Document Sample

                Robert Lobbia, Tony Tan, Brian Wiese, Yoshihiro Noda, Saul Rios, and Cathy Leong
                             The Mechanical and Aerospace Engineering Department
                                    The University of California, Los Angeles
                                             Los Angeles, CA 90095


Laboratory scale hybrid rocket engines are statically tested and performance characteristics including measured
specific impulse and regression rate are evaluated. Rubber, tar, and cellulose fuels are each tested with nitrous oxide
as the primary oxidizer. For the rubber fuel, HTPB, the effects of metallic and oxidizer additives (aluminum powder
and ammonium perchlorate) are related to the measured fuel regression rate. A theoretical model is also developed,
within the standard framework for a turbulent boundary layer heat transfer problem. Fuel regression rates from the
model are compared to our measured regression rates. Three different oxidizer injectors are also tested in an effort
to quantify the influence of mixing and turbulence on the fuel regression rate. Combustion instabilities observed in
the experiments are analyzed and compared to theoretically expected modes. Our measured specific impulse results
show excellent agreement with the theoretical values (for tar-paper, measured 235 sec, theoretical 240 sec).
Measured regression rates also agree well with theory and improvements of 30% and more are attainable with
additives to the HTPB fuel.

                      Introduction                            rates are typically an order of magnitude slower than
                                                              comparable composite propellant rockets and therefore
Background                                                    the addition of multiple port openings or accelerants to
                                                              the fueld grain is often employed. The first hybrid
          The fundamental aspect of a hybrid rocket is        rocket was developed around 1933 by S.P. Korolev and
that one fuel is essentially stored as a liquid of gas and    M.K. Tikhonravov. Working within GRID they used
the other fuel is held as a solid. Different combinations     Gaseoline-collophonium mixture and liquid oxygen
of fuels in both solid and liquid states have been            propellants. Other developments took place with other
utilized. There are some advantages to using a hybrid         propellant combinations such as coal and gaseous
rocket system.       The system is safer than most            nitrous oxide, coal and gaseous oxygen, coal and air,
conventional rockets especially the combustion process        polyethylene and hydrogen peroxide, benzene and
of the propellants is more stable. Typical hybrids tend       potassium per chlorate, benzene and ammonium nitrate,
to be environmentally friendlier. The propellants for         benzene and ammonium per chlorate, and many others.
the hybrid rocket are easily stored and there is a least      Noticeable work continued to about the early 70’s when
likelihood of explosion prior to ignition. The system         development subdued. The study of the hybrid system
also has the capacity to start, stop, and restart important   picked up force again in the early 80’s and continues to
in many applications. The materials used for the rocket       be investigated by different groups worldwide. Today,
are more cost efficient. These inherent safety features       the most common hybrid fuel is HTPB due to the fuel’s
of hybrid rocket motors make them ideal candidates as         desirable manufacturing/structural properties and Isp
replacements of the large solid rocket boosters used in       performance (250-300 seconds, typical). Metallic
many of today’s launch vehicles, including the space          additives, shown to increase the regression rate (see
shuttle. Additionally, these safety features make the         ref#), are often mixed into the HTPB during casting of
study of hybrid rockets well suited for university            the rubber fuel. Recent research in hybrid fuels has
research. A hybrid rocket also has a comparable               included the use of nano-sized energetic particles as
specific impulse than liquid propellants and higher           well as the use of paraffin-based fuels, which have both
density specific impulse than liquid propellants.             been observed to burn with improved regression rates
System complexity is also minimal which hybrid                comparable to solid rockets (#Stanford reference).
rockets, requiring half the lines/pumps as liquid rockets.
There are also a couple disadvantages in employing a          Motivation
hybrid rocket system. The mixture ratio of a rocket can
vary from test to test. This system has a lower density-             Nitrous oxide has several distinct advantages
specific impulse than solid propellants. Solid                when applied to the design of relatively small scale
propellant remains after combustion within the                hybrid
chamber, which reduces efficiency. The fuel regression
                                                              levels, a hybrid fuel grain with multiple combustion
Experimental Setup                                            ports is often employed. This increases the fuel surface
                                                              area needed for higher fuel flow rates.
          The test stand configuration (see fig. 1) for our
experiments consists of a nitrous oxide supply line           Because the oxidizer and fuel ingredients are already
connected to the combustion chamber. An integrated            mixed during propellant manufacturing, the regression
solenoid/check valve controls the flow of the nitrous         rate of solid rockets is controlled by chamber pressure
oxide. Two pressure transducers are used in static test       only. In hybrid rockets, fuel regression rate is mainly
firings: one is near the aft section of the combustion        dependent on the oxidizer mass velocity, or oxidizer
port (just prior to convergent section of the nozzle) and     flux, which is the mass flow rate of oxidizer traveling
the other lies further back on the nitrous oxide supply       through a combustion port divided by the cross
line. Two bolts are attached between the test stand and       sectional area of the port.
the combustion chamber (which slides freely on rails)
and a force sensor is inserted in between one of the          (talk about the fuel injector and its multiple holes
bolts.                                                        increasing regression rate? I don’t know if that is
                                                              correct or not)

                                                              We performed six tests, each with a different solid fuel
                                                              mixed with nitrous oxide as the oxidizer. The solid
                                                              fuels were tar paper, HTPB, and HTPB with additives.

                                                              The regression rate of a hybrid rocket is related to the
                                                              heat transfer properties associated with turbulent
                                                              boundary layer of fuel grain surface. Ignition heat
                                                              vaporizes the fuel particle at the surface of the grain,
                                                              which prepares it for combustion. As combustion takes
                                                              place, it generates more heat that continues the
                                                              vaporization of fuel. This cycle sustain the reaction
                                                              inside the chamber of a hybrid rocket. It is the nature
  Fig. 1. UCLA AIAA hybrid rocket engine test stand           and the extent of the heat transfers from the combustion
                                                              mixture to the fuel surface that governs the regression
All the sensors are linked to a computer where data is        rate of the fuel. A comprehensive analysis of this
recorded for analysis. The inner diameter of the test         behavior is made by Woolbridge and Muzzy.
section is approximately 1.5 inches with an overall           Woolbridge and Muzzy based their model of the heat
length of 11 inches. A small (25 gram) solid propellant       transfer within this boundary layer on convection and
igniter is used to preheat the nitrous oxide and initiate     radiation. They made studies that considered the
the combustion of the hybrid motor. Nitrous oxide             combustion area both the ahead and behind the merge
undergoes an exothermic reaction at about 600 oC (1           point of the boundary layer created by the oxidizer
atm) in which it transitions to diatomic nitrogen and         flow. And for comparable oxidizer mass flux, our
oxygen (2N2O  2N2 + O2 + Heat Release). Initial              experiment exhibited a similar regression rate. (0.038
combustion can only occur if hot oxygen is present near       in/s vs. Woolbridge and Muzzy ’s 0.035 in/s)
the fuel grain. In an effort to maximize the degree of
the nitrous oxide preheating, the injector was designed       The effect of adding of aluminum power to the
to extend into the fuel grain approximately one grain         regression rate was tested. Aluminum powder, sizing
diameter. Placement of the igniter (which has same            45 microns, was added to the HTPB fuel mixture.
port geometry as the fuel grain) about the base of the        Extensive studies have been performed by other
injector ensures the hot ignition gases heat the nitrous      universities on the effect of aluminum additives to
as it flows through the injector and the annular flow of      hybrid rocket fuel grains (lots of REF#s). The
these hot gases past the tip of the injector ensures good     aluminum additive has shown to boost the regression of
mixing and heat exchange between the two flows.               the fuel because it enhances the radiation and
                                                              conduction of heat to the fuel grain hereby vaporizing
Regression Rate Analysis                                      the fuel at a faster rate (REF# old metallized fuel
                                                              studies). In the one study, the addition of 13% UFAl
The fuel regression rate of hybrid engines is much less       aluminum particles (average size of 0.150 microns)
than that of composite solid rocket propellants, usually      achieved an average increase of 42% in regression rate
about one-third its value. To achieve higher thrust           over the baseline fuel. Our experiment shows the
regression rate to increase by 32% from the addition of     measure the secific impulse due to the combustion of
15% Aluminum.                                               the nitrous oxide with different fuels (tar-paper, HTPB
                                                            with Aluminum and Ammonium Perchlorate additives).
The effect of ammonium perchlorate (AP) on the              Using standard fluid-dynamic and thermodyanmic
regression rate was also studied. AP is an oxidizer.        relations along with chemical property tables one can
Because of this it allows the fuel grain to burn            analytically compute the specific impulse of a given
somewhat like that of a purely solid rocket. So while       hybrid rocket engine configuration [ref WILCOX and
AP does not significantly alter the heat transfer           PEP]. For our setup we have designed the inner throat
properties of the turbulent boundary layer, it does add     diameter of our nozzle to provide a chamber pressure of
another mode of reaction to the combustion. The paper       approximately 500 psia. Our nozzle then expands the
published by George, Krishman, Varkey, Ravindran            throat's sonic gases to atmospheric pressure (14.7 psia)
and Ramachandran provided the background for our            with a conical half angle of 15 degrees. The aft
studies. Their studies show a significant increases in      bulkhead of our test stand (which attaches the nozzle to
regression rate with the addition of AP, a result with      the combustion chamber) serves as a thermal reservoir
which our experiment concurs.                               to prevent the nozzle from rapidly heating during each
                                                            test (our ceramic spray coated stainless steel nozzle
Excessively Fast Localized Regression Rates: Failure        would quickly melt/erode if there was nowhere to
Mode Analysis                                               dissipate the heat buildup). Generally, the hotter the
                                                            combustion temperature the higher the Isp, and indeed
75% of our tests with the HTPB fuel experienced             using the our hybrid test stand's specifications we
failures in the 1/8” thick steel fuel casing. These         compute the theoretically optimal Isp vlaues as 235-
breaches were not accidental but the result of what         seconds for tar-paper (O/F=4, T~2800K) and about
appears to be rapid localized burning of fuel. give         250-seconds for the HTPB (O/F=7, T~3300K). Figure
estimates…show photos…suggest design that does this         2 shows these Specific Impulse vs Mixture Ratio
to entire fuel grain….lead into stability                   curves.
                                                                                                            Theoretical Specific Impulse vs Mixture Ratio (N2O/Fuel)


                                                                                                                                                                   HTPB fuel
                                                                                                                                                                   Tar-Paper fuel


                                                               (seco nds)

                                                               Specific Impu lse, I


Fig. 2. Failure analysis due excessively high regression
                           rate.                                                      180

Specific Impulse Performance                                                          170

         Hybrid rockets typically have higher specific                                160

impulse (Isp) ratings than most solid rockets and are
near the high Isp’s of liquid propellant rockets. The                                 150
                                                                                            1.5   2   2.5      3   3.5   4   4.5    5     5.5     6     6.5    7   7.5     8        8.5   9   9.5   10

typical specific impulse range for hybrid rockets is 200                                                                           Mixture Ratio (N 2O/Fuel)

to 350 sec. Modern liquid fueled rockets perform in the
range of 300 to 400 sec and average solid systems lie in    As a comparison to this expected performance we have
the 150 to 250 sec range. A given rocket motor’s            measured the specific impulse from each of our rocket
specific impulse is specified by three parameters: 1)       engine tests. Thrust data (figures 4 and 5) from our
Chemical Propellant Selection, 2) Designed Chamber          force transducer is integrated (this is the impulse) and
Pressure (or nozzle throat), and 3) the Exit Pressure. In   divided by the overall weight of the nitrous oxide and
our experiments with hybrid engines we keep the last        fuel consumed in each test to provide us with a
two parameters constant (same nozzle for all tests) and     measurement of the engine's specific impulse. Since
we used a small (25 gram) section of a potassium            convection on the surface layer of the fuel. The
nitrate based propellant as our igniter/preheat charge      addition of aluminum powder also increases the fuel
the impulse contribution from this is first subtracted      density but a disadvantage is that it reduces the specific
from the integrated impulse prior to division oxidizer      impulse of the engine because the increase of the
and fuel weights.                                           combustion temperature does not compensate for the
                                                            added weight of the fuel grain mixture. (Reference
                                                            Sutton) ((Another reference to put here is the article
                                                            about fuel additives….the title is something like
                                                            “Performance of High-Energy Nano-Particles in Fuels”
                                                            and was off the AIAA site I believe))

                                                            Our HTPB fuel is not an oxidizer, but ammonium
                                                            perchlorate(AP) can be added to the fuel to lightly
                                                            oxidize it and therefore intensify the combustion
                                                            process. This causes the temperature to be higher and
                                                            the regression rate of the fuel to increase. AP is
                                                            popularly used as the oxidizer in solid-fuel rockets such
                                                            as the space shuttle’s SRBs and model rockets.

  Fig. 4. Tar-paper + nitrous oxide thrust and pressure     We observed the effects of the aluminum and AP in our
                         profiles                           tests when the three doped fuel grain mixtures all failed
                                                            partway into the burn as opposed to the HTPB base fuel
                                                            grain which did not fail.

