Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 10.5.4 - Experimental Study of the Droplet-Laden Gas Flow in a Solid Rocket Motor Model
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00263
 Material Information
Title: 10.5.4 - Experimental Study of the Droplet-Laden Gas Flow in a Solid Rocket Motor Model Particle-Laden Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Tóth, B.
Anthoine, J.
Rambaud, P.
Steelant, J.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: droplet-laden gas flow
particle image velocimetry
slag accumulation
solid rocket motor
 Notes
Abstract: In solid propellant rocket motors (SRM) incorporating a submerged nozzle, liquid alumina residues of the combustion can be entrapped in the cavity formed by the casing and the nozzle integration part. This continuous entrapment leads to an accumulation of slag resulting into a considerable mass at the end of the booster operation. Therefore, the aim of a long-term research project at the von Karman Institute for Fluid Dynamics is to better understand the slag accumulation process. Various aspects of the slag accumulation are experimentally investigated using a simplified cold-gas model. This article presents aspects of the internal flow field characterisation by a new two-phase Particle Image Velocimetry (PIV) method in a representative cold-gas, two-phase flow configuration. The technique is based on a two-colour YAG laser and phase separation methodology using fluorescent dyes. This approach permits to separate simultaneously the images of the two phases by optical means. The current focus of attention is the droplet-phase characterisation. Via image processing the droplet images are identified and the droplet sizes are estimated. The results prove the filtering effect of the inhibitor, which implies a considerably lower droplet concentration (especially larger classes) in the downstream recirculation region. Furthermore, a preliminary tracking algorithm shows the increasing deviation of the mean droplet trajectories for increasing droplet diameter classes with respect to the gas-phase motion.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00263
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1054-Toth-ICMF2010.pdf

Full Text



7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


Experimental Study of the Droplet-Laden Gas Flow in a Solid Rocket Motor Model


B. T6th* J. Anthoinel P. Rambaudt and J. Steelant*

European Space Research and Technology Centre-ESA, 2200 AG Noordwijk, The Netherlands
t Centre du Fauga-Mauzac, Office National d'ttudes et de Recherches A6rospatiales, 31410 Mauzac, France
von Karman Institute for Fluid Dynamics, B-1640, Rhode-Saint-Genese, Belgium
balazst@gmail.com
Keywords: droplet-laden gas flow, particle image velocimetry, slag accumulation, solid rocket motor




Abstract

In solid propellant rocket motors (SRM) incorporating a submerged nozzle, liquid alumina residues of the combustion
can be entrapped in the cavity formed by the casing and the nozzle integration part. This continuous entrapment
leads to an accumulation of slag resulting into a considerable mass at the end of the booster operation. Therefore, the
aim of a long-term research project at the von Karman Institute for Fluid Dynamics is to better understand the slag
accumulation process.
Various aspects of the slag accumulation are experimentally investigated using a simplified cold-gas model. This
article presents aspects of the internal flow field characterisation by a new two-phase Particle Image Velocimetry
(PIV) method in a representative cold-gas, two-phase flow configuration. The technique is based on a two-colour YAG
laser and phase separation methodology using fluorescent dyes. This approach permits to separate simultaneously the
images of the two phases by optical means.
The current focus of attention is the droplet-phase characterisation. Via image processing the droplet images are
identified and the droplet sizes are estimated. The results prove the filtering effect of the inhibitor, which implies
a considerably lower droplet concentration (especially larger classes) in the downstream recirculation region.
Furthermore, a preliminary tracking algorithm shows the increasing deviation of the mean droplet trajectories for
increasing droplet diameter classes with respect to the gas-phase motion.


