EUCASS2017-425
EUCASS2017-425
EUCASS2017-425
13009/EUCASS2017-425
Abstract
Hybrid rocket motors have several potential advantages respect to current used propulsion systems (i.e.
solids and liquids) like simplicity, safety, reliability, environmental friendliness, lower cost. A particular
positive feature of hybrid rockets is the possibility to control the thrust level operating only on the oxidizer
mass flow. Thanks to this it is possible to develop a relatively simple propulsion system that is throttleable
on demand without the complex mixture ratio control and related hardware of a liquid system. In the past
University of Padua has developed a lab-scale hybrid rocket motor that can be throttled at few different
discrete levels with the use of parallel feedlines. To give the possibility of having a continuous throttling
capability a new mass flow control has been developed recently. The mass flow control make use of a
cavitating pintle. The cavitating pintle acts as a cavitating venturi in order to choke the mass flow and
make it independent of downstream pressure. The pintle is used to change the venturi throat area and
consequently varying the oxidizer mass flow keeping a constant upstream pressure. The paper presents the
design of the cavitating pintle and the experimental campaign composed by cold tests followed by hot fire
tests of the lab scale hybrid rocket.
1. Introduction
There has been recently a renewed interest on hybrid propulsion thanks to a stronger focus towards its advantages
like affordability, safety, reliability and environmental friendliness. At the same time the research has progressed in
order to overcome the inherent issues of hybrid propulsion like the low regression rate. One promising solution is the
use of paraffin wax1 ’2 ’3 . University of Padua has been working on paraffin wax based hybrids since almost 10 years
ago4 ’5 ’6 ’7 . Another advantage of hybrids is the possibility of throttling and is the focus of this work.
Thrust profile control in hybrid rocket motors is paramount. For launching vehicle applications this means
optimizing the trajectory, hence reducing the total ∆v required to reach orbit. In ascending and descending vehicles
(ADV), another peculiar application for throttleable hybrid rocket motors (HRMs), the thrust control is required in
order to achieve soft landing and trajectory control.
Since most rocket applications use a fixed throat nozzle, and not a variable one, throttling is generally achieved
by varying the combustion chamber pressure and hence the propellant mass flow. In solid rocket motors (SRMs) the
propellant grain is shaped in order to vary the burning area during the burning time, while in liquid rocket motors
(LRMs), oxidizer and fuel flows are simultaneously controlled.
Of course the trust control achieved in SRMs is rigid, can be set only during the design and manufacturing phase,
because directly linked to the shape of the propellant grain. The thrust control in LRMs is far more flexible, and give
the chance to arbitrarily change the level of thrust during the flight, i.e. throttling.
Copyright © 2017 by Ruffin, Barato and Santi. Published by the EUCASS association with permission.
DOI: 10.13009/EUCASS2017-425
To throttle a HRM, the oxidizer flow is the best involved variable that can be directly controlled, while the fuel is
partially linked to the latter. Using the simplified form of the Marxman power law model the fuel mass flow is propor-
tional to ρ f uel Ab a Gnox . Unfortunately this kind of control implies a o/ f shift that inevitably leads to a characteristic
velocity loss, while in liquid and solid rocket motors the ratio between the reactant is kept constant allowing to work
always at the maximum specific impulse. This is the main drawback in hybrid rocket motors. Furthermore, since throat
erosion has been demonstrated to be particularly severe in hybrid rocket motors (HRMs),9 the ability to control the
oxidizer flow to the combustion chamber is required in order to compensate the force coefficient reduction due to throat
erosion in any application where this degrading contribute is not acceptable.
On the other hand the main advantage of throttling hybrid rocket motors is the simplicity of the control. In
fact the control of just a single feeding line is needed and in order to do this a single dedicated flow control valve is
required. Taking the Lunar Module Descent Engine (LMDE) as an example it was realized with two identical flow
control valves, and since the required mass flows and operating fluids were different it used a positioning link from a
single actuator to grant the correct ratio between the mass flows. Of course the complexity and probability of failure of
the assembly increase with the number of movable parts. So HRMs are less expensive and more reliable also from the
throttling point of view with respect to LRMs.
Continuing downstream of the FCVs in the descent engine there is a variable area injector which uses a moving
sleeve to vary the injection area of both fuel and oxidizer so to obtain the optimal atomization for all the operative
mass fluxes.8 This device would be required also in a HRM, and in this case it would be slightly simpler, but if the
oxidizer used is a mono-propellant, such as HT P or N2 O, with the appropriate catalytic bed it is possible to have a
gaseous injection for a wide range of mass fluxes and so further reducing the complexity of the throttling apparatus.
This approach is possible also with LRMs but few are the fuel/oxidizer combinations in which both the reactants act as
mono-propellant.
Currently at the University of Padova we are studying a HT P fed Hybrid Rocket Motor which merge all the
2
DOI: 10.13009/EUCASS2017-425
technical advantages described before and so it is well suited for throttling. HT P has a high density, is green, can be
stored at room temperature, is very versatile and has a low temperature sensitivity that makes it particularly well suited
for throttling applications.7 This article describes the design and static-characterization of a flow control valve based
on the principle of the cavitating venturi to be applied on the motor under study. Before this description, a section is
presented, in which the theoretical limits of HRMs throttleablity are analyzed. The final section of the article presents
the results of a preliminary test campaign focused on characterizing the motor behavior for a wider range of oxidizer
mass flows than the one used in the test motor up to now.
Considering a throttleable Hybrid Rocket Engine in which we vary the oxidizer mass flow to control the thrust, and
assuming a cylindrical port fuel grain that is consumed exclusively in the port, we can describe the fuel mass flow as
using the simplified form of the Marxman power law. As said before this lead to a oxidizer to fuel ratio shift while
throttling:
ṁox ṁox ṁox ṁ1−n
ox D p
2n−1
o/ f = = = n = (1)
ṁ f uel ρ f uel Ab a Gnox 4n π1−n a ρ f uel L p
ρ f uel π D p L p a π ṁDox2 /4
p
where Ab is the burning area, a and n are the Marxman power low coefficient and exponents respectively, D p and L p
are port diameter and length. It is worth noting that to reduce the effect due to a port diameter change it is desirable to
have n as close to 0.5 as possible while to reduce the effect due to throttling, i.e. a change of the oxidizer mass flow, n
should be toward 1. Practically the power law exponent ranges from 0.45 to 0.8. Of course a change in oxidizer to fuel
ratio imply a variation of the characteristic velocity, and since, via throttling, we can not operate at the optimum point
this always result in a c∗ penalty. Another important consideration is how the characteristic velocity for the propellant
formulation changes with the o/ f . Figure 3 show the frozen c∗ trends for paraffin and the currently most used oxidizers
(90% HT P, N2 O and LOX) obtained using CEA.10 As you can see the LOX trend is quite steep, while the HT P and
N2 O are flatter.
1800
HTP
1700 N2 O
LOX
1600
characteristic velocity [m/s]
1500
1400
1300
1200
1100
1000
900
2 4 6 8 10 12 14 16 18 20 22
o/f []
Figure 3: Characteristic velocity trends with the oxidizer to fuel ratio for paraffin (different oxidizers)
Table 1 reports the maximum values of c∗ and c∗ sensitivity 1 as well as two information about the c∗ penalty
which incur with a throttling ratio of 5 and 10. In the first case (balanced), the throttling take place around the o/ f at
which the maximum c∗ is achieved and ranges in both the oxidizer rich and fuel rich regions, in the second case the
motor operates only in the fuel rich region, starting from the optimal mixture ratio. There are two main reasons why
1 Where c∗ sensitivity means the second derivative of the trends around the maximum point.
3
DOI: 10.13009/EUCASS2017-425
the second c∗ penalty is reported, first the quantity of total impulse associated with the low thrust is smaller if compared
with the high thrust, and considering LMDE duty cycle, applicable for a soft landing ADV in general, the motor spend
less time at the lower thrust. Secondly, in the fuel rich region the drop in c∗ is partially compensated by the relative
increase of consumed fuel mass and so the motor behavior results more linear than the one achieved in the oxidizer
rich region. A linear motor would facilitate the throttling control loop, especially if open.
The power law exponents n used to obtain the data presented in table 1 were 0.5, 0.5 and 0.65 for HT P, N2 O
and LOX respectively. As we can see the advantage of having an higher exponent for the paraffin-LOX formulation
and hence a lower o/ f shift associated with throttling is neglected by the high sensitivity of the propellant, for the
maximum achievable characteristic velocity efficiency is comparable with the one for HT P and N2 O.
This being taken into account it is worth evaluate the advantages of having a power law exponent equal to 0.5
which is actually true for the paraffin-N2 O and HT P propellant formulation. First of all this imply that there is no
o/ f shift for a port diameter variation, as explained in equation 1. This means that there is no o/ f shift for a constant
oxidizer mass flow but it also means that, as consequence of throttling, the return to the previous combustion chamber
pressure is granted with the same exactly oxidizer mass flow thus not requiring for the flow control system to know the
port diameter history. For sake of clarity let’s consider the following example:
Figure 4 shows the results of a simplified quasi-stationary numerical simulation using a lumped parameter model.
In figure 4a the three different normalized oxidizer mass flow profiles are presented, these curves are input for the
simulations. The results reported in figure 4b and figure 4c show the difference in the output normalized pressure
profiles. In particular after a variation from the nominal oxidizer mass flow and the consequent restore of the nominal
conditions there is a difference in the restored pressure, which is also dependent on duration and intensity of the throttle
down. Figure 4b shows how this is not the case with an n exponent equal to 0.5, and of course this simplifies a lot the
thrust control, because it guarantees a one-to-one relation between oxidizer mass flow and pressure. Therefore the 0.5
exponent provides the best accuracy for a throttleable hybrid in open loop, accepting, anyway, a slightly higher c ∗ ∗
penalty compared to a higher n.
It is important to note that all this considerations were made under the assumption of cylindrical fuel grain
consumption following the simplified Marxman power law. However there are possibility available to reduce the
negative effects of throttling on the characteristic velocity such as the implementation of consumable fuel thermal
protections and tayloring the post combustion chamber design.
3. FCV design
There are many options in realizing a flow control valve, in practice every valve in which it is possible to control
the position of the port, can act as a dissipative mean to control the flow, unfortunately somtimes this kind of valves
can be strongly non linear. Variable area cavitating venturis (VACV) are a particular kind of flow control valves in
which the flow is controlled due to the cavitation taking place at the venturi throat, in these condition the flow is said
”chocked”. It has been demonstrated12 ’15 that if the ratio between downstream and upstream pressures is lower than
0.8,0.9 depending on the operating fluid and venturi design the flow is chocked, in this conditions the mass flow through
the cavitating venturi is equal to:
q
ṁ = C D Ath 2 ρ p0,up − p sat (2)
where Ath is the device throat area, C D is the discharge coefficient, ρ is the operating fluid density and p0,up and p sat
are the total upstream pressure and saturation pressure of fluid. So the flow through a cavitating venturi depends
exclusively from the upstream total pressure and not the pressure difference between the valve gauge as in a common
dissipative valve. Furthermore the flow can be easily controlled by varying the venturi throat area, equation 2 shows a
4
DOI: 10.13009/EUCASS2017-425
1
mass flow 1
0.9 mass flow 2
mass flow 3
0.8
0.6
0.5
0.4
0.3
0.2
0.1
0
0 5 10 15 20 25 30
Burning time [s]
(a)
1 1
mass flow 1
mass flow 1 mass flow 2
0.9 mass flow 2 0.9
mass flow 3
mass flow 3
0.8 0.8
0.7 0.7
Pressure ratio []
Pressure ratio []
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 5 10 15 20 25 30 0 5 10 15 20 25 30
Burning time [s] Burning time [s]
(b) (c)
Figure 4: Difference between pressure profiles after an arbitrary variation of the oxidizer mass flow for n = 0.5 and
n = 0.8
linear dependence between the two variables, even if, this liner dependence can be mined by the discharge coefficient.
The easiest way to reduce the throat area is to use a linearly actuated pintle or sleeve.
The hybrid rocket laboratory at University of Padova have been using fixed throat cavitating venturi for a long
time, and also performed throttling tests with discrete levels of oxidizer mass flow using parallel feeding lines. But
right now we are willing to improve our test capabilities and characterize throttleable hybrid rocket motors, and in
order to do this a variable area cavitating venturi based flow control valve has been designed and characterized .
In this article a variable area cavitating venturi is presented as a flow control valve. This device is not new
in rocket applications, in,12 Randall describes comprehensively the fixed area cavitating venturi he was adopting at
Curtiss-Wright Corporation as well as a VACV design using a pintle to vary the throat area. In12 Randall also describes
the good practice guidelines in designing a VACV. Afterward TRW Systems developed a series of throttleable liquid
rocket motors adopting a variable area cavitating venturi as a flow control valve, among them the famous Lunar Module
Descent Engine.11 In recent times when a renewed interest started to gather around hybrid rocket motors, VACV started
to be applied to this propulsive technology as well. During the SPARTAN program, focused on the development of
a descending vehicle employing throttleable hybrid rockets,13 MOOG-Bradford realized a VACV based flow control
valve to be applied to the rocket engine. The University of Padova was entrusted to characterize this FCV.14 Short
after the Beihang University developed and realized their own VACV,15 this valve is used in their test bed to study and
characterize throttleable hybrid rocket motors.
Figure 5a shows the VACV design. For sake of simplicity in manufacturing and quality control the pintle was
designed to have a conical shaped end, the pintle was helically moved toward the venturi throat by mean of a drivescrew
machined in the main body. An O-ring sealing was used to separate the wet region of the flow control valve from the
5
DOI: 10.13009/EUCASS2017-425
drivescrew, and in order to relief the pressure caused by a movement of the pintle a venting hole was machined in the
main body. The venturi body, which resemble a nipple like fitting, is screwed in the main body, the sealing is granted
by the O-ring placed in the groove at the end of the thread. This type of connection aside from providing an effective
sealing also grants the precise and reproducible positioning of the venturi in the main body, which is necessary in order
to reduce position uncertainties. The mass flow enters with an angle of 90◦ with respect to the pintle axis and port. On
the opposite side of the entering bore there is a pressure tap to be used with a trasducer.
(a) (b)
For a conical shaped pintle the throat area can be assumed as the lateral area of the troncated cone which gener-
atrix is perpendicular to the pintle surface at the frustum. Under this condition the throat area function with the pintle
stroke is:
!
tan α tan2 α 2
Ath (x) = π 2 Rth + Rup (1 − cos α) x − x (3)
cos α cos α
where α is the apex half angle, Rth is the throat radius, Rup is the upstream throat radius (in the sagittal plane) and x
is the pintle stroke, from the position in which the pintle touches the venturi. A comprehensive demonstration to this
equation can be found in.15
Equation 3 shows how for a conical shaped pintle the throat area function is non linear with the stroke, in
particular it is quadratic. The non-linearity term can be reduced by decreasing the apex angle since tan2 α/ cos α tends
toward 0 faster than tan α/ cos α for α → 0. But, since there are mechanical limits and the maximum pintle stroke
increase for a lower apex angle, making the pintle slender, in practice the VACV is slightly oversized, meaning having
a bigger throat diameter than necessary, so that the operating pintle stroke must be smaller and hence reducing the
effects of the non linear part of equation 3.
The designed VACV has the following requirement and characteristics:
Of course all the geometrical parameters were turned in to quotes on drawings, quotes associated with tolerances.
Because of these tolerances an uncertainty on the throat area exists. Figure 6a shows the trend of the throat area with
the pintle stroke as well as its uncertainty due to the manufacturing and assembly tolerances. Figure 6b shows how
the throat area uncertainty becomes relevant for low pintle strokes where the ratio between area uncertainty and area
6
DOI: 10.13009/EUCASS2017-425
explodes. In figure 6b are also displayed the different contributors to the total uncertainty, in our design the most
evident is the pintle stroke uncertainty.
×10 -6
4 0.1
2.5
0.06
2 0.05
0.04
1.5
0.03
1
0.02
0.5
0.01
0 0
0 1 2 3 4 5 6 7 8 9 10 11 0 1 2 3 4 5 6 7 8 9 10 11
Pintle stroke [mm] Pintle stroke [mm]
(a) (b)
Figure 6: VACV throat area and throat area uncertainty trends with the pintle stroke
The major contributor in the pintle stroke uncertainty are the drive-screw thread tolerances. Since their effect becomes
most evident for very low pintle stokes, we decided to set, during characterization a minimum of 0.5mm to the pintle
stroke at which point the uncertainty relative error is 6%
The characterization consisted in a series of tests to determine the mass flow varying the pintle stroke. The tests were
performed with two different operating fluids, water and 91% H2 O2 aqueous solution, and for three different pressures.
Characterizing the flow control valve is important for two reasons, first we need to precisely know the needed pintle
stroke to achieve the required mass flow for a given upstream pressure, secondly the VACV can be used both as flow
control valve and flow meter hence having a redundant measurement in our experimental setup.
Figure 7 shows the experimental setup used to characterize the flow control valve. From upstream to downstream
the components are a high pressure nitrogen tank, the pressure regulation block consisting of two pressure regulators
serially connected, the tank, a series of automated ball valves and at the end, the variable area cavitating venturi.
Downstream of the cavitating venturi there is a pipe which connects the valve with a collector tank.
The tank design is particular since it also allows us to measure the mass flow. It consists of a precisely manu-
factured cylindrical barrel with two flat bulkhead at its ends. Inside the tank two rooms are separated by a piston the
above volume host the pressurant while the lower volume holds the oxidant or water, the impermeability of the piston
is granted by specific sealing. Connected to the piston and directed upward there is a stem which crosses the uppermost
bulkhead, the steam is connected to a linear potentiometer. Thanks to the potentiometer output both the instantaneous
mass flow and cumulative mass are determined. In order to grant the accuracy of the measure before each test most of
the gas and fluid in the lower volume is vacuum removed before filling up so to reduce the ullage volume during the
test. This is a very good method to measure the mass flow over a wide range of values. The tank capacity is four liters.
There are three pressure sensors in the setup ptank is screwed in the upper bulkhead of the tank, and hence measure
the pressurant pressure, pup and pdown are placed immediately uspstream and downstream of the cavitating venturi. A
thermocouple, placed at the tank outlet, measures the outflow temperature in order to compensate the density data
and also for safety reasons in case of dissociation inside the tank. Valves V05 and Vtank are manually actuated before
loading the tank and pressurization. Valves VC and Vtest are pneumatically and automatically actuated in this order
during the test. Valve V purge is pneumatically and remotely actuated, it allows low pressure nitrogen to flow trough
the last part of the line in order to purge this last volume from any residual HT P, during fire test the valve is also
used to purge the combustion chamber. The downstream volume, between Vtest and the engine injection is kept to a
7
DOI: 10.13009/EUCASS2017-425
Ptank
Pup Pdown
x
VACV
PR1 PR2 V05
Vc Vtes t To collector
Vdump Vpurg e
HTP tank
minimum. Valve V purge is pneumatically and remotely actuated as well but its function is to empty and depressurize
the line volume between VC and Vtest .
4.2 Results
The flow control valve has been characterized for three different upstream pressures, 30, 45 and 60 bar using both bi
demineralized water and 91%HT P. The pintle stroke was varied by hand using the pintle handle and the reference
goniometer on the back of the flow control valve, which are visible on figure 5b. The strokes values set ranged from
0.5 to 10mm with a resolution of 0.25mm.
400 500
30 bar 30 bar
450
350 45 bar 45 bar
60 bar 60 bar
400
300
Measured mass flow [g/s]
Measured mass flow [g/s]
350
250 300
200 250
200
150
150
100
100
50
50
0 0
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Pintle stroke [mm] Pintle stroke [mm]
(a) H2 O (b) 91% HT P
The first important results of the characterization test campaign are reported in figure 8, they are the trends of the
measured mass flow versus the pintle stroke. The mass flow measured with HT P results higher than the one of H2 O
according to the law in equation 2.
Figure 9 shows the discharge coefficient trend computed from equation 2 using the measured mass flows and the
throat area reported in equation 3 and figure 6a. The trends result similar and the data, even if scattered, are quite in
the same range, suggesting no correlation with the upstream pressure. The biggest relative difference in the data are
8
DOI: 10.13009/EUCASS2017-425
1 1
0.99 0.99
0.98 0.98
Discharge coefficient []
Discharge coefficient []
0.97 0.97
0.96 0.96
0.95 0.95
0.94 0.94
0.9 0.9
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
Pintle stroke [mm] Pintle stroke [mm]
for very low pintle strokes probably due to throat area uncertainty. In figure 9 are also reported the fitting functions
obtained with the least square method. These functions are:
Equations 2, 3 and 5, are used to predict for a given upstream pressure the mass flow for the set pintle stroke.
In order to preliminary characterize the engine throttling behavior four test where performed. The tests were aimed to
show how the pressure varied with the mass flow. All four times the engines performed in the fuel rich region, this was
voluntary because, as said in section 2, the motor behavior is more linear due to partial compensation of characteristic
velocity reduction and relative increase in consumed fuel mass. The throat was realized in refractory material which
show no erosion during the four tests. A mixer was added in the combustion chamber to increase the steadiness of
performance, and keeping high efficiency, with the varying mass flow. The motor was tested using four mass flows: 57,
138, 219 and 282g/s, achieving a maximum throttling ratio around 5. The flow control valve upstream pressure was
set to 60bar.
Figure 10 shows the small hybrid rocket motor, currently in use at University of Padova, in particular this picture was
9
DOI: 10.13009/EUCASS2017-425
taken during the forth fire test of this preliminary campaign. The flow control valve is placed immediately upstream
of the catalytic bed. The setup, from the VACV upward, was the same adopted during the flow control valve char-
acterization. In figure 11 are reported the pressure plot of post and pre combustion chamber during test number two.
The flatness of the pressure profiles indicates that for paraffin-HTP propellant formulation the Marxman power law
exponent is 0.5.
30 post catalytic
combustion chamber
25
Pressure [bar]
20
15
10
0
10 12 14 16 18 20 22 24 26 28 30
Time [s]
Figure 11: Pre and post combustion chamber pressure profiles during the second test at 138g/s
Figure 12 represents the motor behavior at different oxidizer mass flows, since the motor was performing in the fuel rich
region the trend is particularly linear. This is a very desirable characteristic in order to simplify the control algorithm
and improve the predictability of the thrust at various levels.
45
40
Combustion chamber pressure [bar]
35
30
25
20
15
10
5
50 100 150 200 250 300
Oxidizer mass flow [g/s]
Figure 12: Average combustion chamber pressure versus oxidizer mass flow
A flow control valve based on the variable area cavitating venturi was presented and characterized. Because of its
peculiarity this device can both control and measure the flow. The functions presented can be used to predict the flow
through the device and also measure it. The static behavior of a small hybrid motor was preliminary characterized.
10
DOI: 10.13009/EUCASS2017-425
In the future some improvement will be applied at the device and the motor will be better studied and charac-
terized for what concern throttleability. Right now the pintle is manually moved but in the future this device will be
controlled by a DC motor with a position feedback loop. Since the independence between mass flow and downstream
pressure was demonstrated, in the future we will measure also the limits to which this independence holds true. Since
the pintle will be automatically controlled, a dynamic characterization of the flow control valve will be possible and
hence the test motor can be characterized and throttling test performed.
7. Acknowledgments
The authors would like to thank Giorgio Massa e Roberto Torri from Solvay (Rosignano-Livorno, Italy) for the support
and the procurement of Hydrogen Peroxide.
References
[1] M. A. Karabeyoglu, , Transient combustion in hybrid rockets, PhD dissertation, Stanford Univ., Dept. of aeronau-
tics and Astronautics, Stanford, CA, Aug. 1998.
[2] M. A. Karabeyoglu, D. Altman, B. J.Cantwell, Combustion of Liquefying Hybrid Propellants: Part 1, Journal Of
Propulsion And Power, 2002, Vol.18; No. 3, pages 610-630.
[3] M. A. Karabeyoglu, Greg Zilliac, Brian J. Cantwell, Shane DeZilwa and Paul Castellucci Scale-up Tests of
High Regression Rate Liquefying Hybrid Rocket Fuels, AIAA-2003-1162, 41st Aerospace Sciences Meeting and
Exhibit, Reno Nevada, January 2003.
[4] M. Lazzarin, M. Faenza, F. Barato, N. Bellomo, and A. Bettella, Numerical Simulation of Hybrid Rockets Liquid
Injection and Comparison with Experiments, Journal of Propulsion and Power, Vol. 31, No. 1 (2015), pp. 352-364.
[5] N. Bellomo, M. Lazzarin, F. Barato, A. Bettella, D. Pavarin, M. Grosse, Investigation of Diaphragm Effects on
the Efficiency of Hybrid Rockets, Journal of Propulsion and Power, Vol. 30, No. 1 (2014), pp. 175-185.
[6] N. Bellomo, F. Barato, M. Faenza, M. Lazzarin, A. Bettella, D. Pavarin, Numerical and Experimental Investiga-
tion of Unidirectional Vortex Injection in Hybrid Rocket Engines Rockets, Journal of Propulsion and Power, Vol.
29, No. 5 (2013), pp. 1097-1113.
[7] F. Barato, A. Bettella, D. Pavarin, Numerical Investigation of Pressure-Fed Solutions for Paraffin-Based Hybrid
Rocket Motors, 49th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit, 14-17 July, San Jose,
CA, USA.
[8] William R. Hammock, Jr., Eldon C. Currie and Arlie E. Fisher, APOLLO EXPERIENCE REPORT - DESCENT
PROPULSION SYSTEM, National Aeronautics and Space Administration, WASHINGTON, D. C. March 1973
[9] B. Evans, K.K. Kuo, E. Boyd and A.C. Cortopassi. Comparison of Nozzle Throat Erosion Behaviour in a Solid-
Propellant Rocket Motor and a Simulator. AIAA 2009-5421. 2-5 August 2009.
[10] Sanford Gordon & Bonnie J. McBride. Computer Program for Calculation of Complex Chemical Equilibrium
Composition and Application. NASA Reference Publication 1311. October 1994.
[11] Jack M. Cherne, MECHANICAL DESIGN OF THE LUNAR MODULE DESCENT ENGINE, TRW System,
Redondo Beach, CA, USA.
[12] L. N. Randall. Rocket Applications of the Cavitating Venturi. American Rocket Society. 11 June 1951. Toronto,
Canada.
[13] Guido Parissenti, Mario Pessana. Throttleable hybrid engine for planetary soft landing. 4th European Conference
for Aerospace Sciences EUCASS. July 2011. Saint Petersburg, RU.
[14] M.Faenza, F.Moretto, R.Tijsterman, R.Popela, P.Dvorak, D.Petronio, D.Pavarin. Experimental Characterization
of a Cavitating Pintle Valve with H2O2. Space Propulsion Conference 2014. 21 May 2014. Cologne, Germany.
[15] Hui Tian, Peng Zeng, Nanjia Yu,Guobiao Cai. Application of variable area cavitating venturi as a dynamic flow
controller. Flow Measurement and Instrumentation 38. 11 May 2014
11