Prediction of Pressurant Mass Requirements For Axisymmetric Liquid Hydrogen Tanks
Prediction of Pressurant Mass Requirements For Axisymmetric Liquid Hydrogen Tanks
Prediction of Pressurant Mass Requirements For Axisymmetric Liquid Hydrogen Tanks
AIAA-95--2964
(NASA-TM-I06973) PREOICTION OF
N95-29642
PRESSURANT MASS REQUIREMENTS FOR
AXISYMMETRIC LIQUTO HYDRCGEN TANKS
(NASA. Lewis Research Center) 8 p Unclas
N. T. Van Dresar*
Copyright _ 1995 by the American Institute of Aeronautics and Astronautics, Inc. No copyright is asserted in the United States under Title
17,U.S. Code. The government has a royalty-free licenseto exerciseallrightsunder the copyrightchimed herein forGovernmental purlmSes.
All other rightsare reserved by the copyright owner.
Pl P2 P3 P4 P5 P6 P7 P8
Table 1 -- Fitted constants for the Epstein and Anderson 2 correlalion for hydrogen propellant.
x eXpL-.st,1--c-
[ (1"_ p6 (1-z-g)
J P7
S ]
Since publication of the revised correlation, additional
experimental data was obtained at the NASA Lewis
where Research Center for the pressurized expulsion of liquid
hydrogen from spherical and nearly spherical tanks 4-8.
The data series and references are listed in Table 2.
Wpo = pOAv (2)
Tank
Data Series Reference Tank Diameter Pressurant Gas
shape
Vail Dresar
I & Stochl 4 2.2 m Oblate _oid GH2
A total of 60 data points are available from the sources ackieved by omitting the liquid residual volume and the
in Table 2. These data points were obtained using a mass of the corresponding tank wall from the
variety of pressurant gas diffusers. Data obtained with analysis--i.e., the appropriate tank volume does not
straight-pipe gas injectors were not included as this include the liquid residual volume and the appropriate
tank mass does not include the mass of the tank wall
injector configuration results in high heat and mass
that remained wetted at the conclusion of the
transfer rates at the liquid surface ? . With a few
exceptions, the data variables fall within the ranges experiments.
specified for the Epstein and Anderson correlation as
shown in Table 3. The most significant differences are Epstein and Anderson state that their correlation may be
used when the initial ullage volume does not exceed 20
some data points having longer total outflow time and
the low ambient heat flux for the NASA data. percent of the total tank volume. For the present data
set, initial ullage volumes were from 5 to 14 percent of
In this work, the Epstein and Anderson correlation is the tank volume after correcting for the liquid residual
volume.
compared to the NASA data and a revision of the
correlation is provided.
The correlation further assumes a uniform wall
Comparison of Data to E_ostein and Anderson thickness and material. All of the NASA data was
Correlation obtained in tanks fitted with lids that were thick
compared to the tank walls, and in the case of dam from
references 5-8, the lid material differed from that of the
Although not stated, the Epstein and Anderson
correlation assumes that the tank is completely expelled tank. The adjusted tank walldensity, wall thickness and
(i.e., liquid residuals are zero). In the NASA wall specific heat capacity were obtained as follows:
experiments, expulsions were stopped at approximately
five percent liquid fill level. Therefore, when comparing t3.,
= mmnk + ml/a (6)
the predictions to the data, adjustments were made to Vta_ + Vna
correct for the liquid residuals. This correction was
4.5
.... I .... I .... I .... I .... I ' ' '_. where AV is the volume of expelled liquid and Asw is
% ! the area of wall surface swept by the liquid free surface
4.0
v." during the expulsion process (i.e., the wall surface area
initially wetted by the propellant that is exposed to gas
3.5 at the end of the expulsion). For an initially full tank
that is completely expelled, this definition is equivalent
to Epstein and Anderson's definition. Otherwise, this
a,_-3.0 definition removes the influence of both liquid residuals
L_
and warm tank walls above the initial liquid level.
2.5 When Epstein and Anderson's correlation form and
constants are used with the alternate equivalent diameter,
_Vda/ . 11I much improved results are obtained as seen in Fig. 2.
2.0 Errors range from -10 percent to +14 percent with a
v mean error of +0.6 percent. The root mean squared error
1.5 ,,,1_,_,1_,,,I,,,,1_,,,I,,,," is 4.7 percent.
1.5 2.0 2.5 3.0 3.5 4.0 4.5
CF 4.0
.... I .... I .... I .... i .... ,r' ' ''/.
exp
3.0
• _+14oovo,o
\ ........-"
• .e,_ S° #°s. _S !
Epstein & Anderson 0.330 0.281 4.26 0.857 1.50 0.312 0.160 0.986
Revised Constants 0.300 0.291 5.71 0.906 1.50 0.312 0.160 0.986
Table 4 _ Comparison of original and revised constants for the Epstein and Anderson 2 correlation
for hydrogen propellant.
1. AGENCY USE ONLY (Leave b/ank) 2. REPORT DATE 3. REPORT TYPE AND DATES COViV--KE_
WU-242-70-04
e. SUJTHOR(S)
N.T. Van Dresar
9. SPONSORING/MONITORING
AGENCYNAME(S)AND ADDRESS(ES) 10. SPO NOORING,MONIT(N_NG
AGENCY REPORT NUMBER
Prepared for the 31st Joint Propulsion Conference and Exhibit COSlX3nsmed by AIAA, ASME, SAE, and ASEE, San
Diego, Califmnia, July 10-12, 1995. Responsible person, N.T. Van Dresar, organization code 5340, (303) 947-7553,
Boulder, Cokrado.
_2.. BSTPJBUT_WAWUUU_:LrrY
STATEMENT 12b. DISTRIBUTION CODE
Unclassified
-Uulimited
Subject
Categoa3,
20
publication is available from the NASA Center for Aerospaze Information, (301) 621-0390.
Experimental data from several test sexies are compared to an existing correlation that predicu the amount of pressurant
gas mass required to expel liquid hydrogen from axisymme_c tanks. It was necessary to use an alternate definition of the
tank equivalent diameter to acconunodate thermal mass in the tank wall that is initially warm and to accommodate liquid
residuals in the tank after expulsion is stopped. With this modifw.ation, the existing correlation predicted mass require-
merits to within 14 percent of experimental results. Revision of the correlation constants using a nonlinear least-squares
fit of lhe current experimental data has a minor effect, thus supporting the vafidity of the original correlation's form, its
fitted constants, and the a]_te definition of the tank equivalent diameter.
O8
Pressurization; Fluid transfer
16. PRICE CODE
A02
17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSiFICATiON 20. uMrrATION OF ABSTRACT