Introduction

The first stage of launch vehicles (e.g. Ariane 5, Vega,
Shuttle, H-II) generally incorporates solid propellant
rocket motors (SRM). These are typically operating dur-
ing the first part of the lift-off providing most of the
thrust to accelerate the vehicle. To shorten the overall
length, the nozzle is submerged in the last segment of
solid propellant. That means that the convergent, the
sonic throat and a part of the divergent are surrounded
by solid propellant. This integration allows orientation
of the nozzle to provide adaptation of the rocket trajec-
tory during the ascent. During the combustion, the re-
gression of the solid propellant surrounding the nozzle
integration part leads to the formation of a cavity around
the nozzle lip. The aluminium, being part of the pro-
pellant grain composition, is oxidized during the com-
bustion. This process goes along with the generation of
alumina droplets, which are then carried by the hot core
flow towards the nozzle. Meanwhile, the droplets inter-


act with the vortices formed by the internal flow and thus
may modify their structure. As a consequence, some of
the droplets are entrapped in the cavity instead of being
exhausted through the throat. The amount of entrapped
droplets in the cavity depends most probably on their in-
teraction with the vortices. The accumulation of these
droplets in the cavity generates an alumina puddle, also
called slag. This slag reduces the performance of the
solid propellant motor due to its dead weight, and ab-
sence of impulse generation. In the case of the Ariane 5
solid rocket motors (MPS P230) the total mass of the ac-
cumulated particles can reach up to 2 tons in each motor
by the end of the launch ( ( )).
Previous studies related to slag accumulation. Slag
accumulation in aluminized SRMs has already been
studied directly or indirectly through various projects
pointing out a number of consecutive mechanisms driv-
ing the entrapment of alumina droplets.
The first obvious steps in this process is the alu-
minium combustion, the alumina formation and the gen-











erated droplet size distribution, which have been stud-
ied by several authors: ( ), ( ),
( )and ( ).
The entrainment and transport once the droplets are
flow-born have been investigated first through quasi-
steady numerical simulations by
( ), ( )and ( ).
However, ( ) and
( ) proved the unsteady nature of the internal flow
field to be important. Also the slag accumulation was
shown by ( ) and ( )
to be linked to the pressure oscillations occurring in the
SRMs. As these oscillations are related to different inter-
nal vortex generation mechanisms, among others
( ), ( ) and
( ) studied these processes. Furthermore,
( )and ( )
assessed the interaction of the droplets with the internal
surfaces of the motor and the behaviour of the eventually
formed liquid film (primarily on the nozzle lip) to eval-
uate whether a droplet contributes to the slag or returns
to the flow.
Finally, once the slag is deposited in the cavity, its
negative contribution to the launch does not finish. Be-
ing liquid, it can slosh inside the cavity as
( ) showed. If the aerodynamic forces are strong
enough (and/or combined with the sloshing), slag may
be ejected through the nozzle. Though the on-board
dead mass decreases, the momentum impulse loss of the
motor cannot be recovered. Furthermore,
( ) assessed the thrust oscillations in function of the
amount of ejected slag as they may lead to control prob-
lems and possible vehicle instabilities. Another conse-
quence of the slag ejection is that the total amount of the
actually accumulated alumina might be higher than what
is observed after a firing test or launch.
In spite of many predominantly numerical investi-
gations of the slag accumulation and the internal flow
in SRMs (e.g. ( ), ( ),
( ) and ( )), the oc-
curring processes are still not fully understood. There
is no known model that could entirely describe the ac-
cumulation mechanism and provide clear suggestions to
designers to minimize the slag mass in future launchers.
In addition, the available experimental observations are
even more limited and in case of a real motor they are
mostly limited to global accumulated slag mass (such
as ( )) or internal pressure evolutions (such as
( )). The lack of a more detailed ex-
perimental database, including the quantification of the
interaction between the two phases, limits the validation
and application of the numerical models.
Objectives There is a clear need for a more fundamental
study of the two-phase flow interaction, the slag accu-


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


mulation process and the generation of a detailed exper-
imental database for numerical validation. To achieve
this goal, the driving parameters of the slag accumula-
tion are characterized in a simplified 2D-like cold-gas
model ( ( )). The interaction of droplets
with the flow and the entrapment process of the droplets
in a stagnant area modelling the nozzle cavity are inves-
tigated primarily experimentally. However, not only a
detailed experimental database is produced, but numer-
ical simulations are also performed and validated to ex-
plore their limitations and strengths.
In the present article, a two-phase particle image ve-
locimetry (PIV) assessment is shown. The statistical
analysis (mean flow field, turbulence quantities) of the
campaign was shown by ( ). Currently, a
more detailed analysis of the droplet-phase is performed,
focusing on the spatial distribution of various droplet
classes and their mean motion.

The experimental conditions

The measurement technique. In the present investi-
gation a two excitation and two emission spectra tech-
nique is used to separate the two phases optically. Dif-
ferent fluorescent dyes are dissolved in the two phases
(Rhodamine B in the tracers and Butyl PBD in the
droplets). These dyes are excited simultaneously by dis-
tinct wavelengths.


Droplets Butyl PBD Tracers Rhodamine B


xi


/ Camera sensitivity\


U


_J Optical filters' ".
A transmittance


860 X [nm]


532


Figure 1: The principle of the two excitation and two
emission spectra technique.

A schematic of the PIV technique of the present study
can be seen in Fig. 1. A Quantel Twins Ultra 180
MiniYAG laser is applied that produces laser pulses at
A 532 nm (about 100 mJ) and at A 266 nm
(about 30 mJ) simultaneously. The two pulses excite
fluorescent dyes dissolved in the liquid of each phase
(droplets and gas-phase tracers), which have high ab-
sorbance at only one of the wavelengths (Rhodamine B
at A 532 nm and Butyl PBD at A 266 nm). The two
phases emit light at very distinct bands of the spectrum.
Therefore, by using two cameras equipped with proper


.


- *


1266run(UV


1 532run (reen


IIllumination












optical filters only the light of the fluorescent emission
of one of the phases is recorded with each camera.
Further details on the technique are provided by
( ).
Measurement parameters. The geometry of the ex-
perimental model simulates the main characteristic fea-
tures of the MPS P230. In Fig. 2 the set-up is shown
in its real (vertical) orientation. However, later the ge-
ometry will be always represented horizontally. The test
section (depicted in Fig. 2(b)) is designed in a way to
allow the use of optical measurement techniques (e.g.
PIV) to characterize the internal flow field. Therefore,
most of the walls are made of transparent material.


(a) The wind tunnel. (b) The test section.


Figure 2: Experimental configuration.

The square i,_' 11 x 200 mm2) test section has a sym-
metric arrangement, separated by a 5 mm thick splitting
plate. In this arrangement the flow in the two sides is
also symmetric and therefore the measurements are car-
ried out by using only one side of the test section.
The inhibitor between the second and the third seg-
ments of the MPS P230 is modelled by a pair of inclined
obstacles that induce vortex generation. The height of
the inhibitor is h 33.5 mm. It is placed at about
Li = 9h (310 mm) from the 2D nozzle.
The model of the submerged nozzle including the
model of the appearing cavity, is installed at the outlet of
the test section. The set-up is designed for subsonic op-
eration. Therefore, the sonic condition of the real case at
the nozzle is not respected. Moreover, it is not the goal
of the present study to model perfectly the real condi-
tions, but rather to focus on basic phenomena such as a
vortex shedding, two-phase flow aspect and fundamental
mechanisms leading to liquid accumulation in stagnant


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


areas.
The experiments are carried out at room temperature
using air to model the gas-phase of the internal fluid of
the SRM and using droplets of mainly water to model
the alumina droplets. A bulk air velocity of Uo 2 m/s
was chosen corresponding to a Reh 5025. This
Reynolds number is substantially lower than in case of
the real motor. Nevertheless, by respecting the turbulent
flow regime, the viscous effects should remain similarly
negligible.
The solution representing the liquid-phase is created
volumetrically from 1/5 part Methanol and from 4/5
part demineralized water, containing 2.67 g/1 Butyl-
PBD fluorescent dye. The droplet-phase is generated
by a spray device (based on a Lechler 156.000.17.13
model), which is mounted in the middle of the stagna-
tion chamber of the wind-tunnel. The atomizer is sup-
plied by pliq. 0.8 bar liquid and pair 1.25 bar
air pressures. With these conditions it provides a flow-
rate of about Qvd 0.133 1/min. The generated
droplet-size distribution is characterized by PDA mea-
surements using water. A mean Sauter droplet diameter
of d32 = 106.2 pm is obtained with a distribution rang-
ing from about 30 to 245 pm. These parameters provide
a Stokes number of about St = 2.33 and a volume frac-
tion of ap 0.4 10 4. The non-dimensional param-
eters in case of the P230 motor are about St 6 and
op = 2.2 .10-4.
The fluorescent tracers are injected just upstream the
spray device. They are produced by a nebulizer using
glycerine and 7.5 g/1 Rhodamine B.


2 50E-07
2 00E-07
1 50E-07
> 100E-07
5 00E-08
0 00E+00
T- V o co 0 co
Dp [plm]


Figure 3: Droplet-size distribution of the spray device.

The laser sheet representing the measurement plane
is generated in the mid-span plane of the test section
perpendicularly to the splitting plate. Image pairs are
acquired at a rate of 3.07 Hz with a separation time
of At 330 ps. Among the settings of the cameras
2 x 2 pixels2 inning is selected to gain sensitivity and
to respect the desired acquisition rate. Therefore, both
of the image pairs have 640 x 432 pixels2 resolution.












As Figure 2(b) indicates, altogether three fields-of-
view (FoVs) are defined to cover the area of the test sec-
tion, which is accessible with the UV laser beam through
a quartz window. In each FoV 1200 to 1700 two-phase
image pairs are recorded with both cameras in series
of 100 samples. Between the series the test section is
cleaned to reduce the effect of the deposited droplets.
After the two-phase acquisitions a series of 1200 to 1700
image pairs are recorded in single-phase flow configura-
tion as a reference, in order to see the effect of droplets
on the flow field.


Experimental results

Once the image pairs are processed with the PIV algo-
rithm, the instantaneous velocity fields are averaged to
obtain the mean flow properties and turbulence quanti-
ties.
As mentioned above, the statistical flow quantities are
not explained in this article. Only the mean streamlines
and the measurement quality are highlighted in Figure 4.
During the evaluation of the results, the regions, where
SNmean < 2 should be handled with great care, as
the observed information might have no physical con-
tent due to the low measurement quality.


2-
1 g,



0- 1
0


123456
1 2 3 4 5 6
X/h [-]
(a) Single-phase flow condition.


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


the behaviour of droplets of various sizes, the acquisi-
tions are further analysed. A droplet image detection al-
gorithm is developed, which identifies the location and
the diameter of particle images using wavelet analysis.
The droplet image diameters are assumed to be propor-
tional to the real diameter. Considering droplets ranging
from some tens of pm to a few hundreds of pm, during
the image formation the Mie scattering law is gradually
overtaken by the laws of geometrical optics. In this latter
domain, the larger an object is, the larger its image be-
comes. In addition, the droplet images are acquired from
the fluorescent light emitted by the droplets. Thus, by
assuming a constant illumination energy and a uniform
fluorescent dye concentration, the intensity of the fluo-
rescent light is proportional to the volume of the droplets
(therefore, ~ D,3). Due to the fact that the wind-tunnel
walls are wet, a halo appears around each droplet image,
which makes the droplets look even larger in diameter
(see e.g. Figure 5). The intensity of the halo depends on
the intensity of the light source: the droplet. Thus, if a
larger droplet image is detected, with a high probability
it should represent a droplet, which is of the larger kind
physically.


SNman [-]
4
3.5
3
2.5
2
1.5
1


0 1 2 3 4 5 6
X/h [-]
(b) Two-phase flow condition, air-phas


3 4
X/h [-1


SNman [-]
4
3.5
3
2.5
2
1.5
1

e.

SNman [-]
4
3
2.5
2
1.5
1


5 6


(c) Two-phase flow condition, droplet-phase.


Figure 4: Measurement quality and mean flow field.

The droplet-phase distribution. In order to separate


Figure 5: Sample acquisition of fluorescent droplets
with the present technique.

Hence, after the detection, the histogram of the
droplet image diameters is plotted, as shown in Fig-
ure 6(a). Although, here the dimension of Dp is indi-
cated to be [mm], it is still an arbitrary magnitude; the
[pixel] diameter value is simply multiplied by the mag-
nification of the camera. In order to determine the phys-
ical quantity, a one-dimensional wavelet-like algorithm
is looking for the best match (the superposition is shown
in Figure 6(b)) between the arbitrary diameter histogram
and the one obtained from PDPA experiments (Figure 3).
Thus, the droplet diameters can be obtained with physi-
cal units (pm), as it is shown in Figure 6(c).
Once the droplet locations and diameters are known,
one can investigate the spatial concentration map of the
droplets and the average diameters as well.
In Figure 7 the mean droplet number concentration








7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


2
2- ^ ^ ^ ^

>-^


X/h [-]


5



5 -


05 1 15 2 25 3 3!
D [mm]


(a) The arbitrary diameter distribution.

Dp [,m]
50 100 150 200 250 300
f Detection 0
o1 | PDPA 0 5
0014


0 05 d -001

5 1 15 2 25 3 39
Dp [mm]

(b) Relating the arbitrary distribution to the
physical quantities.


05
04
03
02


50 100 150 200 250 300
Dp[.lm]
(c) The resulting physical droplet diameters.


Figure 6: Determining the physical diameters of the de-
tected droplets (using the data of FoV#2).



distribution is shown. As it is expected, in the recir-
culation region downstream the inhibitor less than 4-
times lower droplet concentration is found than in the
free stream flow. The inhibitor is likely to filter out the
larger droplets and in the recirculation region typically
smaller droplets should be found, which are capable of
responding quicker to the air-phase motion and thus be-
ing carried by the flow structures.
This expectation is confirmed by the arithmetic mean
droplet diameter distribution of the current experiments
(see Figure 8). In the recirculation region a considerably
lower mean droplet diameter is found than in the free
stream. Furthermore, as the smaller particles enter in
the recirculation zone, in the wake of the shear layer the
mean droplet diameter increases and therefore this re-
gion exhibits even slightly larger average sizes than the


Figure 7: Mean droplet concentration.


free stream region.







X/h1 [i[-]
2 X95



Figure 8: Mean droplet diameter distribution.
77
68
59
50
0 1 2 3 4 5 6
X/h [-]

Figure 8: Mean droplet diameter distribution.


Furthermore, one may notice that in Figure 7 in the
zones corresponding to the three FoVs, the centre re-
gion always suggests a slightly higher local droplet con-
centration. However, this should not be considered to
be physical. The phenomenon should be the conse-
quence of the non-uniform laser sheet thickness. Dur-
ing the laser sheet generation the cylindrical laser-beam
is stretched to form an elongated ellipse cross section.
Therefore, the centre region of the laser sheet is always
somewhat thicker than at the extremities and thus in the
centre more droplets are visualised.
Looking at the zone corresponding to FoV#3 in Fig-
ure 7, one can see that the droplets seem to be pushed
even closer to the splitting plate than in FoV#2. Con-
sidering the fact that in this zone the reference velocity
is about 17% higher than in the preceding zones (as it
is described by ( )), this observation is
probably physical.
Finally, the average droplet diameter in the region of
FoV#2 appears to be about 10% lower than in the two
neighboring regions. The mismatch is certainly not
caused by the flow, but the source is not yet identified.
In any case, the figure still represents well the tendencies
of the droplet diameter distribution.
The droplet-phase motion. In order to separate the
motion of droplets of various sizes, the acquisitions are
further analysed.
In order to be able to determine the displacement
of each detected droplet, a simple droplet tracking ve-
locimetry (DTV) algorithm is developed. DTV is taking
all the detected droplet locations in each of the image
pairs and using the air flow field obtained from the PIV
algorithm as a predictor, droplet-image pairs are asso-











cited. Although the current predictor might bias the
measurements, indication could still be obtained from
the mean tendencies of the droplet motion.
Possessing both the physical diameter and the dis-
placement of each droplet, three different diameter
ranges are defined and the mean displacement is exam-
ined. Figure 9 shows the mean flow field of the air-phase
(Figure 9(a)) and the different ranges of droplets (Fig-
ure 9(b) to 9(d)).

2- -- -


El-


X/h [-]
(a) Air-phase.



Cs'^'^ a S


0 1 2 3
X/h [-]


4 5 6


4 5 6


(b) Droplet-phase (0 < Dp < 80 pm).


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


roughly the air motion. Even the small secondary re-
circulating structure at the downstream corner of the in-
hibitor is well captured. However, the matching is not
perfect, as e.g. the main recirculating structure appears
to be noticeably flatter.
The difference is more pronounced, when examining
the medium-class droplets (80 < Dp < 200 pm, Fig-
ure 9(c)). Here, the small corner vortex is less defined
and the main recirculation bubble shows an even flatter
behaviour, which means that the droplets of this range
have such a long characteristic time that does not allow
them following the air-phase, while it is accelerating to-
wards the vena contract.
Finally, the largest droplets of the domain i '1 ll < Dp,
Figure 9(d)) show that they are almost absolutely insen-
sitive to the air-phase motion. However, it is interesting
to notice that the droplet travelling at the inhibitor tip
between about 1.1 < Y/h < 1.5 are still effected by the
air. As the continuous-phase rapidly accelerates with a
considerable transversal velocity component, these par-
ticles are residing during a sufficiently long time that
their trajectory is slightly deviated towards the splitting
plate. As the air momentum is decreasing in the top
part of the main recirculation bubble, the droplets re-
quire a long time (corresponding to about X/h 4) to
recover this deviation. In this case even the main recir-
culation bubble is not well resolved. In practice, only a
few droplets are detected in this region. This is why a
chaotic mean motion can be observed at Y/h < 1.


Conclusions


Xh2 3[-
X/h [-]


4 5 6


(c) Droplet-phase (80 < Dp < 200 pm).


0 1 2 3
X/h [-]


4 5 6


(d) Droplet-phase (200 < Dp < 250 pm).

Figure 9: Mean motion of the two phases.

As one can see, the pattern drawn by the smallest
droplets (Dp < 80 /m, Figure 9(b)) is rather similar
to the one drawn by the air-phase. Therefore, in mean
these particles appear to be still capable of following


An experimental campaign is shown, which was con-
ducted as a step towards the understanding of the slag ac-
cumulation phenomenon in solid propellant rocket mo-
tors. A two-phase particle image velocimetry technique
was developed, which provides a simultaneous insight to
the gas-phase and the discrete-phase motion of a poly-
disperse droplet-laden flow.
During the present analysis of the database the
droplet-phase properties are extracted and assessed.
Thanks to the image forming principle of the fluorescent
doped droplets, the physical droplet diameters could be
estimated from the images besides the detection of their
position. The plots of the average droplet size distribu-
tion confirms the filtering effect of the inhibitor. Only
small particles, which have a sufficiently small associ-
ated characteristic time are capable of entering in the
recirculation zone. Large droplets remain in the free
stream flow regime.
A simple tracking mechanism is also developed to be
able to determine the displacement (velocity) of the in-
dividual droplets. Based on their size, several classes
are defined and their average velocity distribution is


CV_ ---


2-. -

0-


U I . I . . . . . . . .











compared to the air-phase motion. For the smallest
class a very similar velocity pattern is obtained as for
the continuous-phase. However, the largest class pat-
tern is substantially different, showing a mainly ballistic
droplet behaviour.


Acknowledgements

This study has been supported by the European Space
Agency through the GSTP activity number C51.MPA-
818.


References

M. W. Beckstead. A model for solid propellant com-
bustion. In Proceedings of the 14th Jannaf Combustion
Meeting, 1977.

N. Cesco, L. Dumas, A. Hulin, T. Pevergne, and
Y. Fabignon. Stochastic models to the investigation of
slag accumulation in a large solid rocket motor. AIAA
Paper 1997-3118, July 1997.

N. S. Cohen. A pocket model for aluminum agglom-
eration in composite propellants. AIAA Journal, 21(5):
720-725, 1983.

K. W. Dotson, J. W. Murdock, and D. K. Kamimoto.
Launch vehicle dynamic and control effects from solid
rocket motor slag ejection. Journal of Propulsion and
Power, 15(3):468-475, 1999.

J. Dupays. Two-phase unsteady flow in solid rocket mo-
tors. Aerospace Science and T7. Im. .1. .. 6(6):413-422,
2002.

J. Duterque and D. Lambert. Synthese des travaux
sur l'agglom6ration de l'aluminium dans le propergol
du MPS P230. In 3em Colloque R&T CNES/ONERA
"Ecoulements Internes en Propulsion Solide". Poitiers,
France, March 1998.

Y. Fabignon, J-F. Gu6ry, F. Godfroy, P. Le Helley,
J. Hylkema, L. Jacques, G. Lavergne, and P. Villedieu.
Slag accumulation in large segmented solid motors with
a submerged nozzle. In 2nd European Conference on
Launcher T,. r... 1... Rome, Italy, November 2000.

