Академический Документы
Профессиональный Документы
Культура Документы
|
|
.
|
\
|
+ = +
2
a
2
a
p
p
b
H
2
2
v
1
cos r
u 2
x
S
v
t
S
u
c
c
c
c
(5)
The mathematical model took into account the local propellant burning rate,
b
u
(absolute propellant wall velocity), the burning products emission velocity from burning
surface,
g
u
cos
1
cos P , 1
b
t
A
u x
A
u u
b
b
b g
p
c
c
=
c
c
= = . (6)
The total enthalpy of gas in every slide of the grain core, and , the total enthalpy
of propellant burning products are expressed as
i
H
p
H
p
v
i i
h H
1 2
2
= + = ,
2
2
g
u
p p
h H + = . (7)
3.2 UNSTEADY MODEL
The fundamental equations (1) (5) are nonlinear. However, a linearization process is
possible using the method of regular perturbations.
In this process, the internal flow field is separated into two components, a steady part
and an unsteady or oscillatory part.
The unsteady burning and flow processes generally provide an oscillatory pressure in
the motor chamber. Our study cases pointed out the presence of small pressure oscillations
around the case average pressure ( bar p 5 . 0
max
s A , 1 << ' p p ). Hence, the acoustic linear
analysis can be applied and the pressure oscillations can be considered as harmonic
oscillations.
Since the time dependent part is oscillatory, then each of the independent variables [4, 6]
can be written as
( )
t i t
dependent time steady
e e x F F F F t x F x F t x F
e o
c c
+ = ' + ~ + = ) (
~
Re ) , ( ) ( ) , (
) 1 ( ) 0 (
.
The following non-dimensional variables were introduced:
2
, , , , , ,
r
r
r
L
x
x
T
T
H
H
u
u
m
m
a
v
p
p
b
b
p
p
p
b
b
= =
'
=
'
= '
'
=
'
= '
'
= '
'
= '
'
= '
- -
t v
o , (8)
where: ) cos /( 2 u r u m
b p
= .
Defining each unsteady component ) cos(
~
)
~
Re( t
t
O = = '
O
F e F F
i
, where:
F
~
- amplitude of unsteady component variation,
L t a /
0
= t normalized time,
) 2 (
0
o t i f
a
L
+ = O - normalized pulsation in complex form, in which appear
the frequency and damping parameter o; by adequate transformations of eqs. (1-3, 5) we
obtain:
f
INCAS BULLETIN, Volume 3, Issue 1/ 2011
Doru SAFTA, Titica VASILE, Ioan ION 88
(
+ = + O
- -
-
- -
dx
d
M
dx
r M d
r
M
dx
d
i
L
v
v
v
~
~
) (
2
~
1
~
1
1
(
+ = + + O
- -
-
- - -
-
-
x
M
dx
r M d
r
M
x dx
dr
r
i
p L
c
c
t
c
v c
v
~
)
~ ~ ~
(
) (
~
~
1
~
1
2
1
2
2
2
, (10)
| |
)
`
+ = O
- -
-
-
x
S
M S
dx
r M d
r
M S i
p L
c
c
t
~
~
~
) 1 (
~
) (
~
1
2
1
2
. (11)
The solutions of this system can be expressed in the following form:
.... ....
~ ~ ~
....
~ ~ ~
....
~ ~ ~
1 0 1 0
1 0 1 0
+ O + O = O + + =
+ + = + + =
L L
L L
M S M S S
M M v v v
(12)
where ( )
L L
a v M / = is the perturbation parameter.
According to the perturbations method, the eqs. (9-11) become:
order 1: 0
~
1
~ 0
0 0
= + O
-
dx
d
i
v , (13)
0
~
~ ~ 0
0
2
2
0 0
= + + O
- -
-
-
dx
d
dx
dr
r
i
v
v
, (14)
0
~
0
= S ; (15)
order
L
M :
- -
-
- -
(
+ O = + O
dx
d
M
dx
r M d
r
i
dx
d
i
0
1 0
1
1
1
1 0
~
~
) (
2
~
1
~
v
v
v , (16)
- -
-
- - -
-
-
(
O + = + + O
x
M i
dx
r M d
r x dx
dr
r
i
p
c
c
t
c
v c
v
0
1 0 1
2
1
2
1
1
2
2
1 0
~
~
)
~ ~ ~
(
) (
~
~
~ ~
, (17)
0
2
1
2
1 0
~
) (
1
~
~
~
t
-
-
- |
|
.
