Features
of Axial-Flow
Stage Design Data for Axial-Flow
III
Design Parameters
IV
Materials
Liquid-Hydrogen
Pumps
Page Pumps
4
...................
17
for Blade and Vane Profiles in Axial-Flow
Used for Major Components
...........
on Axial-Flow
X
Pumps
Pumps ...........
........
18 65
LIQUID ' ROCKET AXIAL-FLOW
ENGINE
TURBOPUMPS
1. INTRODUCTION An
axial-flow
pump
the
periphery,
that
fixed
blades
tips
and
(stator
by
fluid;
surrounds Both
axial-
and
By far,
with
and
greater
based
on
the
the small
fluid.
In
with
the
experience
important
to note,
that however,
herein,
axial-flow
pumps.
centrifugal
even Some pumps.
(primarily
volutes)
the
criteria
of the In most
are discussed
saw
of the
of
is
each
set
rotor
velocity
90 ° to the
diffuser
that
acquired
treated
instances,
on centrifugal
as such
reflects
of axial-flow
made
to incorporate
subsequent
technology in this
monograph
generally because
pumps
to the
may
(ref.
applicable the 1).
subject
the
J-2 is
based
hydrogen
extensive
design
on
technology
liquid
particularly
blading,
In five
This monograph
pump
involved
axial-flow
of development
service
V vehicle.
in the
areas herein,
levels
for
use of
of staging.
operational
Saturn
The
of the
has been that
is well suited
engines.
ease
various
technology,
evolved
areas
which
rocket
relative
extensive
of the
propellant-feed
consideration
all of which
no attempt that
to
through
and
this
of
type,
and
Apollo
design
some
led
weight,
five pumps
briefly
headrise
roughly
by the
in rocket-engine
of current
carried
15-F)
stages
technology though
monograph
and
Mark
been
that
fluid
spiral
(headrise)
blade
of a bladed
pumped
a high-speed
centrifugal
logically lower
mechanical
had
and
of
operating
traverses
consists
the
all
shaft, fluid
on
sets
at the
under
pump
pump
utilized
of the
of configurations, the
clearances
as the
increase
requirements
designed
these
been
made
development
and
axial-flow-compressor
discussed
design
been
S-IVB
number
design
applicable
and
(the
and
the
sweeps
imparts
have
however,
pump
small
blades
contains
fluid.
efficiency, were
experience
hydrodynamic
any
has
pumps S-II
action
pressure/flow
higher
one
on
a relatively
pumped
the
with
impeller
that
maintained
produced
to pressure
the
as a propellant,
axial-flow used
This
pumps
use
and
production;
engine
shaft.
blades;
A centrifugal-flow of the
is converted
and collects
its potential
instances,
vanes.
airfoil-shaped
or housing
are
parallel
of pressure
rotation
centrifugal-flow
hydrogen
type
stator
rotor
vanes
nearly
carrying
a casing
the
stator
flows
the
impeller
impeller
propellants
liquid
and
of this velocity the
systems.
on
the
and
of increases
so that
from much
fluid
blades
shaped
outward
dense
pumped
or a cylinder,
within
between
blades
summation
of rotor
(impeller)
speed
positioned
rotor
The
produced
of a set of disks at high
vanes)
between
conditions. (stage)
consists rotates
as the
in
the
area
of
use
was made
of
compressors.
It is
in this document design
be useful
are applicable to both matter
of the
in the
pumps
design
to both
of axial
types
of pumps
is covered
fully
in
The axial-pump designprocessis directed toward achievinga hydrodynamic andmechanical design configuration that will meet the requirements of the engine system within the constraintsimposedby other componentsin the turbopump assembly.Thus, the axial-pump design cannot be divorced from the design of components such as the turbine, inducer, bearings, and seals.In instanceswhere components or systems influence the axial-pump design, appropriate reference is made to other monographs in the Chemical Propulsion series. The hydrodynamic design of an axial pump involves basically (1) the selection of appropriate fluid-velocity diagrams and (2) the design of blading that will achieve fluid turning per the diagramswith the predicted loss. Major problems in hydrodynamic design include (1) failure to achieve pump headrise because of improper fluid turning or low efficiency, and (2) failure to maintain adequatestall margin during transient or steady-state pump operation. Structural adequacy of the axial pump is achievedby keeping operating stresseswithin material limits and by maintaining adequateclearancesbetweenrotating and stationary components. Major problems in structural (mechanical) designinclude blade and vane fatigue failures, excessive housing and rotor deflections, and failure of the thrust-balancesystemto keep rotor axial loadswithin manageablemagnitudes. The monograph beginswith a section that provides a brief background on the axial-pump applications and views the pump in terms of the total turbopump assembly.The remaining sectionstreat the pump designin the order in which a pump designerwould proceed.These sectionsdeal with stagehydrodynamic design,pump rotor assembly,pump stator assembly, pump materials for liquid-hydrogen applications, and safety factors as utilized in state-of-the-art pumps. In each of these areas, the monograph establishesthe basis for successfulaxial-flow pump design.
2
2. STATE OF THE ART
2.1
OVERALL
The
use
of
applications relatively
axial-flow
are
and carried through of these pumps and in figure
abrupt
included on
adding
the.
stages
M-l),
operating short.
integral
with
the
which
on
The
the
*Factors Terms,
been
limited
the
front a
This
drives
for
the
first
9. The highly
head
on
(NPSH)
for converting symbols,
axial
U.S.
materials,
M-l,
of the
and
26,
the
blading
pumps and customary
units
and abbreviations
life the
(size).
requirement
relative
simplicity
requirement
i.e.,
has
pump and
been
in
a significant
pumps
on the Mark to achieve
increased
to operate pressure System
or identified
assembly
of Units in Appendix
the inducer
that
includes
turbine.
overhung
from
on separate
t 5-F was identical
was utilized
sufficient
with
to be developed.
version
designed
in this
turbopumps of the other
a direct-drive mounted
of the
to
(SI Units) B.
the
rear
With
Mark
the
utilized 15-F
in
per stage.
values
prevent
a The
bearings.
to that
basic
headrise
at low
been
a service
discussed
stage
were
25 were
have
sec, with
pumps
an inducer
turbopumps
Mark
axial-flow
to the International
pumps
was 500
an uprated
are defined
the
packaging
15-F and M-1 liquid-hydrogen In general, configurations
26
9 and
blading
develop
and
for
designed the axial
a design
of
axial-flow
assembly,
been
curves pump
have been favoring
axial-flow
was
have to
because
an axial-flow
engine
in turn,
Chief design characteristics
axial
was
for example,
shown;
Mark
rocket
stage,
Mark
loaded
two
housing,
for the Mark
an
and
uprating
favored
fixed-speed that were
characteristic
of
weight,
thrust
short
rear-bearing-housing
15-F,
the
of the
for state-of-the-art
components
bearing
over a wide configurations
condition.
life
to the
liquid-hydrogen
and production. I*. Head-vs-flow
choice
efficiency, was
sec.
slope
The
engine
uprated
10 000
to
a centrifugal pump could pressure. Considerations
in
service
similar
of an additional Mark
inducers
suction
steep
for the M-1 pump, of
are
Mark
9 was
more
incisive; discharge
configuration
and
of
the points.
where
duration
system,
whereas
exception on
are
stall**
necessary
duration
turbopumps
Mark
has
and headrise were required. Additionally, the of axial pumps has restricted their use to
and cutaway views of the Mark in figures 2 and 3, respectively.
thrust-balance
The
axial
development
Cross-sectional are presented
bearing,
engines
advantages
the
overhauls) the
axial-flow
the
Design
(between
factor in monograph.
turbines
the
applications
to achieve
relatively
stator
apparent at
potential in those
Required life
in head
noted in table I was not the flow at the required
Additionally, (e.g.,
rocket
various levels of development their use are tabulated in table
1 ; readily
drop
applications to deliver pump
in
that did not require significant throttling or operation As noted, the state of the art is reflected in five
shown
the
pumps
DESIGN
in which high volumetric flowrate narrow operating-range capability
applications flow range. designed features
TURBOPUMP
of net
cavitation
are given in Appendix
positive in the A.
Table
I. -
Chief
Design
Features
of Axial-Flow
Liquid-Hydrogen
Pumps
Delivered Pump
Speed,
gpm
Headrise,* ft
10 230
51 500
32 800
flow,
Mark 9
Application
Number of stages
rpm Inducer plus six main stages
Phoebus (Development)
Mark 15-F
9 062
28 266
40300
Inducer plus seven main stages
J-2 engine (Operational)
.Ca
18 500
Mark 25
Mark 26
9 000
34 000
62 000
40 000
24 000
Tandem inducer
plus four main
Phoebus
stages
(Development)
Inducer plus seven main stages
J-2 engine (Experimental)
62 300
M-1
13 225
56 500
Inducer plus transition eight main stages
Overall
headrise
-
inducer
inlet to volute
discharge.
plus
M-1 engine (Development)
m nmSTALL LINE _HEAD/FLOW CHARACTERISTIC
2.8
*OVERALL HEAD COEFFICIENT INDUCER INLET TO VOLUTE DISCHARGE
2.6
26
2.4 z w
-
c,.)
2.2
o
J
2.0
.s
r_
I
"r
1.8
K 15-F
MARK 9
1.6
1.4
o_
80-
>_z _J
60-
la_
L
I • 04
I •06
I .O8
I .10
I .04
I .06
I .08
I .10
I
I .06
I .08
I .10
I .12
I .06
I .08
for axial-flow
pumps.
INDUCER INLET FLOW COEFFICIENT Figure
1.-
Head coefficient
and efficiency
vs flow
coefficient
I .10
I .08
I .10
.12
Ol FFUSER PUMP VOLUTE
_
TURBINE
PUMP STATOR
ROTORS
FLANGE REAR BEARING INLET
BOLT
FLUI O FLOW c_
INE NOZZLE | NE STATOR
FRONT IIF,,ARI NG
PUHP ItOTOit-
INOUCER STATOR AND FRONT DF..ARI NG HOU51NG
ilIRUSTDAL,ANCE PISTON
TUItB I NE INLET HANIFOLD REAR BEARING NOUS I NG
I NSULATI ON
Figure
2. -
Cross-sectional
view
of
Mark
15-F
turbopump.
REAR
BEARING
HOUSING EXHAUST
HOUSING
PUMP OUTLET
FRONT HOUS
BEARING
INDUCER HOUSING-_ i: TURBINE -.a
SECOND-STAGE
STATOR NE FIRST-STAGE TURBINE
NOZZLE
INE INLET MANIFOLD
BALANCE PUMP VOLUTE PUMP STATOR
CER
PUMP ROTOR
Figure
3. -
Cutaway
view
of
M-1 fuel
turbopump.
PISTON
ROTOR
ROTOR
following
blade
cooled
by
limits
the
for
rows
(ref.
liquid
2).
Both
hydrogen;
relatively
roller
bearing
short-life
and
ball bearings
DN values
pumps
(ref.
3).
Labyrinth
employed to control internal flow. Face-riding turbine end of pump to prevent propellant conditions
and
operated
control
successfully
2.1.1
minimize
at surface
applications,
initial
design
as possible
with
the
total DN
(ref.
vehicle
are given
phase
design
3),
4).
the
rubbing
of speed
parameters
pump
usual
hydrodynamic These
speed
(ref.
4),
and
at high turbine
include
turbine
stress
(ref.
of speed
seals
speed
and
speed (ref.
pump
in establishing specific
turbopump speed
N_ -
critical
turbopump
and
Ss.
speed These
are pump
parameters
rotational are
related
speed by
NQ1/_
(1)
H_
NQ _ as
where
Q = volume H = headrise,
speed, flowrate,
rpm
(2)
-
(NPSH)
N = rotational
2),
7.
suction
speed,
as
associated
expressions
Ns = specific
to
During
cavitation 5),
been
in order
the design
on the
been at the static
have
constraints
inducer
been
have
efficiencies.
is to set
mechanical
the effects
seals
face-riding
and
practice
and
constraints
restrictions
and have
state-of-the-art
shaft-riding
is operated
turbopump, the
(ref.
turbopump
of a new
seal
and
ft/sec
increased
in reference
Ns,
the
to achieve
with
utilized
seals have been utilized the turbine area during
operation;
up to 400
assembly.
Details
during
the
been
and
and lift-off leakage into
and
turbopump
6).
speed
weight
consistent
(ref.
speeds
specific
speeds
rocket-engine
high
Important
leakage
Speed
turbopump
bearing
the
Turbopump
In normal the
to
have
of 2 x 106 are considered
- (gpm) v2/ft 3/_
rpm gpm
ft
8
3_
N, the
Ss= suction specificspeed,rpm
The
NPSH
= net
design
speed
suction
speed
speed
forcing
functions.
design
constraints
change
in blade the
Mark usually
strict
of the
Mark
sequential
dynamic
for a new
and
stability
aspects
machines
are
of the
shaft
designed
and
so
discussed
the
which
have
Axial
loads
blades
limits of the
blading
26 were
not
and
type
of
rotors
has
An
received
extensive
have
both
to the study
systems
applicable
to
here,
In the
Mark
speeds
operating 9, 15-F, were
centrifugal
however,
of the and
to discuss
axial
briefly
the
with rotor dynamics in the state-of-the-art been used in the design of the rotor/bearing
In the M-1 turbopump, maximum
both
in rocket
discussion
the
speed 25, and
above
and disadvantages
shaft-and-bearing
was below
the
26 turbopumps,
system
criticals
of each
of these
system
first critical the
but
was speed
systems
below
design
(operation
high
radial reacted
below
in order load
and
the
system
to achieve
a high
stiffness
by a triple
first first
characteristics,
set of ball bearings
critical critical
speed), speed.
the
were rotor
approaches
were encased
used
high Thus,
are
bearing
speed
for the
Limited
testing
housing
so-called was
flexibility
M-1 Mod
performed
1 (initial with
9
was this
performed test
(ref.
configuration)
turbopump,
rotor/bearing roller
to support
in a radially
flexible
rear rpm.
akin
6.
were
critical
in a
previously
attention and
rates
16 000
resulted
were
done to ensure that radial shaft support would be at the intended Extensive analysis along with experimental effort to determine
system
by
speed-selection
considerable bibliography
6. It is appropriate
advantages
is desired
(turbine)
in
established
It was noted
this being locations. and
limited
speed-dependent
were could
sequential, the
on a design
were
with
and
to resonance.
precluded
based
26 pumps
design.
rotor
operating
The
approach
stiffness
Mark
programs.
turbopump
in reference
M-1
system
and
cavitation
Mark
pump
proximity
and problems associated Two approaches have
the
mode.
redesign
the
in reference
its supports.
that
first-flexural
since
and
turbopump
turbopumps.
the
of the
in effect
turbopump
presented
so that
15-F
and
operating-speed
of turbopump
for axial-flow
designed
15-F
Rotor Dynamics
overall design concepts axial-flow configurations. systems
sense,
development of
was inducer
Mark
resonance were
relation
behavior
design
ft
The
avoid these
frequency
Turbopump
engine
In
to that
in the
v2
on the M-1 turbopump
order
Note
conducted
head,
Ss of 43 000.
natural
This
2.1.2 The
in
designs
9.
suction
constraint
specific
operating
that
positive
- (gpm) ft3/,
because
bearings, the
shaft.
housing,
roller-bearing bearing spring
8). The
predicted
turbopump the
was
M-1 engine
program was terminated. However, during one of the tests an inadvertent overspeedto 15 500 rpm occurred.At this speedlevel, the bearingandsupport-strut accelerometertraces indicated that the critical speed was being approached, thus lending credibility to the analytical model, which is describedin reference8. The axial turbopumps that have been designedto operate above the rotor/bearing system critical speedshavehad ball bearingsat the support locations, a configuration that achievesa relatively low radial spring rate and low systemcritical speed.A rotor operating abovethe first critical speedmay developnonsynchronouswhirl (ref. 6); this phenomenonoccurred in varying degreesof severity on the Mark 9, 15-F, 25, and 26 turbopumps (refs. 9, 10, and 11). Nonsynchronous whirl and axial oscillations that occurred with the Mark 9 were examined extensively both analytically and experimentally (ref. 10). In the Mark 15-F, nonsynchronous whiff was identified as a major source of alternating stressin the turbine disks and disk-to-pump shaft coupling (the turbine disks resist plane-of-rotation changes associated with the whirl); in addition, rotor radial displacement as high as 0.030 in. peak-to-peakwas measured.These problems were solved by increasingthe axial preload on the ball bearings, which increased the threshold of shaft stability and suppressedshaft deflections to tolerable magnitudes.Severenonsynchronous whirl alsowas observedduring development testing of the Mark 25 pump (ref. 11). In this pump, singleball bearingswere used at the rotor support locations; a design changeto duplex ball bearingseliminated the severewhirl problem in the pump operating-speedrange. In general,the rotor dynamics problems that have occurred in axial-flow pumpshavebeen difficult to diagnose and solve. Suitable analytical models for the prediction of nonsynchronous whirl were not available at the time the pumps treated herein were designed.In the designprocess,an attempt wasmade to avoid whiff by considering those factors then known to be related to whirl problems and by designingso that the operating speedwas not near a critical speed.However, asindicated in the precedingparagraph,these measureswere not always adequate. Effort to solve nonsynchronous-whirl problems in centrifugal turbopumps was exerted subsequentto the design of the axial-flow pumps discussedin this monograph. This effort has provided improved analytical methods that permit a more thorough treatment of rotor instability during the designphase(refs. 12 and 13).
2.2
STAGE
2.2.1 The
DESIGN
Realm of Operation
considerations
discussed the specific
involved
in reference speed
range
7. The
in the
selection
discussion
in which
axial
of the
herein stages
have
10
type
of pump
is limited been
essentially
used.
for a given
applici_tion
to an identification
are of
Figure 4 (adptd. from ref. 14) showsdesign-point performance for various types of pumps in terms of specific speedNs and specific diameter Ds (D_ = DHla/QY2, where D =.rotor diameter, ft). This kind of figure is useful in the pump selectionprocess,in that it identifies specific areas where different types of pumps are suitable and givesan estimate of pump efficiency and diameter. As shown oll the figure, axial-flow-pump stageshavespecific speedsrangingfrom approximately 3200 to 11 000. It should be noted that these are stage characteristics; considerabledifference occurs when the entire pump is examined. For example, the figure indicates that for the M-1 mainstagespecific speedof 4470 an axial flow configuration is suitable; however, the specific speed of the entire M-1 pump is approximately 900, and examination of this region on the N_-Dsdiagram indicates that a centrifugal pump could havebeenselected,somedecreasein efficiency being anticipated.
2.2.2 The
Stage Hydrodynamic
procedures
design that
used
practices exist
that
stations
in the
hydrodynamic
for axial-flow
compressors
in the
assumed
pump
the
for
figure
6. In the condition
flow
are
(i.e.,
Weight and start
and
usually as noted
density are
being
pump
(or stage)
of the
propellant
pumping
rise
used
weight
flowrate
process.
as the
(e.g.,
in
and
are dictated
by the
with
and
pressure
beginning
pressure the
In reference
angles
engine
thrust
rise
end
engine flow
7, a method
from
is described
11
in
average in the
by
applying
propellant
at
the
and
inlet
pressure the
operating
is
the
stage
turbopump
range
and
of the
and
in which
flowrate
density
In and
changes
requirement
of
state
or an increment the
desired
pump. propellant
thermodynamic
process
exit).
requirements,
to volume
headrise
first
total
straightforward, heating
the
weight
of the
inducer
requirements
propellant
pumping
for
at the
are converted
pumps), of the
the
radial-equilibrium
include
with
conversion
When
is shown
model.
associated
rise is determined and
blading
flow
variations
conditions
required
radial
A typical
represent
accounted
and
off-design
factor.
the
it is
at discrete
and
stages
fluid
constraints
propellants, conversion
flow
velocities,
designer
are
axial-flow
high-pressure-hydrogen
for a given at the
the
for
followed conditions
in which
plane.
diagrams
addition,
to the
Additionally, the
plane
streamline-to-streamline)
location
speed,
by
provide incompressible
significant
of
temperature,
previously.
analysis, For
designer
have flow
model
(hub-to-tip)
velocity
energy
efficiences
turbopump
pressure
(i.e.,
streamline
is prescribed
characteristic pump
the
rise,
pressure,
flowrate
headrise.
to
pressure fluid
speed
assembly, the
given
and
Radial
or estimated
flow
5 ; nomenclature 5), the
continuity,
generally
(blade-to-blade)
in the meridional (fig.
streamline.
pumps
three-dimensional
a two-dimensional
in figure
each
of axial
18). The
circumferential
illustration
at
loss data
by
the flow
from
Losses
Specifications
in the
is illustrated
model
at a given
experimental
stage
flow
determined
considerations.
flowrate
stage
flow
design (ref.
approximated
to describe
a pump
flow
are
average
can be used
model
Design
required
the
points of the
headrise
is
N
= NQI/2H -3/4 S
_=
.30
-I12HI14
Ds = DQ 1.0
..... _._ .40 " TIP CLEARANCE _,_¢.", .40 RATIO 2 X 10-4 _ .*"- T..._ - _ _-... ",,._",.. 50 T/=.20
. 30
H -
DISCHARGE MINUS HEAD
.."--_-'_-._.._.. \\r'- 77=. 60 .40 _---. __.._-..'..'_,"
STATIC
INLET
HEAD
TOTAL
-N.-N--N-PITOT
]._.u .... "C "_'_ _ " .""''-"'... _--_,,_ "- "...."''".30._
"_L'-''-- _oRAGRy DISPLACEMENT ..... PARTIAL EMISSION --CENTRIFUGAL OR AXIAL
.._ _ -. :. _.- ....,r-_ .5o"_ v \ __
0.1
__I
I
J
;
/
_O_otoL,
_
_x'-
_,_\\_".6o___
.---_ _._'._\'-__o__-.____
.io
ba
TiP
,, 8o ,'--.
CLEARANCE
\
\ _
_
",
" _'\X\_._
"_O
_
___
P, AT,O _x ,o-3, -,., \\_,,,--:---.--.)_ ___ .80
¢)
M I O M-I <>MARK OMARK QMARK _"A" "B" Q"C"
O.O
i
I
NSITION STAGE TRA MAIN STAGE 9 AND 15-F 25 AND ALSO "E" STAGE 26 AND ALSO "D" STAGE STAGE STAGE STAGE
l f I lO2
\
Ik NASA rH/r •
--\\_'_C_. -_'_'-------"_ -\x _-¢_\\_
T = 0.4
(REF. !.5 )
NASA rH/r T = 0.7
(REF. 16 )
41,NASA
I
\
AXIAL_
I
rH/r T = 0.8
I I
I
_
--'_
-_/)
(REF. 17)
I
I
i I
i
i
104
103
I I I 105
N S
Figure 4. -
Approximate
N s -- D s diagram
for single-disk
pumps and low-pressure-ratio
compressors
(adptd.
from
ref. 14).
ROTOR
MERIDIONALI, STREAMLINES
STATOR TIP
_--
HUB (A) VIEW
N HERIDIONAL
I
PLANE
I
I
I
I
w _2
/Y AXIAL
/
\
TANGENT CAMBER
ul
CL
/
--
"_'\
)
U
Vlu I
(B) VIEW IN CIRCUMFERENTIAL AT A GIVEN STREAMLINE
Figure
5. -
Typical
flow
model
13
PLANE
for an axial pump stage.
TO MEAN LINE
MEAN
CAMBER
CHORD
L INE
!' \/ _
1¢/2 .
AXIAL
O[ STAGGER _
CAMBER
ANGLE ANGLE
_
DEVIATION
Wl
FLUID
ANGLE
INLET
(RELATIVE)
VELOCITY
(RELATIVE) _12 FLUID FLUID i
INLET ANGLE ANGLE EXIT
INCIDENCE
ANGLE
w2 C FLUID CHORD S
EXIT VELDCITY LENGTH
TANGENTIAL
SPACING
Figure 6. - Nomenclature for axial pump blading.
determined
from
point
the
at
the
summation
end
of
of incremental
each
pumping
isentropic-enthalpy
increase
by
then
be determined
and
in sizing
With
inputs
parametric evolve
of volume study
an overall
requirements for flowrate
used
the
flowrate,
usually pump
efficiency. local
normally
densities areas
at
will satisfy
enthalpy.
determined
the both
by at the
in the
speed,
conditions that
studies
Propellant
turbopump
is conducted,
in isentropic is
flow-passage
headrise,
configuration
(ref. 19). The and headrise:
increases
increment
and
off-design
50%
streamline
using
state
state
the
points
can
pump. requirements, being
hydrodynamic
are conducted
The
dividing
the
and
a
used
to
mechanical
dimensionless
forms
Va
= -6
gcH ¢'-
(3)
r/n (V2u
U2
-
-- Vlu)
U
(4)
where _b = flow g a =
fluid
coefficient absolute
U -- blade
tangential
qJ -- head
coefficient
gc = gravitational
H = headrise,
riri = stage
velocity
(axial),
velocity,
conversion
ft/sec
ft/sec
factor,
32.17
lbm-ft* -lbf-sec 2
ft-lbf* -lbm
hydraulic
efficiency
(sec.
2.2.2.1.2)
V2u = tangential
component
of absolute
velocity
at rotor
exit,
Vxu -- tangential
component
of absolute
velocity
at rotor
inlet,
ft/sec ft/sec
lbm-ft *The
use of gc in units
make
equation
of 32.17
(4) and similar
lbf_sec2
equations
instead
of g in units
dimensionally
correct
15
of ft/sec 2 and the use of H in units under
all environments.
of
ft -lb f -Ibm
instead
of ft
Typically the procedure involves a preliminary selection of blade and vane profile type (sections 2.3.1.1 and 2.4.1.1), velocity diagram type (section 2.2.2.2), hydrodynamic loading (section 2.2.2.1.1), and solidity (section 2.2.2.4), all of which areinterrelated with flow coeffcient, head coefficient, and efficiency. Blade tip speed(which is related to rotor diameter) and flow coefficient (which is related to blade tip speedand hub/tip ratio) are then studied over a range of values to determine stagehead coefficients and the number of stagesnecessaryto produce the required headrise.This procedureallows a number of pump configurations with different diameters and lengths to be examined in terms of stress, weight, and turbopump critical speed. Changesin the preliminary selections of profile type, velocity diagram type, etc., may be necessaryin the processof arriving at a suitable configuration. When the configuration has been established, velocity diagrams, hydrodynamic loading, and efficiency at each streamline in the meridional plane are determined, and the blade and vane angles(section 2.2.2.3) are selected to produce the desired diagrams. Off-design performance is then estimated, iterations againbeing madeif necessaryto compromiseproperly the design-point andoperating-rangerequirements. Basic design data for the state-of-the-art stages are given in tables II and III. In the multistage pump configurations, the stageshavebeen essentiallyidentical within the given configuration. The M-1 pump had a lightly loaded "transition" stagebehind the inducer to provide a uniform head distribution to the first mainstageand to provide better cavitation characteristics than were possiblewith the more heavily loaded mainstage.Additionally, the M-1 pump had a linearly decreasing outside diameter between the third-stage stator dischargeand the fifth-stage rotor dischargeto account for hydrogen compressibility. Stages for the Mark 9, 15-F, 25, and 26 pumps were identical within each configuration with the exception that none of the last stagesin thesepumpsutilized a stator. The flow path on the Mark 25 was adjustedlinearly from the first rotor inlet to the last rotor dischargeto account for hydrogen compressibility. Adjustments for hydrogen compressibility were not incorporated in the designof the Mark 9, 15-F,and 26 pumps. 2.2.2.1
BLADE
2.2.2.1.1 Blade High
Blade Loading
loading.
increase
-
in the
blade
in order
LOADING,
The
to
obtain
and Stall
energy
tangential
loading,
STALL
velocity stage
AND
EFFICIENCY
Margin
added
i.e., large high
MARGIN,
to the
fluid by the
of the
fluid
changes
between
in tangential
headrise
and
pump the
velocity
therefore
rotor rotor
(V2u-
a small
blade inlet Vtu
number
an increase in blade loading could turning in the direction of rotation,
be achieved However,
would
relative
(w2),
the
fluid
velocity
at the
16
rotor
exit
and
the
of the
rotor
exit.
in fig. 5), are desirable
observed in figure 5 that that would increase fluid decrease
is a function
of stages.
thereby
It can
be
by a blade shape this configuration increasing
flow
Table
II. -
Stage
Design
Data
for
Axial-Flow
Pumps
Diffusion
Factor
Rotor _T
_T
Ns
Ds
Stator
Hub
Tip
Hub
Tip
0.037
0.392
0.333
rH
rT
6.80
8.00
0.850
0.396
0.126
7640
0.0400
0.916
0.352
6.80
8.00
0.850
0.420
0.258
4470
0.0478
0.894
.448
.598
.477
.477
Mark 9
2.99
3.61
0.829
0.294
0.226
4450
0.0511
.52
.41
.57
.54
Mark 15-F
Same
as Mark 9
Mark 25
Same as "E"
Mark 26
Same as "D"
"A"
stage
3.13
3.63
0.861
0.390
0.279
4000
0.0511
0.87
.58
.48
.58
.56
"B" stage
3.13
3.63
0.861
0.390
0.316
3650
0.0533
0.87
.64
.54
.64
.62
"C"
stage
3.13
3.63
0.861
0.390
0.338
3460
0.0538
0.86
.68
.57
.68
.66
"D"
stage
3.13
3.63
0.861
0.390
0.35
3380
0.0542
0.84
.72
.61
.72
.70
"E"
stage
3.13
3.63
0.861
0.465
0.35
3220
0.0500
0.85
.53
.47
.55
.50
v = 0.4
1.78
4.53
0.393
0.284
0.135
10650
0.0272
0.800
.593
.223
.577
.373
NASA v = 0.7
3.15
4.50
0.700
0.294
0.282
4760
0.0429
0.937
.693
.426
NA
NA
3.60
4.50
0.800
0.466
0.391
385O
0.0436
0.955
.631
.664
NA
NA
Stage
M-1 transition M-1 main
NASA
stage
stage
(rotor
only)
NASA
v = 0.8
(rotor
only)
NA = notapplicable
Table III.-
Design Parameters
Chord, Pump
Profile
for Blade and Vane Profiles in Axial-Flow
Maximum thicknessto-chord ratio
in.
Camber, deg
Stagger,
deg
Profile type Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Hub
Mean
Tip
Blade Vane
C-4 C-4
1.430 0.991
1.315 0.991
1.200 0.991
O. 120 0.08
0.095 0.08
0.070 0.08
1.037 1.19
0.872 1.08
0.745 0.99
19.30 14.60
13.85 17.50
3.40 43.60
47.30
53.10
59.20
33.50
32.72
39.90
M-I mainstage
Blade Vane
C-4 C-4
1.302 1.057
1.176 0.968
1.070 0.882
0.120 0.t5
0.097 0.15
0.075 0.15
1.43 1.42
1.13 1.18
0.915 1.00
27.86 26.04
21.86 29.57
18.50 37.56
36.39 36.83
45.02 35.27
51.17 34.12
Mark 9
Blade Vane
1.37 0.87
........ ........
1.36 0.98
0.09 0.117
........ .......
0.05 0.065
1.21 1.9
........ ........
1.05 1.8
18.5 40.0
........ ........
12.0 40.0
61.9 38.0
........ ........
67.9 35.3
Mark
Blade
M-1 transition stage
15-F
Nonstandard Nonstandard
Vane
Oo
Solidity
Pumps
.......
Same as Mark 9 ...............................................
Mark 25
Blade Vane
......
Same as "E" --
Mark 26
Blade Vane
......
Same as "D"--
"A" stage
Blade Vane
Modified double-circular-arc
0.923 1.00
........ ........
0.892 0.924
0.140 0.120
........ ........
0.051 0.050
1.37 1.4
........ ........
1.14 1.3 l
38.7 36.9
........ ........
21.36 34.8
40.35 37.38
........ ........
54.39 40.35
"B"
stage
Blade Vane
0.912 0.99
........ ........
0.872 0.860
0.138 0.121
........ ........
0.052 0.054
1.34 1.31
........ ........
1.11 1.3
43.1 39.4
........ ........
24.9 39.4
39.05 35.12
........ ........
53.75 38.99
"C"
stage
Blade
0.899 0.980
........ ........
0.852 0.850
0.14 0.122
........ ........
0.054 0.054
1.33 1.30
........ ........
1.09 1.29
46.0 43.0
........ ........
27.1 43.1
38.0 34.1
........ ........
53.15 37.85
Vane Blade Vane
Modified double-circular-arc
0.909 0.975
....... .......
0.836 0.874
0.132
........
0.062
1.33
........
1.06
50.0
........
28.9
36.5
........
52.22
0.123
........
0.058
1.33
........
1.29
47.8
........
46.1
34.1
........
36.85
"E" stage
Blade Vane
Nonstandard Nonstandard
1.667 1.538
....... .......
1.268 1.394
0.104 0.085
........ ........
0.081 0.089
1.9 1.7
........ ........
1.28 1.61
45.9 35.1
........ ........
29.1 36
30.0 41.1
........ ........
41.0 38.9
NASA
v = 0.4
Blade Vane
Double-circular-arc
1.50 1.50
1.50 1.50
1.50 1.50
0.10 0.08
0.085 0.08
0.07 0.08
2.5 2.37
1.44 1.36
1.0 0.95
61.35 62.20
13.84 51.40
5.43 44.15
19.97 20.34
........ ........
70.19 9.61
NASA
v = 0.7
1.52
1.52
1.52
0.085
0.0775
0.070
1.44
1.19
1.01
27.6
19.8
0
52.2
60.5
67.1
1.49
1.49
1.49
0.09
0.08
0.07
1.25
1.11
1.00
43.4
27.0
39.8
46.2
55.0
"D" stage
Blade (no stator)
NASA
v = 0.8
Blade (no stator)
42.7
diffusion large
in the
amounts
in total
blade
row.
Large
of diffusion
headrise.
tend
In reference
velocity
gradients
to produce 20,
thick
a blade-loading
on
the
blade
or
angles
and
was used for correlating loading, i.e., the loading axial-flow head
and
In the over
pump
work,
loading
limits.
notation
the
stage
solidity.
experimental at which the the
of figure length,
blade-row
diffusion
boundary
layers
and
based
on diffusion
5, diffusion
are given
In axial-compressor
total-pressure-loss data boundary layer separates factor
has been
factors
for
associated
parameter
row was developed for axial-flow compressors. The parameter, the index of local diffusion on the blade suction surface in terms velocities
surfaces
used
the
similarly
rotor
and
relatively
diffusion of fluid work,
and from
the
high losses in the blade
factor DF, is an inlet and outlet diffusion
factor
for indicating limiting the suction surface. In
in estimating
stator,
with
with
loss in total
constant
radius
by the expressions
(DF) R = 1 -
w2 Awu -- + -W 1
(5)
2OWl
where (DF)R
= diffusion
factor
Wl
= fluid
inlet
W2
= fluid
exit
Aw u = change ft/sec
for rotor
relative
velocity,
relative
velocity,
in tangential
cr = solidity
ft/sec ft/sec
component
of relative
velocity
--- wl sin/31R - w2 sin/32R,
= C/S (fig. 6), dimensionless
and
V3 (DF) s = 1 -
V--_+ 2aV----_
where (DF)s
= diffusion
VE = fluid
inlet
V3 = fluid
exit
factor absolute absolute
for stator velocity, velocity,
AV_
ft/sec ft/sec
19
(6)
AVu
= change
in tangential
e = solidity
values
given
II.
Values
Stall
is the
table
of absolute
velocity
= V2u -
V3u, ft/sec
as above
Diffusion-factor in
component
for hub
and
tip
for the
generally
fall
state-of-the-art
within
the
axial-flow-pump
range
blading
appropriate
to
are
axial-flow
compressors. Stall
margin.-
stage
when
headrise in
the
flow
drops
with
separation abruptly.
state-of-the-art
blading (table factor reached the
as the
of
in the Three
axial
pumping rotor
or stator
different pumps.
capability
factor,
ratio
the
with
been the
used
Mark
in an
axial-flow-pump
to the
point
to define 9 and
"A"
where
the
is an index
relative
of blade-passage
the
stall point
through
when the rotor-hub or stator-tip factor (RF) dropped to a value
factor
of fluid outlet-to-inlet
occurs
progresses
have
Experience
retardation
that
passage
conditions
II) indicated that stall occurred a value of 0.75 or the retardation
diffusion
defined
loss
"E"
diffusion of 0.5. As
diffusion.
It is
velocity:
W2
(RF)R
-
(rotor)
(7)
(stator)
(8)
Wl
V3
(RF)s
In the
M-1
which
the
equivalent
design,
stall
equivalent diffusion
prediction
diffusion factor
was ratio
blade suction surface ratio of 0.10, the
and stator
constant
radius
COS/32R (DFeq)R
-
based
_
on
or factor
is approximately
velocity on the thickness-to-chord with
- V2
the
DFeq
equal
reported
is used
to the
to the fluid outlet equivalent diffusion
ratio
an
an indicator
of the
relative factors,
in reference
fluid
21, in
of stall.
maximum
The
relative
velocity. For a blade with at minimum loss, for rotor
are
0.61
.12 + -_
[Vla]
Wl sin/31R
(9)
[Wl ]
cosmos{06, (DVoq s-
method
72O
[V uV3u]
IV2] [v2a
(10)
In the M-l, stall was consideredto occur when either DFeq had a value of 2 (ref. 19). Figure 7 showsthe relation of ideal head coefficient, diffusion factor, efficiency, and flow coefficient for an axial pump stagewith a reaction of 0.5 (sec.2.2.2.2) and a solidity of 1.5. The curves are based on analysis at the 50% streamline (pitchline) and, as noted on the figure, the efficiencies do not include tip clearanceloss or other secondary flow losses. Examination of the figure indicates that a design point could be selected to obtain maximum stage efficiency. However, in the state-of-the-art pumps, the design point selection has been made consistentwith maintaining a stall margin. The trend has been to designfor higher flow coefficients to take advantageof the increasedideal head coefficient at a given diffusion factor. However, for a given flow, pump diameter, and speed,a limit is reached, since as the flow coefficient is increased the blade heights become small and the tip leakagelossesbecome increasinglysignificant. Note also from the figure that increased stall margin (decreaseddiffusion factor) for a given flow coefficient, reaction, and solidity can be achieved
only
at the
expense
of ideal
head
coefficient. 7/ .80
1.4
1.2
EFFICIENCIES ARE BASED ON LOSS DATA SHOWN IN FIGURE 8 AND DO NOT INCLUDE TIP CLEARANCE OR OTHER SECONDARY FLOW LOSSES. REACTION R - 0.5 SOLIDITY _1.5 OF - DIFFUSION FACTOR - STAGE EFFICIENCY
.84
.86
1.0 .88
b-
0.8
.9O
¢9
w O ¢9
.e DE
.gT 0.6 .915
OF 0.8
0.4 .91 .90 .88
0.2
.86
FLOW COEFFICIENT
Figure 7.-
Relationship of stage ideal head coefficient, diffusion factor, and flow coefficient.
21
efficiency,
2.2.2.1.2 Stage
Efficiency
hydraulic
efficiency
is defined
as the
ratio
H
Hi
of actual
headrise
H to the
ideal
headrise
Hi:
r/H
where
Hloss
is the
Stage
head
loss
losses,
and
can
flow
have
state-of-the-art estimated, M-1
broken
efficiencies
methods
the
of the head
be
secondary
Design-point Two
sum
the
pump,
reference reproduced
here
losses
in the into
to
Mark
by
pump
were
based
on
as figure
8 (adptd.
the
losses, boundary
25,
and
and
in table
hydraulic
26 pumps,
hub-to-tip
end-wall layers
are given
losses
from the
profile stages
predict
9, 15-F,
(11)
stage.
blade
loss distribution
loss) incidence
Hloss
Hi
produced
used
In the
radial
losses
losses
down
been
(minimum
Hi
for the axial-flow
pumps. and
-
-
was
friction and
for axial-flow
from
18). The
ref.
efficiency
an average assumed
compressors.
total-pressure-loss
clearance.
II. in
efficiency
was In
correlation
at
The
correlation
parameter
COS/_exit
(12a)
2(/
using
the
conventions
= total-pressure
of figures
(head)
5 and
6,
- loss coefficient
Hloss (rotor) w_/2g
(12b)
c
Hloss
(stator) Vz2/2gc
_exit
=
fluid
angle
(12c)
at the rotor
or stator
exit
22
(/3z_ or /33s in figure
5), deg
is
is given
by
where,
the
to be constant.
diffusion-factor/total-pressure-loss
developed
tip
(annulus)
PERCENT
OF
ROTOR
BLADE HEIGHT T P TO HUB
FROM
.14
r_ p-
SHADED .12
AREA
IS REGION
OVER
WHICH
TOTAL-PRESSURE-LOSS PARAMETERS FELL FOR THE IO-PERCENT STREAMLINE
lO
.I0 •08
I r_
oq U
_13 MJ D_ I J
O
•06 , 50, •04
90
O. 02
AND
70, ALL
STREAMLINES
I 0. I
.2
.3 DIFFUSION
I .4
1
I
.5
.6
FACTOR
Figure 8. -- Variation of total-pressure-loss parameter with diffusion factor at reference incidence for NACA-65-series and double-circular-arc blades (adptd. from ref. 18).
.7
In the
design
methods
of
given
determine based
in
the
tip,
than
lower
incidence
portion
of the
for the
streamlines
reference
16
that
parameter
for
axial-flow
correlating
near
minimum-loss
18. The
given
total-pressure-loss 16 indicated and in view
blading,
loss parameters
results
especially
M-1
in reference
the on
the
those
shaded
at and
determined
angles
near
area
the
indicated
were
pump
in figure
rotor
that
by the
8 was used
tip. This
the
rotors
for axial-flow
determined practice
magnitudes generally
compressor
to was
of
were
rotors.
the
lower,
Reference
that a specific explanation for the lower magnitudes was not readily apparent, of limited number of rotors that were tested, no generalization of the results
was attempted. It is important in the
to note
design
the that
correlation likely
care
be
discussed
2.2.2.2 any
the
selection
stage
the
margin. has
also
blade
varies location
parametrically diagram (ref.
and
28).
It
correlation. to
the
22, which,
the
of ref.
the
Effort
the
rotors
total-pressure-loss
in which
flow
to
design
of
are reported
static
headrise
stall
extend the
in 16),
parameter
outlet
angles
are
range
of
the
axial-flow
in references
in the
pumps
23 through
that
maximum
the
axial
26.
would
radius;
(i.e., that
was
concluded in using
the
velocity
fluid
and in
of
stage
by of the
sets In
been since
the
extent
stage. has
desirable,
the
is equal
then
the
stall margin for
ratio
to a great
reaction
is therefore stage
efficiency
of the
selection
stator
attainable
R = 0.5)
profile
This
and
selection
a maximum
been used is constant
a symmetrical pump
design
if a forced-vortex offer
benefit
stage).
established
as the
the stall
equal
can be obtained. occurs
to one-half
the
with
a
velocity
of
27).
selected
determine
in
is in effect
is defined
rotor
magnitudes
factor
diagram
diagram
R (reaction in the
margin
implies
with A
of velocity
reaction
flow pattern has fluid axial velocity
only.
type
stage
stator
(ref.
inversely to
in the
dominant
when
element
at all radii
appreciable
and the
shown
A free-vortex radial free-vortex flow the velocity
the
indiscriminately
in reference
(including
applying blading
to be accomplished
velocity
rotor
diagram
the rotor
for
to the
efficiency
been
symmetrical
radial
rotor
pumps,
in the
when staggered
of this effort
pump,
value
diffusion
A symmetrical
diffusion
be used
is emphasized rotors
subsequent
Results
8 cannot
DIAGRAMS
in the
of velocity
restriction
used
conducted
in an axial-flow
state-of-the-art
It
was
monograph.
in figure
axial-pump
exercised
range
of a design
headrise
influences
be
given
This three
18 to highly the
VELOCITY
For
of
should
data in this
degree
pumps.
results
outside
experimental
static
test
of reference
to
the correlation
of axial-flow
summarizing cautions
that
lower that
the
loading, for
selected
hub/tip
in all state-of-the-art axial pumps. In the from hub to tip while the fluid tangential diagram with flow
better ratios
forced-vortex
24
therefore
a hub-tip
ratio
pattern
that
efficiency, of flow
can be achieved of
imposed or higher
0.8
and
pattern.
higher,
0.8
was
at one studied
a symmetrical head
coefficient
there
Additionally,
was
no
it was
concluded that with the free-vortex flow pattern, the radial location of the symmetrical diagramwasnot critical aslong asit wasnearthe meanradius. Blockagefactors are included in the determination of velocity diagramsto account for the reduced flow areasresulting from end wall and bladesurface(profile) boundary layers.The magnitudes selectedfor these factors are dependent on the particular designaswell as the designmethod being used. A designmethod employing appropriate experimental loss data (usedon M-1) automatically accountsfor profile boundary-layer blockageand it is necessary only to account additionally for end-wall boundary-layer blockage. In the M-1 pump, this blockage was estimated as 4% of the annulus area.A designmethod in which averagestage efficiencies are estimated must utilize blockage factors that take into account the area reduction due to both end-wall and blade-surfaceboundary layers. In the Mark 9, 15-F, 25, and 26 pumps, these factors were estimated at approximately 10%of the annulus areaon the basis of compressor blade-surfaceand end-wall boundary-layer information obtained from reference 18. Analytical and experimental investigations of stages with impulse blading have been conducted (refs. 29 and 30), but to date this type of blading hasnot been utilized in rocket enginepumps.
2.2.2.3 With
BLADE the
ANGLES
velocity
diagrams
established
in the
meridional
plane,
the
blade
angles
and
blade
shape are selected to turn the flow in accordance with the desired diagrams. This selection involves determination of the incidence angle, camber angle, and deviation angle at each of the
hub-to-tip
and,
for
streamlines
the
pumps
considerations. turning prediction For angle
)
rotor 18, 27,
deviation
angles.
were
determined
basis
of experimental
Mark good
15-F selected
26),
the
design
the
6).
angle
the
and
row
M-l,
of the
data a design basis
deviation
would
that
reduce
methods
incidence In the
of achieving angle
the
angles
angle
selected, deviation
stage
stage
established
from
of
work
angle Mark
25
compressors testing
of +3 ° was and
selected
26 pumps,
low loss. In all of these
was determined
from
25
the
loss)
pumps
relationship
Accurate diagrams.
and vane
camber
of deviation
8%. angles
and
and
deviation
in reference in air.
In the
on the
basis
a design
of fluid
6.
desired
blade
by about
designer cavitation
amount
angle
1° in prediction
minimum
for
or
the
with
of incidence
low-Mach-number
by the
the
in achieving
an error
(at
is selected minimum-loss
0 and
for selection
incidence
obtained
angle important
31 give procedures
In the
angle either
of a 50%-reaction
of 1.0 indicated
and
on
the incidence
is extremely
analysis
incidence
based
camber
a solidity stator
The
was with
on
performance. on
6 that,
27,
by use
pumps,
cavitation
was
figure
deviation
of 30 ° and
References
on
in reference
for the
5 and herein,
is dependent
of the
example,
angles
of
Note
(/31 -/32
(figs.
discussed
incidence (Mark
(ref.
prediction angles 18 on the Mark
of achieving angle
9, 15-F, 31)
9 and of 0 °
25, and
0 _6-
(13)
where a = distance
flexit
fluid
=
to the point
exit
angle,
of maximum
camber
from
leading
edge,
in.
deg
flexit
- dimensionless
ratio
.50
Subsequent fluid to
to the
were
conducted
design slotted
2.2.2.4
SOLIDITY
in general,
increases increase the
as the
solidity increasing
desired the
length.
Here,
weight,
and
2.2.2.5
CAVITATION
Rocket
engine
provide high
the
suction
is sufficient
the
turbopumps
capability performance to preclude
cavitation
test
the
flow
skin
data
test
applicable
obtained of
fluid
(high
with
problems
limits
friction
number increased
with
incorporate at low
inlet
the
pressure following
26
the
data.
solidity)
included inlet
angle,
These
values
area).
pressures. of the stage
as possible,
as the
reached,
because coefficient
since for
that
conflict
may
an attempt and
row
a given with
can be made
increasing
length,
losses
the
chord
increased
rotor
rotor.
initial
Inducers pumped (ref.
head
Additionally,
a case,
of blades
turbopump
inducers
hydrodynamic a hydrodynamic
the ideal
are soon
blade
III.
Both
From
a requirement In such
the
in table
coefficient
blades,
standpoint.
in the
as the
multiple-circular-arc,
selection.
blades
and
thin
by reducing
increase
data
range
of available
9). However,
is confronted
and
wide
given
into
(increased
for operation
The
are
as many
(fig.
as a rule
water
of experimental
18.
stages
limits
require
critical-speed
using
double-circular-arc,
a relatively
factor,
a structural
solidity
range
tests
26).
enter
to use
cascade
reference for
over
the
diffusion
stages
designer
potential
within
solidity from
desired
in
axial-pump
is increased
high-solidity
thickness
to achieve
profiles
desirable
of reaction,
with
given
considerations
values
length,
an extended
23 through
selected
herein,
performance
(refs.
design
it is usually
fixed
those and
state-of-the-art
been
mechanical
chord
of
discussed
to provide
as
solidity
(o)
standpoint for
and
solidities
have,
pumps
double-circular-arc
angle,
and
such
characteristics
camber
The
of the
in order
methods
fluid-turning and
design
are designed fluid
2).
pumping
element
to achieve
to a magnitude
In order
to
to achieve
that high
].00 n-
0.90 __
0.6
-JO.80
0.5
0.4
% I-z w
---O.3 w O i_)
_0.2 //
] REACTION R = 0.5
//
0.1
I 0.5
Figure
suction
performance,
therefore
must
pumping
elements
conditions inlet
All the
In
inlet a
flow
The with with
2. Discusssion
9,
design here
initial
more criteria
area
25,
and
were
inlet from
the
inlet loaded
have pumps,
the
necessary. pump
The
inlet
inducer
to the
Additionally, must
and
achieved
inlet
the
be compatible
inducers
headrise, ill the
exit
axial
as the _low
stage of the
exit with
the that
cavitation
for inducers types
initial
area
inducer
A discussion
and mainstage
27
and
stage
better blading.
practices
to transition
the stage
loaded
mainstage
utilized totally
inducer
recommended
will be limited
are
hub-to-tip)
accomplished
a lightly
heavily
and efficiency.
mainstage. from
26
was
mainstage
I 2.5
flow the
mainstage.
between
stage
and
angles
turbopumps
15-F,
stage
for the
at the
higher-flow-coefficient and
I 2.0
on head coefficient
in flow
requirements
transition the
of solidity
coefficients
axial-flow
Mark
conditions
possible
the
of the
"transition"
utilized.
along
the
flow
velocities
state-of-the-art
mainstage
flow
of fluid
Effect
a transition
requirements
element. pump,
low
provide
(i.e.,
flow
9.-
I I 1.0 1.5 SOLIDITY G
pumping
transition, stage.
mainstagc provided
performance or" inducer is presented of axial-flow
and
In the
M-1
inlet
was
acceptable than
that
cavitation ill rc['el-ellCe stages.
Analyses are conducted on the initial axial-flow stagesfollowing the inducer to ensurethat sufficient pressureis availableto prevent headrisedegradationcausedby cavitation. This can be accomplishedby comparisonwith cavitating test results of similar designsor by analysis of fluid velocities on the bladesurface.The cavitation parameter r r NPSH TR
(14)
--
u 2/2go
and cavitation
number
K
Pf
Pv
-
K -
(15)
pfw_/2gc
where NPSH
= net
positive
suction
static
pressure,
lbf/ft 2
Pv = fluid vapor
pressure,
lbf/ft
Pf = fluid
Of = fluid
which
are
density,
commonly
characteristics
lbm/ft
used
of the
references
15 through
in
figure
10
correlation
parameter
represents
the
flat-plate
were
lightly
0.426,
0
highly
32 through
cambered
the
NASA
for
parameter
according
inducer
also
been
to the
little
used
inducers
modified
head
parameter (ref.
magnitudes
to evaluate
the
cavitation
loss data
r r and 35).
at which
The
head
two,dimensional
for these
a cavitation solid
line
breakdown theory
rotors
are
breakdown in the will
figure
occur
of reference
for
35. The
(i.e., head breakdown) data for the NASA v = 0.4 and v with the inducer theory of reference 35. These rotors
or no
As might
v = 0.8 axial
34. Cavitation
cavitation
developed
with
have
on axial-flow blading are relatively limited. Several axial pump blade profiles have been tested and the results are reported in
the high-head-loss compare favorably
loaded
design,
blading.
of
-- 0 °, respectively).
fiat-plate
3
terms Z
inducers
shows that axial rotors
2
17 and
cavitation
figure = 0.7
the
in
ft-lbf/lbm
in inducer
axial-flow
Cavitation performance data rotors with double-circular-arc given
head,
camber
at the
be expected, rotor
(DF
theory.
28
tip (DF data
= 0.664,
from
= 0.223, the more
0
= 5.43°,and heavily
0 = 27 °) did not correlate
loaded well
DF
=
and with
.5
• :Iz
[]
(M
z II
w F,-
. 4
.Z
.3
/
o_
F-
1-2Z
.x9
g
(MOD,FIED TWO-
/
.2
•
DIHENSIONAL THEORY AS GIVEN IN PART 2 OF
/ /
/k/ 0.1 0.02
REFERENCE
.04
.06
BREAKDOWN
.08
CAVITATION
PARAMETER,
10.
-- Comparison for
In view velocities
were
static
pressure
pressure
(_T/2)
_l"
--rH/r T
REF.
17.1 °
0.4
15
• 0% 034%
22.9 °
0.7
16
• 5_ []46%
21.5 °
0.8
17
B T AS USED HERE IS THE BLADE TIP ANGLE MEASURED FROM A PLANE NORMAL TO THE AXIS OF ROTATION.
theoretical
double-circular-arc
and
experimental
cavitation
breakdown
parameter
profiles.
of the lack of cavitation performance data on similar designs, analysis on the blade surface has been used to evaluate the cavitation characteristics
state-of-the-art surface
of
.12
• 3% A40%
NOTE:
Figure
.I0
CORRELATION
Z = _ TAN
HEAD LOSS
35)
blading. on
to ensure
for predicting
In the
determined
Mark
by use
the
blade
that
incipient
the
9, 15-F,
25, and
26 pumps,
of a stream-filament
surface
was then
pumps
cavitation
would
method
determined
be
free
and
from
of airfoil-type
fluid (ref.
compared
cavitation.
blading
was
velocities 18). The
used
of fluid of the
on the blade minimum
with
the
local
fluid
vapor
An approximate
method
in the
method
M-1.
The
utilizes the equivalent diffusion factor DFeq (sec. 2.2.2.1) along with the cavitation parameter TR and the cavitation number K. An approximate value for the ratio of maximum fluid the
velocity equivalent
blade method solidity
on the
blade
diffusion
by use of the cavitation for
blade
= 1.5 are
profiles shown
surface
factor.
to fluid
This number
with in figure
ratio
velocity was then
K and
the
at the related
cavitation
a thickness-to-chord 11. For
a blade
29
blade
inlet
to the
NPSH
parameter ratio
row
with
of
derived
by using
requirement
for the
7R.
0.10,
a given
was
The
results
reaction design
flow
of this
= 0.5,
and
coefficient
1.4 REACTION
(R)
-- 0.5
SOLIDITY (a) NPSH
=
1.5
/ /
"/'R-- "-u2/2gc 1.2
DF
/ /
/
_'R
/ o--
l.O
•88 /
z
/
UJ
/ /
I.¢.
.8 UJ
o
.91 "-r
.6.915 DF
0.8 .4 --
0.2
--
•4 1
,
,
.3
5
"rR= 0"351 0
7
f
0.2
.4
OF
EFFICIENCIES FACTOR
FIGURE END-WALL Figure
11. -
Correlation flow
AT
ARE
8 AND
THE
DO
NOT
l 5
.8
1.0
COEFFICIENT
TOTAL-PRESSURE-LOSS
SION
1.3
I
VALUES FOR CAVITATION PARAMETER ON BLADE WITH THICKNESS-TO-CHORD STAGE
l.l
.6
FLOW NOTES:
.
I
BASED
( _R ) ARE RATIO OF ON
THE
PARAMETER 50%
INCLUDE
CORRELATION
VERSUS
STREAMLINE
BASED O.IO.
AS
DIFFU-
SHOWN
TIP-CLEARANCE
ON OR
LOSSES. of pump
coefficient,
cavitation
stage efficiency,
3O
parameter and
with
diffusion
ideal factor.
head
coefficient,
and
ideal
head
parameter (inducer) 2.2.2.6
coefficient,
rR. stage
required
NPSH
can
With the required NPSH then can be determined.
known,
the
OFF-DESIGN
Analysis
of
predicted
for
engine
the
Reference
19
which
transients,
gives
losses
Mark
and angle
values. Off-design component that the test
are given
be
mixture-ratio
made
the
requirement
so that
excursions,
predictions
was
was
Mark and
the 25
for
36
assumed
and
to
to vary
more
were
blade
the
(2)
be
pump
and
cavitation
of the
previotls
M-1
performance
based
on
the
(1)
a one-dimensional at
to results
can
chamber-pressure
as to how
constant
according
reliable
conditions of predicted
in reference
prediction
predicted
channel
with
losses
loss was assumed resulted from the
off-design flow results. Curves
pump
from
the
excursions.
deviation
the point
angle
and
two-dimensional
mean-line
design
be
analysis*
value
of low-Mach-number
in
and
blade
air tests
at the
at a higher level to account for tip leakage and end-wall boundary-layer of predictions with M-1 test results indicated that the one-dimensional
analysis
15-F
deviation
must
reference
assumed
50% streamline but losses. Comparision mean-line
of
angle were
determined
headrise
methods involve assumptions with incidence angle.
off-design
method
deviation
channel
performance
prediction losses vary
blade-element
be
PERFORMANCE
off-design
In general, blade-element
the
method.
Off-design
a one-dimensional
were
assumed
mean-line
to be constant
to be equal to the velocity variation in incidence angles
(see sketch below). This and measured headrise
method versus
performances analysis at their
of the in which
design
point
head of the normal velocity on the rotors and stators at correlated favorably with flowrate for the Mark 25
9. j_
- /THIS •
_
IS THE NOAMAL VELOCITY C_PO_ENT
'NVt_ i
THE _ REPRESENTS THE
_ ",.
OFF-DESIGN
CONDITION
I\
In general, the off-design performance engine mixture-ratio or chamber-pressure any
problems.
distinct Analysis
However,
operating based
levels
on values
there during
was the
at the 50% streamline
required of the excursion (at a tendency
engine
transients
(fig. 5).
31
for
state-of-the-art design point
the in the
Mark
15-F
J-2 engine
pumps in regard to thrust) has not caused pump start.
to stall
at three
Start problems are not unique to engineswith axial-flow pumps but are related to the interaction of the pump (whether axial or centrifugal) and the thrust chamber.For example, the RL10 engine, which employed centrifugal pumps for both oxidizer and fuel, had start problems similar to those on the J-2. Both the J-2 and the RL10 areregenerativelycooled enginesin which the hydrogen flow from pump dischargeis routed through tubes around the thrust-chamber walls and servesas a coolant before it is injected into the combustion chamber.Start anomalieshavebeen associatedwith reduced fuel flow coincident with rapid increases in thrust-chamber coolant-circuit pressure or in chamber pressure. The consequenceof the reduced fuel flow (which in the axial pump may be a stalled condition) is somewhat dependent on the enginesystem cycle, but in severecasesthe typical result is damageto the thrust chamberas a result of inadequatecooling. Figure 12 illustrates the three potential stall points of the Mark 15-F pump during the J-2 engine start transient; these potential stall points were termed spin-down stall, LOX-dome-primestall, andhigh-speedstall. Spin-down
stall.
discharge) by
and
the
tubes.
prior
pump
and
During
vaporized. zone and
- Spin-down to main was
this
stall
propellant
discharged
interval,
the
undershoot.
stall
initial
ignition.
was
The
fuel-pump
fuel
unrestricted
warmed
by
the
was
flowrate dropped. by an initial
avoided
on
the
acceleration
flowrate
was
thrust-chamber
the
and
accelerated cooling
a portion
was
of the upper combustion The head demand on the
Thus flowrate by
rapidly
chamber,
resistance quickly.
J-2
(start-tank
into thrust
encountered the high pressure drop increased
increased, and was characterized
Spin-down
during
relatively
fuel
The warm fuel then injector, and injector
fuel pump quickly chamber with fuel
occurred
the process overshoot
thermal
of priming the followed by an
preconditioning
of the
chamber. Several
methods
pre-start
purge
development, different a
from
technique
was
the combustion
on
the
cooling
tubes from
chamber stall.when
point
during
On
the
pre-start
purge
and
and
start
fuel J-2X
employed
standpoint
an index
during
dump
cold of the
the
low-level
second
stall
point,
indicated
the
oxidizer
flow
primed
the
dome
degree
some
produced of chill
a
from
preconditioning have
The hydrogen
been
the
is
employed.
high resistance
line
around
the
cooling-tube-bypass as a film coolant
in
operation. as
LOX-dome-prime
manifold
of the
to over 150 psi. This increase in chamber fuel-pump head. Since the speed change
32
to 8 sec.), With
each
a bypass
successfully.
(up
helium.
to avoid
engines,
The
chamber pressure quickly increased an immediate demand for increased
lead
However,
alternatives
of utilizing
and
fuel
Thrust-chamber several
an overboard been
with starts.
so that
uncertain.
J-2S
a long
satisfactory
of view,
the
has the
utilized:
chamber, was
incorporates start.
attractive
12, occurred
and
were
produced
measurement
during
LOX-dome-prime figure
gradient
system
injector
thrust-chamber
nitrogen, methods
an operational
RL 10 engine the
preconditioning
cold
of these
temperature
undesirable of
with each
temperature
single
The
of chamber
stall injector,
on and
pressure caused could not occur
50× 103
FLOW STALL
REQUIREMENT
LINE _/
CONSTANT j___j
AT
O/F
4O
_'_A"_s_° / l\
_o,._
3O w
Q.
/,
a-
2O i,i h
.'/¢1_
"\\ 2s oooRP.
\\\
SP,N_DOWN_J_/_2\_ORIGINAL
lO
S/ALL
_'_.,
2 FUEL Figure
12.-
START
Potential
stall
\
17 000
RPM
1
1
[
1
4
6
8
IO
PUMP points
INLET of
Mark
FLOW, 15-F
33
pump
O3 xl
GPM during
start
of
the
J-2
engine.
instantaneously,
pump
shown
12, the
high
in figure flow
and
Control achieved most
of
fuel-pump
by
regulating
engines,
engine
start.
Initial
J-2 by
two
system
the
pressure
in the
value;
the
in the
if the
main
sufficiently
high
level.
CLEARANCES
As
hub/tip
increasingly
high
ratios
are
clearances
high, to keep
stator-vane running
so
of
Mark
9 pump
tip clearances. tip clearance
effect
changing
higher decreased
than
that
head from
rotor
is not
as the
main
as
speed
line
from
to
stall)
was
prime.
The
dome
valve.
In
flow
during
the
dome
prime
main
start
was
both
oxidizer
the
early
were
J-2 and
portion
relieved
valve
increased
of
on the
was reduced;
by utilizing
the
control valve.
annular
a higher
has been
running
seeks
this O/F
achieved
by
became
thrust
built
up to a
had
area tip
becomes
approximately
was
designed
hub/tip small
on the 0.005
to operate
of
losses
stages
to maintaining
was
an
clearance
tip clearance
clearance
of
stall
the
given radial
M-1 pump
a vane
and
the
(O/F)
was
In state-of-the-art
rotor-blade
clearance
and
to
ratio
high-speed
clearance
area,
work.
nature
the engine
engine
tip
in
mixture
necessary
before
flow
stage
The
in. The
with
similar
so-called
the
attention low.
running
0.015
was
full open
of the
was
buildup,
The
increases,
considerable the
thrust
O/F
total
losses
stall,
for an oxidizer/fuel
oxidizer
pump
in air to determine
showed
the range
amount of was increased
of
of the
of the
12). During
axial
in.
tested
results
within
9 pump
Thus,
0.049
tip
Mark
in.;
the
at a blade
in. (this
was
a
- see fig. 3).
appreciable clearance Mark
at LOX
high-speed
reached
that was
was
The
flow.
nearness
following
oxidizer
oxidizer
is orificed
valve
these
0.020
configuration
point,
percentage
and
clearance
clearance
shrouded The
tip
set
of
pressure
main
opening
paragraph,
of
high
in order was
portion
the
an
percentage
relatively
pump
initial
level (fig.
with
of
an increasingly
the
fuel
a constant
contol
chamber
restrict
this
along
therefore
stall problems
stall
oxidizer
ratio
to
reduced
high pressure.
of the
J-2 engine
previous
flow
2.2.2.7
15-F
third thrust
a problem
become
(1)
The
design-point
a transient
(and
used
of
spin bottle.
The
oxidizer
the
was
during
stall.
and
expense
a throttleable
15-F pump
hydrogen-gas
as noted
flow
magnitude
changes:
restricting
underwent
utilized
speed
at the
coefficient
method
stall.-
achieved
to low
Mark
LOX-dome-prime 5.0 at the
pump
flow
this
pump
High-speed
fuel
technique
RL10
(2)
was
low pressure
effective
and
head
tested
clearance stator.
rotor
tip
the
(0.95%
pump
the
effect
of variations
in rotor
insensitive
to changes
was rather
to 3.25%
of vane
height)
was lost, particularly near the stall 1.58% to 3.57% of the blade height.
symmetrical
in the
that
The
clearance
was
not
unexpected,
and
the
static
air tests was
also
increased.
34
pressure showed (This
since
the
(ref.
the
clearance
stator
in stator
However,
an
point, as the rotor tip The more pronounced velocity
rise in the that
37).
and
stall
diagram
rotor margin
effect
for
the
is substantially of the
on stall
pump
also
was
observedin tests with liquid hydrogen but wasnot systematically investigated.)Attempts at determining the effects of tip clearanceon efficiency during the tests were unsuccessful becauseof the relatively smallrise in air temperature. Axial clearancesbetween blades and vanesin generalhavebeenselectedto minimize overall pump length while maintaining adequate protection against blade and vane axial interference or vibration during pump operation. In the Mark 25 and 26 pumps, the mechanical designof the bladesand vanesfrom a vibration standpoint wasbasedin part on maintaining a specified axial clearance between rotor and stator rows. The analysis for establishingacceptableaxial clearancesconsidersthe forced vibration amplitude of a blade or vane to be a function of the wakevelocity fluctuation of the upstreamrow. The velocity fluctuation in turn is a function of the relation of axial spacingto upstreamcord length and the proximity to resonance of the blade natural frequency with the wakes from the upstreamrow. The analysisis basedon information presentedin references38 through 42.
2.3
PUMP ROTOR
Tile
pump
and
the
rotor
assembly
thrust
(Mark
rotor
15-F
and
clamped consisting
2.3.1
Blades
setting
in section
(i.e.,
Research
and
so-called
series,
and
standard
the
"non-standard" profiles excessive axial
surface
pumps
will
rotor
have
blades,
been
the
Mark blades
rotor
ill the
structure
with
consisting
(Mark 9 and and dovetail
pump
utilized
rotor
assembly
through-bolts rotor structure
be
profiles
double
These arc
C-4) that
achieve
diffusion.
are listed
turn
in table
was were a
Values
design
the
turned
circular
(British to
will
fluid
tile
profiles
of
disks
25), and (M-l).
structure,
design
of the
integral
blading
(with
integral
a builtup
used. either
The
for the
III: typical
remaining
blade significant profile
35
the thickness
modified
prescribed
involves
desired angle
compressors
included
maximum
the
desired
for axial-flow with
procedure
so that
through
of blading
profiles
designed
of the
a one-piece
the hydrodynamic
that
fluid
standard
profile
concepts 13):
a builtup
2.2.2,
profiles
development
the
Three
consists
rotor
TYPES
of blade
achieved on
26),
here
(fig.
together by of a welded
PROFILE
As indicated
system. assemblies
Mark
blading) assembly
2.3.1.1
as discussed
balance
state-of-the-art
ASSEMBLY
with
the
yielded
NACA-65 at
profile
nomenclature
design
loss). data
series,
the
British-C
In the
M-l,
pumps profiles
parameters
a
utilized or
distribution
is illustrated
are
predicted
state-of-the-art velocity
and
diagrams
considerable
midchord.
double-circular-arc surface
the selection
velocity
special
to
avoid
employed in figure
14.
in
A--l
BLAOES .AC.,NEB ,NTE_RALLY W.THROTO. 0.0.----_
_.LANCE /--P,STON
I
THRUST-I_LANCE-__-_=
.
4-.- BEAR ING COOLANT-FLOW
I
RETURN
A-_ (A)
I SECTION
MARK 15-F
BLADES HACHINED INTEGRALLY
WITH
/---
DISK
,__
L/_A_ STUB
SHAFT
SHA T _TIE
BOLT
(B) MARK 9
BLADES, TSI_
IT' ON
DOVETAI LED ---_
BALANCE
_
P' STON
._,,_,_,M_
..
,_1_
e LEAKAGE-FLOW
RETURN
(c) M-i
Figure
13. -- Three
types
36
of rotor
assemblies.
A-A
STAC K I NG I
AXIS
/
ILING LOWER (PRESSURE) SURFACE UPPER
\
(SUCTION) SURFACE
\ CENTER
OF
GRAVITY
\ \
MAXIMUM-MOMENT-OFINERTIA AXIS
/ /
/
LEADING
EDGE SECTION
J_x'_---
M IN IMUM- MOMENT-OFINERTIA
AXIS Figure 14. - Typical profile nomenclature.
3?
A-A
EDGE
Standard
profile
camber
line
(fig.
compressors the
w z
are
thickness
surface
shapes
are
6).
Basic
compared on the
by
a specified
thickness
in figure
distribution
velocity
defined
is that blade
thickness
distributions
15 (adptd. which
for
from
ref.
is required
distribution
some
of
about
the
a mean
profiles
used
18). On the non-standard
to satisfy
the
desired
in
profiles,
distribution
of
profile.
I
65-SERIES .....
C_
BLADE PROFILE I • DOUBLE CIRCULAR ARC
w
I
I
10
ZO
0
1 30
I
I
I
I
40
50
60
70
80 ,
90
100
PERCENT CHORD
Figure
The
15.-
selection
Comparison
of
considerations.
a blade
of basic
thickness
profile
is based
According
to
imposed by distribution
stress considerations, in order to reduce
distribution
would
maintained good
over
diffusion
on
and
the
9, 15-F,
Mark
exceed
the and
times thickness
double-circular-arc to obtain used
for
the
M-1
of the
upper
portion. were
distribution
the
inlet at
blade
with
0.6
for stress
axial-flow
a fiat
(with
the
resulting
the
chord. radius
larger
than
As previously
Research
38
has
had
Mark
26
surface
a thin
been
velocity velocity
velocity
commensurate
method velocity leading
conducted
with used
on
of ref.
18)
did not edge
utilized
the British
is
maximum
profiles
trailing-edge
noted, also
a fiat ideal
distribution,
pump
design
requirements
have the
suction-surface
profiles The
18).
mechanical
non-standard
stream-filament
the maxhnum
The
local
velocity The
fromref.
thickness
as is possible,
well back.
that
reasons). stages.
maximum
surface
achieving
leading-edge
and
maximum
(limited)
designed
of
profiles(adptd,
hydrodynamic
within
(suction)
such
velocity.
about
the In
three
for axial-flow pumps would to a minimum. Further,
are positioned
25 pumps
profile
a stiffer
which
for
both
43,
profiles cavitation
camber
velocity
on
reference
in
rear
maximum
the 1.2
maximum
one
as much
thickness to control
be
distributions
with
a modified
radius
(primarily
C-4 profile on
was
standard
double-circular-arc profiles (refs. 15 through 17, 23, and 32 through 34), on multiple-circular-art and slotted double-circular-arc profiles (refs. 24 and 25) and on variations of profiles similar to thoseusedon the Mark 26 (ref. 44). From a mechanical-designstandpoint, the structural merit of the profile is reflected in the camber angle, the chord length, the thickness distribution, and the maximum thickness-to-chordratio. The camberangleis establishedby the hydrodynamic design.For a given chord and thickness distribution, increasedblade strength is achievedby increasing maximum thickness-to-chord ratio. From the hydrodynamic standpoint, however, it is desirable to maintain low blade thickness. Thus, a mechanically desired maximum thickness-to-chordratio may not be achievable(note in table III that a maximum value of 0.15 was usedin the state-of-the-artblading). Increasedblade strength alsomay be achieved by increasing the chord, maximum thickness-to-chordratio being held constant. However, with a given solidity (sec. 2.2.2.4), this procedure requires a decreasein the number of bladesso that, again,a limit may be reachedfrom the hydrodynamic standpoint.
2.3.1.2
MECHANICAL
DESIGN
The predominant requirement in the mechanical design of axial pump rotor blades the blade withstand the combined steady-state and vibratory stresses for the required the kept
pump.
Basically,
within
vibratory stress stress condition. compromises
blade The
in the
vibratory of
due
to
to
centrifugal
magnitudes
load
Blade
the
are
provide
the
operation diagrams
not
accurately
vibratory
stress
due
is and
is highly
to gyroscopic a
the
recovery
loads
pump
analysis
of
on
and
stacking
the
blade
untwist
line
moment
steady-state
the
to consider only the (fluid force), and
forces
discussed
and
and
design.
dependent
been necessary hydrodynamic
about
to centrifugal
As a result,
in axial-flow
in the blading combined
stress
strength,
hydrodynamic
be estimated.
areas
moment
(to
adequacy,
only
due
utilized
steady-state
or ultimate
stresses
unsteady
is subject
stresses
tilt
the
it usually has steady-state
hydrodynamic
to occur during pump of modified Goodman in general
blade
in which yield
limit corresponding to the steady-state must be met without unacceptable
intractable
shear
has not been
structural
that
at best,
the
and
and
moment)
appraising
predicted the use
normal
magnitudes.
hydrodynamic
assuming
The
can,
blading, centrifugal,
one
on either
steady-state
45);
loads
which
The
(ref.
of the more
state-of-the-art due to the
based
material fatigue considerations design.
to these
stresses
is an iterative limits
predictable
is one
loads.
negligible
In
are
stress
geometry. In the normal stresses vibratory
procedure property
hydrodynamic
loads
of the blading
type
design
material
is kept below the These structural
hydrodynamic response
the
specified
is that life of
have to
forces been
of
counteract
in this monograph. and
vibratory
stresses
are compared with blading material properties by (fig. 16). As indicated above, vibratory stress predictable. was
This
proportional
39
uncertainty to
the
has been hydrodynamic
handled
by stress
(Tcf
= STEADY-STATE STRESS DUE TO CENTRIFUGAL LOAD
mff
= STEADY-STATE DUE TO FLUID
°_s
= COMBINED
STRESS FORCES
STEADY-STATE
STRESS
_alt
= VIBRATORY STRESS - ASSUMED EQUAL TO off k = STRESS CONCENTRATION FACTOR
II.
.a
ALLOWABLE
ALTERNAT
', -N/---
% o'alt
k alt
"
-I
%s MEAN
Figure 16. -
(proportionality blade
root
plotted
factors fillet
on
have
ranged and
Goodman
STRESS
o"m,
PSI
Typical modified Goodman diagram for blade stress.
is estimated
the
ING
from
applied
diagram
0.3
to
1). A stress-concentration
to the vibratory
along
with
the
stress;
calculated
identify
resonant
frequency, for rotor
with blade
designed blade rows
axial-flow
have
axial-flow by
flutter
the stiff
has
been
diagrams
Rotating
no
stall
of the
known
discussed chord
blades checked
margin
magnitude
is then
steady-state
stress.
(plots
of
failure
pump
been used. A typical figure 17. Typically,
on operating
speed
been
instances in this
length with
has
at the
little
observed
axial-pump
blade of
root)
by the empirical
blading
of flutter.
4O
given
not
has
Adequacy in reference
been
low
of the 39.
between
preceding blade of excitation in identified
as an
occurred.
aspect
been
blade
Campbell diagram blades have been
vibration
the
and
vs
was maintained
have
blade
In general,
of the rules
has
that
self-excited
monograph.
likelihood
but
failures
envelope
speed
and known sources of excitation. Wakes from in the flow stream are the predominant sources
for any
been pumps
relatively
a 15-percent
frequencies obstacles
source
There divided
at least
pumps.
excitation
Campbell
forcing function as a variable) have resonant conditions is shown in
so that
natural or other
conditions,
for the
this stress maximum
The design is deemed acceptable if the point falls within the material a blade resonant condition is known not to exist at the design speed.
To
factor
(flutter) ratio and
in
(blade has
blading
the
length
resulted
in
in regard
to
3RD
NATURAL
FREQUENCY
_I
'*zzzJ_Izlz_zltI*_IIzr_tttt_z*_ttt'''zzr'_''r*_z1_/'_-_l_-z'DUE BLADEFREQUENCY TO BAND TOTHISRESONANT BE SPEED FREEcoNDITIONS_.I...oFRANGE /
TOLERANCE
2ND
NATURAL
FREQUENCT,,,
---l
/
f
. ,,,_,_
...............
:_{:_::z _:z_i
_
J
/
RESONANT j
IST
NATURAL
Jill
la_
FREQUENCY
tiill,
//
iii
llllt
J
'_IIIill
•
z:z::__/1
I
_:/,._\_/
CONDITION
J
_2
--D,,.
15_
|5_
MARGIN
-.,,._---MECHANICAL
DESIGN
sP EED ,..._
.,,I-
PUMP
SPEED,
the
effort
blade
aspect
natural
ratio
the
models
compared
reasonably
lower
accurate been
cases,
blade
rather
failures tapering frequency resonant diagram
on
the
the
the
and
all the
natural
Mark
leading
of the condition in figure
15-F edge
were to the
results
frequencies.
of analytically
flutter
analyses.
computer
predicted
(ref.
47).
Other
fatigue
In
pumps has
failures
unpredicted
determining the
solutions natural
approach
Blade in that
that
natural
the
lumped
frequencies
that
have
not
M-l,
of had
been have
blading
sufficiently occurred
frequencies
have were
in
of excitation.
it. For
to failure
problems
has
example,
instances
of
eliminated hub
by cutting
as shown
blade to a magnitude that during pump operation 19. A course
46)
cantilevered-beam
solution
redesign
problem 1, and
(ref.
inadequacy,
sources
expedient to
beams
SPEED
blade resonant conditions.
and
approximately
the
to this
difficult
Campbell-diagram
experimental
ratios,
upstream
than
to the
cantilevered
largely
In most
for was
well with
to predict
attributed with
devoted
blading of
aspect
resonance
been
frequencies
of
parameter with
has
DESIGN
RPM
Figure 17. - Typical Campbell diagram for identifying
Considerable
--NOMINAL
of action
in figure was was
involving
41
back
the
18. This
above the eliminated, modification
been
to modify
first-stage chord change
the
existing
rotor-blade at the
fatigue
tip by ¼ in. and
increased
the
natural
forcing frequency, so that the as shown by the Campbell of existing
parts
rather
than
I
Figure
18. -
(A)
ORIGINAL
BLADE
(B)
MODIFIED
BLADE
Modification
of Mark
BLADE
15-F first-stage
blade to eliminate
resonance.
NATURAL
IFIED
FREQUENCY----,,,.,,_
BLADE
8OOO ORIGINAL BLADE
m
FORCING
6000
FREQUENCY 19 CYCLES/REV--_
Z w D
4000 OPERATING SPEED RANGE 2000
10 OO0 PUMP
Figure
19. -
Campbell
20 SPEED,
diagram
42
000
30 000
RPM
for Mark
15-F first-stage
blade.
redesign
can
be taken
is at a tolerable rotor
blades
the
were
not
modification
pump
when
the
from
tested
performance
the
engine
to determine
described.
the
However,
penalty
system
the
associated
standpoint.
degradation
with
The
in stage
modification
did
FINISH,
AND
not
the modification
Mark
15-F
first-stage
performance
caused
noticeably
affect
by
overall
performance.
2.3.1.3
PROFILE
It is necessary the
profile
has
fairing
in both
fluid
desired
surface
specified the
of the
the
better
63
than
finishes Fillet
as small factor.
Fillet
thickness.
both
of 63 microinches blading. the
been
and (gin.)
In
practice,
tolerances
usually
M-1
blades
to
the
The
strength
of the
specified
methods
as a
necessary
produced
as manufactured
desired
+ ¼°.
has been
the
to
blade
continuous the
within
fatigue
rms
have
and
to achieve
held
the performance
dimensional
order
in order
on the basic
of a smooth In
have
limits
tolerance
requirement
typically
RADII
tolerance
the
directions.
can affect
juncture
of the
consistent
radii
have
ranged
Stress-concentration
2.3.2
the
roughness
example,
strict
Typically,
in. with
angles
fairly
FILLET
to
a surface
typically
had
finish surface
rms.
possible
estimated
on the basis
blade
profile
and
with
maintaining
from
approximately
factors,
(root
a reasonably
applied
of information
its support 30%
to the
such
as that
section) small
to 60%
vibratory given
have
been
kept
stress-concentration
of the
profile
stress
(sec.
in reference
48.
maximum
2.3.1.2),
have
Blade Attachment
2.3.2.1
METHODS
Axial-pump
blading
state-of-the-art attachments
principally
pump
rotor
20). on
machined
carry
centrifugal
required, and
configurations.
The
machined
with
the
The
selection
the
blades
blades.
This
load
it may this
been
cost
of
of the
are
and
However, a lower-cost
benefit
may rotor,
override for
the
example,
43
blades,
if large rotor
weight had
376
drum
which for
had
an axial
will be heavier
to the heavier
attachment.
a rotor
considerations.
attached
is due
or
method
assembly
to produce
M-1 pump
disks
attachment
mechanically
difference
with
M-1 mainstage
a blade
manufacturing,
of the
be possible
integrally
exception
weight,
in which
integrally the
has
pumps (fig.
based
blades,
blade
within For
within
longitudinal
state-of-the-art
gin.
at the as
and
profile
blade
of 32/_in. radii
the
the
profiles performance.
as + 0.002
surface
for
manufacture
been
6),
A maximum
requirement
blade
transverse
(fig.
finish
SURFACE
hydrodynamic
been
angles
blade.
TOLERANCES,
to manufacture
achieve
are
only
magnitude
rotor
benefit mainstage
of
dovetail is
In general,
a
than
one
with
required lots
by using integrally blades
all
pump
structure
manufacturing assembly
on
to
of blades individual machined
in comparison
BEND UP END AFTER
AS_
_
S._A_ _ P,N--I_ /_-
_----------_-'/ \.. "_/
/_
o_SHEAR .'N
ROTOR
ALL IN
BLADE
DIMENSIONS
BLADE
ASSEMBLY
I
INCHES
O. 305
-o. 3o5 F
0.024
O. 2587
R
0.025
0.2748
t 0.025 0.026
R
0.1445
0.2587-1P-
SECTION
THROUGH
SECTION
ROTOR SLOT
Figure 20.-
THROUGH
BLADE
Design details forM-1 dovetail attachment.
44
DOVETAIL
I
with
102
for
the
integral-blade because
design
The
because
MECHANICAL
critical blade
and
strength
attachment
outward
so that
the
situation pins
Stresses beam
the
lock
in the the
kept
2.3.3
an
replacement
in selecting pumps.
centrifugal
load
of the
the
below
could during
operation.
as leaf that
were
calculated
in the
direction
would
however,
were
with
incorrect
by
assuming
parallel
blades a force
assenbly
that
to the
resistance in figure
pin
dovetail
is, the
based
fatigue
on limit
positioning
and
the
blade
radially
in the
same
dovetail
were
met
with
bottom
possible
acted
slot
section). that
limits
on the
was not
the
of the
neck
requirements
to exert
load
material
axial
M-1 at the
airfoil;
property
to position
These
springs
to the
the
the
The
to
analysis
centrifugal
for
desired
be tip ground
Stress
for the
provisions
additionally
20. Note
discharge)
a load,
used
material
include
the
and
structure.
20).
(transposed
as that
specified
also
exist
on
maintained
it was
(designed
loads
rotor
(fig.
was
must
tabs
to the
blade
of the
with
this
as a cantilever
acting
at the
tip
of
to sliding that could occur at the 20 that reverse load (i.e., load in a
be counteracted
by the
bentup
tab. The
source
and
not definable.
CONFIGURATION
Three
rotor-structure
rotor
structure,
flow
made
Rotor
2.3.3.1
pumps
blade
engine
within condition.
would
force
pump
of such
individual
based
manner
stress
M-l,
in figure
blade
toward
probability
of blades
in rocket
loads
was
airfoil
the pin. No allowance was made for frictional blade dovetail and rotor slot interface. Note direction
number
a consideration
method
section)
stress
assembly
attachment
bending
vibratory
was
the
attachment
in the same
vibratory
rotor
as illustrated method.
with
and
In the
that
and
dovetail) attachment
neck
steady-state
blade.
vibratory
(the
configuration
of the
for been
is rare
to carry
blade
stress the
to the
retention contact
and
was achieved
and
corresponding
a large
has not
of damage
designed
dovetail
steady-state
tensile
such
requirement
objects
as the
steady-state
adequacy
maximum
be
dovetail
of the
design
Structural
this type
steady-state
tang
section
The
by foreign
must
airfoil
a single
of machining
DESIGN
attachment the
utilized
cost
expensive.
method,
transmit
shear
very damage
blade
The
9; the
of potential
attachment
2.3.2.2
Mark
(fig. path
machined
13). for
The
M-1
rotor
and
machined
concepts from
A number
thrust ring
was
components
been
a single
of axial
balance
structure
have
system
and
in the
forging,
was
were
machined
holes
a one-piece TIG
utilized
bearing (hollow)
welded
used
coolant
on
the
in the flows
configuration
together
45
state-of-the-art
as shown
pumps. Mark
forging and
15-F
in figure
and
Mark
to provide
to lighten
fabricated
A one-piece
the
from 21. The
26
a return structure.
four Mark
forged 9 and
INCONEL FORGING
718
_-------CONSUHABLE
LLER
INSERT
WELD
(INCONEL
BEFORE
7|8)
(IHCQNEL AFTER
WELDING
ROTOR
718)
WELDING
DRUM STUB
FRONT
STUB
SHAFT
(2
PC)
CIRCUHFERENTIAL. WELD
Figure 21. -
Mark
25
together the
pump with
torque
The
choice
integral
considered. weight
of
configuration,
(fig.
and
design
than
in
concern forging the
welded
cost
from and
construction
rotor
46
shafts
clamped
radial
positioning,
the
turbopump
the
for the
during
such the
weight
forging noted
stub
relative
is made
that (ref.
In In
from and
M-1 was was
manufacturing the
the
M-1
selected
design
pump,
forging
in reducing to eliminate (ref.
with a
(a one-piece
a single
difficulty
sufficient
initial design
standpoint
machined quality
size,
a one-piece
49).
critical-speed
previously
as
selection.
indicated
a rotor
required
and
to attain
influence and
achievable size
disks
tie bolts.
types
desirable
over
used
pump
speed
lowest
the
rotor.
Considerations
rotor
rotor),
of the
in the
builtup
of
with
were
phase. critical
was
a builtup
However, and
or
a comparison result
concept
by shear
design
methods,
of M-1 fuel-pump
13). Rabbets
a one-piece
would
a single
builtup
transmitted
one-piece
is stiffer
the
preliminary 15-F,
blades
lightweight rotor
being
assembly Mark
used bolts
between or
methods, the
through
loads
conceptual of
rotors
Fabrication
(3)
50).
The
was the this two
pumps that have usedthe builtup concept (Mark 9 and Mark 25) weredesignedfor ground application. In both cases,the relative easeof stagingand the capability to test singlestages during development were the primary considerationsin selection.
2.3.3.2
MECHANICAL
Reference
6 presents
various
types
to the
special
Pump
a complete
of rocket
rotors
of
subjected loads
the
designs used
the
pump
less
than
input
have in the
the
superimposed are load
analysis
(transient
the
that
attach
have
was
been
were the
employed, discussion
the
torque
optimized
in
and
been
used
used
along have
above
turbine
since
the
of rotor
Mark
to the
drive
15,F
pump.
for these
couplings
been
rotors
and
Mark
pumps
is presented
was
2.3.4.1
TYPES
Turbopump
Gleason
of the
mounted
has
the torque
power torque shaft
and stress
speed
model
used
25, a ball spline
on separate
the
of the
bearings.
with Mark
tests
analysis
of the inducer. were
and
in defining
determined case of the
evaluations
splines
and
at a mechanical
design
disk
been
estimated
calculated
nominal
from
steady-state
in the
Mark
maximum
unbalanced
attachment and
of all
in the
(ref. of the
Curvic* M-1 to
coupling
was
A complete
6.
OF SYSTEMS
rotors
pressure-times-area
for 9 and
in reference
Axial Thrust Balance System
used
limit since
nonuniform been
typically have been methods. In the
26
In the Mark
with
The
shutdown
has to
been the
of photoelastic have
2.3.4
*Copyright,
used
due
DN,
as determined
from
pump,
and
torque
turbine.
resulting the
Loads
on all state-of-the-art
in the
drive
means
The bearing
and
of 5 percent
10 percent
typically
and
startup
moments
section.
by
methods
for the is limited
bending,
loads.
magnitude
torque,
bending
determined
pump
during
(magnitudes
and
has
the
dissipation
torque
inertial
steady-state
alternating
torque-limited
generally
speeds
therefore
by permissible
Average stresses in one-piece type rotors or thin-shell theory and finite-difference rotor
couplings
An
been
critical
herein
centrifugal,
and
between
been
power
load
have
51). Finite-difference builtup-type rotors. Splines
value). and
torque,
established
support
steady-state
at the
speed
turbopump. finite-element 15-F,
has
forces
condition
design
process
turbine
and
discussion
thermal
been
a bearing
Centrifugal
hydrodynamic
possible
had
on the
typical).
with
design
steady-state the
stresses,
The
simultaneous
has, in effect,
hydrodynamic
from
rotors.
designs. to
along
rotors
of loads,
turbopump
of axial-pump
are
state-of-the-art
torque
discussion
engine
features
differential-pressure the
DESIGN
are forces
Works,
subjected and
Rochester,
fluid New
to
high
momentum
York.
47
axial changes
thrust in the
loads pump
originating and
turbine.
from These
loads
must
be known
accurately,
bearings,
back
pistons,
or some
combination
form
of compensating
and
some
The
Mark
concept balance axial
9, Mark
shown system gap
at the
15-F,
centrifugal
25, and
to counteract drums,
In axial-flow
piston Mark
be made balance
methods.
balance
Mark
must
impellers),
of these
(resulting
on both
piston provide characteristics
(on
provisions
have
been
26 pumps
them
by thrust
compensating
turbopumps,
balance
thrust
bearings
used.
incorporated
the so-called
series-flow
in figure 22. High-pressure fluid from pump discharge is introduced into the and flows through two variable orifices in series to a low-pressure area. Shaft
movement
orifice
that
vanes
and
a are
nominal
preload
designed
package
to minimize
from
sides
changes
of the
in
piston.
pump
Resultant
or turbine changes
thrust)
in pressure
force change to counteract the unbalanced shown in figure 23. All these thrust-balance pump into
operating
the
bearing
or prevent
point
the
package.
rubbing
only
Axial
of the
axial
stops
causes differential
load. Typical systems were
load
on the
were
into
orifices.
Xl
l x2 .
.
BALANCE-PISTON
.
FORCE
THRUST-BALANCE
(F)
FLOWRATE
(WB)
J/I/Z/i
/
¢.
\
/
X I
I
I
I
i
I
i
o5. GAP
RATIO
23. -
Typical
i
_i
I.o Xl X I +
Figure
i
X2
performance
of a series-flow
48
thrust-balance
in
across
the
performance designed such
bearings
incorporated
balance-piston
a change
system.
would the
be the bearing
I VOLUTE
ASSY MANIFOLD ,o
VARIABLE PRESSURE
HIGHORIFICE
VARIABLE
LOW-
PRESSURE
ORIFICE
STATOR
HIGHPRESSURE
_-_
SEAL
ASSY
SPACER
BEARING RETAINING
NUT
;EAL PISTON
BALANCE
/
RING
SECONDARY
PIN LOCK
/I SEAL
PIN MATING
LOW-PRESSURE
RING
SEAL FIRST-STAGE
ROTOR
PRIMARY
ASSY
LABYRINTH
SEAL.
SEAL
SPACER
Figure 22. -
Series-flow thrust-balance system used on Mark 15-F pump.
_
DISK
Balance-piston of the 0.015
such
that
+ .001
spacer load
on the
until
the
bearing
from
0.015
rotor
to
suction)
The
turbine
was
zero
assembly
to
system
tandem direction.
pump
Additional
the
flow
of the
on
system,
is presented
2.3.4.2
MECHANICAL
thrust
magnitude
standpoint, aerodynamic such
are
and
to
load.
was
orifice
was
(fig.
Preload
22).
the
a specified
bearing
exercised
seal).
gaps) machining
in the
preloading)
were
a on axial
retaining
to
Preload
load applied
on
nut
prevent the
load
pump
opposite
end
direction.
accomplished
with
in figure
reversal load was
26)
and
line (Mark
bearing/balance-piston
(i.e.,
axial
bearing of the
Mark
arrangements,
at the support
previously across
the
toward
pump
one
bearing
in the
for load
sharing
in one
to a lower
externally
9 and
turbine
load
of
designed
routed
in figure a single
differential
capability
has been
Mark
the
front
in pressure
to
the
toward
25. As in the case
a change
necessarily
and
load
pump is shown flowed through
to the
Thrust
up
suction
1 and
a bias
transmitted
system
15-F
pump
was
caused
package
(M-l).
pressure The
flow
area
of
also has
25).
including
a parallel-flow
3.
DESIGN
predict
accurately
extremely
high
thus small.
for example,
by applying be
operation
by custom
assembly
torquing
had
axial
shown
only
in references
analyses as the
springs
(Mark
information
involves
end
bearing
load
thrust-balance
to the
usually
support
was achieved
to provide
movement
bearing
internally
to
This
unbalanced
externally
It is difficult
15-F pump, high-pressure
assembly
low-pressure
and
designed
accommodated
because
both
routed
and
proper
temperature.
and
axial the
be
discharge
the
with the
setting
point.
shaft
counteract
set,
Mark
during
(care
the
manner
was
operating
could
and
to achieve
thrust-balance system used in the M-1 hydrogen fluid was introduced from pump discharge and
concept,
piston
seal the
(gap
in order
on the
in fig. 22)
gap
a set of ball bearings
discussed
been
toward
piston
was achieved
2 bearing
orifice
limits
of low-pressure
low-pressure
(no.
in a similar
the
nominal
through
the
at liquid-nitrogen
orifice.
pump
in. setting
procedures
rotor
variable
areas
(sum
the
assembly
pump
The design for 24. High-pressure
the
balance
assembly
was achieved
These
The
The
low-pressure
transmittal
to strict
travel
bearing
rotor
be held
axial
in. The end
must
system. the
to fit between
turbine
the
gaps
thrust-balance
designed
the
orifice
can
be
Thrust
the
significant loads
drops in flow passages, and system and in the pump
of
thrust
pressure-times-area
are in themselves
magnitude
axial
orifice
fluid rotation and turbine
even
though from
5O
and
in the
rotor.
The
analysis
the
variation
in predicted
inaccuracies,
from
a percentage
pump
to inaccuracies.
coefficients effects proper)
a turbopump
forces,
obtained subject
of
hydrodynamic Additionally, thrust
balance
on pressure distribution are involved. In order
and
turbine
assumptions system,
in
pressure
(both in the balance to cope with these
"dLund
lan_ t-lAI aqJ. Joj_ ws;sAs
eOUeleq-_.sm4.
L -- "#E aJnfi!-I
3NI7dS
_07_ 39V)IV37 inN 9NINIVI3_ NOISId 33NV7VB f3S HIN I 'dAgV7 tr_
gVl
N307 1V3S HINI_ABV7
WIHS NOiSId
33NV7VB AlgN3SSV _OIVIS
9NI_V3B
ONISNOH 3NIggfll
9NISflOH 31n70A
THRUSTCOOLANT
BEARING ._/BELLEV
I LLE
SPRINGS
i
-: ....
_--'_-
SEAL
_/
F---
I
-"--
ROTOR
!
SEAL
I
ROLLER BEARING
TRIPLE THRUST BEARING
ROLLERBEARING HOUSING
--
FLEXIBLE
COOLANT
BEARING
SUPPLY
SUPPORT
Figure
25.
-
Thrust-bearing
assembly
52
on M-1
fuel
pump.
problems,
design
correction
capability
balance
pistons
calculated force
margins
the
during
on
axial
of
in
the
the
Mark
thrust;
balance
the
thrust-balance
systems
development
9,
15-F,
i.e.,
if the
piston
would
phase
25,
and
have
been
calculated
by
pressure the
no
either
where
2.3.4.3
SYSTEM
In
state-of-the-art
system
due
or
piston
to
differential
with
and
balance
the
mechanical
pumps,
is unstable,
considerable systems.
can
that
systems
original
cause
make have
contact been
balance-piston
without
instability
Hardware
damage involved
balance-piston the
The
axial
damage
to
during
the
examined
Essentially
in the
due
orifices. of the
to
heavily This
or heavily
cause
the
has
The
loaded
erratic
the
form
oscillations
pump
se.
Stresses
have
maximum speed
the
oscillations
worn
or
condition bearings,
thrust
been
calculated
being also
stator
housings,
assembly the
volute,
to
used
have
the
rotor,
for been
same
bearings
and
Dynamic
characteristics
either
design
by
for
the
was
adopted
changes
normally
broken
analog
to
Mark
resulted behavior
was
for
not
by
and of the of the
simplified
functioned
Mark
15-F,
cavity
of a catastrophic
inserts
that
25,
in bearing
explained
pressure. nature
and
(originally)formed
of the bearing
instances, not
rotating
pump the
if
the Mark 9, 25, and 26 had 15-F, there were numerous
in contamination
in several
9
of
occur
simulation
in balance-piston
was
carbon
stability
which
motion (refs. 52 and 54) or solution (refs. 52 and 55). designed
and
given of the
oscillations.
analytically
of abrupt
balance-piston
PUMP STATOR
bearing
system
rotor
been
operation was achieved through a series of changes involving orifice insert material, and the method of orifice retention.
2.4
of thrust-balance
growths
Aside from scuffing on balance-piston surfaces, problems. During the development of the Mark
primarily overheated
at design
centrifugal
attention
configuration
change.
and 26 pumps. no operational of
the
thrust
the
design
or
the
twice
pressure-times-area
pump
per
techniques,
thermal
nonlinear equations describing balance-piston linear equations programmed for digital-computer
occasions
pistons
example,
to handle
the
permit
STABILITY
surfaces
properly
closed,
calculated
to
For
sized
investigations 52 and 53.
finite-element
the
thrust-balance
thrust-balance
The
the
provided
necessary.
self-compensating stationary
problems
analyses
across
Stresses
included
the
disk
differential
analysis.
the
structural
were
were
to twice
been
turbopump.
pumps
orifice
be equal
operating conditions. Analytical and experimental and pressure distribution are reported in references There
of the
26
high-pressure
have
coolant
failure.
completely,
clearances,
the and
Although trouble-fxee
flow
restrictions,
ASSEMBLY as discussed and
herein
the cylindrical
53
consists housing
of the that
vanes,
encases
the
the vanes.
front
and
rear
2.4.1
Vanes
2.4.1.1
PROFILE
The types considered 2.3.1.1.
of profiles used in their selection
The
2.4.1.2 With
TYPES
significant
VANE the
vibration
for the vanes in the state-of-the-art are the same as those for the blades
profile
design
MECHANICAL
exception
of
of stator
vanes
parameters
are listed
pumps and the factors are discussed in section
and
in table
III.
DESIGN
centrifugal
force
in axial
flow
considerations, pumps
the
is identical
analysis
to the
of load,
rotor
blade
stress,
and
analysis
(sec.
2.3.1.2). Vane
development
During
the
caused
by
with
stress
excitable During design figure
the of
design the
the
vanes
3 that
the
calculated
surfaces
steady-state
the
of the
stress tooling
to the had
TOLERANCES, associated as those
Vane Attachment
2.4.2.1
METHODS 15-F segments
and
Mark that
vane.
This
the
natural
blades. cracking
frequency
by increasing
change
the
decreased
to magnitudes
the
above
larger
those
were
radii
on
the
a profile the
and
on was
portion
of the
shroud
raised
redesign.
pressure ratio
leading
Note
the vane
profile
across
a thickness-to-chord
FINISH,
tolerances,
discussed
26 pumps,
problem.
designed,
the
fabricated,
hydrodynamic
Since
it
and
suction
of 0.15
instead
trailing
edges
and
for
axial
loss for the pump.
SURFACE with
on
pressure
necessitated
been
the
stress
As initially loading
that
to obtain
2 to 3% performance
same
a vane
steady-state
differential
already
required
are the
of
to compromise
hydrodynamic
to a magnitude that
2.4.2
cylindrical
due
action
considerations
Mark
the
stress
necessary
configuration.
This
Design
with
of vane
was corrected
frequencies
a potential
0.10.
PROFILE
In the
alleviate
apart
2.4.1.3
vanes
to
moved
in an estimated
occurred instances
result
of the
natural
it became
were
original
that were
the problem
section
vane
profile
resulted
pump
the
to withstand
the
root
those there
as the
rotor;
was a shrouded
additive
to use
diagnosed
at the
M-1 pump,
in order
the
to
15-F,
frequencies.
M-1 vane
adequate
However,
was desirable of
of the
Mark
off the
increased
forcing
similar
the
was
wakes ratio
and
been
of
cracking
forcing
by known
structurally vane.
The
the
thickness-to-chord
steady-state
have
development
fatigue.
resonance profile
problems
early
for blades
the
AND
surface
were
around
the
54
finish,
in section
vanes
assembled
FILLET and
fillet
radii
2.3.1.3.
machined rotor
RADII
integrally and
encased
on
three
120 °
in a one-piece
volute/stator housing(fig. 26). In the Mark 9 andMark 25 pumps, the assemblyconcept was similar. The vanes, however, were integral with segmentedrings (three 120° segments comprising one stator row) with cylindrical spacersusedbetween stator rows (fig. 27). The M-1 mainstagesincorporated individual vaneswith mounting lugs that, when assembled, were captive in cylindrical retaining rings(fig. 28). The rings and vaneswerethen encasedin cylindrical housings.
ROTOR ASSEHBLY
STATOR SEGMENT (THREE SEGMENTS PER ASSEMBLY)
Figure
As in the
case
of rotor
weight,
manufacturing,
pumps,
the
methods
housings.
These
distortion
associated
2.4.2.2 The loads
attachment
to the
support
Rotor-stator
blades,
the
and
with
assembly
selection
assembly
selected
cylinders
MECHANICAL
vane
26.-
have
Mark
considerations. the
split
15-F
pump.
of an attachment
permitted
preclude an axially
for
the
potential
method
Note use
that,
in
of continuous
propellant
for vanes
leak
the
is based
on
state-of-the-art
cylinders paths
for and
stator thermal
housing.
DESIGN must structure
be
designed and
to transmit
additionally,
55
the
to position
steady-state and
retain
and
vibratory
the vane
both
airfoil axially
ROTOR
STATOR
Figure
SEGMENT
27.-
DISK
(TYP)
(TYP)
Stator
segments
and
rotor
disks
for
Mark
(57
9 pump.
VANES) /_VANE
(TYP)
KEYS (6 PLACES)
_IyNSTNAGEBoSL_ TOR
BLADE
(TYP)
ROTOR (24
RETAINING
Figure
28.--
Rotor-statorassembly
56
for
M-1
pump.
PLACES)
RING
and
circumferentially.
used
in the
has been
There
state-of-the-art
transmitted
Individual
stress
Additionally,
diameter
gives
the
configuration.
Thus, the
length,
the
resulting
and
1 presents
turbopumps.
The
pumps;
associated
HOUSING
stator
and
the
axial
one
of the
two
housings
pumps,
flow
volute
major
of
a
load
and
volute
stress
single-vortex particularly stabilizing
and
of flow
diffusion
on the
design
bending
and
a the
outside
a
suitable
moments over
collectors
is applicable
is limited
to both
essentially
have a as
noted
effective performance.
than
to the
structure
propellant
turbopump
housing
together,
as in the
as in the
Mark
housing
and
weight, housings
diffuse
tie the
volute
some
degree overall
when
an appreciable
in rocket
engine
centrifugal
or axial
particular
features
and
fabrication since
flow
walls
together.
reference
1,
a
folded
the
efficiency
57
design
the
can volute;
of a volute-exit
of
Mark
25
phase
pumps
be
have
had
path
is
vanes
in
radial
of the
9, the
in utilized
this
housing
leak
hydrodynamic
in figure results
involves
Single
propellant
the exception
volute in
of It is
consist
9, and
considerations.
which
vanes line.
26 pumps.
the
as illustrated
the
It may
Mark
as to minimize
envelope,
motion
M-I,
a potential
With
of "foldover" housing
assembly.
state-of-the-art
as well
encases
to the discharge
Mark in
pumps), for the
the
double-vortex
in maximizing
15-F and
configuration
that
flow
bolted
in production
in
the
of the
deflection,
had
delivers
volute) housing,
smaller
rather
pressure-containing
and
stator/volute
to structurally
Additionally,
path to
be distributed
members and
to turn
permits
flow which
present
through
low.
generally
herein
is the
(especially
proper
sections
volute
not
designs.
stator/volute
All of the
volute
in
does
be transmitted
vibratory
can
are relatively
reference
collects
be a single
are preferred
the
and
attachment
assembly
28).
usually
must
length
on the stator (fig.
a continuous
arc
discussion
housing and
stator
or it may
eliminated.
pump)
methods
Housings
discussion
structural
(i.e.,
hydrodynamic,
load
attachment
TYPES
stages
selection
units
the
the axial-flow
M-1
the
in shear
loading
steady-state
therefore
in the
acting
of forming
of the
a complete
in the
appreciable
airfoil
loads
therefore with
2.4.3.1
The
necessity
section
material
that
with
(torque)
by keys
hydrodynamic
Stator and Volute
Reference
The
lug
(e.g.,
an
the
to
2.4.3
the
problems
circumferential
housing
only
designer
transposed
flow
vanes
since
no structural
The
adjacent
of
problem,
attachment.
been
pumps.
to the
attachment
critical
have
kind
conical
Mark
29.
Folding
lower to of
the
weight. obtain motion
diffuser
and
a is in
_SER
MARK
(A)
9
(CAST)
NO
rIARK
FOLDOVER
(B)
Figure 29. -
Two
standard
techniques housings:
cast
is less than
that
obtained
only
housing
which For
can
be
example,
steel
a
flightweight
Inconel
718
structures
was
encountered
core
breakdown
complete
core
completely achieve
in the heat,
the
steel
structures.
original
to the been
axis utilized
together discharge
in the
structurally. section
of
volute
housings
stage
of pump
enters
rotation exit
the
of
core the
Mark
diffuser
metal.
erosion
25,
housing
vanes
and
welded
difficulty
in
because
of
Problems
of
problems
housings
15-F,
vanes
liquid
diffuser
volute/stator
achieved.
volute
(flightweight
of the
was
were
not
necessary
to
for the Mark
26
for a
advantages,
Considerable
of the
However,
grinding for
the
and
diffusion
on axial-flow an exit prior
passage
In the and
and
area
pump
cost
always
and
requirement
pressures
these
not
stator
and
pumps
9 pump
were
builtup
DESIGN
vohlte
the last axial
in the
The
are
M-1
axial
time
in practice
configurations).
fluid
resolved.
lead
weight,
initial
immediate
housing and
passages.
design)
discussion 1. The
volute
considerable
HYDRODYNAMIC
reference
an
for the later
FOLDOVER
the state-of-the-art
However,
the
CONSIDERABLE
fabrication
of additional for
(CAST)
degrees of foldover.
the
structure.
not
erosion, were
and
than
A complete
was
various
in fabricating
expense
planned
(C)
Ideally,
selected
hydrodynamic
welded
from
design
resolved,
(other
2.4.3.2
at the
breakdown
acceptable
employed welding.
were
were
from
FOLDOVER
for a welded
castings
because casting
and
M-I
types showing
been
casting
(WELDED)
LIMITED
Volute
have
stator/volute
15-F
M-l, 12-in.
system pumps
passage
and
to its being to guide
and
additional discharge
58
have
been
is gradually collected
diffuse
diffusion pipe
hydrodynamic
the
designed
turned
in the flow
design
volute and
such
toward
in the
that
to tie other
the
a plane
proper.
was accomplished
attachment;
is given
the
in
flow
normal
Vanes
have
volute
walls
between
the
10-in.
axial
pumps,
the
mean velocity in the volute was the sameas that in the dischargepipe (i.e., there was no conical diffusion section). The diffuser vaneshavebeen designedfor zero incidence at the design flowrate, the vane angle and area distributions being chosen to minimize friction losseswhile maintaining a specifiedvane loading (maximum diffusion factor). The volutes proper havebeen designedon a one-dimensionalbasiswith approximately constant velocity (ref. 56).
2.4.3.3
The
MECHANICAL
assembled
of the
volute,
turbopump.
designed flange
to and
DESIGN
stator From
withstand mount
housing, a stress
internal
reaction
and
bearing
standpoint,
pressure,
loads,
and
the to
pressure
hydrodynamic vehicle
design both
of
inertial
and
loads
volute
magnitude,
growths, housing;
that
usually
is the
an
finite-difference
solved
cause critical
deflections
axisymmetric
The second approach with test results.
Proof-pressure
testing
state-of-the-art
volutes
designs generally have deflection analyses.
from
a structural
has
been
and been
shell
model
subjected
at In the
revolution, are used
loading
analysis
is used
be sized
59
standpoint. first,
the
part
Because
of
structural
rotating
most
of rings,
and
the
and
manufacturing
volute
complexity, to verify
to
have section
beams
on
and
the
volute
shell
is
/'or either
stresses
of the
testing
cross
programs
accuracy
are
designed
matched,
the volute
may be stator
in volutes
plates,
approach,
line
components.
critical
are
computer
(an
to react
generally
be
Stresses
the
junctions
second
loads
must
and
reasonable
as a normal to complete
Mount
stationary
due
a volute
misalignment,
points
to detennine
demonstrated
housings.
assembly. mount
from
surges,
must
installation
consisting
rotations of
has
specified
stator
line
In the
stresses.
analysis
and 58, resp.). were compared
pressure
so
orifice
are determined
hydrodynamic
be
loads,
be provided
and
connections
the
between
approaches.
and
thin
or finite-element
at
rubbing
and
for moments
must
oscillations
the
housing
structure
by a simplified
The
internal
line
must
balance-piston
for increased
possible
on the
the
by two
is represented
are as
more
analytically
foundations.
equations
could
rigidity
loads
from
assembly
circumferential
alignment,
pressure
foundation
and line inertia forces. Turbopump mounts in this case, rotor thrust loads and turbine
imposed and
Sufficient
for propellant from
and
circumferential
and
determined loads
are additionally
determined
of the volute
treated
that Flanges
asymmetric
deflections
to account
structural
housing
axial
bearing
and
the
volute/stator
affect
axial
form
assembly loads.
excursions,
differential thermal the volute/stator
minimize
elastic
than
is typical).
considerable
been
In order
on
assembly
stator
engine
greater
20%
symmetric
The
pump
analysis.
pressure
pressures, located
the
acceleration,
increase
of
and
the
stator
that housing deflections do not adversely clearance, or blade and vane tip clearances. Internal
housings
when
(refs. solutions
process new the
57
ol]
volute
stress
and
2.4.4
Bearing Housings
2.4.4.1
TYPES
The
primary
function
positioning
and
turbopumps
have
cradled
between
and
turbine
the
have
assemblies
for
either
(i.e.,
type,
with
state-of-the-art design
2.4.4.2 As to
axial
and
can
or turbopump
be
severe
and blade
desired
clearances
and
6; the
its effect
on
9,
loads
tip the
the
reference
turbopump
25,
degree itself
being
mounts
attained
and or
axial axial
pump
end)
stage.
is
bearing
bearing
housing
Additionally,
axial
bearing
for the
volute
radial
locally
their
stops,
M-1 bearing
at
be the
bearing
been
bearing
speed.
utilized
housing,
stator
the same
carrier.
on critical
have
and
stiffness
in the
influence
M-1 rear
M-1
of
will in general
forgings
of the the
assembled
of the
end
In the
housings.
proper that
in
which
housing,
all was
flightweight
housings.
also
bearing be
includes
rotating-assembly
and deflections
achieved. piston
26
pumps
was
the
and,
external
structural
are subjected
thermal loads
is given
gradients
reacted
to radial
be minimized
rigidity
in those
the
housings
pressure,
must
Axial
form
bearing
attention
so that
is necessary
designs
with
at the and
axial the
to maintain
axially
preloaded
load. reacted
by
discussion critical
the
such
6O
that
bearing
of bearing speed.
housings are determined during the flight inertia loads of the turbopump and
components the internal
particular
Radial are
balance
must
rotor,
housing),
design,
clearances thrust
stationary
of this assembly,
turbopump
turbine
of the
vane
loads
by the include 15-F,
the
mounts.
for
radial
reference
Mark
and
axial
front
inducer
of the
exception
As a part
in the
to maintain
be reacted 2.3.4) and
the (pump
seal packages,
housing
castings
requirement
2.4.3,
deflection
desired
rotor
that front
All of the
of the
radial
state-of-the-art
DESIGN
radial
flanges
The
row
packages,
stiffness
As with
turbopump.
(which
bearings,
such
bearing
CRES
the
casting.
in section of the
alignment
All
in terms
of bearing
with
a single
an immediate
indicated the
maintain
of the
bearing.
stator
spring
The
design
300-series
MECHANICAL
foundation
locations
outboard end)
the
classified
desired
housings not
be
or flexible).
the
of
from was
for
bearing
and
assembly.
bearing
inducer
(turbine
vanes
a discussion
structures
machined
can
rigid
6 presents
Welded
two
the
rear
included
types
mount
Reference
the
is to provide rotating
mounts.
housing
bearing
at
with
of the
have
housings
turbopump
supported
incorporated
turbopump
bearing the
bearings,
outboard
housing
Bearing
the to
been the
designs and
of
support
axial
Rotor
housings housing axial
radial thrust
thrust-balance-system rotor assembly. thrust
toward
are
discussed
spring loads
The pump
rate that
in and must
analysis (sec. design of the suction
was
transmitted to the turbopump housing assemblyby the front bearinghousing and thrust toward the turbine was transmitted by the rear bearing housing. Axial thrust in the M-1 pump was reacted only at the front bearinghousingthrough a triple set of ball be/_ringsand spring system,asshown in figure 25. In general, stress
the
complexity
analysis
calculating
of the
local
of
total
stresses
inducer-vane stresses flange stresses).
the
bearing
structure
that,
iv, the
in the
front
housing
difficult. opinion
bearing
structure
Stress of the
has
analyses designer,
housing,
stress
Housing Interfaces and Static Sealing
2.4.5.1
INTERFACE
The
components
Structural
making
continuity
assembly
and
SEAL up
must
operation.
positioning
the be
pump
housing
maintained
Particular
in general
In
this
concentricity mainstage
was stator
machined
been
accomplished
types
59
of seals
design at the manufacturer.
the
has
been
each
must
by using
presents that
been
with
a monitoring
port
M-l,
leaks
developed
in some
prevent problems
seal hydrogen were
and with possible period of time.
leaked,
a helium
leakage
to
associated creep
with of
the
the
final
in (e.g.,
and
the
at
the
local
interfaces.
interfaces
during
diametral
tolerances
pump
components
an
and
was
interference
are
difficult
volute
and
(2)
fit on
the
seals
for
61
support
and
housing
stator
in general
rocket
engines.
in figure
and secondary
was
and
rear bearing
liquid
illustrated
cases 60
stackup subjected
30. The
has
seals).
During both
the indicates of the to high
The
interface
is specified by seals (i.e., double
where
through
Reference
which
and
to ease and rear
clearance
bearing
to the
and
assembly.
in those
tolerance
stator
front
relative
pump static
and
the
the
rotor
introduced
atmosphere.
material,
rotor
seal type and in general pump utilized redundant
joints,
purge the
of pumps
the primary
of the
the
to (1)
using
the and
of the
in axial
between
bolted
to stationary
(i.e.,
as a unit,
discussion
used
seal is dictated by the Note that the M-1 axial
seals
secondary
during
a detailed
have
affected
positioning
shims
of
by
subassemblies
machined
Axial
are
be given
accomplished
that
on
were
as a unit).
reaction,
the parts assembled at different temperatures three housing interfaces between the front
machining
accomplished
housing
were
Reference
case,
be of significance
to mount
assembly
across
attention
diameters of the mating flanges with the buildup problems. The M-1 had bearings.
analytical primarily
TYPES
concentricities that provide proper alignment of rotor sealing that must reliably prevent propellant leakage. Radial
a precise
consisted
could
due
2.4.5
AND
made
have
the seal conical
testing
of the
the primary
monitoring that
and
port
to
the
leakage
double-sealed
joints
loading
for
a long
•sdLund le!xe u! pasn sleos op, eJ,s _.o sadA_L -
'OE oJnS!-I
_;z:_!vw (G)
6 _w
(3)
ONl_dS 3177V13N G31Vfl13V-3_nSS3_d
9N I _1-0
>
3177V.l.3
(]3xvnJ.3V-3_lnss3ua C',l ',D
9z :w_vw (]NV3-St :_VW
////////__!_ (]3,:IVHS-FI 31llVl3W
(B)
_-. (v)
.... ,_
_ii_ 7V3S
A_IVONOg3S
I
_///"_'/
031vn13v-3_nss3_a 7V3S
1_10(:1 9N I _OJ. I NOW "
I_VWI_a
2.4.5.2 A
MECHANICAL
tight
interface
assembly
in the been to
separating the
the
loads
at any
the
that
to the
joint
is analyzed
a spring
model
Stresses
and
is used
the
at
under
all
allowable
loads
typically
steady-state
interface the
loads
preload
preload
Interface
in
inertia
bolt
exceeding
pump
Each
is included and
system.
The
environmental
in determining
expansion
joint
without
combinations. for
The
operating
must
be
analysis
that
examined
(e.g.,
must
pressure
be transmitted
mount).
as a spring
for the
adequate
effects
conditions.
misalignment,
turbopump
calculated
probable
loading
conditions.
an
thermal
conditions,
all potential line
all load
to ensure and
temperature
propellant
are
for
of the
assembly
transient
the
or contractions
elastic
stiffness
temperature loads
and
resulting
of the
condition stresses
from
in each
thermal
parts
being
of the
effects
making
investigated interface
are included
in
analysis.
Measurement of applied required bolt preload normally each
result
part
in the preload.
Preload
range
parts.
This
been
joint
can
method
necessary
2.5
torque (torque wrench) at assembly. However,
in a preload
required
that
assembly
can must
be minimized is time
vary
by a factor
be capable
by
measuring
consuming
to use this method
is the most common the uncertainties
but
may
method of achieving in friction coefficients
of 3 or 4. This
of withstanding
deflection
of
be warranted
in the stationary
condition
a load
the
the
or compression
assemblies.
of axial
that
3 or 4 times
tension
in critical
components
means
a
It has not
pumps.
MATERIALS
The
materials
stated the
enough
be
interface
interface
parts.
great
of separating for
ensure
the joint
and
maintained
or flange
for
load,
through
up
bolt
and
carefully
be
must
examined
conditions,
The
must
combinations
stresses have
joint
therefore
probable
DESIGN
utilized
previously, selection
Relative
of
strength-to-weight susceptibility
steady
steady-state
major
components
pumps
were
based
on
materials
ductility,
ratios must
conditions. and
designed
has,
fatigue
in order
have
adequate
ductility
vibratory-stress
components
such
conditions,
been
expansion
in order
the
to avoid
as blades
that
strength
the
basis
characteristics,
weight
alloys low.
fracture-type are
Thus
temperature. on
designs,
pump
IV. As
applications.
evaluated
In flightweight to keep
in table
at liquid-hydrogen
thermal
fatigue
63
are noted
for liquid-hydrogen
in general,
failure.
desirable
pumps
properties
strength,
are In
of axial
material
to hydrogen-embrittlement
additionally stress
was
candidate ratio,
strength-to-weight alloys
all of these
of materials
merit
in the
exposed
is a consideration.
and
with
high
Candidate
failure to
of
under
significant Material
thermal-expansionrates must be consideredin those components having a critical interface (e.g., bearings in a bearing housing), since a prescribed fit at both assemblyand operating conditions is required. Consideration additionally is given to fabrication processesand operating environment in which hydrogen could be absorbedinto or otherwisecontaminate the material and result in hydrogen embrittlement andsubsequentfailure. Considerablematerial property data have been obtained during rocket enginedevelopment programs. Much of the work done in support of cryogenic pump and other component development is reported in references61 through 66. Thus, the discussionherein is limited to someof the more significant material problemsthat haveoccurred with axial pumps. As noted previously, the M-1 rotor structure consisted of four forgings that were TIG welded to form a one-piecerotor. Extensive development work wasconducted to establish welding and inspection procedures for the weldments (ref. 64). One pump rotor was fabricated and utilized in liquid-hydrogen turbopump testing. This rotor had known weld defects prior to the test program. Post-test examination of one of the rotor weldments indicated that nearly all of the defects propagated during testing (ref. 60). The effect that thesedefects would have on long-time operation of the rotor wasnot established,however, becausepump testing wasdiscontinued when the M-1 engineprogram was terminated. As indicated in table IV, titanium alloy A110-AT-ELI was utilized for the M-1 transition rotor. Mechanical-property testsconducted on the forgingsindicated the elongation to be an unacceptably low 1 percent at liquid-hydrogen temperature. It was determined that excessivehydrogen content in the forgings wasresponsiblefor the low ductility. Elongation of 10 percent at liquid-hydrogen temperature was achievedby degassingthe forgings in a hard vacuumto lower the hydrogen content (ref. 67). Carbon was used initially as the orifice insert material on the Mark 15-F thrust-balance system.This choice was madein order to avoid galling during contact of the balancepiston and orifice insert. As noted previously, contact during pump operation was sufficient in someinstances to break the carbon. The carbon particles contaminated the bearingcoolant flow and at times resulted in bearing failures. To prevent these impact-type failures, the material waschangedto leadedbronze, asnoted in table IV.
2.6 The
SAFETY various
organizations
propulsion design either
system the
design
of
for safety
policies
pumps
therefore factors
for
employ on
organization
occasionally
is appropriate values
and
design
but
responsible
component
instructions
structural practice,
FACTORS
the
definitions
to define that
have
the been
structural
individual
safety
or in general
the
the
factors.
manuals Values
responsible have
been
have terms
64
of
that
for safety
contracting consistent
differed
as used
utilized
design
from
with
contain
of axial
or
other
comprehensive
factors
are specified
agency.
Terms
aeronautical
organization
in this document
in the design
a turbopump
by
used
engineering
to organization. and
to indicate
pumps.
in
typical
It
Table
IV.
-
Materials
Used
for
Major
Components
Pump Component
Mark 9
Mark
15-F
on
Axial-Flow
Pumps
Configuration
Mark
25
Mark
26
M-1
Material Rotor
K-Monel
310
Blades
310
K-Monel
K-Monel
Same
as
Mark
15-F
lnconel
K-Monel
718
Mainstage: Inconel 718; transition: Ti A110-AT-ELI
Volute
310
Stator
310
310
housing
(Integral with
310
304
310
304
ELC
volute) Vanes
Front
bearing
310
310
310
Inconel
310
310
310
347
310
310
310
304
718
housing Rear bearing housing Balance
piston
Balance orifice
piston
K-Monel
Inconel
Flame-plated
Leaded
tungsten carbide on
bronze
Silver-plated 310
A1 2024
718
A1 7075-T73 304
310
Limit
load.
or service under
the
limit
operating
there
all
factor
maximum
load.
operating
The
specified
hydrostatic-proof-test
environmental
limits
(a
maximuln
limits
defined
(3-sigma) physical
of the
by
in the
multiplying following
and
engine
a combination
specified factor
factors
have
or calculated
pressure)
3-standard-deviation
is uncertainty
limit-load
is the
(excluding
maximum
vehicle limits.
load
lnaximum
including
specified
When
The
pressure
(1)
vehicle
-
load >
1)
been
65
that that
or vehicle,
or
lack
used
of
limits
limits
3-sigma to
in axial-pump
the
of the
influence
or (3) the
of 3-sigma
of a service
can be expected
operating
variables
is applied
value
loads,
maximum
and
specified
data specified design"
on
load
to occur engine (2)
or the
engine
conditions, or
or
operating
calculated
a
Type Centrifugal
load
to rotational
speed;
speed)
Load
on blades
Structural
load
induced load
Design
safety
than
to fluid
due
to internal
due
factor.
load
in
-
of the
the
stress.
the
load
slightest,
may
be
prevents sometimes
thrust
design
mechanical
design
safety
factor for
design
design
load
multiplier
(or divider)
uncertainties,
e.g.,
within
structure.
distributions
(or pressure)
is the
stress,
or combination
-The
failure
is an arbitrary the is the
variations
product
greater
in material
of the limit
load
(o1
from
the
factor.
stress
load
stress).
produces
defined
- The
1.05
gimballing
and load
safety
1.0
contraction
to account
design
(or
1.2
and and
design
quality,
The
application
Allowable
called
is applied
pressure
- The
the design
stress.
highest
factor
1.1
expansion
(or pressure).
and
Design
load
forces
to engine
fabrication
pressure)
(limit
is sometimes
due
1 applied
Design
result
by thermal
Inertial
properties
to rotation
factor
1.1
design
Load
due
Limit-load
of load
allowable
of the
as buckling,
structural
of design
loads,
load
pump
yielding,
the component from referred to as criterion
in any
(or
stress)
structural or
resulting
whichever
is the
element
ultimate,
performing its load or stress;
element,
fatigue
intended allowable
condition
load
that,
results
if exceeded
under
consideration.
failure,
whichever
in
ill
Failure condition
function. Allowable load is stress is equivalent to material
strength. Margin stress
of safety. exceeds
the
- The
margin
design
load
of safety or stress.
(MS) The
MS-
where
R is the
Material maximum
ratio
endurance alternating
of the design limits. stress
load
is the
margin 1
-The
material
that
the material
of safety
to the
the
is defined
allowable
endurance can
66
by which
allowable
load
or
as
1
R
or stress
fraction
sustain
limit
(also
load
or stress.
called
for an infinite
fatigue number
limit)
is the
of cycles.
Proof
pressure.
prove
the
-Proof
adequacy
pressure of design
is the and
pressure and the proof-pressure temperature is not feasible, the difference
in material
strength
test
pressure
quality.
The
factor. proof-test
in the
Proof-pressure
design
factor.
of
at operating
-
the
temperature
pressure to obtain the proof pressure. state-of-the-art axial pump components. In practice
in the 0.2%
design
offset)
of axial
factor
pumps,
of 1.2 typically
a typical
on
factor stress,
for fatigue, expressed as a ratio of material typically has had a value of 1.33 ; the fatigue
values
of 4X
predicted
cycles
typical
for low-cycle
of cycles
to failure
the
at to
to partially
product
of the
the design compensate
temperature
to ensure
is a multiplying
A value
1.1 ; the
and
fixture
(based
fatigue; i.e., the number operating cycles.
has been
a component is the
limit
operating for the
at which
the
are to be subjected to proof-pressure tests are do not occur during the proof test. Care is
proof-pressure
Proof-pressure
to
pressure
When proof testing pressure is adjusted
proof testing is conducted. Components that designed such that detrimental deformations exercised simulated.
applied
proof
fatigue, should
67
value
factor
that
factor has
is properly
applied been
for design
for ultimate
loading
to the
used
in testing
safety-factor has
been
design
for
1.5. The
the
yield safety
endurance limit to allowable alternating factor based on cycles to failure has had and
10X predicted
be 4 or 10 times
the
cycles number
for high-cycle of predicted
3. DESIGN
CRITERIA
Recommended 3.1
OVERALL
3.1.1
Practices TURBOPUMP
Turbopump
Design
criteria
presented
and
unacceptable
critical
occur
exist
13 provide
between
operating
3.2 3.2.1
turbopump
rotor
or speeds
(for
rotors
at which
speed
are
dynamics
shall
self-excited
verify
that
nonsynchronous
whirl
range.
designed
recommended
to operate
analytical
and critical
above
modeling
speeds
are given
the
first
techniques. in reference
critical).
References
Recommended
6,
margins
6.
DESIGN
selection
o.f a pump
requirements speed
shall
verstts
capabilities, Reference
O,'pe to satisfy based
on
configuration
7 presents
engine
be
upratotg
recommended
a complete
practices
and stage
head
and ./low
and
pump
speco"ic
and probable
of the design the
off-design total
relationships,
weights,
discussion with
of
efficiency
probable
associated
design
examination
and
potential,
given
selection
operating
considerations, of
the
ralzge
costs. design
various
types
criteria, of
pumps
and for
application.
It is recomlnended above
turbopump
Realm of Operation
The
rocket
speeds
speeds
STAGE
selecting
in
of the complete turbopump rotating assembly and support system and utilized to predict critical speeds and the threshold speed of
whirl
12, and
of
in the operating-speed
An analytical model should be formulated non-synchronous
involved
Rotor Dynamics
predictions
do not
practices
7.
Turbopump
Analytical
DESIGN
Speed
recommended
in reference
3.1.2
and
approximately
that
an axial
3000
configuration
and when
be considered
throttleability
68
and
wide
when
stage
fixed-speed
specific flow
speeds
range
are
are not
required.
Additionally,
it is recommended
that
a detailed
capabilities be made whenever pump uprating is a design is appropriate to both the axial pump and the centrifugal the
addition
of stages
centrifugal
3.2.2
pump
is relatively
requires
straightforward,
an additional
stage
Stage Hydrodynamic
The
stage
design
requirements,
shall
of the
requirement and the specific speed pump. The axial pump, in which
is recommended
to satisfy
axial-pump
the
when
uprating
the
competing
requirement.
Design
reflect
an
mechanical
examination
acceptable
compromise
requirements,
and
among
hydrodynamic
pump
configuration
overall
requirements. The the
flow
model
used
in the
three-dimensional
axial-flow
real
compressors
blade-to-blade to final
flow
coefficient,
flow
be
planes
Prior
selection hub/tip
guides
in the parametric
ratio,
of axial-flow
blade
should
approximate
that
design
practices
is recommended approach,
average
flow
flow in the hub-to-tip
hydrodynamic
design,
tip speed
is achieved.
stages
should
conditions
plane
(ref.
a parametric
be made
It is recommended
that
the
18).
study
so that the
for in
involving
an acceptable
tbllowing
be used
as
should
be
study:
•
The
stage
design
•
The
stage
hub/tip
For
stages
less
than
with
flow
high
1700
coefficient
ratio
ft/sec
for high-strength
3.2.2.1.1
It this the
stage
and
configuration
BLADE
in
to represent
of the detail
mechanical
design
condition.
followed;
are used
pump
3.2.2.1
hydrodynamic
LOADING,
should
should
hub/tip
not
be greater
ratios alloys.
STALL
MARGIN,
be less than than
(_> 0.8),
for high-strength
nickel-base
not
the
AND
0.9. blade
titanium
0.25.
alloys
design and
tip
speed
less than
1500
ft/sec
EFFICIENCY
Blade Loading and Stall Margin
Design-point
blade
loading
shall
reflect
an acceptable
compromise
of stall
margin
and efficiency. It is recommended loading. influence
For the
that
a given selection
the
diffusion
application, of the
factor the
(eqs.
(5) and
stall-margin
design-point
diffusion
69
(6))
requirement factor.
be used of Stages
as a measure the
pump
designed
will
of blade greatly
for optimum
efficiency should have a maximum design-point diffusion factor between0.45 and 0.55 (at any radius on either the rotor or stator). For pumpsin which a minimum number of stages is desired, a design-point diffusion factor between 0.55 and 0.60 may be selected if stall-marginrequirementspermit.
3.2.2.1.2
Stage Efficiency
Predicted
stage
friction,
efficiency
and secondary
It is recommended predicting particularly to highly
with
high
stagger
Predictions
3.2.2.2
radial
due
to profile,
end-wall
8. Data
care
experimental for
should
in references
be estimated
and
results
be
used
with
shown
be exercised
on
highly
in applying
8,
figure-8 blades
26.
secondary-flow
loss to obtain by methods
in
figure
and multiple-circular-arc
23 through
loss
to the profile
data
axial-flow-pumps
with
for double-circular-arc
(annulus)
can
pump
the
loss,
overall
presented
stage
including
tip
efficiency.
in references
The
68 and
69.
DIAGRAMS
pattern
suitable
losses
favorably
therefore
Data
are reported
losses
and
compare
and
blading.
be added
VELOCITY
account
in figure
not
region,
should
of these
The
do
end-wall-friction
loss,
magnitude
tip
angles
of
clearance
22)
staggered
into
compressor,
as illustrated
(ref.
in the
data
cascade,
losses,
blades
take
flow.
that
profile
staggered
shall
of flow
and
compromise
of
flow
with
the
stage
type
of velocity
headrise,
diagram
stage
shall
efficiency,
be based
and
on a
stall-margin
requirements. A
free-vortex
recommended
for designs
desirable
to use
alternate
flow
In
the
that,
patterns of and
due
whose magnitude The factor should
offer
the
ratios
pumps,
0.8,
a more
velocity
blade-surface
by linear taper by the method
hub/tip
less than
coolant
for liquid-hydrogen
reduction
a symmetrical
having ratios
might
of bearing
to end-wall
preferably estimated area
hub/tip
preparation
recirculation due
pattern
and
velocity greater
the
suitable
diagrams,
than
design
boundary density
layers increases
layers
normally
depends on the particular be selected by the designer
0.8.
should
at
the
mean
If it becomes
be examined
radius
is
necessary
or
to determine
if
compromise. the
compressibility
thrust-balance-system
in the flow path; the presented in reference
to boundary
diagram
should
flow,
and
of
be considered.
in excess
the
channel
of 6 percent
pump area
fluid,
reduction
It is recommended be accounted
for,
magnitude of the density change can be 7 (pp. 99-102). As indicated previously, is accounted
for by using
a blockage
factor
design as well as the design method being used. on the basis of experience with similar designs.
70
Caution should be exercisedbecause,as pointed out in reference 18, indiscriminate use of correction factors can lead to a design that is as poor or worse than one in which boundary-layercorrections are ignoredcompletely.
3.2.2.3
BLADE
Fluid
The
turning,
deviation
angles,
selection
of
incidence
angle
from
Thus,
a cavitation
and
guides
References
23
minimum-loss for
Accurate
prediction
design.
calculating
In view
that
deviation-angle provide
and
available
or which
exact have
solidity are
rule been
3.2.2.5
for
be the angle
Mark in
extend
the
references
range also
and
slotted
range
of
range
68,
and
be
III.
used
The
for
and
may
23 through
deviation-angle and
it
deduced
profiles
performance
70.
26 pumps,
2.2.2.3
References
double-circular-arc
of
an acceptable
25, and
in table
deduced
provide
the
31,
NACA-series
profiles.
of
18, 27, 68, and
27,
9, 15-F, section
for
C-series
Acceptable
extend
18,
optimum
for minimum
in achieving
to those
recommended
e.g.,
as that
that
important
given
diagrams.
in references
in references
similar and
same
blade
is inappropriate.
are given
of the
rule
parameters 18 are
these
be
within have
selecting
the been
be within for
achieving
degradation
for
a range
can
ideal with
headrise, increasing
0.75 should solidity
CAVITATION
The pump
mainstage
shall not
which
26
rules
for
turning-angle
profiles.
be subject
to cavitation.
71
experimental
in axial-pump
be specified.
applications
of approximately
high
¢br
demonstrated
solidity
in axial-pump
associated
values
successiVEly
a value
demonstrated
selected
desirable
efficiency
design
velocity
application;
correlations
given
incidence,
SOLIDITY shall
No
necessarily
are
double-circular-arc
profiles;
specific
is extremely
deviation-angle design
the
angles
in the
the
incidence
18.
angle
that
design
in reference
for multiple-circular-arc
Solidity
that
for
angles
reference
reflect
and
use
correlations
double-circular-arc
3.2.2.4
of
for standard
data
correlations
the
by fluid
data
deviation
having
rules
be used
deviation
of its successful
profiles
will not
provide
given
as defined
on
incidence
26
of the
recommended
nonstandard
is dependent
for selecting rules
and
properly
magnitude
through
incidence
Methods
shall
standpoint
a recommended
procedures
also
of radius
and
loss.
is
as a function
camber,
incidence
70.
ANGLES
(table
On the Ill),
data applications.
basis
of magnitudes
it is recommended
to 1.9. High-solidity be
are
analyzed
is at a tolerable
to
that
stages, ensure
level (fig.
that 9).
which the
It is recommended that the inducer be designed to provide sufficient head to avoid cavitation in the initial mainstagefor all anticipated pump operating conditions. Adequacy of the initial mainstage as free from cavitation should be determined from cavitation-test data for similar designsor from analysisof fluid velocities on the bladesurface.
3.2.2.6
OFF-DESIGN
The
pump
PERFORMANCE
stall
anticipated
point
during
at any
either
operating
transient
speed
shall
or steady-state
be at a flowrate pump
less than
that
operation.
For designs with hub/tip ratios greater than 0.8, it is recommended that a diffusion factor 0.75 or a retardation factor of 0.50 at any radius on either the rotor or stator be assumed
of as
the
of
condition
0.70
or
a
at which retardation
corresponding
3.2.2.7
to the
will occur.
factor
of
It is further
0.55
be
recommended
used
minimum-flow-coefficient
as
a
that
a diffusion
permissible
requirement
factor
operating
condition
of the pump.
CLEARANCES
3.2.2.7.1
Radial
Radial
tip clearances
It is recommended vane height of deflection rotor
stall
and
thermal
that
Axial
blade
centrifugal
growth,
and
clearances
rotor
shall that
succeeding
vane
row.
clearance
(or
stackup,
blade
tilt
vibratory
loads,
rotor
component
tip
and vanes clearance
shall minimize
of not
more
head
than
losses.
2 percent
of the
clearance. The clearance analysis should (and hydrodynamic pressure imbalance frame
and
dynamics
housing
deflections,
blade
or
include effects if applicable),
component
differential
effects.
Axial
It is recommended The
a radial
be used as an operating due to rotor imbalance
contractions,
3.2.2.7.2
on the blades
differential
minimize
an operating
blade)
row
analysis in
wake
axial
axial
be at least should direction
thrust-bearing thermal
effects
clearance effects used),
deflection,
contractions.
72
blade
between
10 percent
include (if
on adjacent
tip frame
a blade
of the of
or vatze rows.
chord
(or vane) length
assembly
deflection deflections,
dimensional due rotor
to
row
of the
and
tolerance
steady-state
Poisson
the
upstream
effect,
and and
The axial clearancerecommendedaboveis consistentwith and is a necessarycondition in the blade designpractice outlined in section 3.3.1.2.3. Deviation from this clearancevalue may be desirableif, for example,the initial mainstagebladerow is precededby a long-chord inducer stator. A smaller axial clearance(i.e., lessthan 10%upstream chord) would increase the amplitude of the load fluctuation and would require that appropriate methods referenced in section 3.3.1.2.3 be utilized in designing the blade from the vibration standpoint.
3.3
PUMP ROTOR
3.3.1
ASSEMBLY
Blades
3.3.1.1
PROFILE
The
blade
TYPES profile
blade-surface
shall
(1)
velocities
produce
and
(2)
the
desired
provide
the
fluid
blade
turning
with
with
required
adequate
structural
strength. The
selection
cavitation
of
a
profile
to a minimum,
high
blade-surface
velocities;
the
relative
velocity
profiles
inlet
designed
mainstages standard the
to
where
(the
thickness-to-chord
As
a
general
with
ratios
should held
of 0.10
and
imposed
by
trailing-edge
a maximum the
Thus, be
maxim,
or 0.11 by
(within
trailing-edge structural thickness
preferred.
the
best
the
chord
the
length
are
and
is
bending with
the
the
thinnest
of 0.13
be used
stresses
British
chord, maximum
by
exist,
maximum
1.25
times
nonstandard for
Of the common
If the NACA-65-series defined
reduce
recommended
the
camber,
are utilized
profile
or
consideration.
same
that
than
profiles
double-circular-arc),
ratio
If excessive
To
excessively
avoid
C-4 offers
and
maximum
blade
strength
profile
is utilized,
standard
thickness
(ref.
43).
as an upper structural
It
is
limit,
adequacy
thickness-to-chord
ratio
limits). radii
and up
over
thickness-to-chord
increasing solidity
profiles
is recommended. thickened
a maximum
having
application. losses, no greater
distribution
predominant
C-4, and
profiles
if standard
C-4 profile
velocity
velocity
is the
British for
profile
Double-circular-arc
prescribed
of cavitation 65,
particular
suction-surface
is recommended.
should
be achieved
Leading-
reduce
aerodynamic that
constant
the
modulus
ratio. edge
recommended
on and
NACA
the British
the trailing distribution.
depend margin,
achieve
section
is required,
will stall
avoidance
profiles
maximum
type improve
should
be kept
manufacturing to
about
as small
considerations.
one-quarter
73
of
the
as possible At blade
within
subsonic maximum
the
limitations
speeds, thickness
a total (i.e.,
trailing-edgeradius equal to one-eighth of maximum thickness) should havelittle effect on aerodynamic performance (ref. 43). It is cautioned that a specified fluid outlet angle is demanded,and excessivetolerance within the above limit may not fulfill the outlet-angle requirement.
3.3.1.2
MECHANICAL
3.3.1.2.1
DESIGN
Structural
The
Strength
mechanical
effects
design
of centrifugal,
It is recommended mechanical
design be
(fluid
force)
load
(as
of
magnitude
in accordance
pump)
should
stress
is at least with
the
determined
the
stress
due
an
the criteria
above
and
of
be determined
nominal stress
maximum
the
design
due nominal
speed.
and stress
this maximum
in section
at a This
to hydrodynamic
steady-state
with
practices
loads. loads
the
examination
combined
on the combined
vibratory
steady-state
the
and
be based
to centrifugal
percent
establish
be determined with
10
and
maximum
from
to
shall
off-design magnitude.
steady-state
stress
3.3.1.2.3.
Stress Distribution
The stress distribution Stresses
analysis shall in the blade.
should edge,
blades,
the
root
and
at the
include
outermost
steady-state
radius.
The
for a particular
the
point stress
stresses design
on the usually
listed because
the
stress
due
to centrifugal
•
Normal
stress
due
to hydrodynamic
•
Shear
•
Normal (on
and
twisted
to direct
shear airfoils)
stresses
the
stress
surface
(see
section
where
vary
with
and
the
stacking
stress
axis)
at the blade
sections,
leading
edge,
fig. 14); for cantilevered
be considered,
magnitudes
the blade although
blade
is tangent some
to
may
be
geometry:
load bending
hydrodynamic due
convex should
condition
(along
of the
is at the
below
Normal
due
stress
of longitudinal
an examination
•
stress
maximum
at a number
should
greatest
fillet
negligible
identify
be determined
this determination
trailing the
that
blades
hydrodynamic,
steady-state
combined
Vibratory
and
the
speed
should
requirements
steady-state
that
stress
3.3.1.2.2
of axial-flow-pump
to
and hydrodynamic
74
moment
load untwist
forces
moment
resulting about
the
from airfoil
centrifugal stacking
load line.
• Normal stressdue to bending moments resulting from blade stacking-line tilt or offset. If the direct and torsional shear stressesfor the particular design are appreciable, it is recommendedthat the principal stressesbe determined and that the Mises-Henckytheory of failure (distortion-energy theory) (ref. 71) be used to calculate an "effective" stressfor comparision with uniaxial material property data.
3.3.1.2.2.1
Blade
The
Tolerances
stress
analysis
magnitudes
and
The blade maximum
stress stress.
tolerance
and
recommended
shall
natural
the
tip
section
that
the
frequency
be used
3.3.1.2.3
Vibratory
Stresses
The predicted stress.
diagrams strength,
stress
steady-state should
data
state
be
as
based
endurance
is
at
the
range
of the
vibratory defined on
limit.
by
adequate
(1)
of the
uncertainties
could
blade
shall
stress
on
Assume
a
safety
stress-concentration blade. design
Section of fillets.
load.
tolerances
otz
stress
factor 3.3.1.3.2
result
the
vibration
analysis,
from
maximum
and
the allowable
alternating
the
blade
should
Goodman
diagrams.
applied
that
the
31, with additional
safety factors conditions
equal
to
the
should to
the and
is
of
Goodman ultimate
stress
line
of
of 1.33 on fatigue, 1.5 set forth in sections
proceed
as follows:
steady-state
then fillet
limits
strength,
alternating
stresses,
the
Modified
to yield
allowable
vibratory
criteria
be within
it
minimum
be less than
appropriate
75
In
blade.
magnitude
provides
tolerance condition that gives hub section is at the minimum
for a nominal
magnitude This
that the
factors
in predicting
vibratory-stress
hydrodynamic
blade
frequencies
modified
It is recommended
involved
for when
natural
the diagram be constructed as shown in figure on ultimate, and 1.1 on yield. Note the 3.3.1.2.3.1 through 3.3.1.2.3.5. In view
of
maximum.
that
in lieu of the
and
property and
effects
analysis should be conducted This condition usually occurs
conditions
material
the
frequencies.
tolerance
Predicted
consider
be
at the
recommended
stress multiplied
root
section
practices
due
to
by
a
of
the
for
the
F = MATERIAL ENDURANCE e F = MATERIAL ULTIMATE tu
LIMIT STRENGTH
B
Fty
= MATERIAL
YIELD
STRENGTH
(0.2%
OFFSET)
Ftu
F
a-
(SF) e = FATIGUE
SAFETY
FACTOR
4-J
Fe
(SF) u = ULTIMATE
_
(SF)y
SAFETY
FACTOR
b
w e¢I--
_
_
ALLOWABLE
__'_ z
: YIELD
SAFETY
FACTOR
ALTERNATING
STRESS
(_F)e
_
_
orm
_-
Z
=-'
7_
__
_
MEAN STRESS O-m, PSI
Figure
The
31.
stress
forces)
-
state
stress
Designs are
considered
The
preceding
vibratory
can
(1)
(2)
the
diagram
stress
illustrating
predicted
stress
falls
If the
factors.
steady-state
the
falls
predicted
(centrifugal
in (1) above
in figure
below
point
the
safety
determined
as shown
state
to reduce
and
the
the
ratio.
neglects be
load,
row,
the
vibratory
acceptable.
practice
stress
frequency met:
by
diagram
Ftu
plus
should
fluid
be plotted
16.
allowable
above
the
alternating line,
steady-state
the
stress
stress
blade
until
line
geometry
an acceptable
is achieved.
hydrodynamic blade
the
be changed
design
Goodman
Goodman
in which
should
Modified
defined
and
on the modified
(3)
(SF) Y
_o j
(2)
ty
LOWER 0F(-T TT '
hi b--1 ,:K
determined
vibration
axial
row
should
The
first
wakes
spacing
natural
from
the
is based
the
between
the
frequency upstream
of blade
of
row
than the
at
outlined,
blade
row
analyzed
design between
forcing
frequency
natural
76
of the
occur
15 percent above the mechanical on speed should be maintained of the blade
given
be
the
steady-state from
and
below
the
due
the
upstream
that
resonance
a pump
(fig.
17).
speed
to
natural must
upstream blade
speed. Additionally, the second harmonic frequencies
stress
conditions
upstream
such
nonresonant
wake-to-blade
following
should
will not
that
to wakes
the
being
at least margin
and any
the
the
10 percent
blade
premise
due
factor
practice
to or greater
the
fluctuation
magnification the
on
product
of load
for
be equal
and by
amplitude
Specifically,
The
damping
be
blade
chord. due
to
which
is
a 15-percent of the wake
It is recognized that it will not always be possible to apply the above practice- for example, in an axial pump with a wide operating range. In such cases,it is recommended that vibration amplitudes (and stresses)be estimated by the methods outlined in reference 41 or 42. Additionally, it is recommendedthat the designerconsult references72 and 73 to assist in the solution of designproblems that might arise in a specific application. These referencescontain extensivebibliographieson the subjectof blade stressandvibration.
3. 3.1.2.
3.1
The
Fabrication
stress
Effects
analysis
shall
include
the
effects
of manufacturing
processes
on material
properties. The
material
blades
ultimate
should
processes
3. 3.1.2.
be
and
3. 2
surface
Calculations
as
the
models
following
should
Base
used
for the
natural
include
in
The of
degree
the
for
of the
results
hydrogen
shall
include
(i.e.,
be
employed
74
the
M-1
Mark the
Fluid
75.
15-F
of the
depends
be
used
dovetailed
to nominal
Methods
and
fixity
should
fluid:
of vibration.
references
10-times-size the
used
in the
effects
of
design
of the
manufacturing
blade.
effects
should
of base
designer
data
mass
mode
the
Effects
frequencies
force
that at base loads equivalent "builtin" at the base.
the
limit
reflect
the
effects
of
blade
taper,
pre-twist,
(e.g.,
ref.
and 72).
camber)
as well
Additionally,
the
be considered:
Experimental
Virtual
endurance
that
manufactured
geometric
centrifugal
fixity:
judgment
blade
and
environment. that
of
strength, specimens
and Environmental
of and
effect
yield from
finishes
Geometric
geometry Analytical
strength, obtained
In
stator
virtual-mass
will
effects
vane
33,
be small,
in air,
with
and
(from water
of reference denser
fluids
geometry
of this effect
data
75. the
32.
be assumed
blade
ref.
frequency
that
Vibration natural
Verification
of Natural
reduction
testing
of" prototype
Frequencies or actual
frequencies.
77
blades
shall
verify
the
calculated
to be
and for
a
with
in liquid
significant. 3.3.1.2.3.3
Note
are given 74)
are compared Note
the
analyzed.
in figure
could
on the
and
being
shown
magnitude oil,
attachment,
design
are
depend the
method
of
the blade
experimental
vibrating but
type
specific
blades
speed,
for determining figure
the
the
rotor
design
analytical-prediction
effect
on
for
blade
is
I
F
I
_FREQUENCY
FIRST
FLEXU_L
sooo__t__.__----l---.--------/
NOTE:
SEE
FIGURE
20
FOR
M-|
--_
DOVETAIL .O5
25001_
DIMENSIONS
_2ooo_
,--LOGAR.T. .C 4
iooo-
"___ -CENTRIFUGAL AT NOMI_L
5°° t
Effect
.03
°
• 02
_
z
.01
LOAD DESIGN
SPEED
500
I
I
IOOO
1500. BASE
Figure 32. -
D-
I_-IE° °'VALENT I
O
.04
LOAD,
of base load on blade natural
I
I
2000
2500
I 3000
LBF
frequency
and damping
(M-1 dovetail).
TEST MODE FLUID
O
1.0--
O
\
•
WATER
\
OIL
A D
r-
PREDICTED
FIRST
a.
'*0.5 =, u.
w
_v
'
500
NATURAL
Figure
33.-
Effects
of fluid
I • 1500
1000 FREQUENCY,
virtual mass on Mark
78
HZ
15-F vane natural
frequency
(data from
ref. 74).
In view
of the
uncertainties
blading,
it is recommended
through
a frequency
within
the
effect,
3. 3.1.2. 3.4
Resonance
Campbell frequencies
with
for
frequencies
in the and
of low-aspect-ratio
on prototype
known
or actual
potential
vibration
operational
of elasticity,
shall
under should
and
forcing analysis,
blades
frequencies experimental
environmental
effects
fluid virtual-mass
effects).
potential
frequencies
(i.e.,
be used
to forcing for other ports).
3.3.1.2.3.5
Self-Excited
from
forcing
by
the proximity
frequencies.
Figure
to resonance 17
of blade
shows
the
natural
recommended
applicable.
frequencies sources
to determine
forcing
margins
be examined return-flow
be separated
all conditions.
potential
In addition
size shall
the
frequencies
to account
in modulus
natural
be conducted
to encompass
of these
frequencies
margin
diagrams
testing
the
Margin
proximity-to-resonance
Blade
use
change
natural
adequate
bench
be modified
centrifugal
in predicting
sufficient
In the
should
Blade
that
range
pump.
magnitudes
involved
due
to wakes
of excitation
from
(e.g.,
adjacent
blade
thrust-balance
rows,
system
the
pump
should
or bearing-coolant
Vibration preclude
self-excited
vibration.
It is recommended that the empirical frequency-parameter ref. 39) be used to avoid self-excited vibration:
2zr
rule
noted
below
(adptd.
from
ft C
_t-
_> 1.6
(16)
_>
(17)
Wl
2rr fb C _b W1
where
_t = torsional ft = first
frequency
torsional
parameter
frequency,
Hz
79
0.33
fb = first flexural frequency,Hz C = blade chord length, ft wl = fluid relative velocity at stall mid-radius, ft/sec _b = flexural frequency parameter
3.3.1.3
PROFILE
3.3.1.3.1
TOLERANCES,
Tolerances
Profile
and Surface
tolerances
performance
the
longitudinal The
that of
direction.
specification
used
polish
"Out-of-spec" that
local
rotor
stator
hydrodynamic
be held
finish
should
consider
Surface
-There that
blades,
A fillet radius equal information (ref. 48) for be used
FILLET
RADII
affect
blade
hydrodynamic
finishes
do
in. be specified fairing
within
in
on the
both
the
basic
profile
transverse
and
¼°.
the manufacturing
of 63 /_in. rms
technique
or better
that
will be
are recommended.
be permitted.
not
meet
On
the
especially In no
in case
from
the
J-2
specifications other the
hand,
engine ("out
should
(Mark
of spec")
small
trailing-edge
program
usually
deviations region,
an out-of-spec
15-F
that can
seriously
condition
are
prevail that
pump) of small
in all the affect
would
the affect
be accepted.
Fillet radii shall considerations.
outlined
continuous
is evidence
performance.
Fillet Radii
should
should
not
3.3.1.3.2
practice
angles
should
integrity
of -+ 0.002
Blade
blades.
structural
applicable
tolerance and
hydrodynamically.
or
not adversely
smooth
the
conditions
consequence
shall
a
marks
parts.
AND
adequacy.
of surface
in producing
Transverse
finish
a maximum
restriction
FINISH,
Finish
and surface
or structural
It is recommended with
SURFACE
the
be as small
within
to the maximum thickness indicates a stress-concentration
recommended in
as possible
section
in assessing
fillet-to-blade 3.3.1.2.3.
The
of
the
stress-concentration
thickness factors
80
imposed
by structural
the blade is recommended. factor of approximately
reference
ratios.
limits
ratio noted for
for use above
other
in the
or other fillet-to-blade
Available 1.1 would be blade suitable
design data
thickness
3.3.2
Blade Attachment
3.3.2.1
METHODS
The
blade
attachment
manufacturing, An
appropriate
method
designed.
A
study
which
in
single
should
be made mechanically
large
practice
during
3.3.2.2 3.3.2.2.1
Single-tang
corresponding in sizing
that
determined
The
method
retention
of
considerable will be reacted
The
data
to retain blade
the
particular
be recommended. design
the
pump
phase
are
of the
evaluated
turbopump.
in applications
expense
of
to prevent
being
A configuration
considerations
consideration
be made
those
in pure
must
attachment
airfoil
weight).
If
incorrect
The
requiring mechanically
assembly.
on
in the
airfoil
axially
all
probable
be
applied,
withstand
loading
M-1
pump
are
recommended
and vibratory-load that
condition in section
blade
the
A steady-state
stress
shall
failure.
used
state
defined
under
the
cause
This
the
margin
would
is shown
cause
on figure
if
condition airfoil
failure
16 and
is the
at the should same
as
3.3.1.2.3.
in the load
dovetail
slot
conditions.
because
should If
it is difficult
provide
shear
pins
to ensure
that
positive are the
used, load
shear.
predicted
stress
as defined
to
practice
Stress
predicted
blades,
to a stress
Vibratory
alternating The
the
at
on
(cost)
receive
cost
are utilized.
dovetail.
by the
safety
properly
should
wouM
blades
the
used
lower
similar
attached
be used
3.3.2.2.2
which
dovetails
dovetail
depends
manufacturing
should
provision
attached
to that
mechanically
of weight,
Strength
mechanically
equivalent
compromise
DESIGN
Structural
For
blades
or preliminary
blades
(potential
MECHANICAL
an acceptable
cannot and
conceptual
are selected,
reflect
the
therefore weight,
the
lots
blades
shall
considerations.
attaching
attached
production
attached
for
assembly,
use
of
method
and assembly
state
in the
attachment
shall
be
less
than
the
allowable
stress. steady-state by
modified
and vibratory Goodman
stresses diagrams.
81
should
be compared
Modified
Goodman
with
material diagrams
property should
be
constructed with adequatesafety factors applied to yield strength, ultimate strength, and endurancelimit. It is recommendedthat the diagrambe constructed in accordancewith the practice defined in section3.3.1.2.3. It is recommended that the maximum stress in the neck section of the dovetail be determined by methods basedon the photoelastic test results(ref. 76). The vibratory stress magnitude should include an appropriate stress-concentrationfactor (ref. 48). Generous fillets should be used.
3.3.3
Rotor
3.3.3.1
CONFIGURATION
The
basic
rotor
compromise
of weight,
A recommendation cannot
configuration
for
properly
be
size,
a basic
made.
(one-piece critical
or builtup)
speed,
configuration
Both
builtup
and
shall
cost,
and assembly
that
would
one-piece
interface, builtup
from
a single
tie bolt,
and
concpet,
one-piece
3.3.3.2
welded
forging bearing
and
MECHANICAL
recommended
Axial Thrust
3.3.4.1
TYPES
should
be examined
phase and a suitable choice Size permitting, a one-piece
this
construction
problems
problems
that
that
may
may
be
made rotor
precludes
(1) disk
be associated
with
associated
with
a a
practices
for mechanical
design
of the
rotor
are presented
Balance System
OF SYSTEMS
thrust-balance
It is recommended type
weldment-quality
for all applications
DESIGN
3.3.4
system
because
misalignment
optimum
configuration.
Design criteria and in reference 6.
The
(2)
is preferred, journal
an acceptable
considerations.
configurations
during the turbopump conceptual or preliminary design after evaluation of assembly methods, weight, and cost. machined
be
reflect
(e.g.,
system that
shall preclude
excessive
a self-compensating
so-called
series-flow
thrust
thrust-balance or
82
double-acting)
loads system
on the bearing. be used.
depends
on
The
choice
the
particular
of
turbopump design. Each type should be examined during the conceptual or preliminary design phase of the turbopump to determine compromisesin terms of recirculating-flow requirements(pump performancepenalty), net thrust load magnitude anddirection over the pump operating range,and potential instability.
3.3.4.2
MECHANICAL
3.3.4.2.1
Design Basis
The
design
the the In view both
of the
pump and turbopump. of the
at
balance
DESIGN
thrust-balance
turbine
over
uncertainties
design pistons
point
system
the
total
involved
and
over
be designed
system,
3.3.4.2.2
the axial
Structural
The
excess
load
design
effects
It is recommended design This
speed stress
of
that
that and
3.3.4.2.2.1
Balance
the
10 percent
the
development not possible
the
thrust-
and
net
axial
thrust
operating
and
range,
of
range
turbine
of
axial
thrust
it is recommended
with
provision
in the
that
system
program. During the initial to counteract thrust with
by thrust
stress
Piston
due
with
caused
balance
differential
to
10 percent
stress
Axial deflection differential shall
than
capability
the
pump
operating
be reacted
centrifugal,
be combined
the piston interface.
It is recommended
of
is at least
should
should
on
transient
to
phase of a thrust
bearings.
Strength
mechanical
combined loads.
load
and
predicting
turbopump
permit trimming during the turbopump the start transient, where it is normally balance
be based
in accurately
the
with
shall
steady-state
the
by
pressure,
centrifugal
above
the
stress
of the
and
load
nominal
due
differential
based
on
dijJ'eretztial
be determined design
to maximum thermal
be
thermal
at a mechanical
speed
of the
differential
contraction
the
turbopump.
pressure
at the
across
piston/shaft
Deflection
at the outer diameter of' not adversely affect the flow that
shall
piston
the
total
piston axial
be
piston
sized travel.
83
the balance system.
so that
outer
piston
diameter
due
axial
to
pressure
deflection
is less
3.3.4.2.3
Balance Piston/Pump
Contact
at the
operating
practices
are presented
piston
and
that
stops
It is recommended rubbing
during
interface
in reference
the
Orifice
start
stationary
or gall with
SYSTEM
STABILITY
3.3.4.3.1
shaft
shall be positive
under
all
orifice
the
shall not
Reference
should
mating
make
in the bearing
transient.
orifices
6.
Contact
be incorporated
on impact
3.3.4.3
pump
stationary
the turbopump
a precaution,
shatter
and
Balance Piston/Stationary
The balance condition.
As
piston
Contact
conditions.
Recommended
3.3.4.2.4
balance
Shaft
contact
package
3 presents
be fabricated
rotating
surfaces
from
(sec.
at any
to avoid
operating
balance
recommended a material
piston
practices. that
will not
3.5).
Range of Stable Operation
The
thrust-balance
Dynamic
analysis
that
will be stable
given
in references
system
should for
shall
be conducted
all turbopump
52, 54, and
for liquid-hydrogen
systems
at high pressures
•
Increasing
cavity
•
Decreasing
cavity
•
Increasing
total
The
Two-Phase thrust-balance
operating
increased
Operating
over
the
to establish
55. From
•
3.3.4.3.2
be stable
these stability
(increased
turbopump
conditions. is achieved
Methods by
modulus)
volume drop.
Flow system
shall
not
be sub/ect
84
to two-phase/'low.
configuration
of dynamic
it can generally
area
pressure
range.
a thrust-balance-system
references,
bulk
operating
analysis
be concluded
are that
It is recommended that the thrust-balance-systemreturn flow be introduced into the pumping system at a point where the pressurelevel is greaterthan the vapor pressureof the recirculating fluid. Conditions within the thrust balancesystemshould be examinedand the local static pressureof the fluid kept abovefluid vapor pressureat all points within the flow circuit.
3.4
PUMP STATOR
3.4.1
Vanes
3.4.1.1
PROFILE
Design profile
TYPES
criteria and recommended types (section 3.3.1.1).
3.4.1.2 With
MECHANICAL the
for vane
3.4.1.3 Design
criteria
fillet
of
radii
3.4.2
3.4.2.1
are the
design
TOLERANCES,
and
for
centrifugal-load
mechanical
PROFILE
practices
vane
profile
types
are
the
same
as for blade
DESIGN
exception
practices
SURFACE
recommended
same
effects,
design
as for blades
FINISH,
practices
as for blades
the
are the same
for vane
(section
AND profile
criteria
and
recommended
(section
3.3.1.2).
FILLET
RADII
tolerances,
surface
finish,
and
3.3.1.3).
Vane Attachment
METHODS
The
vane
attachment
manufacturing,
particular
pump
configuration assembly
method
and assembly
As is the case with
weight,
ASSEMBLY
the
blades,
being
study
difficulty,
reflect
in and
an acceptable
compromise
of weight,
considerations.
an appropriate
designed,
made
shall
and selecting cost.
method
a
single a suitable
The
use
85
for attaching practice method
of individual
cannot should vanes
the vanes be
depends
on the
recommended.
The
include
evaluation
will in general
require
of a
stator housing with a greater envelope diameter than that required for vanesmachined integrally on segmentedrings or cylinders. Thus, from a weight standpoint, the latter method is preferable. Individual vanesshould be considered when largeproduction lots are required.
3.4.2.2
MECHANICAL
3.4.2.2.1
Structural
The
vane
cause
shall
withstand
loading
equivalent
to that
which
would
failure.
and vibratory load condition at the attachment that would cause airfoil failure should be used
exception
figure
Strength
attachment
airfoil
A steady-state in the airfoil the
DESIGN
of centrifugal
load
considerations,
corresponding to a stress in sizing the attachment.
this condition
is the
same
as that
state With
shown
on
16.
3.4.2.2.2 The
Vibratory
Stresses
predicted
stress
alternating The
attachments in
should
a blade
The
practice
with
include
3.4.2.2.3
attachment
shall
dovetail
be below
to the
vane
Goodman
It is recommended
the
allowable
process
should projection (sec.
should diagrams.
factors that
compared
Modified
applied the
be
Goodman
to yield
diagram
With
strength,
be constructed
3.3.1.2.3. be fillet
3.3.2.2.2).
stress-concentration
platform
stress
safety
in section
to the
design
an appropriate
modified
projections
applied
vibratory
adequate
limit. defined
lug-type
be
and
by with
endurance
the
should
the projection
as defined
constructed
and
with
concentration used
be
strength,
in accordance
vane
steady-state
data
should
ultimate
Vane
attachment
property
diagrams
in the
stress.
predicted
material
state
factor
should
be used.
stator
assembly
examined
to determine
steady-state The (ref.
stress
vibratory 48).
if a stress
similar
stress
A generous
to that
magnitude fillet
Load Transmittal method
the housing conditions.
used shall
to transmit provide
positive
load
86
circumferential
transmittal
under
and all
axial probable
loads
to
load
from
Keys acting in shear have been successfullyutilized to transmit stator torque loads on all state-of-the-art configurations and are therefore recommended.With regard to axial load, the stator assembly should be designed to be captive in the pump housing assembly. Differential thermal contraction of the housing and stator assemblyin the axial direction should be matched to the extent that excessivelooseness(or conversely excessivestress)is avoided.
3.4.3
Stator and Volute
3.4.3.1
HOUSING
The
TYPES
stator/volute
housing
hydrodynamic, Reference and
be examined
avoid
should
excessive
diffuse
the
and
castings carefully
flow
cost have
are not
and
considered.
3.4.3.2
HYDRODYNAMIC
3.4.3.3 Design
criteria
and
and
portion
compromise
considerations. discussion
of that
tie the
from
walls
axial
passage together
noted,
achieved
either
last
flow
a weight
along
stator/volute would
or the
cast
stage
radially
the Thus,
cast
all
lead the
Welded time if the
and
proper
vanes
structures cost
guide should
total
pump
be an integral
are presented
in reference
1.
practices
are presented
in reference
1.
lead
advantages
methods
practices
87
volute
if fabrication
fabrication
housing
standpoint.
these
DESIGN
recommended
For for
design phase, considerations
the
structures
structures,
stator/volute
into
DESIGN
recommended
design
housings.
are recommended; and
however,
in practice.
welded that
with
be optimum
and hydrodynamic
structurally.
consideration,
As
been
both
the
volute
of
configuration. from
in the
it is recommended
MECHANICAL criteria
flow vanes
With
permits,
Design
the
predominant.
concept
state-of-the-art configuration
at a suitable
is a predominant
always
acceptable
and fabrication
volute
housing
is preferred
Diffuser
if weight
the
an
be made. During the conceptual or preliminary and deflection, weight, and fabrication (cost)
29)
losses,
be gradual.
be selected time
(fig.
reflect
weight,
for
basic
in arriving
volute
shall
for a detailed
practice
cannot properly load, stress
A "folded"
and
be consulted
a recommended
applications hydrodynamic
To
design
and deflection,
recommended
pumps,
should
stress
1 should
criteria axial
Housings
should assembly unit.
of be
3.4.4
Bearing Housings
3.4.4.1
TYPES
The
bearing
turbopump
rotor
requirements
imposed
Radial
and
axial
conceptual themselves if
specially same
required
by the
satisfy
radial
stiffness
critical
speed
considerations
thrust
balance
requirements
from
requirements
should
bearings
provided by
a housing
piston
orifice
(ref.
be used
housing. with
housings
preload
axial
stiffness
The Low
should
be
achieved
limit
radial
to
axial
should
will be established
assembly. constants.
spring
rate
be achieved
Rotor
axial
movement
axial
spring
constant
bearing
housing
is desired.
beyond
the
stop
the
housings for radial
locally
movement
by spring
to preclude
during
bearing values
in a
(ref.
10).
If, for example, loading
with
position
contact
axial
should
with
the
be
balance
DESIGN
Strength
mechanical
design
of
provided
if a specified
the
high
Structural
effects
standpoint,
stops
and
by
3).
MECHANICAL
The
with
are used,
in the
reacted
3.4.4.2.1
carrier
imposed
system.
of the bearing
a critical-speed
bearing
principal ball
3.4.4.2
system
stiffness
designed
preloaded stops
shall
or preliminary design phase of the turbopump should be rigid structures with high spring
stiffness, The
housing
pressure
of
the
loading,
internal
shall
loading,
be
based
external
on
the
loading,
combined
and
thermal
gradients. Internal
and
dependent (common
external on
to
the
loads specific
all designs)
determined
from
determined
in accordance
loads
at
the
flanges
thermal
may
be
turbopump propellant due
contractions,
assembly from
to thermal
be
to mount
a hot-gas
to be
and points). must
by
It
is recommended to
Rotor
will be subjected
1.2
loads
(common
practices
defined
line
installation
inertia
forces
turbopump Turbine-end be designed
gradients.
88
bearing
the
internal
to all designs) in reference components
reaction housings
to withstand
pressure
maximum
misalignment,
of attached mount
will be partially
that
times
and
criteria
caused
housing
equal
analysis. the
consider
turbine
bearing design.
assumed
with can
necessary
the
turbopump be
the hydrodynamic
differential also
to which
loads that
as
should
be
6. External line
pressure,
(sec.
3.4.5).
(dependent
separate
the stresses
pressure
and
cryogenic deflections
It on
3.4.4.2.2
Clearances
The
bearing
housing
turbopump The
pump
housing)
and Tolerances design
shall
preclude
rubbing
housing
assembly
(i.e.,
be examined
the
as a unit
front
and
rear
in establishing
bearing
the
radial clearance. As a part of this assembly, the bearing and stator alignment. Interference joints with mating tolerances
and
concentricities
suitable
stator clearances. Absolute magnitudes will dimensional tolerances should be controlled can conveniently The
radial
rotor
be achieved
deflection
(high
Adequate
the
safety
be utilized. and
spring below
factors
that
shall
The
safety
movement to
points
should
would
the
stator
axial
"built-up"
due
limit
cause
bearing
radius
the
stress
and
rotor
and
design. Axial axial clearance
to radial
radial
load
on the
deflection).
be sufficiently
rigid
The to limit
rubbing.
housing
against
total
3.4.5
Housing Interfaces
3.4.5.1
INTERFACE
AND
SEAL
design
If
analysis
will dictate
webs
ultimate
and
are
are
and the
1.1 on
method
recommended
used
in
the
0.2
percent
to be used for analysis structure,
structural housing
testing
of
bearing
housings
assembly.
and Static Sealing TYPES
Alignment housing-to-housing
and axial
alignment
interfaces of" the rotor
and
seals
relative
shall
to the
89
in of
web
are recommended.
by
turbopump
strength
methods
of 1.0 or greater
deflection of the
1.5 on ultimate
Finite-element
structures.
ratios
and
or as part
of
of the specific
analysis.
thick-shell
The
rotor
desired
protect
factors
complexity
deflection
thickness-to-fillet
3.4.5.1.1
desired
on the particular that the desired
is that
and
stator
contribute strongly to rotor should be dimensioned with
the
mount those
and
by shimming.
the
rate
volute,
rotor
housings housings
to achieve
assembly
housings,
probable
be dependent to the extent
consider
at turbopump
complex-shape,
component
stationary
failure.
stress
Verify
and
Factors
It is recommended yield
pump
should
to magnitudes
Safety
yield
analysis
structure
deflection
3.4.4.2.3
during
bearing-housing
bearing-housing local
rotating
components.
must
diametral
of
provide
turbopump
for
and
housing
maintain assembly.
radial
as a
It is recommendedthat interference-fit pilot diametersbe usedon housing-to-housingjoints and that some degree of interface fit be maintained under all interface environmental conditions. Note that the recommendedpractice here does not apply to those interfaces where an extreme temperature differential may exist (e.g., between the pump housing and turbine manifold).
3.4.5.1.2
Leakage
The
housing-to-housing
throughout Available of
seal
the
types
sealing
possible.
should and
throughout
the turbopump
up the
joint
The
In critical used.
assembly determining
type
the
conceptual
selected.
with
the
guidance
exist
in
operating
or preliminary
A seal
same
type
fluid
that
should
design
has
phase
demonstrated
be utilized
wherever
on seals.
such
steady-state
the required
bolt
loads
the joint
interface
and
remains
for symmetric operating
be calculated
in determining
housing-to-housing
that
be examined
pump
should
all
joints
range.
be preloaded should
the
assembly used
leakage
and
conditions,
preload,
the
for each
of the
tight
all operating
asymmetric and
elastic
under
loads
at the
transient-temperature
stiffness
of the parts
environmental
conditions
reliably
the preload
making and
a
stresses.
Bolt Preload method
without
probable
interface
In determining
model
3.4.5.2.2
shall
should
conditions,
conditions. spring
joint The
propellant
Strength
continuity
assembly
prevent
DESIGN
Structural
conditions.
during
seal
design
reliably
range.
applications
59 provides
Structural
interface
operating
a suitable
MECHANICAL
3.4.5.2.1
and seal shall
be examined
in previous
Reference
3.4.5.2
be
turbopump
turbopump
reliable
The
the
interface
for
preloading
exceeding joints,
In those torque
bolt
joints and
where
the
that torque
minimum
stress. of friction
at assembly
shall
induce
stresses.
it is recommended
maximum coefficient
the
allowable
The
bolt
elongation
measurement
or other is specified,
probable
coefficient
minimum
permissible
should
be used
90
of friction assembly
in determining
positive the
preload
maximum should
torque
minimum
indicators permissible
be employed and
joint
the
maximum
preload.
in
3.4'5.2.3
Safety
Design yield.
Since
the
Factors
safety
joint
factors
design
shall
is based
assembly, it is important factors of 1.5 on ultimate However,
these
safety
allowable
stresses
are
on
that
the
the joint
elastic
will
stiffness
factors within
should the
be
linear
stress
examined
range.
The
not
of
to stay within the linear range strength and 1.1 on 0.2 percent
used in comparing bolt design failure (ref. 71) is recommended.
3.5
ensure
fail
the
in either
ultimate
components
or
making
up
the
of the stress-strain diagram. Safety yield generally are recommended.
for
each
"effective"
to material
property
given
are applicable
design stress
data.
The
to
ensure
that
of the
bolt
should
be
theory
of
Mises-Hencky
the
MATERIALS
Criteria and recommended practices with liquid-hydrogen propellant.
3.5.1
Property
Selection
successfully are given treatment
3.5.2
to axial-flow
pumps
for use
Data
of materials
for
components
be based on guaranteed minimum reflect probable minumum property Typical property references such
here
in liquid-hydrogen
properties values.
or typical
data at liquid-hydrogen temperature as 61 through 66. Recommended
in liquid-hydrogen
axial
pumps
principally in references 62, for the heat-treatable alloys.
noted
64,
and
63,
are
axial-flow property
pumps
for various materials materials that have in Table 66
and
IV. Data include
shall
data adjusted
are been
for these
the
to
given in utilized materials
appropriate
heat
Ductility
Materials It is recommended
shall possess that
adequate
materials
ductility with
at liquid-hydrogen
an elongation
at liquid-hydrogen temperature be utilized yielding under steady load conditions.
for
91
temperatures.
of at least
components
that
4 percent may
in four be
subject
diameters to local
3.5.3
Impact Strength
Materials
for
adequate
impact
If impact
components
loading
strength
that
strength
be subject
at liquid-hydrogen
is anticipated,
(Charpy
may
to
of
at
least
loading
shall
possess
temperature.
it is recommended
V-notch)
impact
12
that
ft-lbf
the
(or
materials
possess
equivalent)
at
an impact
liquid-hydrogen
temperature. In
particular,
materials
rubbing should combinations
for
thrust-balance-system
components
not shatter on impact have demonstrated
or gall with non-shattering
thrust-balance-systems
and
liquid-hydrogen
mating and
that
may
surfaces. The anti-galling
hydrostatic
be
subject
to
following material characteristics in
bearings
and
are
therefore
Stationary
Component
recommended: Rotating
Component
K-Monel Inconel
718
Titanium
(tungsten-carbide
plated)
(Ti-A 1 10-AT-ELI)
3.5.4
Endurance
Materials
for
vibratory
stress
Experimental
data
components
defining
the
only
it may
be necessary
turbopump.
However,
blades
or
subject
shall possess
selected, of
on a few selected
vanes,
experimental
data
comparable
to those
bronze
Leaded
bronze
Leaded
bronze
(ref.
77)
Limit
obtained of the
Leaded
for with
to
adequate
combined
endurance
endurance
limit
at liquid-hydrogen
alloys.
If endurance-limit an endurance
the
significance
specimens
of the production
clearly that
of the indicates
reflect
92
data ratio
appreciable
that
temperature
have
available
for the
are not
in the preliminary
endurance-limit
manufacturing
component.
and
limit.
to assume
example,
steady-state
final
magnitude designs
should
processes
and
been alloy
design
phase
in the
design
be surface
based
on
finishes
APPENDIX Conversion
Physical
of U.S. Customary
A Units
Conversion
U.S. customary unit
quantity
to SI Units
SI unit
factor a
Angle
deg
rad
1.745 x 10 -2
Flowrate
gpm
m3/sec
6.309x10
Force
lbf
N
4.448
Headrise
ft
m
3.048x10
ft-lbf/lbm
J/kg
2.989
ft-lbf
J
1.356
ft
m
3.048x10-1
in.
cm
2.54
Load
Ibf
N
4.448
Mass
Ibm
kg
4.536x10
-1
NPSH
ft
m
3.048x10
-1
ff-lbf/lbm
J/kg
2.989
N/cm 2
6.895x10
N/m 2
4.788x101
rpm
rad/sec
1.047x10
-1
psi (lb f/in. 2)
N/cm 2
6.895x10
-1
/_in.
/_m
2.54x10
oF
K
ft 3
m 3
2.832x10
-2
gal
m 3
3.785x10
-3
Impact
energy
Length
Pressure
psi (lbf/in. psf(lbf/ff
Rotational
speed
Stress Surface
finish
Temperature
Volume
aExcept conversion quantities, Revision,
for
temperature, factor see
NASA
to
where obtain
Mechtly, SP-7012,
the
equivalent
E.
A.:
The
2) 2)
conversion value International
is in
SI
made unit.
System
as For of
1973.
93
shown,
Units.
Physical
-1
-1
-2
5 K
multiply
a complete
-s
value
listing
of
Constants
given
in
9 (°F +459'67)
U.S.
conversion
factors
and
Conversion
customary for
basic
Factors.
unit physical Second
by
APPENDIX
B
GLOSSARY Definition
Term
allowable
load (or stress)
the load that, if exceeded in the slightest, structural element under consideration. buckling, yielding, ultimate, or fatigue
produces failure of the pump Failure may be defined as failure, whichever condition
prevents the component from performing its intended Allowable load is sometimes referred to as criterion load allowable
to material
ratio of blade height (or length)
aspect ratio balance dram (balancing
stress is equivalent
drum)
strength.
to chord length
special balancing device used to balance axial thrust pumps; it can be used in combination with an automatic or alone (seldom)
base fixity
index of the relative tightness in the mounting or the vane in the vane support
blockage
decrease in effective and end wall
blockage
factor
function. or stress;
in multi-stage balancing disk
of the blade in the rotor
flow area due to the boundary
layer on the blades
the fraction or percentage by which design flow area is increased to account for blockage; conversely, the ratio of flow area corrected for blockage to design flow area
cavitation
formation
of vapor
bubbles
in a flowing
liquid whenever
the static
pressure becomes less than the fluid vapor pressure chord length
linear distance between the end points of the blade-profile leading and trailing edges as measured on the chord line (a line joining the points of intersection of the blade profile leading edge and trailing edge with the mean camber line)
creep
permanent deformation than the load necessary
of material caused by a tensile load that is less to yield the material; some time is required to
obtain creep critical speed
cryogenic
shaft rotational
speed
system coincides
with a possible forcing frequency
fluids or conditions
95
at which
a natural
at low temperatures,
frequency
of a rotor/stator
usually at or below -238°F
Definition
Term
trade
curvic coupling
name
generated
of the
Gleason
in a manner
Works
for a face-gear
type
of coupling
similar to that used for bevel gears
design load (or pressure)
product
of the limit load (or pressure)
and the design safety factor
design stress
the stress, in any structural element, that results from the application of the design load or combination of design loads, whichever condition results in the highest stress
deviation
angle between fluid outlet direction camber line at the trailing edge
angle
an index of local diffusion
D-factor (diffusion
and the tangent
on the blade suction surface:
factor) w2
end wall
(DF)R
=
1--
(DF)s
=
1 -
surface of the housing
endurance (fatigue
limit limit)
forced-vortex
free-vortex
flow
flow
hydrogen
ratio embrittlement
W1
V3 -V2
Awu +
2owl
AVu +
2oV2
and rotor hub between
maximum alternating stress at which an infinite number of cycles flow in which the fluid tangential other than inversely with radius
adjacent
a material
presumably
can endure
is constant
velocity varies inversely
from hub to tip while
with radius
ratio of rotor radius at blade hub to rotor radius at blade tip loss of ductility newly formed
in a metal
as a result
of the exposure
of the metal
to
gaseous hydrogen
impulse stage
stage in which there is no change in static headrise
incidence
angle between
angle
blades
velocity is forced to vary in a manner
flow in which the fluid axial velocity the fluid tangential
hub/tip
to the blade mean
fluid-inlet
direction
and tangent
across the rotor to blade mean camber
line at leading edge limit load (or pressure)
maximum expected load (or pressure) that will occur in a structure under the specified conditions of operation, with allowance for statistical variation
96
Definition
Term
Mach number
ratio of the speed of fluid flow to the speed of sound in the fluid
magnification
ratio of the deflection produced by an alternating produced by a steady load of the same magnitude
factor
the fraction by which load or stress
margin of safety (MS)
the allowable
MS
net positive
suction
head
NPSH =
(NPSH) proof pressure
=
1 --R
load to the deflection
load or stress exceeds
the design
1
-
total fluid pressure - fluid vapor _uid density
design pressure multiplied by the proof-test is the reference from which the pressure
pressure
]
at inlet
safety factor (proof pressure levels for acceptance testing
are established) radial equilibrium
flow condition in an annular passage in which there is no radial velocity component; i.e., the fluid pressure forces in the radial direction.are in equilibrium
reaction recovery
with the centrifugal
the ratio of static headrise moment
retardation
factor
bending caused radial line
in the rotor to static headrise
by centrifugal
an index of blade-passage
forces
force in a blade
that is tilted
from
a
diffusion: w2
(RF)R
in the stage
wl
V3 (RF)s
V2
root
juncture
of blade and rotor hub
safety factor
an arbitrary multiplier (or divider) greater than 1 applied in design to account for uncertainties in design, e.g., variations in material properties, fabrication
quality, and load distributions
within the structure
solidity (blade)
ratio of blade chord length to blade spacing
stacking
imaginary line on which the centers of gravity of the prone stacked to form the blade or vane shape from hub to tip
axis (or line)
97
sections
are
Definition
Term
stagger angle
the angle between the chord line and a reference is the axis normal to the plane of the blade row
stall
loss of pumping capability surface of the blades
stall margin
margin between pump operation at the design-point flow coefficient and operation at the flow coefficient at which the pump will stall
thrust-balance
untwist
forces
flow
direction
as a result of flow separation
flow through the thrust balance system that area) force necessary to balance axial thrust that
provides
on the suction
the (pressure
forces acting on a twisted reduce the blade twist
blade
virtual mass
mass of fluid near a vibrating
blade that vibrates with the blade
volute
spiral-shaped portion stage of a pump
of the housing
produce
that usually
a torque
tending
X
to
that collects the fluid from the last
Definition
Symbol C
chord
C-4
designation
D
diameter
Ds
specific
diameter,
DF
diffusion
factor
DN
index to bearing
length for a family
of airfoil
shapes
D s = DH¼/Q _
speed capability,
in mm and rotational
the product
speed (N) in rpm
ELC
extra low carbon (content)
F
material
f
frequency
g
acceleration
due to gravity
ge
gravitational
constant,
strength
Ibm-ft
98
32.17
lbf-sec 2
of bearing
bore size (D)
Definition
Symbol H
headrise,
i
fluid incidence
k
stress-concentration
MS
margin of safety
N
pump rotational
H = H 2 -H
1 (stage)
angle factor
speed
specific speed, Ns = NQY_/H 3A NPSH
net positive suction head
O/F
ratio of mass flowrate
Q
volume flowrate
P
pressure
R
(1) reaction (2) ratio of design load or stress to allowable
RF
retardation
r
radius
S
blade tangential
of oxidizer
to mass flowrate
of fuel
load or stress
factor
spacing
suction specific speed, S s = NQ'A/(NPSH) 3A SF
safety factor
TIG
tungsten-inert-gas
T73
designation alloys
U
blade tangential
V
fluid absolute
velocity
W
fluid velocity
relative to blade
(welding method)
for a heat-treating
and
velocity
\
99
tempering
process
for aluminum
Definition
Symbol Z
cavitation-breakdown
correlation
parameter,
Z = q_tan (/3T/2)
65 series
NACA designation for a family of airfoil shapes stagger angle fluid angle deviation angle efficiency blade camber angle
P
hub-to-tip radius ratio, p = rH/r T frequency parameter
o.
_-R
(1) stress; (2) solidity, cavitation
parameter,
o = C/S z R = NPSH/(u2/2gc)
flow coefficient,
_ = Vm/u
head coefficient,
_ = gcH/u 2
total-pressure-loss
coefficient:
Hloss
Subscripts a
axial
alt
alternating
b
flexural
cf
centrifugal
e
endurance
eq
equivalent
exit
outlet
forces
"
100
c_R =
w12/2gc
H loss ;_s=
V22/2ge
Subscripts
f
fluid
ff
fluid forces
H
hub; hydraulic
i
ideal
1
liquid
m
meridional;
R
rotor
S
stator
SS
steady
T
tip
t
torsional
tu
tensile
ultimate
ty
tensile
yield
u
tangential
v
vapor
1
rotor
inlet
2
rotor
outlet
3
stator
(fluid) mean
state
outlet
or stator
inlet
or second
rotor
Material
Identification
CRES
corrosion-resistant
helium
pressurant
Inconel
inlet
718
trade
name
steel
helium of
(He)
per MIL-P-27407
International
5597A)
101
Nickel
Co.
for
nickel-base
alloys
(AMS
Identification
Material K-Monel
trade name
of International
alloy containing
Nickel Co. for a wrought,
age-hardenable
Ni, Cu, and A1
leaded bronze
copper alloy containing
LH2
liquid hydrogen
LOX
liquid oxygen,
polyurethane
any of various thermoplastic polymers that contain-NHCOO-linkages; produced as fibers, coatings, flexible and rigid foams, elastomers, and resins
Ti-A 110-AT-ELI
zinc and lead
(H2), propellant propellant
an extra-low-interstitial interstitial controlled
grade per MIL-P-27201
grade per MIL-P-25508
(ELI)
grade
notch toughness, and ductility - 423°F (LH 2 temperature) 300 Series
of Ti-5A1-2.5Sn
in which
elements O, N, and H and the substitutional element at lower-than-normal contents; strength-to-density
series of austenitic
remain
at acceptable
the
Fe are ratio,
levels down
to
stainless steels
(e.g., 304, 310,347) 304L (304 ELC)
extra-low-carbon variety of 304 austenitic steel; used in weldments for corrosive conditions where intergranular carbide precipitation must be avoided
2024
wrought
aluminum
alloy with Cu as principal
alloying element
7075
wrought
aluminum
alloy with Zn as principal
alloying element
ABBREVIATIONS Identification
Organization AF
Air Force
AIAA
American Institute
ASME
American
NAA
North American
NACA
National
Society
for Aeronautics of Mechanical
Aviation,
Engineers
Inc.
Advisory Committee
102
& Astronautics
for Aeronautics
(now NASA)
Identification
Organization NREC
Northern
PWA
Pratt & Whitney
WADC
Wright Air Development
Research
and Engineering
Aircraft
103
Center
Corporation
REFERENCES 1. Anon.: Liquid Rocket Engine Centrifugal Flow Turbopumps. Monograph, NASA SP-8109, December 1973. 2.
Anon.: Liquid Rocket Engine Turbopump NASA SP-8052, May 1971.
3. Anon.: Liquid Rocket Engine Turbopump NASA SP-8048, March 1971.
Inducers.
Turbines.
Bearings. NASA Space Vehicle
NASA Space
8. Severud,
L. K.; and Reeser, H. G.: Analysis
Rocket
Vehicle
Engines.
Design
Criteria
Monograph,
NASA
NASA Space Vehicle Design Criteria
NASA
Space
of the M-1 Liquid Hydrogen
Whirling Speed and Bearing Loads. NASA CR-54825, 9.
Design Criteria Monograph,
Seals. NASA Space Vehicle Design Criteria
6. Anon.: Liquid Rocket Engine Turbopump Shafts and Couplings. Monograph, NASA SP-8101, September 1972. 7. Anon.: Turbopump Systems for Liquid Monograph, NASA SP-8107, August 1974.
Design Criteria
NASA Space Vehicle Design Criteria Monograph,
4. Anon.: Liquid Rocket Engine Turbopump Rotating-Shaft Monograph, NASA SP-8121, February 1978. 5. Anon.: Liquid Rocket Engine SP-8110, January 1974.
NASA Space Vehicle
Aerojet-General
Vehicle
Turbopump
Design
Criteria
Shaft Critical
Corp., Dec. 20, 1965.
Gunn, S. V.; and Dunn, C.: Dual Turbopump Liquid Hydrogen Feed System Experience. presented at the Ninth Liquid Propulsion Symposium (St. Louis, MO), Oct. 25-27, 1967.
10. Graham, R. D.; Rowan, B. F.; and Shen, F. A.: Axial Pump Rotordynamic Study, NASA CR-72302, R-6982, Rocketdyne Div., N. Am. Aviation, Inc., Oct. 1, 1967. 11. Gunter,
E. J., Jr.: Analysis
Lab. for Engineering
of Rocketdyne
Nuclear Feed System Turbopump
Paper
Final Report.
Failure. W-12-229,
Res.
Sciences, Univ. of Virginia, June 1966.
12. Childs, D. W.: The Space Shuttle Main Engine High Pressure Fuel Turbopump Rotordynamic Instability Problem. Paper 77-GT-49, Gas Turbine Conference (Philadelphia, PA), March 27-31, 1977. 13. Choy, K. C.; Gunter, E. J.; and Allaire, P. E.: Rotor Transient Analysis 1 -Theory; Vol. 2 -Computer Program. Rep. 528144 (MAE 77/105), 1977. 14. Balje, O. E.: A Study on Design Criteria ASME, Series A, vol. 84, 1962, pp. 83-114.
and Matching
105
by the Modal Method. Vol. Univ. of Virginia, Sept. 8,
of Turbomachines.
J. Eng. Power,
Trans.
15. Crouse, J. E.;Montgomery, J.C.;andSoltis,R. F.: Investigation of thePerformance of anAxialFlow PumpStageDesigned by the BladeElementTheory- DesignandOverallPerformance. NASATN D-591,February1961. 16. Crouse,J. E.; and Sandercock, D. M.: Blade-Element Performance of 0.7 Hub-TipRadiusRatio Axial-Flow-Pump Rotorwith Tip DiffusionFactorof 0.43.NASATND-2481,September 1964. 17. Miller,M. J.; andCrouse, J. E.:DesignandOver-AllPerformance of anAxial-Flow-Pump Rotorwith aBlade-TipDiffusionFactorof 0.66.NASATND-3024,September 1965. 18. Johnsen,I. A.; andBullock,R. O., eds.:Aerodynamic Designof Axial-FlowCompressors. NASA SP-36,1965. 19. Farquhar,J.;
and Lindley, Aerojet-General
CR-54822, 20.
Lieblein,
S.; Schwenk,
B. K.: Hydraulic Design of the M-1 Liquid Hydrogen Corp., July 15, 1966.
F. C.; and
Broderick,
Limiting Blade Loading in Axial Compressor 21.
R. L.: Diffusion
Blade Elements.
Factor
for Estimating
NACA RME53D01,
Lieblein, S.: Analysis of Experimental Low Speed Loss and Stall Characteristics Compressor Blade Cascades. NACA RME57A28, 1957.
22. Miller, M. J.; Crouse, Axial-Flow 589-599.
J. E.; and Sandercock,
Pump Rotors
23. Taylor, W. E.; Murrin, 1 -Double-Circular-Arc December 1969.
D. M.: Summary
Turbopump.
NASA
Losses and
1953. of Two-Dimensional
of Experimental
Investigation
of Three
Tested in Water. J. Eng. Power, Trans. ASME, Series A, October
1967, pp.
T. A.; and Colombo, R. M.: Systematic Two-Dimensional Cascade Tests, Vol. Hydrofoils. NASA CR-72498, UARL-H910254-50, United Aircraft Corp.,
24. Taylor, W. E.; Murrin, T. A.; and Colombo, R. M.: Systematic Two-Dimensional Cascade Tests. Vol. 2 -Multiple Circular-Arc Hydrofoils. NASA CR-72499, UARL-J910254-56, United Aircraft Corp., April 6, 1970. 25. Colombo,
R. M.; and Murrin,
Double-Circular-Arc
Hydrofoils.
T. A.: Systematic NASA CR-72870,
Two-Dimensional United Aircraft
Cascade
Tests. Vol. 3 - Slotted
Corp., May 1, 1972.
26. Murrin, T. A.; and Taylor, W. E.: Systematic Two-Dimensional Cascase Tests. Vol. 4 - Cascade Test Data. NASA CR-121101, United Aircraft Corp., March 15, 1973. 27. Shepard, _28.
D. G.: Principles
Anon.: Parametric Inc., unpublished,
of Turbomachinery.
The Macmillan
Studies. Part V of Pump Hydrodynamic pp. 33-40.
Co. (New York),
Design. Rocketdyne
Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
106
1956. Div., N. Am. Aviation,
Unpublished. Collected
*29. Campbell,W. E.; Robertson,C. F.; and Yoshikawa,D. K.: The Designand Evaluationof an Axial-FlowImpulsePump.Aerojet-General Corp.,December 1962. 30. Bissell,W. R.; King,J.A.; Umemoto,G.A.; andWong,G. S.:Investigation of LiquidHydrogen Axial FlowImpulsePump(U). Bull.7th LiquidPropulsionSymp.,vol. 2,JohnsHopkinsUniv.,November 1967,pp.233-258.(CONFIDENTIAL) 31. Howell,A. R.: FluidDynamicsof Axial Compressors. Proc.Inst.Mech.Engrs.(London),vol. 153, 1945,p.445. 32. Crouse,J. E.; Soltis, R. F.; and Montgomery,J. C. Investigationof the Performanceof an Axial-Flow-Pump StageDesigned by the Blade-Element Theory-Blade ElementData.NASATN D-1109,December 1961. 33. Crouse, J.E.;andSandercock, D.M.: DesignandOver-AllPerformance of anAxial-Flow-Pump Rotor with aBlade-TipDiffusionFactorof 0.43.NASATND-2295,May1964. 34. Miller,M. J.; andSandercock, D.M.: BladeElementPerformance of Axial-Flow-Pump Rotorwith Blade-Tip DiffusionFactorof 0.66.NASATND-3602,September 1966. 35. Stripling,L. D.; andAcosta,A. J." Cavitationin Turbopumps. Parts1 and2, J. BasicEng.,Trans. ASME,SeriesD,vol.84, 1962,pp.326-350. 36. Serovy,G. K.; and Tysen,J. C.: Predictionof Axial Flow Turbomachine Performance by Blade ElementMethods. Paper61-WA-135, ASMEWinterAnn.Mtg.,Nov.26-Dec.1,1961. 37. King.J.A.: TestingPumpsin Air.Paper67-WA/FE4, ASMEWinterAnn.Mtg.,Nov.12-17,1967. 38. Lieblein,S.;andRoudebush, W. H.: Low SpeedWakeCharacteristics in Two-Dimensional Cascades andIsolatedAirfoil Sections. NACATN-3771,1956. 39. Armstrong,E. K.; andStevenson, R. E.: SomePracticalAspectsof Compressor BladeVibration.J. RoyalAeron.Soc.,vol.64, no.591,March1960,pp. 117-130. 40. Kemp,N. H.; andSears,W. R."Aerodynamic Interference BetweenMovingBladeRows.J. Aeron. Sci.,vol. 20,no.9,September 1953,pp. 585-597,612. 41. Smith,J. E.: Vibrationof Bladesin AxialTurbomachinery. Paper66-WA/GT-12, ASMEWinterAnn. Mtg.,NewYork,NY,Nov.27-Dec.1,1966. 42. Traupel,Walter:ThermischeTurbomaschinen. ErsterBand,SpringerVerlag(Berlin;Gottingen; Heidelberg), 1958. 43. Carter,A. D. S.: BladeProfilesfor Axial-FlowFans,Pumps,Compressors, etc.Proc.Inst. Mech. Engineers (London),vol. 175,no. 15,1961,pp.775-806. Dossier source
for
material
the
design
available
criteria
monograph
for inspection
"Liquid
at NASA
Lewis
Rocket Research
107
Engine Center,
Axial-Flow Cleveland,
Turbopumps." Ohio.
Unpublished.
Collected
44. Anon.:J-2Program,QuarterlyProgress Report,PeriodendingAugust31, 1963.NAA-RR-2600-12, Rocketdyne Div.,N.Am.Aviation,Inc. 45. Sawyer,J. W.,ed.: GasTurbineEngineering Handbook.GasTurbinePublications, Inc. (Stamford, CT),1966. 46. Severud,L. K.; and Chinn,T.: AnalyticalandExperimentalVibration Analysisof the Turbine Bucketsfor theM-1LiquidOxygenTurbopump.NASACR-54830, Aerojet-General Corp.,1965. 47. Regan,P. J.: MechanicalDesignof the M-1 Axial Flow Liquid HydrogenFuel Pump.NASA CR-54823, Aerojet-General Corp.,Feb.15,1966. 48. Peterson, R. E. StressConcentration Factors.JohnWiley& Sons,Inc.(NewYork),1953. 49. Anon.: J-2 Program,QuarterlyProgress Report,PeriodEndingOctober,1960.NAA-RR-2600-1, RocketdyneDiv., N. Am. Aviation,Inc., November,1960. 50. Anon.: Development of a 1,500,000-1b-Thrust (NominalVacuum)LiquidHydrogen/LiquidOxygen Engine.QuarterlyTech.Prog.Rep.4014-02Q-2, Aerojet-General Corp.,July25, 1963. 51. Lynn, E. K.: ExperimentalStressAnalysisin the Designof a Liquid HydrogenPumpRotor. Experimental Mechanics, vol.2,no. 12,December 1962,pp. 19A-23A. 52. Watters,W. E.; andLuehr,L.: Development of Steady-State andDynamicPerformance Prediction Methods for Turbopump Self-Compensating Thrust Balance Systems.NASA CR-72630, AGC-9400-19, Rev.1,Aerojet-General Corp.,March30, 1970. 53. Young,W. E.; and Due, H. F.: Investigationof Pressure PredictionMethodsfor RadialFlow Impellers,PhaseII, FinalReport.PWAFR-1276,Pratt& WhitneyAircraft Div. of UnitedAircraft, March8, 1965. 54. Connelly,R. E.: DesignStudyof the Mark9 Pump.WADC-TN-59-122, Rep.R-1354,Rocketdyne Div.,N. Am.Aviation,Inc.,March1959. *55. Anon.:DynamicStabilityStudyof a Series-Flow ThrustBalanceSystem.AppendixB, R6809P-1, Rocketdyne Div.,N.Am.Aviation,Inc.,1968,unpublished. 56. Stepanoff,A. J.: Centrifugal andAxialFlowPumps.Seconded.JohnWiley& Sons,Inc.(NewYork), 1957. 57. Radkowski,P. P.; Davis,R. M.; andBodul,M. R.: A NumericalAnalysisof the Equationof Thin Shellsof Revolution.ARSJ.,vol.32,no.1,January1962,pp.36-41. 58. Becket,E. B.; andBrisbane,J. J.: Applicationof the Finite-Element Methodto StressAnalysisof SolidPropellantRocketGrains.Rep.S-76,voi. I (AD 474031),November1965;vol. II, part 1 (AD 476515)andvol.II, part2 (AD 476735),Rohm& HaasCo.,January1966. * Dossier for thedesign criteriamonograph "LiquidRocketEngine Axial-Flow Turbopumps." Unpublished. Collected source material available forinspection atNASALewisResearch Center, Cleveland, Ohio. 108
59. Anon.:LiquidRocketDisconnects, Couplings, Fittings,FixedJoints,andSeals.NASAStihee Vehicle DesignCriteriaMonograph, NASASP-8119, September 1976. 0.
61.
Blakis, R.; Lindley, B. K.; Ritter, Ji A.; and Watters, W. E.: Initial Test Evaluation Hydrogen Turbopump Including Installation, Test Procedures, and Test Results. Aerojet-General Corp., July 20, 1966. Sessler, J. G.; and Weiss, V.: Aerospace Structural Metals (Syracuse University), Air Force Materials Lab., March 1967.
62. Williams, L. R.; Young, J. D.; and Schmidt, Thermal Expansion R-6981, Rocketdyne
Handbook.
Elevated
Janser, G. R.: Summary Corp., July 22, 1966.
64.
Inouye, F. T.; Hunt, V.; Janser, G. R.; and Frick, V.: Application Components. NASA CR-788, Aerojet-General Corp., June 1967.
of Alloy
65.
Inouye, F. T.: Properties of Large 7079 Aluminum NASA CR-513, Aerojet-General Corp., July 1966.
in a Cryogenic
Matin, Craig: MPR-3-251-369,
Technology
Engineering
and
63.
*66.
of Materials
2 vols. ASD-TDR-63-741,
E. H.: Design and Development
Properties of Aerospace Materials at Cryogenic Div., N. Am. Aviation, Inc., March 30, 1967.
of theM-1 Liquid NASA CR-54827,
of M-1 Engine. NASA CR-54961,
Alloy Forgings
Typical Low-Temperature Mechanical Properties of Several Rocketdyne Div., N. Am. Aviation, Inc., 1963, unpublished.
Handbook
of
Temperatures.
Aerojet-General
718 in M-1 Engine
Environment.
Materials.
Memo
67.
Anon.: Development of a 1,500,0004b-Thrust (Nominal Vacuum) Liquid Hydrogen/Liquid Oxygen Engine. Quarterly Tech. Prog. Rep. 2555-04Q-1, January - March 1965, Aerojet-General Corp., April 20, 1965.
68.
Carter, A. D. S.: The Axial Compressor. Sec. 5 of Gas Turbine Principles ed., D. Van Nostrand Co., Inc. (New York), 1955.
69.
Vavra, M. H.: Aero-Thermodynamics 1960.
70.
Horlock,
71.
Roark,
72.
Anon.: The Vibration of Blades in Axial Turbomachinery, Part 1: Theory and Practice of Design and Development. Rep. 1088-1 (AD 645156), Northern Research and Engineering Corp. (Cambridge, MA), April 29, 1965.
73.
Anon.: The Vibration of Blades in Axial Turbomachinery, Part 2: Design and Development Handbook. Rep. 1088-2, Northern Research and Engineering Corp. (Cambridge, MA)' April 29, 1965.
J. H.: Axial Flow Compressors. R. J.: Formulas
and Flow in Turbomachines.
Butterworth
for Stress and Strain.
Fourth
Scientific
and Practice,
John Wiley & Sons, Inc. (NY)
Publications
ed., McGraw-Hill
(London),
1958.
Book Co. (New York),
Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
109
Sir H. R. Cox,
1965.
Unpublished. CoUected
*74.
Turner,
J. D.: The
Rocketdyne 75.
Bisplinghoff, 1962.
76.
Heywood,
77.
Young, Pressure
Div.,
Effect
of Fluid
N. Am. Aviation,
R. L.; and Ashley,
R. B.: Designing W.
E.;
and
Cryogenic
AFRPL-TR-67-130,
Inc.,
On Stator
1963,
H.: Principles
Turbopumps, &
J.M.: Final Whitney
Blade
Resonant
Frequency.
SM 3111-8073A,
unpublished. of Aeroelasticity.
by Photoelasticity.
Reddecliff, Pratt
Density
Chapman
and Hall,
Investigation
of
Report,
March
Aircraft
John
Div.,
Wiley & Sons,
Ltd. (London),
Hydrostatic 1, United
Bearings
1966
-March
Aircraft,
Inc. (New
York),
1952. for 31, May
use
in High
1967 15,
(U). 1967.
(CONFIDENTIAL)
* Dossier for the design criteria monograph "Liquid Rocket Engine Axial-Flow Turbopumps." source material available for inspection at NASA Lewis Research Center, Cleveland, Ohio.
110
Unpublished.
Collected
NASA SPACE VEHICLE DESIGN CRITERIA MONOGRAPHS ISSUED TO DATE
ENVIRONMENT SP-8005
Solar Electromagnetic
SP-8010
Models of Mars Atmosphere
SP-8011
Models of Venus Atmosphere
SP-8013
Meteoroid Environment March 1969
SP-8017
Magnetic Fields-Earth
SP-8020
Surface Models of Mars (1975),
Revised September
SP-8021
Models of Earth's
(90 to 2500 km), Revised
SP-8023
Lunar Surface Models, May 1969
SP-8037
Assessment
and Control
of Spacecraft
SP-8038
Meteoroid Environment October 1970
Model-1970
SP-8049
The Earth's
SP-8067
Earth Albedo and Emitted
Radiation,
July 1971
SP-8069
The Planet Jupiter
December
1971
SP-8084
Surface Atmospheric Revised June 1974
SP-8085
The Planet Mercury
SP-8091
The Planet Saturn (1970),
June 1972
SP-8092
Assessment June 1972
of Spacecraft
Radiation,
111
(1974),
Revised December
(1972),
(Near Earth
and Extraterrestrial,
Atmosphere
1974
Revised September
Model-1969
Ionosphere,
and
Revised May 1971
1972
to Lunar Surface),
March 1969
Magnetic
1975 March 1973
Fields, September
(Interplanetary
1970
and Planetary),
March 1971
(1970),
Extremes
(1971),
Control
(Launch
and
Transportation
Areas),
March 1972
Electromagnetic
Interference,
SP-8103
ThePlanets Uranus,Neptune,andPluto(1971),November 1972
SP-8105
Spacecraft Thermal
SP-8111
Assessment
and Control
SP-8116
The Earth's
Trapped
SP-8117
Gravity Fields of the Solar System,
SP-8118
Interplanetary
SP-8122
The Environment
Control,
May 1973
of Electrostatic
Radiation
Charges, May 1974
Belts, March 1975 April 1975
Charged Particle Models (1974), of Titan (1975),
March 1975
July 1976
STRUCTURES SP-8001
Buffeting
SP-8002
Flight-Loads
SP-8003
Flutter,
SP-8004
Panel Flutter,
Revised June 1972
SP-8006
Local Steady
Aerodynamic
SP-8007
Buckling of Thin-Walled
SP-8008
Prelaunch
Ground Wind Loads, November
SP-8009
Propellant
Slosh Loads, August
SP-8012
Natural Vibration
SP-8014
Entry Thermal Protection,
SP-8019
Buckling of Thin-Walled
Truncated
SP-8022
Staging Loads, February
1969
SP-8029
Aerodynamic May 1969
SP-8030
Transient
SP-8031
Slosh Suppression,
During Atmospheric Measurements
Ascent, Revised November During
Buzz, and Divergence,
Launch
and Exit, December
1964
July 1964
Loads During
Launch
Circular Cylinders,
and Exit, May 1965
Revised August 1968 1965
1968
Modal Analysis,
September
1968
August 1968
and Rocket-Exhaust
Cones, September
Heating
Loads From Thrust Excitation,
112
1970
May 1969
1968
During Launch
February
1969
and Ascent,
SP-8032
Bucklingof Thin-Walled DoublyCurvedShells,August1969
SP-8035
WindLoads
SP-8040
Fracture
SP-8042
Meteoroid
SP-8043
Design-Development
SP-8044
Qualification
SP-8045
Acceptance
SP-8046
Landing 1970
SP-8050
Structural
SP-8053
Nuclear and Space Radiation
SP-8054
Space Radiation
SP-8055
Prevention 1970
SP-8056
Flight Separation
SP-8057
Structural 1972
SP-8060
Compartment
SP-8061
Interaction
SP-8062
Entry Gasdynamic
SP-8063
Lubrication,
SP-8066
Deployable
SP-8068
Buckling Strength
SP-8072
Acoustic
SP-8077
Transportation
During Ascent, June 1970
Control
of Metallic Pressure Vessels, May 1970
Damage Assessment,
May 1970
Testing, May 1970
Testing, May 1970 Testing, April 1970
Impact
Attenuation
Vibration
for Non-Surface-Planing
Prediction,
Protection,
of Coupled
October
Design Criteria Applicable
Venting, November with Umbilicals
Instability
(Pogo), October
1970 to a Space Shuttle,
Revised March
1970
and Launch
Heating,
January
Stand,
August 1970
1971
and Wear, June 1971 Deceleration
of Structural
Loads Generated
Systems,
June 1971
Plates, June 1971
by the Propulsion
and Handling
113
June 1970
June 1970
Structure-Propulsion
Aerodynamic
April
June 1970
Effects on Materials,
Mechanisms,
Friction,
Ganders,
Loads, September
System, 1971
June 1971
SP-8079 SP-8082
StructuralInteractionwith ControlSystems, November 1971 Stress-Corrosion Crackingin Metals,August1971
SP-8083
DiscontinuityStresses in MetallicPressure Vessels, November 1971
SP-8095
PreliminaryCriteria for the Fracture Control of SpaceShuttle Structures, June1971
SP-8099
Combining AscentLoads,May1972
SP-8104
Structural InteractionWith Transportationand HandlingSystems, January1973
SP-8108
Advanced Composite Structures, December 1974
GUIDANCEANDCONTROL 8P-8015
Guidance andNavigation for EntryVehicles, November1968
SP-8016
Effectsof StructuralFlexibilityon Spacecraft ControlSystems, April 1969
SP-8018
Spacecraft Magnetic Torques,March1969
SP-8024
Spacecraft Gravitational Torques,May1969
SP-8026
Spacecraft StarTrackers, July 1970
SP-8027
Spacecraft RadiationTorques,October1969
SP-8028
EntryVehicleControl,November1969
SP-8033
Spacecraft EarthHorizonSensors, December 1969
SP-8034
Spacecraft MassExpulsion Torques,
SP-8036
Effects
of Structural
February
Flexibility
December
on Launch
1969 Vehicle
Control
Systems,
1970
SP-8047
Spacecraft
Sun Sensors, June 1970
SP-8058
Spacecraft
Aerodynamic
SP-8059
Spacecraft 1971
Attitude
114
Torques, Control
January
During
1971
Thrusting
Maneuvers,
February
SP-8065
TubularSpacecraft Booms(Extendible,ReelStored),February1971
SP-8070
Spaceborne DigitalComputerSystems, March1971
SP-8071
Passive Gravity-Gradient LibrationDampers, February1971
SP-8074
Spacecraft SolarCellArrays,May1971
SP-8078
Spaceborne ElectronicImagingSystems, June1971
SP-8086
Space VehicleDisplaysDesignCriteria,March1972
SP-8096
Space VehicleGyroscope Sensor Applications, October1972
SP-8098
Effectsof StructuralFlexibility on Entry VehicleControlSystems, June1972
SP-8102
Space VehicleAccelerometer Applications, December 1972
CHEMICAL PROPULSION SP-8089
LiquidRocketEngineInjectors,March1976
SP-8087
LiquidRocketEngineFluid-Cooled Combustion Chambers, April 1972
SP-8124
Liquid RocketEngineSelf-Cooled CombustionChambers, September 1977
SP-8113
Liquid RocketEngineCombustionStabilizationDevices,November 1974
SP-8120
LiquidRocketEngineNozzles, July 1976
SP-8107
Turbopump Systems for Liquid Rocket
SP-8109
Liquid
SP-8052
Liquid Rocket
Engine Turbopump
SP-8110
Liquid Rocket
Engine Turbines,
SP-8081
Liquid Propellant
SP-8048
Liquid Rocket Engine Turbopump
Bearings, March 1971
SP-8121
Liquid Rocket
Rotating-Shaft
Rocket
Engine Centrifugal
Gas Generators,
Engine Turbopump
115
Engines, August 1974
Flow Turbopumps, Inducers,
January
December
1973
May 1971
1974
March 1972
Seals, February
1978
SP-8101
Liquid 1972
Rocket
SP-8100
Liquid
Rocket
Engine
SP-8088
Liquid
Rocket
Metal
SP-8094
Liquid
Rocket
Valve Components,
SP-8097
Liquid
Rocket
Valve Assemblies,
SP-8090
Liquid
Rocket
Actuators
SP-8119
Liquid
Rocket
Disconnects,
September SP-8123
Liquid
SP-8112
Pressurization
SP-8080
Liquid Disks,
Engine
Turbopump
Shafts
Turbopump Tanks
Gears,
and Tank
and
March
Couplings,
1974
Components,
August
1974
1973
May
Couplings,
May
1973
November
and Operators,
September
1973
Fittings,
Fixed
Joints,
and Seals,
1976
Rocket
Rocket
Lines,
Bellows,
Systems
Flexible
for Liquid
Pressure
and Explosive
Rockets,
Regulators,
Valves,
Selection
Relief
March
SP-8064
Solid
Propellant
SP-8075
Solid 1971
Propellant
SP-8076
Solid
Propellant
Grain
Design
SP-8073
Solid
Propellant
Grain
SP-8039
Solid
Rocket
SP-8051
Solid
SP-8025
Hoses,
and
October
1975
Valves,
Check
Factors
June
in Rocket
Motor
Design,
Ballistics,
March
Structural
Integrity
Analysis,
June
Motor
Performance
Analysis
and Prediction,
Rocket
Motor
Igniters,
Solid
Rocket
Motor
Metal
SP-8093
Solid
Rocket
Motor
Internal
Insulation,
SP-8115
Solid
Rocket
Motor
Nozzles,
June
SP-8114
Solid
Rocket
Thrust
Vector
Control,
SP-8041
Captive-Fired
Testing
116
March
of Solid
Valves,
1977
Burst
1971
and Internal
Cases,
April
1973
and Characterization,
Processing
Filters,
1971
April
1970 December
1976
1975
Rocket
December Motors,
1974 March
1971
October
1972 1973 May
1971
Related Documents
Liquid Rocket Engine Axial Flow Turbo Pumps
November 2019 12
Liquid Rocket Engines Centrifugal Flow Turbo Pumps
November 2019 8
Liquid Rocket Engine Turbines
November 2019 15
Liquid Rocket Engine Turbopump Gears
November 2019 12
Liquid Rocket Engine Turbopump Inducers
November 2019 13