Y. Fabignon, J. Dupays, G. Avalon, F. Vuillot, N. Lu-
poglazoff, G. Casalis, and M. Pr6vost. Instabilities and
pressure oscillations in solid rocket motors. Aerospace
Science and T,. ... 1.,, 7(3):191-200, 2003.

F. Godfroy and J-F. Gu6ry. Unsteady eulerian flow anal-
ysis of solid rocket motor slag. AIAA Paper 1997-2859,
July 1997.


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


T. L. Jackson, F. Najjar, and J. Buckmaster. New alu-
minum agglomeration models and their use in solid-
propellant-rocket simulations. Journal of Propulsion
and Power, 21(5):925-936, 2005.

P. Le Helley, S. Coste, and T. Pevergne. Influence of the
unsteady flow on the alumina flux impinging the nozzle
and the aft dome of the Ariane 5 booster. In 2nd Euro-
pean Conference on Launcher T,. In. ... \. Rome, Italy,
November 2000.

N. Lupoglazoff and F. Vuillot. Parietal vortex shedding
as a cause of instability for long solid propellant mo-
tors. Numerical simulations and comparisons with firing
tests. AIAA Paper 1996-0761, January 1996.

R. X. Meyer. Coning instability of spacecraft during pe-
riods of thrust. Journal of Spacecraft and Rockets, 33
(6):781-788, 1996.

F M. Najjar, A. Haselbacher, S. Balachandar, and R. D.
Moser. Simulations of droplet nozzle impact and slag
accumulation in the RSRM. AIAA Paper 2006-4588,
July 2006.

T. Pevergne and P. Le Helley. Simulation numerique des
impacts de particles et du d6p8t dans le MPS P230. In
3e" Colloque R&T CNES/ONERA "Ecoulements In-
ternes en Propulsion Solide". Poitiers, France, March
1998.

M. Salita. Deficiencies and requirements in modeling
of slag generation in solid rocket motors. Journal of
Propulsion and Power, 11(1):10-23, 1995.

M. Salita. Predicted slag deposition histories in
eight solid rocket motors using the cfd model evt.
In J. Anthoine and P. Kuentzmann, editors, RTO
AVT/VKI Special Course "Internal Aerodynamics in
Solid Rocket Propulsion". von Karman Institute, Rhode-
Saint-Genese, Belgium, May 2002.

M. Telara, F. Paglia, F Stella, and M. Giangi. Ariane 5
P230 SRM frontal thermal protection evolution: numer-
ical simulation. AIAA Paper 2006-5242, July 2006.

B. T6th, M. R. Lema, J. Anthoine, and J. Steelant. As-
sessment of slag accumulation in solid rocket boosters:
Summary of the vki research. AIAA Paper 2007-5760,
July 2007.

B. T6th, J. Anthoine, and M. R. Riethmuller. Two-phase
piv method using two excitation and two emission spec-
tra. Experiments in Fluids, 47(3):475-487, September
2009a.

B. T6th, J. Anthoine, and J. Steelant. Experimental char-
acterization of a two-phase flowfield in a solid rocket







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


motor model. Journal of Propulsion and Power, 25(4):
914-920, July-August 2009b.

J. C. Traineau, M. Pr6vost, F. Vuillot, P. Le Breton,
J. Cuny, N. Preioni, and R. Bec. A subscale test pro-
gram to assess the vortex shedding driven instabilities in
segmented solid motors. AIAA Paper 1997-3247, July
1997.

M. Vardelle, C. Escure, P. Fauchais, B. Platet, and
G. Lavergne. Impact of alumina droplets on cold and
hot surfaces. In 2nd European Conference on Launcher
T,. IA1. 1. .. ome, Italy, November 2000.

F. Vuillot. Vortex-shedding phenomena in solid rocket
motors. Journal of Propulsion and Power, 11(4):626-
639, 1995.

H. Wirzberger, Y. Macales, and S. Yaniv. Prediction
of slag formation in a solid rocket motor. AIAA Paper
2005-4488, July 2005.

S. Zaleski and D. Gueyffier. Simulation de l'impact
de gouttes sur des films liquides. In 3"m Colloque
R&T CNES/ONERA "Ecoulements Internes en Propul-
sion Solide" Poitiers. France. March 1998




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs