Experimental and analytical studies of cryogenic propellant tank

Transcription

I
NASA TECHNICAL NOTE
EXPERIMENTAL AND ANALYTICAL
STUDIES OF CRYOGENIC PROPELLANT
TANK PRESSURANT REQUIREMENTS
by M . E. Nein und J. F. Thompson
George C. Murshull Spuce Flight Center
Huntsville, Ala.
N A T I O N A L AERONAUTICS AND SPACE ADMINISTRATION
WASHINGTON, D.
d
NASA T N D-3177 EXPERIMENTAL AND ANALYTICAL STUDIES OF CRYOGENIC
P R O P E L L A N T TANK PRESSURANT REQUIREMENTS
By M. E. Nein and J. F. Thompson
G e o r g e C. M a r s h a l l Space Flight C e n t e r
Huntsville, Ala.
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
For sale by the Clearinghouse for Federal Scientific and Technical Information
Springfield, Virginia 22151
Price $4.00
-
TABLE OF CONTENTS Page
................................................ INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PRESSURIZATION REQUIREMENTS AND LAUNCH VEHICLE DESIGN . . . . . . . .
SUMMARY
....................................
Test Facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Instrumentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Test Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
EXPERIMENTALPROGRAM
PRESSURIZATION ANALYSES
................................... i
4
8
8
......................................... 8
Summary of Analytical Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Modifications in the Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Evaluation of Program P a r a m e t e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Comparison with Test Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions from Comparisons with Test Data . . . . . . . . . . . . . . . . . . . 1 3
Previous Work
THE EFFECTS OF SYSTEM PARAMETERS ON PRESSURANT REQUIREMENT
.......................... 15
.............................................. 87
............................................. 88
CONCLUSIONS AND RECOMMENDATIONS
REFERENCES
. . 14
BIBLIOGRAPHY
iii
I I
I 1 11 U S T OF ILLUSTRATIONS Figure
Title
Page
I.
Comparison of the Weights of the Propellant Feed Systems of
Two Flight Vehicles
. . . .. .. .. . . .
19
IO.
.. . . . . . . . . . . . . . . . . . .
Weight of LOX Tank P r e s s u r a n t Versus Vehicle Thrust . . . . . . . .
Saturn I, S-I Stage; Saturn I, S-IV Stage. . . . . . . . . . . . . . . . . . .
Test Facility for Tank Configuration C . . . . . . . . . . . . . . . . . . .
Interior of Tank Configuration D. . . . . . . . . . . . . . . . . . . . . . . .
Test Facility for Tank Configuration D . . . . . . . . . . . . . . . . . . .
Location of Temperature Probes. . . . . . . . . . . . . . . . . . . . . . . .
P r e s s u r a n t Distributor, Tank Configuration D. . . . . . . . . . . . . . .
Temperature Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Comparison of Temperature Sensor Response Time . . . . . . . . . . .
Copper Plate Calorimeter . . .. . . . . . . . . . . . . . . . . . . . . . . . .
11.
Temperature Response of Copper Plate Calorimeter
. .. . .. .. . .
30
12.
Calculated Free Convection Heat Transfer Coefficients, hf, on
... .... .. .. . .. .. ... .. ....
Vertical Wall
31
Comparison Between Experimental and Computed Heat Transfer
Coefficients, Configuration C, Tests 130-9 and 130-10,
Oxygen as P r e s s u r a n t . . . . . . . . .
. . . . .
32
2.
3.
4.
5a.
5b.
6.
7.
8.
9.
13. ..
.
.
.
14. . . . . . . .. . . . . . . . .
Comparison Between Experimental and Computed Heat Transfer
Coefficients, Configuration C, Test 130-15, Helium as
Pr e s s u r ant . . .
. . ... . . . . . . .. .. .. .. ....
.. . .
15. . .
.. .. .
. . . .
. .
..
Comparison Between Ullage P r e s s u r e Loss for He and GN, PrePressurants Under Liquid Slosh and Nonslosh conditions in Tank
Configuration C . . .
.. . . . .. . . . . .... .
. .. . . . . . . . . . .
iv
. ....... ...
.
..
I
.. .
20
21
22
23
24
25
26
27
28
29
33
34
LIST OF ILLUSTRATIONS (Cont'd)
Figure
16. 17. 18. Title
Page
Measured and Computed Ullage Gas Concentration Gradients
in Tank Configuration C.
35
Experimentally Determined Mass Transfer Mt /Am
(lb/lb) Versus Time t (sec)
36
Liquid Surface Conditions During Pressurization Test in Tank
Configuration C
37
.............................
..........................
...................................
.......
38
......
39
19.
Liquid Surface and Ullage Conditions During SA-5 Flight.
20.
Experimentally Determined Radial Temperature Gradients
21a.
Comparison Between Experimental and Computed Ullage
Temperature Gradient, Tank Configuration C y Test 130-6,
Oxygen as P r e s s u r a n t .
40
Temperature Gradient, Tank Configuration C , Test 130-6,
Oxygen as P r e s s u r a n t .
41
Comparison Between Experimental and Computed Pressurant
Flowrate, Tank Configuration C, Test 130-6, Oxygen as
Pressurant..
42
Ullage Pressure and Pressurant Inlet Temperature Histories ,
Tank Configuration C, Test 130-6, Oxygen as Pressurant.
43
Temperature Gradient, Tank Configuration C, Test 130-7,
Oxygen as Pressurant.
44
Comparison Between Experimental and Computed Ullage Temperat u r e Gradient, Tank Configuration C , Test 130-7, Oxygen as
Pressurant..
45
Comparison Between .Experimental and Computed Pressurant
Flowrate, Tank Configuration C , Test 130-7, Oxygen as
Pressurant..
46
..............................
21b.
..............................
21c.
....................................
21d. 22a.
......
..............................
22b.
....................................
22c.
....................................
V LIST OF ILLUSTRATIONS (Cont'd)
Figure
22d. 23a. Title
Page
Ullage Pressure and Pressurant Inlet Temperature Histories, Tank Configuration C , Test 130-7, Oxygen as P r e s s u r a n t .
47 Comparison Between Experimental and Computed Ullage Temperature Gradient, Tank Configuration C, Test 130-9, Oxygen as Pressurant.
48 Comparison Between Experimental and Computed Pressurant Flowrate, Tank Configuration C y Test 130-9, Oxygen as Pressurant..
49 Comparison Between Experimental and Computed Pressurant Flowrate, Tank Configuration C, Test 130-9, Oxygen as Pressurant
.....................................
50 Ullage P r e s s u r e and Pressurant Inlet Temperature Histories , Tank Configuration C y T e s t 130-9, Oxygen as Pressurant
51 Comparison Between Experimental and Computed Ullage
Temperature Gradient, Tank Configuration C, Test 130- IO,
Oxygen as Pressurant.
52 Comparison Between Experimental and Computed Ullage Temperature Gradient, Tank Configuration C y Test 130-10, Oxygen as Pressurant.
53 Comparison Between Experimental and Computed Pressurant Flowrate, Tank Configuration C, Test 130-10, Oxygen as Pressurant
54 Ullage Pressure and Pressurant Inlet Histories, Tank Configuration C y Test 130-10, Oxygen as P r e s s u r a n t .
55 Comparison Between Experimental and Computed Ullage Temperature Gradient, Tank Configuration C, Test 130-15, Helium as pressurant
56 Comparison Between Experimental and Computed Ullage Temperature Gradient, Tank Configuration C y Test 130-15, Helium as P r e s s u r a n t .
57 ......
..............................
23b. ....................................
23c. 23d. 24a.
.......
..............................
24b. ..............................
24c. ....................................
24d. 25a. .........
...............................
25b. ..............................
vi LIST OF ILLUSTRATIONS (Cont'd)
Figure
Title
Page
25c. Comparison Between Experimental and Computed Pressurant
Flowrate, Tank Configuration C, Test 130-15, Helium as
Pressurant..
58 Ullage Pressure and P r e s s u r a n t Inlet Temperature Histories, Tank Configuration C y Test 130-15, Helium as P r e s s w a n t .
59 .....................................
25d. 26a. ......
Comparison Be tween experimental and Computed P r e s s u r a n t
Flowrate, Tank Configuration D, Test C 003-7a, Oxygen as
Pressurant
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 26b. Comparison Between Experimental and Computed Tank Wall Temperatures, Tank Configuration D, Test C007-7aY Oxygen as P r e s s u r a n t . .
61 Comparison Between Experimental and Computed Tank Wall Temperatures, Tank Configuration D, Test C003-7aY Oxygen as P r e s s u r a n t . .
62 Ullage P r e s s u r e and P r e s s u r a n t Inlet Temperature Histories, Tank Configuration D, Test C003-7aY Oxygen as P r e s s u r a n t
63 Comparison Between Experimental and Computed P r e s s u r a n t Flowrate, Tank Configuration D, Test COO3-12, Oxygen as Pressurant..
64 Comparison Between Experimental and Computed Tank W a l l Temperatures, Tank Configuration D, Test COO3-12, Oxygen as P r e s s u r a n t . .
65 Comparison Between Experimental and Computed Tank Wall Temperatures, Tank Configuration D, Test COO3-12, Oxygen as P r e s s u r a n t . .
66 ...................................
26c. ...................................
2 6d. 27a. ......
.....................................
27b. ...................................
27c. ...................................
27d. Ullage P r e s s u r e and Pressurant Inlet Temperature Histories , Tank Configuration D, Test COO3-12, Oxygen as P r e s s u r a n t
28a. Comparison Between Experimental and Computed P r e s s u r a n t Flowrate, Tank Configuration D, Test COO3-IO, Helium as . . . . . . 67 PreSSUraIIt....
...................................
I
vii
68 LIST OF ILLUSTRATIONS (Cont'd)
Figure
Title
Page
28b. Comparison Between Experimental and Computed Tank Wall Temperatures, Tank Configuration D, Test COO3-10, Helium as P r e s s u r a n t . .
69 Comparison Between Experimental and Computed Tank Wall Temperatures, Tank Configuration D, Test COO3-10, Helium as P r e s s u r a n t .
70 Ullage P r e s s u r e and Pressurant Inlet Temperature Histories, Tank Configuration D, Test COO3-10, Helium as Pressurant
71 Comparison Between Experimental and Computed P r e s s u r a n t Flowrate, S-I Stage LOX Tanks, SA-6 Static Test, Oxygen as Pressurant..
72 ..............................
28c. ...............................
28d. .....................................
29a. .....................................
29b. 30. Ullage P r e s s u r e and Pressurant Inlet Temperature Histories, S-I Stage LOX Tanks, SA-6 Static Test, Oxygen as Pressurant
....
Comparison Between Experimental and Computed Pressurant Flowrate, S-I Stage LOX Tanks, SA-5 Flight Test, Oxygen as Pressurant..
74 Comparison Between Experimental and Computed Pressurant Flowrate, S-IV Stage LOX Tank, Helium as P r e s s u r a n t .
75 .....................................
3ia. 73 ........
31b. Ullage P r e s s u r e and Pressurant Inlet Temperature Histories S-IV Stage LOX Tank, Helium as Pressurant . . . . . . . . . . . . . . . . 76 32a. Comparison Between Experimental and Computed Pressurant Flowrate, S-IV Stage LH, Tank, Hydrogen as Pr,essurant
77 Ullage P r e s s u r e and Pressurant Temperature Histories, S-IV Stage LH, Tank, Hydrogen as Pressurant.
78 Comparison Between Experimental and Computed Ullage Temperature Histories, Tank Configuration D, T e s t 187260, Nitrogen as P r e s s u r a n t .
79 32b. 33. ........
..................
.............................
viii
LIST OF ILLUSTRATIONS (Concluded)
Figure
34.
Title
Page
Temperature Gradient,Tank Configuration C, Test 130-10, Oxygen as P r e s s u r a n t .
80 Comparison of Pressurant Flowrate Predictions by Two Computer Programs with Experimental Results.
.............
81 Comparison of Ullage Mean Temperature Prediction by Two Computer Programs with Experimental Results. . . . . . . . . . . . . .
82 Comparison Between Experimental and Computed Pressurant
Flowrate History, Tank Configuration C y Test 130-6, Oxygen as P r e s s u r a n t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
83 Schematic of Heat and Mass Transfer Conditions in a Propellant Tank. . . . . . . . . . . . . . . . . . . . . . . . . . .
84 ..............................
35. 36. 37.
38. 39. 40. ........
Comparison Between Free Jet Velocity Decay and Forced Heat Transfer Coefficient Decay
...........................
85 The Effects of Various Design Parameters on the Mean Temperature at Cutoff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
86 ix LIST OF TABLES
.
II.
III.
I
Page
Title
Table ....................
16 ..............................
17 Tank Configurations and Test Parameters
Summary of Test Conditions
Parameters for Heat and Mass T r a n s f e r Calculations . . . . . . . . . . . . . 18 X
~
Symbol
A
:"i
AD
b3
DEFINITION OF SYMBOLS
Definition
Tank total surface area
Pres s u r ant distributor a r e a
Constants used in calculation
of gas to wall forced
coefficients (Table
m)
Constants used in calculation of f r e e
convection heat and m a s s transfer
Coefficients (Table III & Computer
Program)
Tank wall thickness
Constants used in calculations of gas to
liquid forced convection heat and m a s s
transfer coefficients
D
Tank diameter
(ft)
D
Diffusion coefficient
( f t2/h r )
EK
Modification factor f o r thermal conductivity caused by mixing of fluid
(Btu/hr f t OR)
Modification of diffusion coefficient
caused by mixing of gas
( ft2/hr)
-
ED
gC
Constant = 32.17
( l b ft/lbf
m
xi
seC2)
DEFINITION OF SYMBOLS (Cont‘d)
Definition
Symbol
gw
Ullage gas-to-wall. heat transfer
coefficient
( B t u / h r f t 2 OR)
C
Ullage gas-to-wall . free convection
heat transfer coefficient
(Btu/hr f t 2 OR)
0 Ullage gas-to-wall forced convection
heat transfer coefficient at tank top
( B t d h r f t 2 OR)
Gas-to-liquid heat transfer coefficient
(Btu/hr f t 2 OR)
Gas-to-liquid free convection heat
transfer coefficient
( B t d h r f t 2 OR)
h
h
h
h
h
S
sc
hL Liquid- to-wall he at transfer
coefficient
K
Gas thermal conductivity
L/D
Tank length to diameter ratio
ni
P r e s s u r a n t flowrate
Mt M a s s transfer
Am
P r e s s u r a n t m a s s accumulated P
Ullage p r e s s u r e r
Tank radius t
Time T
Temperature Th
Vehicle thrust V
Gas velocity V
Tank volume
xi i
DEFINITION OF SYMBOLS (Cont'd)
Definition
Symbol
vd V
Od
X
Y
Y
Y
(ft3/sec)
Reference volumetric pressurant
flowrate
Radial distance from tank wall
S
Gas-to-liquid m a s s transfer coefficient
sc
Gas-to-liquid f r e e convection mass
transfer
so
Gas-to-liquid forced correction m a s s
transfer coefficient at tank top
(ft/hr)
Axial distance from tank top
Z
Z
Volumetric pressurant flowrate at
distributor
i
Axial distance of gas-liquid inter
face from tank top
Dimensional decay coefficient of
ullage forced heat transfer coefficients
(Table III)
PLP (ft-1)
Thermal expansion coefficient of
liquid
Total time of pressurization
(set)
P
Gas viscosity
( l b / f t hr)
m
P
Gas density
(lb /ft3)
m
@
Molefraction
(-)
*T
xiii
DEFINITION OF SYMBOLS (Concluded)
Definition
Symbol
Subscripts
a
Ambient C
Calorimeter f
F r e e convection Ullage gas Interface L
Liquid m
Mean 0
Reference, pressurant inlet U Ullage W
Wall ACKNOWLEDGEMENT
The contribution of the MSFC Test Laboratory in providing the test facilities and
complex instrumentation and obtaining the experimental data is gratefully acknowledged.
Invaluable contributions in program definition and analysis of experimental results were
made by J. Moses, T. Stokes, L. Worlund, and G. Platt of the Fluid Mechanics and
Thermodynamics Branch.
NOTE:
M r . J. F. Thompson is currently Assistant P r o f e s s o r , Mississippi State
University, Aeronautical Engineering Department. He was formerly associated with
Propulsion Division, MSFC.
x iv
I
EXPER IMENTAL AND ANALYTICAL STUD IES OF CRYOGEN I C
PROPELLANT TANK PRESSURANT REQU IREMENTS
SUMMARY
The extensive requirement for pressurization of cryogenic propellant tanks of
launch and space vehicles has directed attention to the need for accurate methods of
analysis of propellant tank thermodynamics. This paper presents the results of experi
mental and analytical studies of pressurization gas requirements for cryogenic liquids.
Experimental r e s u l t s are analyzed for cylindrical and spheroidal tanks ranging in s i z e
over four o r d e r s of magnitude. A parameter study of the controllable variables of a
pressurization system design illustrates their effect on ullage gas temperature.
Pressurization data are provided for use in the development o r checkout of analy
tical pressurization models and for design of pressurization systems f o r future launch
and space vehicles. A tank pressurization computer program , using recommended coef
ficients, can be used to predict total and transient pressurant requirements and ullage
temperature gradients within 10 percent accuracy.
INTRODUCT ION
Determination of the pressurant gas weight for cryogenic propellant tanks is com
plex and defies exact analytical treatment because of the interdependent transient
phenomena of heat and mass transfer that occur simultaneously in a propellant tank.
Mathematical models describing the internal thermodynamics of tank pressurization have
been developed by various investigators.
The analysis by Clark [ i] represents an analytical solution of the governing equa
tions that predict the transient temperature, the response of the pressurant gas, and
container wall. However, the solution requires assumptions, such as constant tank pres
sure and zero initial ullage, that are not always met with real systems. The studies by
.Coxe and Tatum [ 21 are based on analysis of a system in which the ullage is thermally
mixed a'nd heat transfer between the gas and the wall is independent of time and space.
Gluck and Kline [ 31 used dimensional analysis to express gas requirements as a function
of known system parameters; they determined, experimentally, quantities of interfacial
m a s s transfer and gas phase heat transfer.
Epstein [ 41 presented a numerical method for calculation of pressurant gas re
quirements that contains a number of phenomena absent from previous analytical methods.
However, empirical data are required to evaluate many constants and physical parameters.
To provide a reliable method f o r determination of pressurant gas requirements,
the experimental data on pressurization obtained by the Marshall Space Flight Center
during the Saturn launch vehicle development were applied to the method of Epstein. The
physical p a r a m e t e r s and the previously indeterminate constants were developed. After
modification, this numerical method is capable of accurate prediction of pressurization
gas requirements f o r cryogenic propellant tanks.
PRESSURIZATION REQU IREMENTS AND LAUNCH VEHICLE DES IGN
The increasing s i z e and complexity of space launch vehicles necessitates optimi
zation studies of the many subsystems involved in launch vehicle design. The propellant
tank pressurization system is of particular importance because its weight is large in
comparison to the weight of other subsystems. Weight optimization studies of propellant
tank pressurization syste*msfor the Saturn V, S-IC stage, were used to establish the lo
cation of the oxidizer and fuel tanks within the over-all vehicle configuration (Fig. 1 and
.Ref. 5). Even the pressurization system components such as heat exchangers, pres
surant lines, and controls, weigh considerably l e s s than the pressurizing gas.
A further indication of the need f o r optimization of pressurant requirements is
illustrated in Figure 2. The pressurant-mass/tank-pressure ratios of typical launch
.vehicles is given as a function of vehicle thrust, t h r u s t being representative of vehicle
size. Although t h e r e is a g r e a t deal of difference between the propellant tank configura
tions of tactical missiles and space launch vehicles, a near linear increase occurs in
pressurant-masdtank-pressure as vehicle s i z e increases. Considering only pressurant
gas weight, it appears advantageous to use helium as a pressurant. If, however, the
weight of the p r e s s u r a n t storage containers is included in the weight of the pressurization
system, the use of helium as a pressurant in most instances r e s u l t s in a weight penalty.
F o r vehicles with high acceleration and low turbo-pump NPSH requirements, it
is possible to eliminate the pressurization system, relying only on the self pressuriza
tion of the saturated propellant (flash boiling) . F l a s h boiling pressurization, however,
r e s u l t s in high p r e s s u r a n t weight and can only be justified if it significantly simplifies ve
hicle design. Because of the infant knowledge of cryogenic tank pressurization at the
initiation of the Saturn launch vehicle development program, a long series of pressuriza
tion experiments was conducted at MSFC to obtain system design information and scaling
laws for the large propellant tanks of the Saturn I vehicle. Results of this experimental
program and correlations with analytical studies are presented in the following sections
of this report.
2
I
. _
I
---
EXPER IMENTAL PROGRAM
Test Facilities
The experimental work was conducted on five tank configurations at the Marshall
Space Flight Center:
A.
Saturn I, S-I Stage, Multiple Interconnected LOX Tanks (Fig. 3a)
B. Saturn I, S-IV Stage (Fig. 3b) LOX and LH, Tanks
C. A 6. 5 by 39-ft (DxL) cylindrical LOX tank (Fig. 4)
D.
A 13 by 26-ft (DxL) cylindrical LOX tank (Fig. 5a, Fig. 5b)
E.
A 1 by 3-ft (DxL) cylindrical LOX tank.
The test parameters for these tank configurations are compared in Table I.
Configurations A and B were flight vehicles and thus contained the standard test instru
mentation of the Saturn propellant feed system, including continuous liquid level sensors,
tank p r e s s u r e , pressurant flowrates, and supply temperature measurements. Configura
tions C, D, and E were equipped with many thermocouples along the tank axis. Thermo
couples, mounted at several r a d i i at three elevations in these tanks, allowed measure
ment of radial temperature gradients. Wall temperatures were measured in Configura
tions C and D by thermocouples on the inside and outside surfaces of the tank at several
locations. The locations of the temperature sensors in these tanks are shown in Figures
6a and 6b. Special calorimeter plates were mounted in both tanks f o r determination of
gas-to-wall heat transfer coefficients. Finally, gas sampling devices were placed at
several locations to measure ullage gas concentration gradients.
Configurations C, D , and E were equipped with heat exchangers that provide a
variable pressurant inlet temperature up to 1000 OR. The pressurant gas was introduced
at the top of the container through a distributor ( e i t h e r a deflector plate-Configuration
C and E , o r a screen arrangement-Configuration D) to minimize inlet velocities and
disturbances of the liquid surface by impinging gas jets. Figure 7 shows a typical distri
butor configuration. P r e s s u r a n t velocities at the distributor periphery are given in
Table I f o r the five t e s t configurations.
The tank Configurations C, D, and E could be sloshed at rotational o r translatory
oscillation in excess of the first critical frequency of the tank. Configurations A , C, and
D were equipped with c a m e r a s so that the conditions inside the tank could be observed.
The r e s u l t s of tests conducted with the five tank configurations are presented in Figures
13 through 38. The conditions of these tests are summarized in Table II.
3
I.
Instru mentation
Analysis of ullage gas temperature history required a temperature probe with
fast response characteristics and good accuracy. A fast response temperature probe
(Fig. sa) consisting of a fork-like support with a 30 gage CuCo welded thermojunction
was designed at MSFC. The length-to-diameter r a t i o of the thermocouple w i r e and its
distance from the fork base were determined using an analog computer representation
of the heat transfer conditions around the probe assembly. Figure 9a shows the response
time; 63.2 percent of the total temperature change w a s attained in eight seconds when
the probe was extracted from liquid oxygen into a gas circulating a t a velocity of about
.three feet per second [ 61. The response of the probes during a pressurization test w a s
.also determined (Fig. 9b) ; the fork-type thermocouple has a good response characteristic.
A thermocouple mounted on a long, rod-like support (Fig. 8 b ) , which was de
signed for liquid measurement in the high vibration environment of static and flight
testing, exhibited an extremely poor response in the gas phase as indicated in Figure
9a. Response time to 63.2 percent of total temperature change was in excess of 1 0
minutes. Commercial temperature probes of the resistance thermometer type (Fig. 8 c )
were also investigated under these conditions. Although their response was considerably
better than the flight type thermocouple (63.2 percent temperature change in approxi
mately 50 seconds), it was too slow for the pressurization studies.
P r e s s u r e measurements in the ullage space, pressurant supply lines, and liquid
discharge lines were made with close-coupled p r e s s u r e transducers to a s s u r e good
response characteristics. The pressurant flowrate and liquid discharge flowrate meas
urements were obtained with turbine type flowmeters. Liquid level before and during
the tests was measured by capacitance discrete level probes and continuous delta P
measurement of the liquid column.
Test Resu I t s Heat Transfer
.Coefficients. Heat transfer between pressurant and tank side walls
was measured during pressurization tests in Configuration C by two plate calorimeters.
Each calorimeter was a 12 by i2-inchY 30-gage copper plate mounted from teflon spacers
parallel to and at a distance of four inches from the tank wall (Fig. I O ) . Three thermo
couples, spaced to represent equal calorimeter areas and connected as a thermo-pile,
provided a temperature/time history of the copper plate before and during the tests. The
local ullage gas temperature was measured in the vicinity of .the calorimeter (Fig. 11).
The calorimeters were located I1 and 3 0 feet from the top of the test tank.
~
For determination of heat transfer coefficients, it was assumed that heat transfer
to the back side of the plate (toward tank wall) was by free convection because of the
shielding effect of the plate-to-wall arrangement. The free convection coefficients for a
one component gas were evaluated by the equation of Jackson and Eckert [ 71 ; the r e s u l t s
are plotted in Figure 12. The free convection heat transfer coefficient was also
4 calculated for two component mixtures based on the time and space dependent heliumoxygen concentration in the tank The total heat transfer to the calorimeters was +en
corrected using the calculated free convection effect on the back side. The heat transfer
coefficients. to the front of the calorimeter plates measured in Tests 130-9, -10, -15 are
presented in Figures 13 and 14 using gaseous oxygen and helium as pressurants. Ullage
gas-to-wall heat transfer was also evaluated from wall temperature measurements at -a
location 3.5 feet from the top to the tank. Wall measurements a t locations initially below
the liquid surface produced erroneous readings and were discarded. These coefficients
were corrected by subtracting the effect of external heat flux from the measured wall
temperature rise. During a flash-boiling test, which did not require pressurant flow,
the wall temperature rise indicated an external heat flux of 13 B f d m i n ft'; this compares
very favorably with a calculated f l u x of 15 Btu/min f t 2 [ 81 and confirms the method used
for correcting w i l l measurements.
Inspection of Figures 13 and 14 shows very good agreement between measured
and calculated heat transfer coefficients. It is noted that the gas-to-wall heat transfer
coefficient is definitely within the forced convection regime f o r 'the oxygen tests, but in
the f r e e convection regime f o r the helium test. AltEiough the heat transfer coefficient by
forced convection diminishes with increasing distance from the pressurant distributor ,
the free convection contribution (Eq. I ) compensates f o r this decayto such a degree that
a nearly constant heat transfer coefficient is obtained along the tank bulkhead and side
wall.
Sloshing Effects. Pressurization studies conducted at MSFC have shown that
there is little benefit derived f r o m the use of helium as a main pressurant for cryogenic
propellants. However , it was determined experimentally that prepressurization with
helium reduces p r e s s u r e decay during liquid sloshing near the critical frequency. It is
assumed that the helium acts as a buffer zone between the splashing cryogenic liquid and
th.e condensable pressurant, suppressing excessive m a s s transfer.
Figure 15 shows a typical tank p r e s s u r e history for a stationary liquid oxygen
t e s t tank as compared to a p r e s s u r e history in which the liquid sloshes n e a r the first
critical mode of oscillation [ 91. The tank was prepressurized, with either helium or
nitrogen, followed by main pressurization .during liquid expulsion with super-heated
oxygen. The. tank p r e s s u r e history during the slosh test (using helium as a prepres
surant) is nearly identical to the p r e s s u r e history of the nonsloshing expulsion test. In
contrast, prepressurization with gaseous nitrogen resulted in a marked p r e s s u r e decay
during the sloshing of the liquid, which was not evident during a nonsloshing expulsion
test with gaseous nitrogen prepressurization.
Ullage Gas Concentration Gradients. Gas flow conditions and the concentration of
helium gas in a cryogenic propellant tank during pressurization discharge were studied
in test Configurations C and D. Spectrographic analyses were made of gas samples taken
at various positions in the tanks. Samples taken at various elevations in tank Configura
tion C just before the end of the tests yielded the r e s u l t s shown in Figure 16. In the test
5
i n which helium was used for prepressurization and oxygen as the main pressurant, the
helium concentration is maximum at 12 feet above the liquid, and gradually decreases
in both directions.
The concentration of oxygen near the liquid surface is probably caused by accumu
lation of t$e gaseous oxygen that is initially in the ullage before prepressurization. F o r
comparison, Figure 16 also shows the concentration of helium above the liquid oxygen
for the case in which helium prepressurization is followed by pressurization with helium
during liquid expulsion. The oxygen concentration at 10 feet above the liquid interface
was only six percent by volume. The total amount of gaseous oxygen i n the ullage was
only slightly l a r g e r than the amount of oxygen in the ullage before prepressurization
(0.77 moles versus 0.73 moles). This indicates that interfacial m a s s transfer, although
s m a l l under these conditions, was in the form of evaporation.
Mass Transfer. A comparison of m a s s transfer r e s u l t s obtained in Configuration
C with results obtained by C l a r k [ I ] is shown in Figure 17. Condensation in excess of
30 percent of the pressurant flow was found by Clark during liquid nitrogen expulsion
tests with a I by 3-foot cylindrical tank. Similar results were obtained with the MSFC
test Configuration E , also shown in Figure 17. The mass transfer measured in test
Configuration C indicates that condensation was 5 to 10 percent. Condensation in the
l a r g e r facility is less because of the smaller wall-aredvolume ratio of a l a r g e r tank.
Comparing the condensation in the small tank with that in the large tank on t;he
basis of wall-aredvolume ratio, the values a r e approximately equal. During tests at
high pressurant inlet temperature, initial evaporation noted i n Configuration C diminished
a s the test proceeded. However, Clark had found increased condensation at higher pres
surant s e t temperatures in small tanks. These conflicting r e s u l t s point out the incom
plete knowledge of m a s s transfer.
Condition of Liquid Interface. The condition of the liquid interface in Configura
tion C and during the launch and flight of SA-5, are shown in several f r a m e s from a
movie taken inside these tanks (Figs. 18 and 1 9 ) . Violent boiling occurred during venting
of the tank before prepressurization. A s the vents w e r e closed and prepressurization
proceeded, the liquid surface became nearly quiescent before discharge. After discharge
began,disturbance of the liquid surface caused by pressurant flow and acceleration of the
liquid surface were observed; the disturbance diminished a s time and distance between
the surface and the pressurant inlet increased.
Radial Ullage Temperaiture Gradients. Radial temperature gradients obtained
with Configurations C and D a r e shown in Figure 20. In both cases the radial gradients
w e r e small, and there apparently exists little difference between the gas flow conditions
in the two tanks, even though the gas distributors, baffling, and tank diameters a r e not
comparable.
6
The temperature probes at X/ D - 0.025 in Configuration D, which are located be
tween the antislosh baffles ( F i g . 5a),recorded virtually the same temperature as probes
at smaller radii. It was concluded that the gas circulation in the tank is not appreciably
affected by the antislosh baffles, and subdivision of the tank into volume elements perpen
dicular to the tank axis is permissable for the pressurization analysis.
Axial ullage Temperature Gradients. The axial ullage temperature gradients
obtained-in tests 130-6 and 130-7 with Configuration C (Fig. 21a, 21b, 22a, and 22b)
became approximately linear as the test proceeded. These two tests were conducted
with oxygen as pressurant at about 550"R. There was a rapid increase in temperature
of about 30"R immediately above the liquid interface. in these tests, indicating that
m a s s transfer was small. In tests 130-9 and 130-10 (Figs. 23a, 23b, 24a, and 24b)
with the same Configuration with oxygen pressurant at a lower temperature, the ullage
temperature gradients are much flatter; the rapid increase in temperature immediately
above the liquid interface is still in evidence. The ullage temperature gradients in this
same configuration with helium as pressurant ( T e s t 130-15; Fig. 25a, and 25b) are con
cave, r a t h e r than linear as in the tests with oxygen as pressurant, and the increase in
temperature just above the liquid interface is very gradual. The concave shape is to be
expected in this case because the m a s s transfer is in the form of evaporation with an
ullage that is predominately helium. The linear ullage temperature gradients in tests
with oxygen as pressurant indicate that the mass transfer is very small with an ullage
that is predominantly oxygen.
Other Test Results. Tests are being performed with Configuration a, but so far
only threktests have been completed. The pressurant distributor in this configuration
was designed to minimize the gas circulation in the tank, reducing forced convection
heat transfer. While this is the desired condition for optimum pressurization system
operation, it is detrimental to the response time of the temperature probes as the liquid
interface passes. P r e c i s e ullage temperature gradients will not be available until this
instrumentation is improved. However, preliminary data, with very hot GOX used as
pressurant, indicate that the temperature gradients are concave r a t h e r than linear as
was the case in the tests with Configuration C using colder GOX as pressurant. The con
cave temperatme gradients found in the helium pressurant tests with Configuration C
were also in evidence with Configuration D. Pressurant flowrates and wall temperature
gradients from these tests are presented in Figures 26a, 26b, 26c, 27a, 27b, 27c,
28a, 28b and 2%.
Pressurant flowrates in the LOX tanks of the Saturn I, S-I stage, during static
test and flight are presented in Figures 29a and 30. Figures 31a and 32a show pres
surant flowrates in the LOX and LH, tanks of the Saturn I, s-IV stage, during static test.
Finally, ullage temperature histories obtained in a very s m a l l tank, Configuration E ,
containing LN2 pressurized with nitrogen are given in Figure 33.
7
I
PRESSURIZATION ANALYSES
P rev ious W o rk
Pressurized discharge from cryogenic liquid containers was studied ana-jtically
u nder sponsorship of the Army Ballistic Missile Agency
and experimentally by Clark [I]
and later MSFC. The analytical solutions obtained by Clark were applied to test data
obtained for Configuration C. In Figure 34 the axial temperature gradient through the
ullage gas is shown as a fuilction of distance from the tank top or gasdistributor. Excel
lent agreement with test results w a s obtained for an assumed gas-to-wall forced convec
tion heat transfer coefficient of 10 B t d h r ft2"R. Agreement for a coefficient of 2 Btu/hr
ft2"R, approximately in the range of free convection, was poor. This illustrates one
limitation of analytical solutions in which the gas-to-wall heat transfer coefficient enters
as an independent variable.
In spite of this restriction and the assumption of initial zero ullage volume, the
method by Clark was successfully applied in design analyses of the Saturn I pressuriza
tion system. While Clark's analysis assumed stratification of the ullage gas and con
stant heat transfer coefficient, the analysis by Coxe and Tatum [ 21 was based on the as
sumption of a complete thermally mixed ullage gas and constant heat transfer coefficient.
Figures 35 and 36 compare test results obtained with MSFC Configuration C with the
analytical predictions by the method of Coxe and Tatum. Toward the end of the test,
agreement is good possibly because the conditions of constant heat transfer coefficients
Gre approached in the large ullage near the end of the run.
A comparison of the pressurant flow requirements with predictions by an analog
computer simulation developed by MSFC, is shown in Figure 37. Representation of the
pressurization thermodynamics by analog method was difficult because of scaling problems
and the extreme sensitivity of the equations to tank pressure fluctuation. In Figure 35,
36, and 37 pressurant flow requirements a r e also compared with a digital computer pro
gram developed by Rocketdyne [ 41 and modified by MSFC [ IO]. This program closely
matches test data. However, the p r o g r a m insufficiently describes mass transfer and is
sensitive to fluctuations of ullage pressure. These fluctuations do not appear in the
ineasured flowrates because they are apparently counteracted by the effects of evaporation
and condensation [ I 1]
.
S u m m a r y o f Analytical Program
Since the Rocketdyne program makes maximum use of the techniques of digital
computer calculations and is not subject to the restrictive assumptions that are made in
other programs, this method was chosen by MSFC f o r pressurization system analyses.
However, extensive comparisons of the program with test data were required to evaluate'
the physical parameters and constants initially contained in the program as indeterminate
identities. The equations were modified when necessary.
8
I
This program includes in its calculations a pressurant gas storage tank, heat
exchanger, and flow control valve. It considers a propellant tank with o r witho.ut outside
insulation and pressurized with either evaporated propellant o r with a gas stored under
p r e s s u r e i n a storage tank in which the gas expands nonadiabatically. The ullage pres
s u r e is controlled by a pressurant flow control valve that has finite maximum and mini
mum a r e a s and may be either the on-off o r the continuously regulating type. In the pro
pellant tank the ullage gas may be a two component mixture of evaporated propellant and
another gas. The ullage gas temperature, composition, and properties a r e considered
functions of time and of axial, but not radial o r circumferential, distance. Liquid and
wall temperature and properties a r e treated i n the s a m e manner. The heat transfer
modes considered are shown in Figure 38. Mass transfer within the ullage and at the
gas-liquid interface is considered. The effects on heat and mass transfer caused by
gas circulation, as influenced by pressurant gas inlet velocity, is also taken into account.
Modifications in the Program
In the course of the comparisons with test data, it was necessary to make s e v e r a l
modifications in the program to obtain good data correlations. These modifications are
discussed in reference IO. The ullage gas-to-wall heat transfer coefficient, which
decreases exponentially from the tank top, is written a s the sum of a free convection
coefficient and a forced convection coefficient.*
where ho is a n input constant.
Thus the forced convection coefficient at the tank top is a linear function of the
pressurant volumetric flowrate ( e d ) from the distributor. The free convection coef
ficient (h,) is calculated by the free convection equation,
In the same manner the g.as-to-liquid heat transfer coefficient at the gas-liquid
interface is written
* Schmidt [ 121 also writes the total heat transfer coefficient a s the sum of the f r e e and
forced convection coefficients
.
9
(8)
-Ps
h s = hs c + h s o
where h
so
zi
e
is an input constant.
The free convection coefficient h
sc
is calculated by the equation
T
g
(4)
It was found that both forced convection coefficients a t the tank top could be
calculated more accurateiy by a forced convection equation of the standard form expres
sing the Nusselt number a s a function of the Reynolds and Prandtl numbers:
h
r
so = dl
k
(e)
(2)
d3
dz
Thus , the ullage gas-to-wall heat transfer coefficient and the gas-to-liquid heat transfer
coefficient at the gas-liquid interface are better calculated to equations (7) and ( 8 ) .
-PwZ
= h +hoe
�F c
h
-P,
h =h +h e
s
sc
so
zi
Y
where ho and hso are calculated by equations (5) and (6) , rather than being input as
constants, and he and hsc are calculated by equations (2) and (4).
It was also found that the liquid-to-wall heat transfer coefficients could be better
calculated according to a f r e e convection equation r a t h e r than being taken as constank
10 A s in the case of gas-to-liquid heat transfer coefficient at the gas-liquid inter
face, the m a s s transfer coefficient at the interface was written
where Y
so
is an input constant.
The free convection coefficient (Y ) is calculated by the equation
sc
The forced convection mass transfer coefficient at the tank top can be better calculated
by a forced convection equation expressing the Sherwood number as a function of the
Reynolds and Schmidt numbers:
Y
r
($d3
.
D
Thus, the mass transfer coefficient at the gas-liquid interface is calculated by
where Yso is calculated according to equation (12) rather than being input as a con
stant, and Ysc is calculated by equation (11).
Evaluation of Program Parameters
All pressurization analyses contain numerous parameters that must be known
before pressurization requirements can be predicted. These parameters determine the
heat and mass transfer coefficients and the distribution of these coefficients over the tank.
Therefore, studies were conducted to determine the relative importance of each of the
parameters involved in the program, and extensive comparisons with the results
of the tests were made to obtain ,numerical values for these parameters.
A summary of the test conditions is given in Table II, and the values of the
important parameters are given in Table III. The exponential decay coefficients Pw and
ps in equations (7) ( 8 ) and (13) are scaled by the equation:
I
p
= 0.00117
r2
.
The parameters not listed in this table are of small importance and may be taken as
zero.
Comparison with Test Data
The pressurant flowrate and ullage and wall temperature gradients predicted by
the computer program using the calculated constants from Table 111 are compared with
test data [ 13, 14, 15, 16, and 171 in Figures 21 through 30. In all comparisons the
ullage p r e s s u r e , liquid drain rate, ambient heat transfer coefficients, and ambient
temperature were input to the computer as functions of time. Either the pressurant inlet
temperature o r the heat exchanger performance curve was also input.
Figures 21 through 25 show comparisons with t e s t data obtained with Configura
tion C described in Table I and shown in Figure 4. A s can be seen from these figures,
the agreement between the computer predictions and the test data is generally good. The
irregularities in the computed pressurant flowrate, particularly marked in Tests 130-6
and 130-7 (Fig. 21 and 2 2 ) , are caused by the over-sensitivity of the program to changes
in the slope of the ullage pressure curve. Both ullage p r e s s u r e curves of Test 130-6
and 130-7 have depressions in the latter half of the runs, while the slopes of the ullage
pressure curves of the other tests were nearly constant. The agreement between the
computed and measured ullage temperature gradients was good throughout the run for
all the tests using oxygen as pressurant. In the test with helium as pressurant (130-15,
Fig. 25) the pressurant flowmeter failed. Storage bottle pressure and temperature
history were used f o r calculation of a n average flowrate. Therefore, it was not unex
pected that the computed flowrate was somewhat below this value. However, the agree
ment between computed and measured ullage temperature gradients was not as good in
this test as in the test with oxygen as pressurant. This was probably caused by deficien
cies in the program's m a s s transfer calculations from the assumption that all heat trans
fer from the ambient to the propellant is converted to sensible heat rather than latent
heat. In Tests 130-9 and 130-10 the ullage heat transfer coefficients were calculated
from calorimeter measurements anu were compared with those calculated by the com
puter. Although the assumption of exponential decay of the ullage heat transfer coeffic
ient with distance from the tank top (Eq. 7) s e e m s a r b i t r a r y , the r e s u l t s were in excel
lent agreement with the measured heat transfer coefficients (Figs. 13 and 14).
In comparing the velocity decay of a free jet (Fig. 39, discussed in Ref. 18) it
was found that the exponential decay of the forced convection heat transfer coefficient
expressed as a velocity decay (v,/vo)
is bracketed by the velocity decay of a free jet
discharging from a circular opening and that of a free jet discharging from an infinite
slit. This is analogous to the pressurant entering the tank through the gas distributor.
Comparisons with data from the LOX tanks of the Saturn I, S-I stage during static
test and flight are presented in Figures 29 and 30. The agreement between computed and
12
measured pressurant flowrate and pressurant inlet temperature is excellent. Ullage ..
temperature measurements w e r e not available in these tests because instrumentation on
flight vehicles is limited. Figure 29 shows a comparison of the computed and measured
flowrate from the flight of SA-5. The agreement was generally good, though not as good
as in the static test of SA-6. Evaluation of SA-5 p r e s s u r a n t requirements was compli
, .
cated by the complex air flow pattern around and between the propellant tanks of the
Saturn I, S-I stage during flight. The aerodynamic heating was difficult to evaluate; the
only possible approach was to use average values f o r all propellant tanks. Figures 31 and
32 show comparisons with data from the LOX and LH, tanks of the Saturn I, S-IV stage
during, static test. These tanks are not of ordinary cylindrical shape, a s can be s e e n in
Table I; the LOX tank is an oblate spheroid and the LH, tank contains a convex inward
lower bulkhead. By computer variation of the characteristic tank radius used in equa
tions ( 5 ) ,(6),and (12) , it was determined that the proper characteristic value should be
about two-thirds of the maximum radius f o r the LOX tank. This assumption is theoreti
cally justified because a cylinder having the same volume and surface a r e a as a n oblate
spheroid h a s a radius equal to 0.63 t i m e s the maximum radius of the oblate spheroid.
The agreement between computed and measured pressurant flowrate in the LOX tank is
excellent, as shown in Figure 31. Because the pressurant flowrate in the LH, tank w a s a
step function, it could not be matched a t all times. However, the general range of flowrate, a s computed and measured, is the same, and there is excellent agreement between
the computed and measured total pressurant weight.
Test r e s u l t s with Configuration D a r e shown in Figures 26 through 28. This tank
is an approximate one-third scale model of the Saturn V , S-IC stage, LOX tank. It is
the largest single cylindrical LOX tank from which test data is currently available. Com
parisons of the computer predictions with data obtained from three tests with this configu
ration is good for p r e s s u r a n t flowrates and tank wall temperatures.
The final comparison presented is with data from a very small cylindrical tank
(one foot in diameter and three feet long) with flat bulkheads (Configuration E ) . Although
pressurant flowrate measurements were not available in this test, the computed and
measured ullage temperature histories are compared in Figure 3 3 . The agreement is
not as good as obtained in Configuration C, probably because equation (14) for the scaling
of the exponential decay coefficients w a s developed for tanks with rounded r a t h e r than
flat bulkheads.
Conclusions from Comparisons with Test Data
These comparisons with test data cover a range of conditions, using oxygen,
helium, and nitrogen as p r e s s u r a n t s and liquid oxygen, liquid hydrogen, and liquid
nitrogen as propellants in tanks ranging in s i z e over four o r d e r s of magnitude. The tank
shapes were representative of those commonly used in space vehicles, namely cylinders
with various bulkhead shapes and oblate spheroids. A s a r e s u l t of the evaluation of the
many physical p a r a m e t e r s and constants involved in the equations, this program can be
used to predict total and transient pressurant flow requirements, ullage and wall
13
temperature gradients , and gas-to-wall heat transfer coefficients with an accuracy of
The numerical values of p a r a m e t e r s recommended by MSFC f o r use in the
program are given in Table III. There are presently no other values available in the
literature. The characteristic dimension used in the calculation of the exponential decay
coefficients was taken as the radius of the cylindrical section f o r cylindrical tanks. F o r
tanks of other shapes, some comparison with t e s t data was necessary to determine the
proper choice for the characteristic radius. A value of two-thirds of the maximum radius
appears acceptable for oblate spheroids.
=k5 percent.
The comparison with test data indicates a sensitivity of the program to sudden
changes in ullage p r e s s u r e . However, in most c a s e s vehicle design p r e s s u r e s a r e
either constant or v a r y in a monotonic manner. It was further found that considerable
experimental experience with pressurization s y s t e m s is required before this method of
analysis can be applied reliably to evaluate a new system.
THE EFFECTS OF SYSTEM PARAMETERS
ON PRESSURANT REQUIREMENT
Weight optimization of propellant tank pressurization systems demands that a low
pressurant density be maintained in the ullage space; this is analogous to using a gas of
low molecular weight and maintaining a high ullage mean temperature. Therefore, 30
pressurization t e s t s and 120 computer predictions were used to separate the relative
significance of various controllable p a r a m e t e r s of pressurization systems and to de
termine their influence on mean 'ullage temperature. Figure 40 presents a graphical
illustration of the relative influence of these parameters.
From a central origin, representing a reference condition (Saturn V , S-IC Stage)
for all p a r a m e t e r s , the increase (+Y) and decrease (-Y) , of the ullage mean tempera
ture at cutoff is shown as a function of variation of the p a r a m e t e r s on the abscissa. The
parameters were varied over a range expected f o r vehicle design. Thus, p r e s s u r a n t
inlet temperature c a n increase o r decrease by a factor of two from the reference condi
tion, p r e s s u r e by a factor of three, tank radius by a factor of two, expulsion time by a
factor of three, etc. It was indicated that the p r e s s u r a n t inlet temperature exerts the
greatest influence on the ullage mean temperature. Diminishing r e t u r n of this effect did
not exist within the range of investigation (530"R to 1200OR). The mean temperature
increased as the ullage p r e s s u r e was increased and also as the tank radius was increased.
Increasing the tank wall thickness, heat capacity, o r density caused a decrease in the
mean temperature. The p r e s s u r a n t distributor flow a r e a (AD) that controls the gas-to
wall forced convection heat transfer coefficient had a significant effect on the mean
temperature when the a r e a was reduced, but no effect a t all when flow a r e a was increased.
This indicates that the p r e s s u r a n t inlet velocity f o r the reference systems was chosen
at an optimum point. Figure 40 also indicates that helium pressurant must be introduced
I times higher than oxygen p r e s s u r a n t to obtain the same
into a tank at a temperature I.
14 ullage mean temperature. This confirms the results of other studies (Fig. 2) indi
cating that the benefits derived from a helium pressurization system are not based on
weight optimization.
CONCLUSIONS AND RECOMMENDATIONS
a. Pressurization data from cylindrical and spheroidal tanks ranging in s i z e
over four o r d e r s of magnitude are available for development or checkout of analytical
pressurization models and for design of pressurization systems for future launch and
space vehicles.
b. The Rocketdyne tank pressurization program, modified as described herein
and utilizing recommended coefficients, can be used to predict total and transient pres
surant requirements and ullage temperature gradients with an accuracy of h5 percent.
c. No significant radial ullage temperature gradient occurs, even in tanks with
anti-slosh baffles. This permits the assumption of one-dimensional stratification of
the ullage gas for analytical representation of pressurant requirements.
d. Heat transfer between pressurant and tank walls can differ significantly from
f r e e convection, depending on tank geometry and distributor design.
e. The strongest influence on pressurant weight is exerted by pressurant inlet
temperature, for which no diminishing return occurs within a temperature range com
patible with tank materials. Other important influencing factors are tank radius, distrib
utor flow a r e a , expulsion time and aerodynamic heating. The effect of wall heat capacity
is not as significant as might be expected.
f.
M a s s transfer for large tanks is less than previously reported.
g. Additional work is necessary to develop better techniques f o r measuring
gas concentration gradients and m a s s transfer.
George C . Marshall Space Flight Center,
National Aeronautics and Space Administration,
Huntsville, Alabama, July 12, 1965.
15 A
CONFIGURATION
C
SATURN
E
SIC 1/3
1x3 MODEL
SIV (LOX)
14.7 60
-
20-40
14.7 - 60
46
I50
150- 300
-
4 78
9 HEAT EXCHANGER
PARAMETER
D
I
CTL 114
B
TEST
PREPRESSURANT
PRESSURANT
TANK PRESSURE ( p s i a )
TIME OF DISCHARGE (sec.)
PRESSURANT TEMP (" R)
.
I
150
800
I
TOTAL VOLUME
FT3
I@l05 4 @ 70
DIAMETER (in.)
L/D
I
156
(APPROX.)
TANK MATERIAL
INSULATION
ALUM.
1 1 I
I
O:8
(FT~)
I
ss
I
;2
: : :3
ss
ss
:
I
,
0.45
ALUM.
COMMON BLKH
2.5
DISTRIBUTOR FLOW
AREA
I50 400
510
I
I
TABLE I.
TANK CONFIGURATIONS AND T E S T PARAMETERS TABLE 11. SUMMARY OF TEST CONDITIONS
I
TEST; FACILITY I ULLAGE PRESSURE (psia) INLET TEMPERATURE PR) PRESSURANT PRE-PRESSURANT PROPELLANT
65
C-003
-7h
c-003
- '21)
c-003
- 101),
D
I
450
I
GOX
'I
He
I
He
LO2
750
GOX
20
900
G OX
He
40
530
He
He
20
1) Tanks not sloshed
2) Sloshing during SA-5 flight unknown
I
1
LO,
'02
.
IPARAMETER 1 I
bl
I
VALUE
b 2
b3
dl
I
I
I
I
d 2
d3
CI
c4
I
C6
I
C8
0.06
0.8
I
I
I
I
I
I
0.333
0.06
0.8
0.333
0.I3
0.333
0.I3
I
0.333
pw=o.oo117 r
r IN FEET
0 T A B L E 111. P A R A M E T E R S F O R H E A T AND MASS
TRANSFER CALCULATIONS
18 PRESSURIZATION GAS
HARDWARE
FIGURE I. COMPARISON O F THE WEIGHTS O F THE PROPELLANT
F E E D SYSTEMS O F TWO FLIGHT VEHICLES
WEIGHT OF PRESSURANT, Wlp (Ib
-
l-r
Gadpsia)
0
0
0
1I !
.
L1
I
-I
'
CENTAUR AC-7
.
.
..
ir4
$11
I
T.
I
I
.+
;r;
.O
.... -.
JUPITER, THOR
SATURN 18/
S lV,B STAGE
r+
1II
-
i
I
--I
SATURN V /
S II STAGE
SATURN I /
S-l STAGE
- -
-
L
c)
..
r
-I
,SATURN V /
S-IC STAGE
1
m
0
FIGURE 2.
20 WEICiH'I' OF LOX TANK PRESSURANT VERSUS VEHICLE THRUST
FIGURE 38.
SATURN I, S-IV STAGE
FIGURE 3 A .
SATURN I, S-l STAGE
FIGURE 3. SATURN I,
s-I STAGE;
SATURN I,
s-IVST.AGE
21
.
.. . -
FIGURE 4.
22 TEST FACILITY FOR TANK CONFIGURATION C FIGURE 5a. INTERIOR OF TANK CONFIGURATION D
23
FIGURE 5b.
24 T E S T F A C I L I T Y FOR TANK CONFIGURATION D
-
9+
t
l
4'
'7
a
d
.a
13
0.0442
13
1.64
a
a
1
1
a
t,?
fl
7
a
Ring
0
16.65
437
431
a
a
a
-
Tank Diameter (ft) Tank Wall Thickness (ft) Number of Baffles Baffle Weight(lb/ft2) Baffle Spacing a (ft) Baffle Length r (it) Baffle Width (ft) Perforation % Cylindrical Height (ft) 39.3
Top Bulkhead Volume (ft3) 34.3
Bottom Bulkhead Volume (ft3) 48.1
a
a
a
.1
a
a
a .
a
a
CONFIGURATION D
CONFIGURATION C
FIGURE 6A
FIGURE 6 B
FIGURE 6. LOCATION OF TEMPERATURE PROBES Calorimeter
FIGURE 7.
26
PRESSURANT DISTRIBUTOR, TANK CONFIGURATION D
ELEMENT LOCATION
INSULATED WIRE
TYPICAL RESISTANCE BULB ( R T B 149AA)
(CERAMIC COATED ELEMENT ON RTB 144)
MSFC HIGH RESPONSE
THERMOCOUPLE
e; JM
-
93 - r
cu co SS TUBING 6 WALL= 35/1000
TEFLON SEAL PLUG
ROD THERMOCOUPLE
(EARLY PRESSURIZATION TESTS)
FIGURE 8. TEMPERATURE SENSORS
f.3 M
W
2
I
0
ACTUAL PRESSURIZATION TEST
CONTROLLED LABORATORY CONDITIONS
MSFC
MSFC HIGH RESPONSE T C.
T.C.
350
-3
c
. a
0
UYSHIELQED
I
W
c 40
0
c /
PESISTYNCEBULB
R ~ 1B44
r
300
SHIELDED
1
E$RLY MODEL TC.
I-
*W o
LL
0
a
IO
W
a
20
30
40
TI ME, f (sec.)
FIGURE 9.
50
60
160
70
0
B
20
40
I
1
60
80
TIME, t (sec.)
COMPARISON O F TEMPERATURE SENSOR RESPONSE TIME L
100
120
I
FIGURE IO.
COPPER PLATE CALORIIVLFTER c3 0
lor ime ter
50
FIGURE 11.
i
TIME, t ( s e c . )
IO0
L
L
TEMPERATURE RESPONSE OF COPPER PLATE CALORIMETER
I
I
I
I
I.
I
5
w
c
4
I-"
-uwz
I
I
LL
LL 3
w
0
0
a w
LL
v)
z
e
I
I '
a
I
s
W
I
I
GAS TO WALL TEMPERATURE DIFFERENCE AT(OR)
FIGURE 12. CALCULATED F R E E CONVECTION HEAT TRANSFER
COEFFICIENTS, hf, ON VERTICAL WALL
-
MEASUREMENT
COMPUTfR
TEST To (oA)P(psia)
130-9 370
62
130-10
65
450
a
a
W
bW
W
E
a
E
a
5
I-
W
sa.
,
h TOTAL
*-<-
s+
-0
130-10
c
a w
0
I - 0
TOP
n
IO
I
I
1 - I
20
30
AXIAL DISTANCE FROM TANK TOP,Z (ft.)
40
BOTTOM
FIGURE 13. COMPARISON BETWEEN E X P E R I M E N T A L AND C O M P U T E D H E A T
TRANSFER C O E F F I C I E N T S , CONFIGURATION C , TESTS 130-9
A N I 130-i0,OXYGEN AS PRESSURANT
I + NEASUREMENT
-
COMPUTER
TEST To('R)
130-15 560
w
0
01
0
5 0 1P
W
w
IO
20
AXIAL DISTANCE FROM TANK TOP,
Z (f t.)
FIGURE 14. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
HEAT TRANSFER COEFFICIENTS, CONFIGURATION C ,
T E S T 130-15, HELIUM AS PRESSURANT
P(pria)
30
-. 70
0
tn
n Y
2
60 . -
He
GN2
W
a
-I---
3
50
cn
SLOSHING LIQUID I- - - -+
w
115 e 40I
.-
tn
n
0
50
100
TIME, t (sec.)
150
70 - 0
a
Y
s
e
60 -
w
a
2
cn
50-
\
NON- SLOSH ING LIQUID
w
a FIGURE 15. COMPARISON BETWEEN ULLAGE PRESSURE LOSS FOR H e
AND GN, PREPRESSURANTS UNDER LIQUID SLOSH AND NONSLOSH CONDITIONS IN TANK CONFIGURATION C
-
100
W
e9
Q
J
80
z
02 FOR MAIN PRESSURIZATION
0
0"
l~
0
60
W
L
3
d
>
40
=.
W
I-
g
20
W
AND MAIN PRESSURIZATION
e -0
LIQUID LEVEL
IO
AXIAL DISTANCE FROM LIQUID LEVELpi(ft)
FIGURE 16. MEASURED AND COMPUTED ULLAGE GAS CONCENTRATION
GRADIENTS IN TANK CONFIGURATION C
w
-I-I
Q,
0.4
z
w
0.2
0
0
0
To
0.2
6.5' x 40'
6.5' x 40'
0.4
-
I
I
I
I
0
I ' x3'
0.6
130-7 30 130-8 MSFC 5 5 7260
42 59-D 50
5 9 - C CLARK 5 0 59-J 50 7
h
7

50
0
1G NIT1ON
IO0
I50
TIME, t ( s e d
FIGURE 17. EXPERIMENTALLY DETERMINED MASS TRANSFER Mt /A,
( LB/LB) VERSUS TIME t ( S E C )
(OR)
510
310
510
532
530
810
V i o l e n t B o i l i n g During
Venting
S t a r t of Draining
Vents C l o s e d , S t a r t of
Prepressurizat ion
End o f P r e p r e s s u r i z a t i o n
During D r a i n i n g
End of D r a i n i n g
FIGURE 18. LIQUID SURFACE CONDITIONS DURING PRESSURIZATION TEST
IN TANK CONFIGURATION C
w
Ignition
During Holddown
Cutoff
R e s i d u a l Liquid R i s i n g During The F i r i n g of R e t r o Rockets
SA-5
FIGURE 19. LIQUID SURFACE AND ULLAGE CONDITIONS DURING SA-5 FLIGHT
0
TEST
130-6
TANK
C (6.5'x40')
0
COO3-7A
D (13'x 26')
HEGHT ABOVE LIQUID LEVEL (in]
1
400 I98
n
a 350
O

.
W
a => sa W
e
E W
+
01
0 0.2
0.4
0.5 WALL
CENTER
DISTANCE FROM TANK WALL, X / D
0.1
03
FIGURE 20. EXPERIMENTALLY DETERMINED RADIAL TEMPERATURE GRADIENTS
w
c9 600 500
n
15:
' 0
Y
400
a
W
a
3
t 300 W
Q.
=
W
c
W
200
2.
iI
.
I
3
IO0
* - l - l , . - - u0 l
0
TOP
IO
20
30
AXIAL DISTANCE FROM TANK TOP, I(ft.)
FIGURE 2ia. COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED ULLAGE
TEMPERATURE GRADIENT, TANK CONFIGURATION C y T E S T 130-6,
OXYGEN AS 'PRESSURANT
40
BOTTOM
l
V
600
m
a 2 100 I
I
I
FIGURE 2ib. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
ULLAGE TEMPERATURE GRADIENT, TANK CONFIGURATION C ,
TEST 130-6, OXYGEN AS PRESSURANT
5
4
3
I
z
a
2
e
0
3
-
v)
v)
W
TEST
COMPUTER
a I
e
0
I
I
IGNITION AT t = 0
FIGURE 21c.
I
I
TIME, t (SeC.1
COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED
PRESSURANT FLOWRATE , TANK CONFIGURATION C y T E S T
130-6, OXYGEN AS PRESSURANT
I
70
60
- 50 0
.-
cn
c
PRESSURE
w
Q
3 400
40 --a
c
c
I
W
/
a 3
m
m
W
30 -a
E 300
~
a
Q W
(3
2
3
I
W
28
4
d 2001
IC
- COMPUTER
a
R
c
I
010
P
I
I
50
I
I
I
I
coo
I
1
TIME, t (sed
FIGURE 21d. ULLAGE PRESSURE AND PRESSURANT INLET TEMPERATURE
HISTORIES, TANK CONFIGURATION C y T E S T 130-6, OXYGEN
AS PRESSURANT
I
1
60
Ip
Ip
w
\
W
2
3
I
I !
I
m
FIGURE 22a. COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED
ULLAGE T E M P E R A T U R E GRADIENT, TANK CONFIGURATION C y
T E S T 130-7 , OXYGEN AS PRESSURANT
6001
1
7
FIGURE 22b. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
5
4
I
I
I
31
2
I
I
0
IGNITION AT t = 0
FIGURE 22c.
50
IO0
TIME, t ( s e c . )
COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
PRESSURANT FLOWRATE, TANK CONFIGURATION C, TEST
60
I:
h
0
.
.
.
I
600
;
t I
I
I
I
I
I
I
50 -j500 Y
I
Q.
I
c I
W
a
40
3
cn
cn W
a e 30
300 t
z
4
Q
I
a
200
3
20 '3
cn
3
cn
ac: w
IO
0
,
' = 100
Jog
1-
00 TEST
COMPUTER
1
1
I
I
1
50
1
100
I
I
1
1
150
TIME, t (sec)
HISTORIES, TANK CONFIGURATION C , T E S T 130-7, OXYGEN
AS PRESSURANT
600 500
a c
0
Y
400
L
W
a 3
!az 300
W
n
=E
W
I
200
W
c3
4 -I
3
IO0 I
1L
0
0
TOP
FIGURE 23a.
10
20
AXIAL DISTANCE FROM TANK TOP, 3 (ft.)
30
40
BOTTOM
COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED ULLAGE
T E M P E R A T U R E GRADIENT, TANK CONFIGURATION C , T E S T 130-9,
OXYGEN AS PRESSURANT
IT
0
TOP
*
W
IO
AXIAL
20
30
DISTANCE FROM TANK TW, 3 (ft.)
PRESSURANT FLOWRATE , TANK CONFIGURATION C ,
40
BOTTOM
6
cn 0
5
I
I
4
3
I
1
I
i
I
I
0 TEST
-
2
COMPUTER
I
I
FIGURE 23c. COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED
PRESSURANT FLOWRATE , TANK CONFIGURATION C y T E S T
I'
./
700
70
I
- 600
I
60 0a
.-E 50
h
0
Y
e m
G
a
3
40
i
e
z 200
v)
W
R
IO
0
Tr
I
TEMPERATURE
e
100
0 I
-
COMPUTER I
50
100
TIME, t (sec.)
FIGURE 23d. ULLAGE PRESSURE AND PRESSURANT I N L E T TEMPERATURE
HISTORIES, TANK CONFIGURATION C, TEST 130-9,
OXYGEN AS PRESSURANT 150,
600 500 n
ae:
4
0
Y
3 480,
w"
a 3
c w
e 1E w
I-
200
w
(3
a 0
0 TOP
IO
20
30 40 BOTTOM t (ft) FIGURE 24a. COMPARBON BETWEEN E X P E R I M E N T A L AND COMPUTED
U L I A G E T E M P E R A T U R E GRADIENT, TANK CONFIGURATION C y
TEST 130-10, OXYGEN AS PRESSUFtANT
60( 500 n
a
0
Y
+
3
400
a I
0 3
w
a o
t = 100 sec.)
-
COMPUTER
3
2 300
Q:
w
e
I 200 w
0-C
(3
a
4
w
100
0
0
TO P
cn
w
30
10
20
AXIAL DISTANCE FROM TANK TOP, Z ( f t . )
40
BOTTOM
5
4
3
2
0
TESTT I
0
PRESSURANT FLOWRATE, TANK CONFIGURATION C ,
T E S T 130-10, OXYGEN AS PRESSURANT
700 n
n
60- 0: 600
0
Y
a

es 50
e 500
W"
I
3
I-
I
a
d'40
se 400
W
30
E
W
I-
300
I
z
a 200
20 a
3
O
I
0
#
#
W
a
e
I 0 0 TEST
I
1
IO0 1
- COMPUTER I
I
00
I
100
zoo
300
TI M E, t (sec.)
FIGURE 24d. ULLAGE PRESSURE AND PRESSURANT INLET HISTORIES, TANK
CONFIGURATION C y TEST 130- 10, OXYGEN AS PRESSURANT
600
500
n
a
r
0
A t = 20 scc.
Y
$
c
)TEST1
t = I20 sec.
400
W
a
I-
=> s 300 COMPUTER
I
~
~
e
3i W
I
w
200
I,
3
I
c
e9
a
4 100
0
0
TOP
IO
20
30
Z (ft.)
40
BOTTOM
FIGURE 25a. COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED ULLAGE
TEMPERATURE GRADIENT, TANK CONFIGURATION C , T E S T
130- 1 5 , HE LIUM AS PRESSURANT
600 I
500
cs a 0
I
Y
-
400
I
a
Ia
Q:
W
I
0 t'70 stc.
a t= 170 sec.
c
-COMPUTER
300
Q.
E
W
I-
J
3
C 200
100
0 0
TOP
IO
20
30
40
BOTTOM
ANAL DISTANCE FROM TANK TOP, 3 (f t.
FIGURE 25b. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED ULLAGE
TEMPERATURE GRADIENT, TANK CONFIGURATION C , T E S T
130-15, HELIUM AS PRESSURANT
0.5
0.4
0.3 0.2
0.I
0
0
F
IGNITION AT t = O
0
-I-I
o TEST
450 COMPUTER
100
TIME, t ( s e c . )
PRESSURANT FLOWRATE , TANK CONFIGURATION C y
TEST 130-15, HELIUM AS PRESSURANT
I50
70
700
c
60 0a 600
Y
e
c
0
‘3;
n.
500
50
3
+
a
Y
0.
we
40 E
e 400
3
v,
W
a
s
I
v)
2
30 + 300
z
e
a
a
W
3
(3
e
20
J
g 200
A
W
3
a
e
IO
0
100 - no TEST
- COMPUTER
0
0
1
I50
100
I50
TIME, t (sed
FIGURE 25d.
ULLAGE PRESSURE AND PRESSURANT INLET TEMPERATURE
HISTORIES, TANK CONFIGURATION C , T E S T 130-15, HELIUM
AS PRESSURANT
- COMPUTER
TIME, t(mc 1
FIGLXE, 26%. COMP-4RSSON E;E:T\’I’EEN EXPGRIMENTA L AND COMPUTED
PRESSbXANT FLOWR4TE, TANK CONFIGURATION D, TEST
C 003-7a, OXYGEN AS PRESSURANT
4 tS50sec
0 t = I50sec
}
TEST - COMPUTER
20
AXIAL DISTANCE FROM TANK'TOP, E ( f t )
25 BOTTOM
FIGURE 2Gb. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED TANK
WALL TEMPERATURES, TANK CONFIGURATION D , T E S T
C007-7a, OXYGEN AS PRESSURANT
30
I
OO
TOP
i
l
5
IO
i
15
L
A X I A L DISTANCE FROM TANK TOP,
FIGURE 26c.
20
Z
I
I
25
BOTTOM
(ft)
COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED TANK
WALL T E M P E R A T U R E S , TANK CONFIGURATION D , T E S T
C003-7aY OXYGEN AS PRESSURANT
30
t
----
-
~ 4 0
u)
Q
Y A

TE MPI ER ATU RE
la
a
W
a
I ’ PRESSURE
W
II
z
a 400 a
3
I
00
(0
v)
W
1
TEST
I
2001
0
0
I
- COMPUTER
I
I
L
50
100
TIME, t ( s e c 1
150
HISTORIES, TANK CONFIGURATION D , T E S T C003-7a,
200
5
1
I
II
0
-
4
I
TEST
COMPUTER
-
3 2
T
I
50
IO0
TIME, t (sec
150
PRESSURANT F L O W R A T E , TANK CONFIGURATION D ,
T E S T COO3-12, OXYGEN AS PRESSURANT
200
5001 I
I
1
I
-I
$
100
TANK WALL TEMPERATURES, TANK CONFIGURATION D ,
TEST COO3-12, OXYGEN AS PRESSURANT
500 2
cis
cis 0 t = 100sec
A t = 191 sec
}
TEST
- COMPUTER
h
a
0
Y
3
300 W
a 3
a
t!
200
2 W
I
1
1
9
100
TOP
AXIAL DISTANCE FROM TANK TOP, Z ( f t ) BOTTOM
FIGURE 27c.
COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED
TANK WALL T E M P E R A T U R E S , TANK CONFIGURATION D ,
T E S T COO3-12, OXYGEN AS PRESSURANT
50
;IO00
TEMPERATURE
0
Y
n
2 40 u)
n
n
Y
.
E
30
3 a
W
E puuI!
I
v)
v)
W
I-
a
20
W
3
a
v)
v)
4-I =
22 400
W
IO
a 0 a 200 n
"0
1
I
50 1
1
100
1
1
150 TIME, t (sec.) FIGURE 27d. ULLAGE PRESSURE AND PRESSURANT INLET TEMPERATURE
HISTORIES, TANK CONFIGURATION D , T E S T COO3-12,
1
200
2.0
n
1
-
I
LL 1
:
r
I
0 TEST
a
a 3
g
0.s
W
a
a
- COMPUTER
1
1
0
0
1
<
I
I
I
:
I
50
100
TIME, t (sec 1
PRESSURANT FLOWRATE, TANK CONFIGURATION D,
TEST COO3-10, HELIUM AS PRESSURANT
I50
500
0 t=50sec
a
IJ
A t=14lsec
400
COMPUTER
a
CI
a 300 0
2. W
a 3
I- 200
Q:
a
~
W
n
LLL
L W
k A X I A L D I S T A N C E FROM TANK TOP, Z ( f t )
FIGURE 28b.
TANK WALL TEMPERATURES, TANK CONFIGURATION D ,
-2
0
500
I
I
I
3
A t=lOOsec T E S T
400
-
A
n
COMPUTER
e 0
Y
-
3
I-
300
W
a 3
IQ
e 200
W
e
IO 0
C
TOP
I
5
IO
15
20
B O 25
TTOM
A X I A L D I S T A N C E F R O M T A N K T O P , Z (ft)
TANK WALL T E M P E R A T U R E S , TANK CONFIGURATION D ,
30
60
1200
I-
50 -
PRESSURE
---P-@
a n
40
.0

+
2 600'
w
--=-
a
I
w
-
'J
L
w
a
a
4J
2
. . -
.
I0l
z a ,
I
I
I
I
0
0
1
I
TEMPERATURE
I
N
0
7
1
1
50
100
1
TIME, t ( s e c )
HISTORIES, TANK CONFIGURATION D , TEST COO3-10,
HELIUM AS P R E S S U M N T
I50
4
N
50
0
0
50
IGNITION AT t = O
FIGURE 29a.
100
TIME, t (sec.1
PRESSURANT FLOWRATE, S-I STAGE LOX TANKS, SA-6
STATIC TEST, OXYGEN AS PRESSURANT
I50
1400
n
1200
a
0
Y
c
c” n
.O 50-
I
W
v)
n
Y
a
n
3
1
c
a
2 40 -
a
3
U
. 1000 ,
E
TEM~ERATURE
800
Q
I
v)
v)
W
W
c
W
I2
Q
a
E 30-
600
a
a
2 20-
3
400
a W
Q.
10
-
0-
-
200 - 00 TEST
COMPUTER
0
FIGURE 29b.
ULLAGE PRESSURE AND PRESSURANT INLET TEMPERATURE
HISTORIES, S-I STAGE LOX TANKS, SA-6 STATIC TEST,
70 W
a cn
60
w .E
Eg
I
-
I-
COMPUTER
I
50
100
RANGE TIME, t (sed
FIGURE 30. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
PRESSURANT FLOWRATE , S-I STAGE LOX TANKS, SA-5 FLIGHT
TEST, OXYGEN AS PRESSURANT
I50
L
I
/
I
I
I
I
I
I
0
;
I
/
TOTAL PRESSURANT
WE1OHT:
TEST
COMPUTER
100
200
3 '0
TIME, t (sec 1
400
FIGURE 3ia. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
PRESSURANT FLOWRATE, S-IV STAGE LOX TANK, HELIUM
AS PRESSURANT
88.0 Ib.
88.41b.
SO0
-
W
a
a 20W
W
=
I-,
3
a J
J
!z
200
v)
v)
W
a
IO- a 100
I
t
I
OL
0
0
100
FIGURE 31b.
200
TIME, t (sec 1 I
I
300
400
ULLAGE PRESSURE AND PRESSURANT I N L E T T E M P E R A T U R E
HISTORIES, S-IV STAGE LOX TANK, HELIUM AS PRESSURANT
500
0.3 \
I
I
I
I
J,,,-
I
I
1
100
I
I
200
I
1
300
I
TOTAL PRESSURANT
WEIGHT:
85.8 Ib.
TEST
85.1 Ib.
COMPUTER
400
TIME, t ( s e d
FIGURE 32a. COMPAREON BETWEEN EXPERIMENTAL AND COMPUTED
PRESSURANT FLOWRATE, S-IV STAGE LH2 TANK,
HYDROGEN AS PRESSURANT 51
-
600
--
- -a5 0 0
L
P
L
-
-=+l
E400

a a b S S U R E
3
k
Q
L
- ~W 3 0 0 -
--.
A
A
A
-I
TEMPERATURE
/ e
ik
+=+dL
*A
I-
C
z - ,4200
3
COMPUTER
v)
v)
W
a
- p
-
100
OO
IO0
200
300
40 0
- 500
o n TEST
COMPUTER
a
-
0
Y
400
CI
I
I
I
1
-
w
a 3
I-
a 300
a e w
z 200
w
c-
W
c3
a
IO0
J
J
=>
0
50
100 TIME, t ( s e d
FIGURE 33. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED ULLAGE
TEMPERATURE HISTORIES, TANK CONFIGURATION D , T E S T
187260, NITROGEN AS PRESSURANT
M
0
DISTANCE FROM TANK TOP, Z(ft.)
FIGURE 34. COMPARISON BETWEEN E X P E R I M E N T A L AND COMPUTED ULLAGE
T E M P E R A T U R E GRADIENT,TANK CONFIGURATION C , T E S T 130-10, OXYGEN AS PRESSURANT -
l //
I
I
I
I
i
i
I
i
,
-
1
0
-
,
--00
50
IGNITION AT t = 0
TEST
ROCKETDYNE
PROGRAM
~
LOCKH EED
PROGRAM
IO0
TIME, t
(sec)
FIGURE 35. COMPARISON O F PRESSURANT FLOWRATE PREDICTIONS BY
TWO COMPUTER PROGRAMS WITH EXPERIMENTAL RESULTS
150 .
>
I
.
-
&
400
n
a
0
Y
E 350
T
,
I
I-
I
C.
w
a 3
I
,s300
W
e 0
s
w
I-
2
2
----
250
E
w
TEST ROCKETDYNE
;
PROGRAM
LOCKHEED
PROGRAM
c
9
a
4 50
FIGURE 36.
TIME, t ( s e d
100
COMPARISON O F ULLAGE MEAN TEMPERATURE PREDICTION
BY TWO COMPUTER PROGRAMS WITH EXPERIMENTAL RESULTS
I50
'
I o TEST
c
o ROCKETDY NE COMPUTER
-=-ANALOG COMPUTER
.
I
50 100
I50
TIME, t(sec) FIGURE 37. COMPARISON BETWEEN EXPERIMENTAL AND COMPUTED
PRESSURANT FLOWRATE HISTORY, TANK CONFIGURATION C y
T E S T 130-6, OXYGEN AS PRESSURANT
200
PRESSURANT SUPPLY
dT d w
INTERNAL INSULATION
EXTERNAL INSULATION
CONDUCTION
CONVECTION
CONVECTION dQ dQ
RADIATION --I-#O
dt dx
FIGURE 38. SCHEMATIC O F HEAT AND MASS TRANSFER CONDITIONS
IN A P R O P E L L A N T TANK
84
0.0II
10
100
DISTANCE FROM OUTLET, 310 OR Z/a
FIGURE 39. COMPARISON BETWEEN F R E E JET VELOCITY DECAY AND
FORCED HEAT TRANSFER COEFFICIENT DECAY
Tm /Tmo
rill la
r
(I-
P
2)
/
po
ADO
= 6 0 ( p s i 4)
= 17.8 (ft2)
eo
=
ha
=
=
150 (sec)
2 ( b t u / h r f t 2 OR)
Ta
530 (OR)
c p g 6 = 1 . 5 1 (btu/ft20R)
Tmo/Tm
FIGURE 40. THE E F F E C T S OF VARIOUS DESIGN PARAMETERS ON THE
MEAN TEMPERATURE A T C U T O F F
REFERENCES I.
Clark, J. A. et al, Pressurization of Liquid Oxygen Containers, The University
of Michigan under contract with the Department of the Army No. DA-20-018- o r
D-254, March 1961.
2.
Coxe, E. and Tatum, John, Mai&?ro_pe!la$
Tank -Pregsurization System Study
and Test Report, Lockheed-Georgia Company, ContractAF 04(611)6087.
3.
Timmerhaus, K. D., Advances in Cryogenic Engineering, Volume 7 and
Volume 8 (1962 and 1963).
4.
F o r t r a n P r o p a m - f o r the Analysis of a Sinffle Propellant Tank Pressurization
System, Rocketdyne, Division of North American Aviation, Inc. , R-3936-1,
September 1963.
5.
Platt, G. K . , Nein, M. E . , Vaniman, J. , and Wood, C. C . , Feedsystem
Problems Associated with Cryogenic
Propellant Engines SAE National
Aeronautical Meeting, Washington, D. C. , 1963.
6.
High Response Gas _Tem_p_er&tureProbe f o r Pressurization Studies, NASA MSFC
Memorandum R - P&VE - PTF- 64-M- 7 0 , I964.
7.
F r e e Convection
Eckert, E. R. G. , and Jackson, T. W., Analysis of Turbulent
Boundary.
8.
Atmospheric H e a t Transfer to Vertical Tanks Filled with Liquid Oxygen, A. D.
Little, Inc. ,- Special Report No. 50, Contract AF-04-(647) -130, November 1958.
9.
Moses, J. L. , and Nein, M. E. , Evaluation -of Propellant Sloshing on Pressur
- Cryogenic Containers,
- - - presented at the 1962
ant Requirement for Large Scale
Cryogenic Engineering Conference, Los Aiigeles, California.
IO.
Modification of Rocketdyne Tank- Ee_ssurization C o m p u t C P r o g r a m , NASA
MSFC Memorandum R- P&VE- PTF-64-M-83, -1964.
11.
P r e s s u r e Decay in Pressurized Cryogenic Tanks of
Liu, C. K . , Sloshing and~~Launch Vehicles, NASA MSFC Internal Note MTP-P&VE-P-G1-23, December
1961.
12.
Schmidt, E., Experiments of H e a t T r g s f e r by Natural Convection, Chem. Ing.
Tech. 28/1956/No. 3, pages 175-180, OctoberT955.
13.
MSFC Test Laboratory Reports CTL 2 , 36, 34, 37, and CTL Flash Reports
dated August 17, November 20, and December 11, 1962.
87
-
REFERENCES (Concluded)
14.
SA-I5 Static Test Report, Preliminary Summary Data, Confidential.
15.
SA-5 %light Data.
16. Results of Pressurization Tests on-LN2 Facility, NASA MSFC Memorandum
R-P&VE-PE- M-526.
17. S-IV Stage Pressurization Data, NASA MSFC Memorandum R-P&VE-PTF-64-M
38, 1964.
18. Aerospace Applied Thermodynamic Manual, Society of Aeronautical Engineers,
Section 2.
BIBLIOGRAPHY
J. E. Myers, C. 0. Bennet: "Momentum, Heat and Mass Transfer," McGraw Hill,
New York (1962).
R. B. Scott. "Cryogenic Engineering,
(1959).
D. Van' Nostrand Company, Inc., Princeton
V. J. Johnson: "A Compendium of the Properties of MaLerials at Low Temperature
(Phase I) Part I Properties of Fluids, ' I National Bureau of Standards Cryogenic
Laboratory (July 1960).
J. H. Hargis: "Thermophysical Prop-exties of Oxygen and Hydrogen..
Internal Note In R-P&VE-P-63-13 (October 1963).
88
..
NASA MSFC
NASA-Langley, 1966
M329
.._,,.-
“The aeronautical and space activities of the United States shall be
conducted so as to contribute . . . to the expansion o f human knowl
edge of phenomena in the atmosphere and space. The Administration
shall provide for the widest practicable and appropriate dissemination
of information concerning its activities and the results thereof .”
-NATIONALAERONAUTICS
AND SPACE ACT OF 1958
NASA SCIENTIFIC AND TECHNICAL PUBLICATIONS
TECHNICAL REPORTS: Scientificand technical information considered
important, complete, and a lasting contribution to existing knowledge.
TECHNICAL NOTES: Information less broad in scope but nevertheless
of importance as a contribution to existing knowledge.
TECHNICAL MEMORANDUMS: Information receiving limited distri
bution because of preliminary data, security classification, or other reasons.
CONTRACTOR REPORTS: Technical information generated in con
nection with a NASA contract or grant and released under NASA auspices.
TECHNICAL. TRANSLATIONS: Information published in a foreign
language considered to merit NASA distribution in English.
TECHNICAL REPRINTS: Information derived from NASA activities
and initially published in the form of journal articles.
SPECIAL PUBLICATIONS: Information derived from or of value to
NASA activities but not necessarily reporting the results .of individual
NASA-programmed scientific efforts. Publications include conference
proceedings, monographs, data compilations, handbooks, sourcebooks,
and special bibliographies.
Details on the availability o f these publications may be obtained from:
SCIENTIFIC AND TECHNICAL INFORMATION DIVISION
NAT1 0 NA L A E R 0 NA UTI CS A ND SPACE A D M I N I ST RAT I 0 N
Washington,
D.C. PO546