                                                            Combustion Stability of Nitrous Oxide Injectors

                                                            We analyzed the effects of two types of nitrous oxide
                                                            injectors on combustion performance and stability. Our
                                                            testing of different injectors is performed in an effort to
                                                            study how various injection schemes affect the
                                                            combustion and the flame-holding instability. Early
                                                            hybrid engine tests by the UCLA Hybrid Rocket
                                                            Research Project were plagued with flame-holding
                                                            instabilities (often leading to early termination of the
                                                            tests) and low thrust levels, so two new injectors were
                                                            developed as potential solutions to these problems. To
 Fig. 5. HTPB (with Al and AP) + nitrous oxide thrust       increase the average thrust level we simply added
                and pressure profiles                       additional injection orifices (four ports altogether).
                                                            Now, since the hybrid regression rate of the fuel is
                                                            related to the oxidizer mass flux [ref this…] we expect
((Include test results, comparisons of experimental to      that the propellants are consumed faster and at a higher
theory, calculation of specific impulse))                   chamber pressure (provide the nozzle through is set
                                                            properly). It is desirable to operate our hybrid rocket
                                                            engine only for short burning times (less than 10
                                                            seconds) since longer burns result in excessive heating
Hybrid Fuel Additives                                       of the test stand components. Other studies have
                                                            observed relationships between injector design and
Fine aluminum powder (45 micron) is added in small          flame-holding instabilities [1,2]. Thus far, we have
amounts to the fuel grains in an effort to increase the     performed experiments on a flat head axial fuel injector
combustion temperature inside the rocket engine             (fig.1)
thereby increasing the regression rate of the fuel grain.
The aluminum powder works by absorbing heat                 and a cone-shaped fuel injector which has combined
radiation from the combustion and heating up the fuel       characteristics of an axial and radial fuel injector
grain making it easier to vaporize in the boundary layer.   (fig.2).
Without the aluminum, the only heat transfer process is
                                                            Notably, the higher frequency modes are more
    Fro        Sid                  Fro          Sid       attenuated than the lower frequency modes. This
                                                           attenuation corresponds to thrust/chamber pressure
    nt         e                    nt      4    e         oscillations at those higher frequencies that have
    vie        vie                  vie          vie       smaller amplitudes. What we find from these plots is
    w          w                    w       d    w         that there appears to exist distinct modes which could
           F                                e
                                            Fi             potentially contribute to the various types of
           i                                g.
                                            .              instabilities. For each of these tests we did achieve
           g                                2              fairly successful and marginally stable combustion (in
           .                                               that flame blow-off did not occur).
           1                                               We saw the flat head axial fuel injector produced more
Visual observations (video/sound) of our tests show        stable combustion than the combined one, which is in
that indeed the flat head straight injector results in a   agreement with the paper of Mr. Pucci whose tests
more stable exhaust plume (fig. 3).                        showed that axial fuel injectors produced stable
                                                           combustion and radial fuel injectors produced unstable
                                                           combustion.(2) Fig 6-a and 6-b show the time domain
                                                           and thrust domain plots of our two tests. It is common
                                                           to take pressure versus time, but since we have a more
                                                           accurate thrust sensor, we took thrust versus time,
                                                           which is basically fine because the pressure is related to
                                                           the thrust. The average regression rate of axial fuel
                                                           injector was about ___ in/s at an average chamber
                                                           pressure of ____ psig and for combination of axial and
                                                           radial fuel injector, it was ___ in/s at average chamber
                                                           pressure of ____ psig.

               Fig. 3. Rocket Test Image

As a quantitative measure of stability, we have applied
Fourier transforms to convert our time domain data into
frequency space. The spectrums for the two injectors
are shown in figures 4 and 5.

  Fig. 4. Stability of flat and cone type nitrous oxide
   injectors on tar-paper hybrid rocket combustion         1.   Boardman, T.A., Brinton, D.H., Carpenter, R. L.,
                                                                and Zoladz, T. F. “An Experimental Investivation
                                                                of Pressure Oscillations and Their Suppression in
                                                                Subscale Hybrid Rocket Motors.” AIAA Paper No.
     95-2689. AIAA/SAE/ASME/ASEE 31st Joint
     Propulsion Conference and Exhibit, July 10-12,
2.   Justin M. Pucci “The Effects of Swirl Injector
     Design on Hybrid Flame-Holding Combustion
     Instability” AIAA Paper No. 2002-3578. 38th
     AIAA/SAE/ASME/ASEE            Joint     Propulsion
     Conference and Exhibit, July 7-10, 2002

      I   T  dt

I sp 
         W fuel    Woxidizer

Shared By: