THE AMERICAN SOCIETY OF MECHANICAL ENGINEERS

345 E. 47 St., New York, N.Y. 10017

90-GT -323

CS
The Society shall not be responsible for statements or opinions advanced in papers or in dis-
cussion at meetings of the Society or of its Divisions or Sections, or printed in its publications
M Discussion is printed only if the paper is published in an ASME Journal. Papers are available
^^L from ASME for fifteen months after the meeting.
Printed in USA.

Copyright © 1990 by ASME

RAMSCRAM—A Flexible RAMJET/SCRAMJET Engine

Downloaded from http://asmedigitalcollection.asme.org/GT/proceedings-pdf/GT1990/79054/V002T02A028/2399271/v002t02a028-90-gt-323.pdf by guest on 17 January 2022

Simulation Program
LEO A. BURKARDT and LEO C. FRANCISCUS
National Aeronautics and Space Administration
Lewis Research Center
Cleveland, Ohio 44135

ABSTRACT This paper presents a general description of the

calculation methods used in the code and a demonstra-
With the resurgence of interest in high super- tion of some of its capabilities through the use of
sonic and hypersonic flight there is a need to simu- examples.
late airbreathing engines which may be used in this
flight regime. To meet this requirement the RAMSCRAM FLOW PATH DESCRIPTION
code was developed. The code calculates one-dimensional
flow properties at each component interface and the Figure 1 is a schematic of the sittrplest flow path
overall performance of the engine. It uses equilibrium used in the RAMSCRAM program. Calculations are per-
thermodynamics which accounts for dissociation and formed at each flow station shown in the schematic.
allows for any fuel or combination of fuels. The pro- Station 0.0 is the beginning of the forebody (external
gram can simulate ramjet, scramjet, rocket, and ducted compression surface) and has free stream conditions
rocket engines. associated with it. Station 1.0 defines the beginning
of the inlet (internal compression portion). Station
INTRODUCTION 2.0 defines the throat of the inlet or the end of the
mixer if a mixer is to be included. Station 3.0 is the
In the past few years, there has been a resurgence diffuser exit/combustor entrance location. Station 4.0
of interest in high speed flight. Such programs as the is the combustor exit. Station 4.5 is the nozzle
High Speed Civil Transport and the National Aerospace throat. Station 5.0 is the nozzle exit.
Plane have highlighted this interest. Ramjet and scram-
jet engines perform very favorably in the high super- THERMODYNAMIC ROUTINE
sonic and hypersonic flight regimes. Ramjets can be
effectively used from about Mach 2 to Mach 6 and scram- Due to the high temperatures that can be reached
jets give reasonable performance at Mach 6 and above. in ramjet and scramjet engines, we felt that equilib-
The RAMSCRAM program described herein was deve- rium thermodynamics would be needed to maintain good
loped at NASA Lewis Research Center to meet the need accuracy. At the time we were developing this code,
to calculate the performance of ramjet and scramjet the CET86 (1) general equilibrium code was being modi-
engines. The goal in developing this program was to fied to run as a subroutine to. the NNEP (2) turbine
obtain high fidelity simulation while maintaining ease engine simulation code. It was convenient therefore
of use, versatility and reasonable calculation times. to use this version of the CET86 code for the RAMSCRAM
In pursuit of this goal, performance is calculated by program also.
marching from front to back through the engine flow
path in a stepwise fashion. One-dimensional calcula-
tions are performed at each step with efficiency fac-
tors to account for flow path losses. Program input
is made easy through the use of Fortran Namelist input.
Equilibrium thermodynamics is included to account
for high temperature real gas effects. The general
nature of the thermodynamics subroutine, together with
the flexible mixer and combustor components, gives the
RAMSCRAM code the ability to simulate a large variety 11.0 .0 2.0 3.0 4.(1 4.5 5.(1
of engine/propellant combinations. Some examples
include: any chemical rocket or ducted rocket, multi- FIG. 1 - RAMJET STATION LOCATIONS.
fuel ramjet engines, oxygen enriched scramjet engines.

'Presented at the Gas Turbine and Aeroengine Congress and Exposition—June 11-14, 1990—Brussels, Belgium
 In addition to the modifications made to CET86 for this point using the efficiency factor. Two of three
use with NNEP, further modifications were made to allow parameters, airflow, Mach number or flow area, are
more than a single fuel/oxidizer combination to be input to permit calculation of the flow conditions and
specified. Up to 24 reactants can be specified as pos- geometry at this location. If only airflow or flow
sible fuels or oxidizers during a single run with the area are given, then the Mach number is assumed to be
appropriate combination of those reactants specified the terminal Mach number from the forebody calculation
at each location in the engine. This permits calcula- or the free stream Mach number if no forebody is
tion of propulsion systems such as ducted solid rocket. included. If flow area and airflow are input, then the
subsonic or supersonic solution is chosen to match the
Fore body condition of the air just upstream of this station. It
can also be forced to either solution by entering a
A conical, two-dimensional wedge or no forebody subsonic or supersonic Mach number which will be used
can be specified. Figure 2 presents a schematic of the as a first guess in the calculation procedure.

Downloaded from http://asmedigitalcollection.asme.org/GT/proceedings-pdf/GT1990/79054/V002T02A028/2399271/v002t02a028-90-gt-323.pdf by guest on 17 January 2022

forebody. Station 2.0 can be used to determine the flow
In the case of a cone, the conical half angle is field at an intermediate point in the inlet, such as
specified and a conical shock is calculated. The coni- the throat of a ramjet inlet. The same efficiency fac-
cal flow field is integrated across the station 1.0 tor and either Mach number or flow area are input to
inlet height and a mass weighted average is calculated calculate the flow conditions. If desired, a normal
to obtain one-dimensional flow conditions at that shock can be included in the calculation at this point
location. when the inlet is started.
If a two-dimensional wedge is specified, up to 20 The flow conditions at station 3.0 are calculated
ramps can also be included. This allows an isentropic in the same manner as described for station 2.0, with
compression surface to be approximated. The shock off the exception that there is no provision for a normal
each ramp is calculated. If the combination of Mach shock calculation.
number and ramp angle at any ramp will not support an
attached oblique shock, then a terminal normal shock is Mixer
calculated at that point and the subsonic flow field The inlet can be configured to perform as a mixer
is assumed to persist down the rest of the forebody. in the case of a ducted rocket or ejector ramjet simu-
The inlet can be specified to be either started or lation. In this case, station 1.0 becomes the air-
unstarted. If it is started, the conditions after the stream entrance to the mixer and station 2.0 becomes
oblique shocks specify the flow field at station 1.0. the mixed plane. If it is not convenient to use sta-
If the inlet is specified as unstarted, then a further tion 1.0 as the mixer entrance, then another station
normal shock calculation is performed as part of the can be included to follow station 1.0. This station
forebody calculation to give the properties of the flow is designated station 1.5 and it has the same charac-
a station 1.0. teristics as described for station 2.0 in the inlet
section above. The rocket stream enters the mixer at
Inlet station 1.2.
In the simplest case, the internal inlet consists The mixer model is defined by a control volume as
of everything from station 1.0 to 3.0. The flow char- shown in Fig. 3. Streams A (air stream) and B (rocket
acteristics at each station are based on those calcu- exhaust) enter the control volume at flow station 1.5
lated at the previous station and the flow area and and 1.2, respectively, and mix inside the control vol-
efficiency terms given for the current station. The ume. The flow is assumed to be fully mixed and departs
efficiency terms are always relative to conditions at the control volume at station 2.0.
the previous station. If no efficiency factor is pro- The mixed flow conditions at station 2.0 are cal-
vided an isentropic calculation is performed. culated by simultaneously solving the equation of state
The flow field at station 1.0 defines what is com- and the conservation of mass, momentum and energy equa-
ing into the inlet. An efficiency factor can be applied tions. This leads to a quadratic equation. The two
to the flow at this point to account for any losses not solutions represent a low and a high entropy gain. The
accounted for in the forebody calculations. The effi- high entropy solution always results in subsonic flow
ciency factor can be in one of three forms: total pres- and the low entropy solution normally results in super-
sure recovery, kinetic energy efficiency or process sonic flow. The code allows the choice of either solu-
efficiency. Process efficiency is similar to kinetic tion. Normally the subsonic solution would be chosen.
energy efficiency, except that it is referenced to dif-
fuser exit static conditions rather than total condi-
tions. If desired, it is entirely possible to leave
the forebody undefined and account for all losses at
CON'I'KC)L

ANGLE OF ATTA('K
1.5
Tee

RAMI'
RAMPS
SHOCKS

FORLBUUY
SHOCK - INL.EI'

NORMAL
SI IU('K

FIG. 2 - FOREBODY DIAGRAM. FIG. 3 - MIXER MODEI.

 The mixer calculation allows the inclusion of 1000
efficiency factors to account for losses in the mixing
section. These factors include: a momentum loss term aun
for either incoming stream, a wall friction loss term In
and a total or static pressure loss term. The pressure SPI.('I I] C'
loss term is referenced to mixed pressure before the euu
Isp H (i'APTIlRl
loss is calculated. (sec) ARIA
The mixing section does not have to be constant 4nu
(c iii -' /kN)
area. If a diverging or converging area is defined,
then a pressure-area term must be accounted for on the 2uo
walls. This term is approximated by assuming that the
pressure acting on the wall is the average of stream A
I

Downloaded from http://asmedigitalcollection.asme.org/GT/proceedings-pdf/GT1990/79054/V002T02A028/2399271/v002t02a028-90-gt-323.pdf by guest on 17 January 2022

and mixed stream static pressures. 0
The use of equilibrium thermodynamics also accounts
for any burning occurring in the mixing process by main- (AIR I 1 (OW) /(PROPRI.I,AN1' HOW)
taining equilibrium chemistry at the mixed plane.
FIG. 4 - DUCTED ROCKET PERFORMANCE.
Combustor
The combustor model is essentially the same as the
mixer model. In this case, the fuel stream replaces 4UO(I 11)
the rocket exhaust stream of the mixer. Two fuel
streams are permitted to allow the injection of two 3000 volwnc(ric 8000
different fuels simultaneously. Also, up to ten
Isp 6000
sequential fuel injection stations are permitted along is ' , 2000
the length of the combustor. A ramjet combustor is (s ee) N—) 4000
represented by specifying the subsonic solution and a /sec

'lit
10( 1(1 m3

scramjet combustor is represented by specifying the

supersonic solution. Using this model for the combus-
u 0
tor allows the momentum of the fuel to be accounted for JP4 CII4 112
JP4 C1I4 F12
as well as its energy. The fuel momentum can be a sig-
nificant contributor to scramjet thrust. (a) (I,)
Incomplete combustion is represented by including
"inert fuel" in the fuel composition. Currently "inert FIG. 5 - PERFORMANCE COMPARISON OF CANDIDATE RAMJET FUELS.
hydrogen" and "inert methane" are available, but others
can be easily added as needed. Other combustor effi-
ciency factors are similar to those used for the mixer. Ramjet Fuel Comparison
A ramjet was operated at Mach 5 using three dif-
Nozzle ferent fuels. The general thermodynamics routines used
The nozzle calculation is an isentropic expansion. in this program make this type of comparison a very
A throat calculation is provided whenever the velocity simple procedure, requiring only a change in the speci-
coming out of the combustor is subsonic. The nozzle fied fuel to air ratio and the fuel type. Figure 5
can be allowed to expand fully to free stream pressure illustrates the performance of the engine with each
or it can be terminate at a given area. No distinction fuel.
is made between internal and external expansion sections Figure 5(a) presents I s p and Fig. 5(b) presents
of the nozzle. Two efficiency factors are provided, a volumetric I s p, which is defined as thrust divided by
velocity coefficient and a thrust coefficient. volumetric flow rate. The three fuels represented are
JP4, liquid methane and liquid hydrogen. As shown,
EXAMPLE SIMULATIONS hydrogen gives by far the best performance based on
mass flow rate, but it also gives by far the worst per-
Ducted Rocket formance based on volumetric flow rate because of its
A ducted rocket was simulated and an analysis of very low density. The structural weight of the vehicle
its performance versus air flow to propellant flow required to accommodate the fuel and the larger drags
ratio was completed. The rocket burns a composite associated with the greater volume of hydrogen or meth-
solid propellant mixture of ammonium perchlorate oxi- ane play an important part in the tradeoff between
dizer and ethylene oxide fuel. The net thrust was fuels. Here again, the performance provided by the
held constant while the air to propellant bypass ratio engine code would be used in a vehicle analysis to
was varied. As the bypass ratio was varied, the mix- determine the best fuel for specific mission
ture ratio of the propellant was also varied to give requirements.
enough excess fuel to provide a stoichiometric fuel to
air ratio. The specific impulse and air capture area Scramiet Sensitivity Analysis
of the rocket versus bypass ratio is plotted in Fig. 4. A sensitivity analysis was performed for a
The capture area is normalized with thrust. Mach 10 scramjet. The inlet kinetic energy efficiency
A bypass ratio of 6 was arbitrarily chosen as the (nKE combustion efficiency (ng), and nozzle thrust
),

maximum simulated. At some point, the solid propellant coefficient (CFG) were varied. The baseline values
mixture ratio becomes such that it cannot support com- used were: 0.97 for ,IKE, 0.95 for ng, and 0.96 for
bustion. The specific impulse increases linearly with CFG. Figure 6 presents the results of this analysis.
bypass ratio. However, as shown, the air capture area Each parameter was varied over a range of from
also increases with bypass ratio. This offsets the approximately 0.9 to 1. Kinetic energy efficiency was
gain in performance, as the inlet and combustion duct varied with two different downstream constraints. In
will increase in size and weight. The ultimate compro- one case, the diffuser exit/combustor inlet geometry
mise between the two offsetting trends must be accom- was held constant. In the other case the diffuser
plished by doing a vehicle analysis using the informa- exit/combustor inlet Mach number was held constant. In
tion provided by the engine performance code. the first case fKE had only a small effect on engine


2500 CONCLUDING REMARKS

231)0 There has been a resurgence of interest in hyper-
sonic propulsion. Air breathing flight in this regime
2100
requires the use of ramjet and scramjet propulsion sys-
OMBUS I ION I.l FF JC1ENCY
19))0 tems. Vehicle and engine performance and tradeoff
studies require a reasonably accurate, versatile and
1700 LON5"f AN'I'A K6A
Isp
fast running propulsion simulation program.
I5oo The RAMSCRAM code fulfills this requirement. The
(Sec)
1300
general form of the geometry inputs allows the simu-
y^r lation of many engine configurations. Component
efficiency inputs make it easy to study component per-

Downloaded from http://asmedigitalcollection.asme.org/GT/proceedings-pdf/GT1990/79054/V002T02A028/2399271/v002t02a028-90-gt-323.pdf by guest on 17 January 2022

formance effects on overall engine performance. And
the general nature of the equilibrium thermodynamics
routines allow the specification of nearly all conceiv-
.9 .91 .92 .93 .94 .95 .96 .97 .911 .99 I able propellant combinations.
I7FFICl1:NCY PARAMI3FFR
REFERENCES
FIG. 6 — SCRAMJET SENSITIVITY ANALYSIS AT MACH 10.
1. Gordon, S. and McBride, B.J., 1976, "Computer
performance because there was a self compensating Program for Calculation of Complex Chemical
effect of the burner Mach number decreasing as fKE Equilibrium Compositions, Rocket Performance,
decreased. This resulted in lower Rayleigh losses in Incident and Reflected Shocks, and Chapman-
the burner. Combustion efficiency also had only a Jouguet Detonations," NASA SP-273.
small effect on the performance over its range of vari-
ation. Both nKE with constant burner Mach number and 2. Fishbach, L.H. and Caddy, M.J., 1975, "NNEP -
CFG had large effects on the engine performance. This The Navy NASA Engine Program," NASA TM X-71857.
is because they both directly affect the gross thrust
on about a one-for-one basis. Net thrust is only about
15 percent of the gross thrust and therefore any varia-
tion in gross thrust is magnified in the net thrust and
specific impulse values.