www.rasaero.

com

G THE LIQUID PROPELLANT ROCKET ENGINE

Preaaure drop in the combustor due to bat rekase. The analyeis of the
ideal rocket motor b baaed on the postulate that combustion takes place
at constant pressure. In praotice, the release of heat in the gas stream in
the combustor is accompanied by a deorewe of both static and stagnation
pressures. The decrease is dependent on the Mach number at the hot
end of the combustion chamber, and beoomes larger aa the ratio of
combustor cross-sectional area A. to throat area A& approaches unity.
The loss in stagnation prwure represents an undesirable inorease in
entropy, a decrease in the effeotive pressure ratio for expansion, and
henoe a loss in speoifio impulse. The following analysis deals with these
effects.
As before, the subscripts i, c, th, and e refer to states a t the injector
end of the combustor, the hot end of the combustor, the nozale throat,
and the nozzle exit. Except for the nonzero velocity in the oombustor,
the oonditions of the analysis are the same as for the ideal rocket motor.
Since the expansion in the nozsle ia isentropic (heat release completed
in the chamber) the Mach number at c ie obtainable from Eq. 210:

The integrated momentum equation, for frictionlw flow in a con-

stant area duct, provides an expression for the dearwe in statio pressure
in the chamber. It ia wumed that Mic! 0.

The simplest way to treat the performance for the owe of nonzero
M, is first to calculate the equivalent plenum chamber pressure and
temperature oorresponding to the oonditiow a t c, and then to oaloulate
the flow proparoiee in the nozzle on the assumption that the gw flow
originatee from this plenum chamber. The temperature T,O and pres-
sure p: of the equivalent plenum ohamber are really 'the stagnation
temperature and pressure at station a. (See UI,+4for disouaaion of atag-
nation quantities.)

Ah cp(T: -
TI) (3-18)
Eq. 3-18 is simply a statament of the con~ervationof energy, with
the kinetio energy term d;Vi absorbed in the stagnation enthalpy c,Ti.

www.rasaero.com
 C,3 DEPARTURES PROM IDEAL PERFORMANCE
Therefore, Ti is the same as T, in the ideal rocket motor of Eq. 2-7, that
is, the so-called adiabatia flame temperature, or simply the chamber
temperature. The static temperature T. is lees than the adiabatio flame
temperature, by aa much a9 10 or 15 per cent for a throatlees motor, but
this is of no consequenoe for the e x h ~ w velocity.
t Only the stagnation
temperature matters, and this depends only on the thermoohernioal heat
of reaction Ah.
Fig. 0,3d oompares the conditions of the equivalent plenum chamber
with thorn of the aotua ohamber by prmenting the ratio8 p&,, p:/p,,

Fig. 0,Sd. Compuiaon of conditions in equivalent plenum

ohsmber with those in aotual combustion ohamber.
M., and T,/Tf aa funotiona of A./A,L. This figure ie useful as a guide in
carrying out the following computational prooedure recommended for
the performanoe analysis of a rocket motor.
In general, the particular propellant oombhation, the feed pressure,
the nornle l e s t area, the external pressure, and the ohamber dimeneiana
are speoibd in advance. The problem ie to oompute the effective exhawt
velocity c, taking into aooount the effect of the pressure drop in the
oombuation ohamber.
1. From the specified AJAla and 7 , oompute Hafrom Eq. 3-14.
2. With fhis M. and the specified pi, oalculate pa and then pO
, from Eq,3-16
and 8-16.
8. From the apeoified e of the node, and the deotive presme ratio
 G . THE LIQUID PROPELLANT ROCKET ENGINE
p p determine C$ for the equivalent plenum chamber, from Fig.
G,2d and (32%.
4. The thrust is then oaloulated as follows:

5. The oharaoteristic velocity c* depends on the stagnation temperature,

which ia exactly equal to the adiabatio flame temperature. Therefore,
(c*)Qis known, either by thermoohemioal calculation or by experiments
with plenum-type oombustors.
6. The mass flow is then oaloulated in terms of (c*)O:

-
7. Finally, the deotive exhauet velooity is calculated:
c Cg(c*)O
In rooket testing, the chamber preasure is frequently measured at
(3-21)

the forward end of the chamber, through a pressure tap drilled into the
injeotor face. Clearly, this is pi and not p:. Unlaas the measured p1 is
converted by Eq. 3-16 and 3-16 to p:, the reported value of c* will be too
high by as much as 26 per cent (Pig. G,8b) for a throatleaa motor, and
the reported value of CF (experimental)/C~ (theoretioal) will be low
by a comparable error. There effeots ore signifioant for a rocked motor
with a throat diameter equal to 4 or more of the chamber diameter.
If the combustor is not oylindrical, the preceding analysis must be modi-
fied. In partioular, Eq, 8-15 holds only for oylindrical combustors; for
other shapes, it is necessary to know the axial diitribution of the heat
release in order to integrate the momentum equation.
It may be noted that the Bame theory can be used to determine the
axial distribution of heat release, i.e. the over-all kinetics of combustion.
Thus, in a cylindrioal ohamber, the variation of M with axial position x is
obtainable from the pressure distribution by means of Eq. 3-16. Then,
from the equation of continuity (Eq. 3-22), the point-by-point static
temperature oan be computed, and from thia, the stagmiion temperature
and the heat release.

G,1. Theoretioal Sped50 Impulse Calculations. The calculation

of spec%o impulse in Ohe theory of the ideal rooket motor in Art. 2 is
deficient in two very important reapeots: the oombustion gas mixture is
chemically reactive, and the specifio heats vary strongly with tempera-
ture. Both these factorrr were ignored there. It is the purpose of this
article to develop the complete theory, with both effeots present. Inek
 G,8 . BIBLIOGRAPHY
with respect to its oprating parameters. The most important of these
is the combustion pressure-the higher the pressure, the lower the
propellant consumption rate for (L given tbruet; but at the same time,
the greater the weight of the pressurization system. For the com-
pressed gas type, the optimum pressure usually works out to lie be-
tween 260 and 400 lb/in.'; for the turbopump system, the optimum
is usually near 1000 lb/in.n, but a lower pressure is always employed
to ease the cooling problem. It is clear that the turbopump type i s the
most suitable one for long range, large dse rockets with long firing
durations.
3, simplicity and ruggedness: The great simplicity of the compressed gas
type makes it the moat appropriate system for such applicatione as
emall anti-aircraft guided mid=, which have to be ready for imtan-
taneous firing with high reliability with a minimum of pre6ring inspec-
tion and adjustment. 'l'he same consideration applies to detachable
JATO units.
3. Low cost: The consideration of cost ie important mainly for civilian
applications such as the take-off of heavy cargo plrtnes, and perhaps
rocket travel in the future, For military purposes, cost ia s i g n i h n t
mainly as anindication of manufacturing difficulty, and, therefore, for
mimiles or JATO8 required in very large numbers, the.gas-preasuriaed
type is preferred.
6.8. Bibliography.
Arlick 1.
Diationery of guided missile terms. Compiled by Ordnnnae Dept., U.S. Army.
Anli-Air& J., 1849.
Dloaaaxy of guided miwile terms. Prepnred by Cmtltscr on (ha'ded M~&~.ss,
Readaroh md Devdop. Board, U.B. Depf. of Defaud, UM 61/8 and OM 61/8,
sept. 1848.
Layton, 3. P., nnd Youngquint, R. Proposed A.S.A. letter ~ymbolsfor rocket
propulsion. JetPropuleion 86,684 (1861).
A r t i l d.
Malina, F. J. Ghnr&okristosof the roaket motor unit baed on the theory of
perfeat gaoes. J. FrmkUn lmt. 880, 488 (1840).
Wort, H. S., m d Crum, J. Thru6t aoeBaient nnd expansion ratio tablee.
RawWoddddgs Corp., Feb. 1860.
Wert, H. S., Millrr, 1M. M.,and Summerfield, M. The phyaian of roaketo,
Part I. Am. J . Phw. 16, 1 (1847).
h p i r o , A. H. Dytrclmics and T h d m c o d ~ ' c 0of Comprdssiblo Fluid F k ,
Vol, I , Chn 4. Ronald PMM,1868.
Sutton, a. P. gookal ~ropukionBllmum~lr,2nd ed., Chap. 8. Wiley, 1968.
ArlicD 8.
&n&, a. Rooket psrformnnoe with h e ~ added t to inlet propellante by
regenerative nnd external m e w . J d Propcel. 86,712 (1866).
Beakwith, I. E., nnd Moore, J. A. An aaourats md rapid method for the design
of auperaonio nosdw. NACA TmA. Noh IS#, Feb. 1866.

( 617 )
 G THE LIQUID PROPELLANT ROCKET ENGINE
Dillon, P., and Line, L.E., Jr. Heat t r a d e r between aolid particleu and gas in
a rook& nos&. Jet Prqztl. 88, 1091 (1966).
Durham, B. P. T M ohareoteriitics of underexpanded noaalaa. Jet Propul.
66, 696 (1055).
Qflbert, M., Davie, L., and Altmm, D. Veloaity lag of pwtioles in linearly
aooelerated oombustion gases. J . Am. Rocket Soc. $6,28-30 (1966).
Qhsman, I. Impulse expremiom for rocket systams eon-g a nolid phase.
JBt Propul. 87,542 (1967).
Kogan, k Boundmy layer correction in superaonio norzla scaling. 3. Awonazct.
Bd. W ,64 (1968).
Puokett, A. E. Super~onionor& design. J . Applied M&. 18,AZ6bAd70 (1946).
Beifert, H.S., and Attmlur, D, A comparison of ediabatio and iaothermal ex-
pdanion procwss in rocket noadea. J . Am. Rocket Boc. $8, 160-162 (1962).
Bhapiro, A. E. The Dynam'w and Thsrmodpam4c.a qf Cowrssmble Fldd F b ,
Vol. I, Chap. 7 and 16. Ronald fiw, 1983.
Bummerfield, M., Foater, C. R., and Swan, W. C. Flow separation in overex-
panded supeb80aio exhaut nosden. Jef Prqul. 84,819 (1954).
Button, Q. P. Roake$Pr@ulalm, E h m t a , 2nd ed., Chap. 8. Wiley, 1956.
drbick 1.
Battelle Manorial kmt. Phymod properties m d thermodynamic functions of
fuek, o x i d h s , and products of oombustion. Project Rand Rgta. R-187,
188, 186. Columbu, Ohio, 1949.
Ohm% of theoretical periormmoe of several rocket propellant oombinationa.
ROblosrol$m, Dh. North Antw. Auintim, Inc., 1956.
Donegan, A. J., m d Wber, M. Solution of thermochemioal propellaat oalcuh-
tiona on a high speed digiCal computer. Jet Propl, 88, 164-172 (1966).
Fickett, W., and Oowan, R. D. Valuen of thermodgnamio funotions to ~ ~ , C I K ~ K
for aaverd aubstanoes. Lon Alamim Sd.Lab. Rwt. 1787, Bept. 1964.
Hottel, E.C., Wiluame, G. C., and BatterUd, 0. N. Thrrnrodwmic Chatts
jor C o m b w h Prwedtae. Wiley, 1949.
Huff, V. N., and C s l v d , C. 8. Oharts for computation of equilibrium comporG
tion of chemical maotiona in the 0-II-0-N afrptem from W)OO•‹Kto 600O0K.
YACA Teoh. Note 1668, July 1948.
Huff, V. N., and ndrdon, 8. Tables of thermodynamic ffuntionn for analyei. of
airorafb propulsion syatems. NACA Tech. No& 8181, Aug. 1960. (Imluda.
light metals and halogem, as well an O, H, 0, and N oompounda.)
Huff, V. N., and Momell, V. E. Qenerd method for wmputation of equilibrium
oompdtion m d temperature of chemical reaotions. NACA T d . Note dilS,
June 1950.
Krieger, F.J. Chemiod kinatics and rocket nosale design. J . Am. Rocket BOG.$1,
179-186 (1961).
Lewia, B., and von Elbe, G. Combusdim,Flame8 and E2plosioc~ofUasse,Chap. 13.
Aoademio Prw, 1951.
Netl. Bur. h d a r d s . Beleoted d u e s of properties of hydrocarbons. Cirwtar 461,
aov't. fiintiner OBO~,1947.
NatL Bur. Standardu. T a b h of selected values of ohemioal thermodynamio
properties, 8 vol. Cirwlar 600,Oov't. Printing OfBce, 1862.
P m e r , 8. 8. Thennodynamios and chemical kinatics of one-dimerwionalnon-
Viscous %owthrough a h v a l nozzle. J . Chem. Phya. 18, July 1961.
Pock4 Data Book for Rockt Ev@e. Bell Airoraft Oorp., Bufido, N.Y., 1964.
(Pro ertiss of f u l , midisen, and propellant oombimtiaud.)
8.
Button, P., Rocket Propla'm Elemsnta, 2nd ed., Ohnp. 4. Wiley, 1956.
Viohnimky, R., M e , B., m d M~sroldst,J. Combustion temperaturea and ~a.3
oompoaition. 3. Am. Rocket Boc. 86, 105 (1955).
Xrf(els6.
Baker, D. I. Mixture ratio and temprat& BUrVe of n m m w t ~ r ~ x y procket
n
motor combuation ohmbers. Jef Prowl. WBb,
215 (1956).
Additional Material on Nozzle Flow Separation:

Performance Analysis of the Ideal Rocket Motor, Section G,3, Page 458,
Equation (3-5); gives the nozzle static pressure (wall pressure) for flow
separation for severely overexpanded conical nozzles as a function of
atmospheric pressure, for determining the flow separation point for severely
overexpanded conical nozzles.

The additional material which follows from NASA SP-8120, Liquid Rocket Engine
Nozzles, provides the equivalent information for bell nozzles; the nozzle static
pressure (wall pressure) for flow separation for severely overexpanded bell
nozzles (contoured nozzles, parabolic contour nozzles) as a function of
atmospheric pressure and chamber pressure, for determining the flow separation
point for severely overexpanded bell nozzles.
 rex

Nozzles designed for vacuum operation have large expansion area ratios in order to achieve
high specific impulse. It is desirable to ground test engines in the course of the development
program. During ground testing, most altitude engines are overexpanded, often to the extent
that the exhaust gas separates from the nozzle wall. This flow separation can result in
serious problems. For example, a nonoptimum (parabolic) contour was selected for the
nozzle of the 5-2 engine in order t o raise the exit wall pressure. The high-exit-pressure nozzle
was supposed to run unseparated at an area ratio of 27 with a chamber pressure of 700 psi.
A wall-pressure minimum that occurred between area ratio of 14 and the nozzle exit
produced an unstable condition that caused unsteady asymmetric separation, especially
during the startup. The large loads that occurred caused various thrust-chamber structural
failures. A short bolt-on diffuser was developed to eliminate separation during mainstage
operation of the J-2 (ref. 23). Restraining arms were attached from the test stand to the
.
nozzle skirt to absorb the separation loads at startup.

Separation of flow occurs when the gas in the boundary layer is unable to negotiate the rise
to ambient pressure at the end of the nozzle. The exact atmospheric pressure at which flow
will separate from the wall of a nozzle cannot be predicted accurately. Various rules of
thumb to predict separation have been suggested; however, general agreement on one of
these methods has , v t been reached. An early rule stated that a danger of separation existed
when the ratio of exit pressure t o ambient pressure was equal t o 0.4. Later methods based
on fitting of experimental results accounted for the increase in overexpansion that can be
obtained with increasing Mach number. A fit of experimental data for short contoured
nozzles over a broad range of nozzle area ratios (ref. 24) indicates that separation will occur
when

where

Pw, = exhaust-gas static pressure on the wall at separation

Pam, = ambient pressure

PC = chamber pressure = exhaust-gas total pressure

, The method of reference 25 was the basis for a separation-prediction criterion that includes
the effects of gas properties and nozzle shape on separation. A recent and fairly complete
treatment of flow separation in nozzles is presented in reference 26. The results of the
various prediction methods have shown agreement with experimental data in many cases,
but in general predictions are used only as a guide.
 -8120 Overexpanded ozzle References:

Anon.: 5-2 Bimonthly Progress Report, November-December 1965. R-6300-3, Rocketdyne Div.,
North American Aviation, Inc., January 1966.

Schilling, M. T.: Flow Separation in a Rocket Nozzle. M. S. Thesis, University of Buffalo, June 1962.

Crocco, L.; and Probstein, R.:,The Peak Pressure Rise Across an Oblique Shock Emerging from a
Turbulent Boundary Layer Over a Plane Surface. Princeton University (princeton, NJ), March 1964.

Schmucker, R. H.: Status of Flow Separation Prediction in Liquid Propellant Rocket Nozzles. NASA
TM X-64890, November 1974.