|
\
|
+ = O
dx
r M d
r
S i
p
. (18)
The boundary conditions are:
: 0 =
-
x 0
~
, 0
~
, 0
~
, 1
~
1 0 1 0
= = - - = = - v v , (19)
( ) ] ) cos[(
~
Re , 1
~ ~
1 0 0
t
t
O + O = = ' = =
O
L
i
M e
( ) 0
~
Re , 0
~ ~ ~
1 0
= = ' = + =
Ot
v v v v v
i
L
e M
: unsteady flow continuity condition at the nozzle entrance: 1 =
-
x
)
~
~
(
~
S w v M
L L L
+ = v ,
L L
w v , - normalized complex coefficients/ nozzle admittance and coadmitance [3, 4], thus:
0
~
0
= -v , (20 a)
v ,
~
L 0 1
~
v
L
v = - - computed for
0
O . (20 b)
The order 1 solution is provided by eqs. (13) and (14) with boundary conditions (19)
and (20 a). We can obtain
0
~
v depending on
0
~
and
0
O [4 - 6], and from eq. (14) we can
write a unique differential eq. in
0
~
.
The normalized pulsation
0
O can be computed based on a transcendent eq. built in this
stage. As first evaluation of this parameter, we have:
) 0 (
0
O
INCAS BULLETIN, Volume 3, Issue 1/ 2011
89 Regarding the evaluation of the solid rocket propellant response function to pressure coupling
t
o
t
k
b
a k
a X + |
.
|
\
|
+ = O
cos
) 1 (
sin ) 1 ( 2
) 0 (
0
(21)
The Eqs. (16) and (17) with boundary conditions (19) and (20 b) provide the solution of
L
M order. In this case we can express
1
~
v depending on
0
~
and
1
~
, and a unique
differential eq. in
1
~
. The pulsation
1
O can be computed as solution of the following eq.:
) 1 ( I
1
~
~ ~
I 2
3 2 1 1 L
p
v I i
t
+
|
|
.
|
\
|
+
= O , (22)
where:
,
- -
}
= dx r I
~2
0
1
0
2
1
( )
-
-
-
}
= dx
dx
r M d
I
~2
0
1
0
2
1
2
,
( )
}
-
- -
-
|
|
.
|
\
|
O
=
1
0
2
0
0
2
1
3
~
1
dx
dx
d
dx
r M d
I
. (23)
Analyzing these integrals a physical significance can be identified: is proportional to
the chamber acoustic energy, expresses the interdependence/ coupling between the
pressure oscillations and the combustion phenomenon, shows the acoustic energy spent
to get an axial unsteady velocity of burning products flow turning losses. The last term
of eq. (22) expresses the acoustic energy losses in the nozzle.
1
I
2
I
3
I
Considering the steady-state conditions, the eqs. system solutions may take the form:
..... *) ( *) (
..... *) ( *) ( *) ( *
2
2
L 1
2
2
1
+ + =
+ + + =
x M M x M M M
x P M x P M x P p
L
L L o
, (24)
with the following boundary conditions:
0 ) 1 ( , 1 ) 1 ( ,... 0 ) 0 ( , 0 ) 0 ( ,... 0 ) 0 ( , 0 ) 0 ( , 1 ) 0 (
2 1 2 1 2 1
= = = = = = = M M M M P P P
o
.
The variation of the chamber static pressure p , and the evolution of Mach number M in
the grain core for various X parameter values (implicitly its time variation) can be obtained
by integrating the system:
2
2 *
2 * *
0
2
2
1
1
*
) (
) (
1
*
cos
4
2
1
1 *
*
M
dx
r d
r
M p
r a
u b
M M p
dx
d
s
s s
b p
+ =
|
|
.
|
\
|
+
u
, (25)
( ) | |
*
) (
) (
1
* 1 *
*
2 *
2 *
2 2
dx
r d
r
M p M p
dx
d
s
s
= + , (26)
where * , , *
0
*
p
p
p
L
r
r
L
x
x
s
= = = .
According to this model, the variables p and v can be computed as functions of time:
( ) } ]
~ ~
Re[ 1 { (x*)] P M (x*) [P ) 1 (
) (
1 0 1 L 0 0
1 0
t
c c
O + O
+ + + = ' + =
L
M i
L
e M p p p , (27)
( ) ]
~ ~
Re[ *) (
) (
1 0 1
1 0
t
v v c v c
O + O
+ + = ' + =
L
M i
L L
e M a x M M a a v v . (28)
3.3 PROPELLANT RESPONSE TO PRESSURE COUPLING
In order to evaluate the propellant response it is necessary to know its dependence on the
oscillations amplitude and frequency, the nature of flow oscillations, on average pressure
and the propellant composition [1 - 3].
In the case of a motor working considered as a harmonic one and assuming that the
propellant response is dependent of the steady pressure and pulsation and taking into account
INCAS BULLETIN, Volume 3, Issue 1/ 2011
Doru SAFTA, Titica VASILE, Ioan ION 90
only the pressure coupling and neglecting, the admittance corrections, we can write [2, 4 -
6]:
p p
u u
R
b b
up
/
~
/
~
~
~
= =
t
, response in burning temperature to the pressure coupling. Thus:
t
~
~ ~
p
Tp up pc p
R R R R
+
= + = = , (29)
is the propellant response in combustion to the rocket motor pressure coupling, named the
linear propellant response.
Taking into account the normalized pulsation, O, [4, 6, 7],
|
|
.
|
\
|
O + = O O + O ~ + = O
0
0 0
1 1 0
0
2
) 2 (
a
L f
M
i
M a
L
i M i f
a
L
L L
L
t o
o t ,
the real part and the imaginary one of the combustion response will be:
c
R
2
1
0
3
) (
) (
2
1
1
I
I
M a
L
I v
R
L
r
L
r
p
+ +
+
= ,
2
0
0
1
) (
) (
2
2
I
a
L f
I
M
v
R
L
i
L
i
p
|
|
.
|
\
|
O +
= . (30)
4. RESULTS AND CONCLUSIONS
In the frame of developed model the variations of unsteady pressure and flow velocity
amplitude (
0
~
,
0
~
v ), for longitudinal oscillation mode (k=1), can be easily evaluated. Thus,
one can observe that the parameter variation of *) (
~
0
x is very appropriate of cos (tx*),
Fig.8.
Fig. 8 Unsteady pressure amplitude
0
~
vs. no
dimensional length x*
Fig. 9 Unsteady velocity amplitude
0
~
v vs. no
dimensional length x*
The flow velocity amplitude reaches a maximum value at the middle of the chamber and
this parameter value at the beginning of the combustion process is higher by 30% compared
to its value at the end of the engine operation, Fig. 9.
INCAS BULLETIN, Volume 3, Issue 1/ 2011
91 Regarding the evaluation of the solid rocket propellant response function to pressure coupling
We can assert, on the basis of the theoretical model, that the real part of propellant
response- , is a function of the oscillations damping, acoustic energy in the motor
chamber and various losses in the burning chamber. The imaginary part of propellant
response - , mainly depends on the normalized pulsation, on the burning chamber gas
column and on the pressure oscillations frequency.
) (r
p
R
) (i
p
R
The computing procedure of the propellant response evaluation model basically
demands. 3 types of data: a). combustion products characteristics (thermodynamical
calculus); b). dimensions of the grain, case and nozzle; c). experimental results the average
pressure, burned thickness, frequency and damping oscillations as time functions.
Applying our model, the main steps of the computing procedure are:
calculus of perturbation parameter
L
M based on the rate (
cr L
A A / ),[5];
iterative calculus of normalized pulsation
0
O ,[4, 5], (first evaluation (21), Fig.10);
determining of the integrals
1
I ,
2
I ,
3
I and of
) (r
L
v ,
) (i
L
v , [4 - 6];
calculus of the global response in combustion
p
R , eq. (30), using the experimental
data obtained by jet intermittent modulating techniques.
Regarding the evaluation of the integrals , , , for the considered grain shape, we
can use suitable analytical formula. Thus, figure 11 depicts the variations of these integrals
depending on the relative burning grain thickness for the first oscillation mode.
1
I
2
I
3
I
Its important to highlight that the grain core flow unsteady velocity calculus, eq.(28),
requires parameters v , or M to be known and for this reason the evolution was
computed by numerical integration, corresponding to t
*) ( M
1
x
e{0,
, }, (Fig. 12).
b
t ) 5 ( , / 2 / 1
b
t ) 5 ( ,
b
t ) 5 / 3 ( ,
b
t ) 5 / 4 (
b
t
Fig. 10 Normalized pulsation ratio to t k
vs. non-dimensional thickness ratio to (1- )
1
b
Fig. 11 The integrals vs. non-
dimensional thickness ratio to (1- )
3 2 1
I and I , I
1
b
Laboratory and analytical presented tools were utilized to characterize the acoustic
response properties of a solid propellant (in a first stage a no metalized propellant, based on
ammonium perchlorate and polybutadiene, presented in chapter 2) as a function of
frequency, for different values of case average pressure.
The real part values of the propellant response, eq. (30), were obtained taking into
account the dependence time - damping parameter, Fig. 7. The results concerning the
variation of the real part of the propellant response in respect of the frequency is illustrated
in Fig. 13.
INCAS BULLETIN, Volume 3, Issue 1/ 2011
Doru SAFTA, Titica VASILE, Ioan ION 92
Fig. 12
The magnitude is acceptable and we can observe its increasing variation until a
maximum value corresponding to a frequency level of about 10 KHz. An explanation of this
variation trend may be found in the unsteady behavior of the burning zone to high frequency.
) (r
p
R
Fig. 13 Real part of coupled pressure response
function vs. frequency at 3 case average pressure
values composite propellant
Fig. 14 Real part of coupled pressure response
function vs. frequency at 2 case average pressure
values DB propellan
The absolute value of doesnt exceed in magnitude some units and it is very
sensitive with (e.g. -5% induces the response decreasing at about zero) and can
influence the average pressure at various frequencies (e.g. -2 at 7KHz, -30 at 9 KHz).
) (i
p
R
0
a
0
a
Our paper pointed out that during all the tests with the subscale test solid rocket motor,
for average pressures of 3 - 8 MPa, the excited mode was the fundamental longitudinal mode
(first and eventual second harmonics).
We have extended the application of the presented model also in the case of a DB
propellant [5], (based on NG/NC/DNT, , m/s 1380 * = c
598 . 0
675 . 0 p u
b
= ), used for a 122
mm solid rocket motor. Thus, the figure 14 depicts the influence of the mean pressure on the
propellant response function, for frequency values of 1 3 KHz. This result is important for
INCAS BULLETIN, Volume 3, Issue 1/ 2011
93 Regarding the evaluation of the solid rocket propellant response function to pressure coupling
DB propellant rocket motors design, especially with a higher length to diameter grain ratio,
and a higher volume charge rate of the motor case.
As an important conclusion for the two case studies, the most unfavorable
propellant response is obtained at smaller pressures, in the proximity of the stable working
limit of the rocket motor. Moreover, the unsteady motor working changes could be
accompanied by abrupt rise of average chamber pressure.
Taking into account the researches made by Kuentzmann and Traineau [3, 4, 6] for
several composite propellants we have extended the application of this propellant response
evaluation indirect method for the DB propellant mentioned before. Thus, during the tests
made with the same solid rocket motor (MR-01, [5] , 122mm D
R
= ), at two different initial
temperatures, we observed that rather high level frequency combustion instabilities can
appear in certain conditions and they can induce an important variation of the average
pressure evolution (Fig. 15, 16). The most disturbed time-pressure evolution (nonlinear
instabilities) was recorded for smaller pressure values, corresponding of negative propellant
initial temperature C T
i
= 40 (Fig. 16).
Fig. 15 Pressure time history, at , C 40 =
i
T
[5] , 122mm
for DB
propellant motor ( D
R
= )
Fig. 16 Pressure time history, at
for DB propellant motor ( )
, C 40 =
i
T
[5] , 122mm D
R
=
This experiment can sustain the conclusion concerning the prediction property of
propellant response function towards the motor behavior. Hence, at smaller chamber mean
pressures, the high frequency combustion instabilities may have a significant influence on
the rocket motor working.
Our linear acoustic analysis cant be applied if the pressure oscillation magnitude grows
larger than about 10 % of the mean pressure [1, 2].
Research towards predicting and quantifying undesirable axial combustion instability
symptoms necessitates a comprehensive numerical model for internal ballistic simulation
under dynamic flow and combustion conditions. The study model of the present paper, based
on the perturbations method was included in the mathematical model of a complex
simulation computer program of solid rocket motors disturbed working, which operates with
a large termogasdynamical and engine construction data basis. This is also a benefit of using
an indirect evaluation method of coupled pressure propellant response function.
Our next step in the combustion instabilities study will be to carry on the improvement
of the mathematical model and to use the jet modulating technique to keep on the
investigation of the DB propellants and to explore the metalized propellants response
function in various working conditions.
INCAS BULLETIN, Volume 3, Issue 1/ 2011
Doru SAFTA, Titica VASILE, Ioan ION 94
INCAS BULLETIN, Volume 3, Issue 1/ 2011
5. REFERENCES
[1] F. S. BLOMSHIELD, Lessons Learned In Solid Rocket Combustion Instability, American Institute of
Aeronautics and Astronautics, Missile Sciences Conference, Monterey, California, November 2006.
[2] F. E. C. CULICK, Combustion Instabilities in Solid Rocket Motors, Notes for two lectures of special course
Internal Aerodynamics in Solid Rocket Propellant, von Karman Institute, 2002.
[3] A. DAVENAS, Technologie des Propergoles Solides, Masson, Paris, 1989.
[4] P. KUENTZMANN, A. LAVERDANT, Dtermination exprimentale de la rponse dun propergol solide
aux oscillations de pression de haute frquence, La recherche arospatiale, 1984.
[5] D. SAFTA, Propulsion Thermodynamics and Fly Dynamics Elements, Military Technical Academy,
Bucharest, 2004.
[6] J. C. TRAINEAU, M. PREVOST, P. TARRIN, Experimental low and medium frequency determination of
solid propellants pressure-coupled response function, AIAA 94-3043, 1994.
[7] M. SHUSSER, F. E. C. CULICK, N. S. COHEN, Combustion response of ammonium perchlorate composite
propellants, Journal of Propulsion and Power, vol. 18, no. 5, 2002.